pax_global_header00006660000000000000000000000064147457677160014542gustar00rootroot0000000000000052 comment=12a22de74a74003ae231849f2fcb5063e9f928de beanquery-master/000077500000000000000000000000001474576771600143725ustar00rootroot00000000000000beanquery-master/.github/000077500000000000000000000000001474576771600157325ustar00rootroot00000000000000beanquery-master/.github/workflows/000077500000000000000000000000001474576771600177675ustar00rootroot00000000000000beanquery-master/.github/workflows/checks.yaml000066400000000000000000000030371474576771600221160ustar00rootroot00000000000000name: checks on: - push - pull_request jobs: ruff: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - uses: actions/setup-python@v5 with: python-version: '3.11' - run: python -m pip install ruff - run: ruff check beanquery/ coverage: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - uses: actions/setup-python@v5 with: python-version: '3.11' - run: pip install -r requirements.txt coverage - name: Run coverage # Check tests coverage. Instead of checking project coverage of # all the tests we check that each module is exaustively tested # by the dedicated units tests. run: | set -x echo '{ "beanquery/parser/*": "beanquery/parser_test.py", "beanquery/query_render.py": "beanquery/query_render_test.py" }' | jq -rc 'to_entries | .[] | (.key + "=" + .value)' | while IFS='=' read src test do python -m coverage run --branch --include "$src" --omit beanquery/parser/parser.py -m unittest "$test" python -m coverage report --fail-under=100 -m python -m coverage erase done build: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - uses: actions/setup-python@v5 with: python-version: '3.11' - run: pip install -r requirements.txt wheel build - run: python -m build --no-isolation - run: python -m pip install dist/beanquery-*.whl beanquery-master/.github/workflows/docs.yml000066400000000000000000000012111474576771600214350ustar00rootroot00000000000000name: docs on: push: pull_request: jobs: build: runs-on: ubuntu-latest steps: - uses: actions/setup-python@v5 with: python-version: '3.11' - uses: actions/checkout@v4 - run: python -m pip install .[docs] - run: python -m sphinx -W -b html docs/ build/html/ - uses: actions/upload-pages-artifact@v3 with: path: build/html deploy: needs: build permissions: pages: write id-token: write environment: name: github-pages runs-on: ubuntu-latest if: github.ref == 'refs/heads/master' steps: - uses: actions/deploy-pages@v4 beanquery-master/.github/workflows/release.yaml000066400000000000000000000011571474576771600222770ustar00rootroot00000000000000name: release on: push: tags: - 'v[0-9]*' jobs: build: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - run: python -m pip install build - run: python -m build - uses: actions/upload-artifact@v4 with: path: dist/* upload: needs: build runs-on: ubuntu-latest environment: upload permissions: id-token: write steps: - uses: actions/download-artifact@v4 with: merge-multiple: true path: dist - uses: pypa/gh-action-pypi-publish@release/v1 with: attestations: false beanquery-master/.github/workflows/test.yaml000066400000000000000000000016621474576771600216370ustar00rootroot00000000000000name: test on: - push - pull_request jobs: test: runs-on: ubuntu-latest strategy: fail-fast: false matrix: python: - '3.8' - '3.9' - '3.10' - '3.11' - '3.12' - '3.13' beancount: - '~= 2.3.6' - '~= 3.0.0' - '@ git+https://github.com/beancount/beancount.git' exclude: - python: '3.8' beancount: '@ git+https://github.com/beancount/beancount.git' - python: '3.9' beancount: '@ git+https://github.com/beancount/beancount.git' steps: - uses: actions/checkout@v4 - uses: actions/setup-python@v5 with: python-version: ${{ matrix.python }} allow-prereleases: true - run: pip install 'beancount ${{ matrix.beancount }}' - run: pip install -r requirements.txt - run: python -m unittest discover -p '*_test.py' beanquery-master/.gitignore000066400000000000000000000000151474576771600163560ustar00rootroot00000000000000__pycache__/ beanquery-master/CHANGES.rst000066400000000000000000000042541474576771600162010ustar00rootroot00000000000000Version 0.1 (unreleased) ------------------------ - The ``HAVING`` clause for aggregate queries is now supported. - The ``empty()`` BQL function to determine whether an Inventory object as returned by the ``sum()`` aggregate function is empty has been added. - Added the ``round()`` BQL function. - ``NULL`` values in ``SORT BY`` clause are now always considered to be smaller than any other values. This may results in rows to be returned in a slightly different order. - It is now possible to specify the direction of the ordering for each column in the ``SORT BY`` clause. This brings BQL closer to SQL specification but queries written with the old behaviour in mind will return rows in a different order. The query:: SELECT date, narration ORDER BY date, narration DESC used to return rows in descending order by both ``date`` and ``narration`` while now it would order the rows ascending by ``date`` and descending by ``narration``. To recover the old behavior, the query should be written:: SELECT date, narration ORDER BY date DESC, narration DESC - Type casting functions ``int()``, ``decimal()``, ``str()``, ``date()`` have been added. These are mostly useful to convert the generic ``object`` type returned by the metadata retrieval functions but can also be used to convert between types. If the conversion fails, ``NULL`` is returned. - The ``str()`` BQL function used to return a string representation of its argument using the Python :py:func:`repr()` function. This clashes with the use of ``str()`` as a type casting function. The function is renamed ``repr()``. - The ``date()`` BQL function used to extract a date from string arguments with a very relaxed parser. This clashes with the use of ``date()`` as a type casting function. The function is renamed ``parse_date()``. Another form of ``parse_date()`` that accepts the date format as second argument has been added. - The ``getitem()`` BQL function return type has been changed from a string to a generic ``object`` to match the return type of function retrieving entries from metadata dictionaries. The old behavior can be obtained with ``str(getitem(x, key))``. beanquery-master/LICENSE000066400000000000000000000432541474576771600154070ustar00rootroot00000000000000 GNU GENERAL PUBLIC LICENSE Version 2, June 1991 Copyright (C) 1989, 1991 Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. Preamble The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software--to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation's software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Lesser General Public License instead.) You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things. To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it. For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software. Also, for each author's protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations. Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all. The precise terms and conditions for copying, distribution and modification follow. GNU GENERAL PUBLIC LICENSE TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION 0. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term "modification".) Each licensee is addressed as "you". Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does. 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change. b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License. c) If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.) These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program. In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License. 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.) The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code. 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. 6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. 7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system, which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License. 8. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License. 9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. 10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. NO WARRANTY 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. END OF TERMS AND CONDITIONS How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the "copyright" line and a pointer to where the full notice is found. Copyright (C) This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. Also add information on how to contact you by electronic and paper mail. If the program is interactive, make it output a short notice like this when it starts in an interactive mode: Gnomovision version 69, Copyright (C) year name of author Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. This is free software, and you are welcome to redistribute it under certain conditions; type `show c' for details. The hypothetical commands `show w' and `show c' should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than `show w' and `show c'; they could even be mouse-clicks or menu items--whatever suits your program. You should also get your employer (if you work as a programmer) or your school, if any, to sign a "copyright disclaimer" for the program, if necessary. Here is a sample; alter the names: Yoyodyne, Inc., hereby disclaims all copyright interest in the program `Gnomovision' (which makes passes at compilers) written by James Hacker. , 1 April 1989 Ty Coon, President of Vice This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Lesser General Public License instead of this License. beanquery-master/README.rst000066400000000000000000000004171474576771600160630ustar00rootroot00000000000000beanquery: Customizable lightweight SQL query tool ================================================== beanquery is a customizable and extensible lightweight SQL query tool that works on tabular data, including `Beancount`__ ledger data. __ https://beancount.github.io/ beanquery-master/beanquery/000077500000000000000000000000001474576771600163655ustar00rootroot00000000000000beanquery-master/beanquery/__init__.py000066400000000000000000000032661474576771600205050ustar00rootroot00000000000000import importlib from urllib.parse import urlparse from . import parser from . import compiler from . import tables from .compiler import CompilationError from .cursor import Cursor, Column from .errors import Warning, Error, InterfaceError, DatabaseError, DataError, OperationalError from .errors import IntegrityError, InternalError, ProgrammingError, NotSupportedError from .parser import ParseError __version__ = '0.1.dev1' # DB-API compliance apilevel = '2.0' threadsafety = 2 paramstyle = 'pyformat' def connect(dsn, **kwargs): return Connection(dsn, **kwargs) class Connection: def __init__(self, dsn='', **kwargs): self.tables = {'': tables.NullTable()} self.options = {} self.errors = [] if dsn: self.attach(dsn, **kwargs) def attach(self, dsn, **kwargs): scheme = urlparse(dsn).scheme source = importlib.import_module(f'beanquery.sources.{scheme}') source.attach(self, dsn, **kwargs) def close(self): # Required by the DB-API. pass def parse(self, query): return parser.parse(query) def compile(self, query): return compiler.compile(self, query) def execute(self, query, params=None): return self.cursor().execute(query, params) def cursor(self): return Cursor(self) __all__ = [ 'Column', 'CompilationError', 'Connection', 'Cursor', 'DataError', 'DatabaseError', 'Error', 'IntegrityError', 'InterfaceError', 'InternalError', 'NotSupportedError', 'OperationalError', 'ParseError', 'ProgrammingError', 'Warning', 'apilevel', 'connet', 'paramstyle', 'threadsafety', ] beanquery-master/beanquery/__main__.py000066400000000000000000000001111474576771600204500ustar00rootroot00000000000000from beanquery import shell if __name__ == '__main__': shell.main() beanquery-master/beanquery/compiler.py000066400000000000000000001017161474576771600205570ustar00rootroot00000000000000import collections.abc import typing from decimal import Decimal from functools import singledispatchmethod from typing import Optional, Sequence, Mapping, Union from . import types from . import parser from .errors import ProgrammingError from .parser import ast from .query_compile import ( EvalAggregator, EvalAnd, EvalAll, EvalAny, EvalCoalesce, EvalColumn, EvalConstant, EvalGetItem, EvalGetter, EvalOr, EvalPivot, EvalQuery, EvalConstantSubquery1D, EvalRow, EvalTarget, FUNCTIONS, OPERATORS, SubqueryTable, ) # A global constant which sets whether we support inferred/implicit group-by # semantics. SUPPORT_IMPLICIT_GROUPBY = True class CompilationError(ProgrammingError): def __init__(self, message, node=None): super().__init__(message) self.parseinfo = node.parseinfo if node is not None else None class Compiler: def __init__(self, context): self.context = context self.stack = [context.tables.get('postings')] @property def table(self): return self.stack[-1] @table.setter def table(self, value): self.stack[-1] = value def compile(self, query, parameters=None): """Compile an AST into an executable statement.""" self.parameters = parameters placeholders = [node for node in query.walk() if isinstance(node, ast.Placeholder)] if placeholders: names = {placeholder.name for placeholder in placeholders} if all(names): if not isinstance(parameters, Mapping): raise TypeError('query parameters should be a mapping when using named placeholders') if names - parameters.keys(): missing = ', '.join(sorted(names - parameters.keys())) raise ProgrammingError(f'query parameter missing: {missing}') elif not any(names): if not isinstance(parameters, Sequence): raise TypeError('query parameters should be a sequence when using positional placeholders') if len(placeholders) != len(parameters): raise ProgrammingError( f'the query has {len(placeholders)} placeholders but {len(parameters)} parameters were passed') for i, placeholder in enumerate(sorted(placeholders, key=lambda node: node.parseinfo.pos)): placeholder.name = i else: raise ProgrammingError('positional and named parameters cannot be mixed') return self._compile(query) @singledispatchmethod def _compile(self, node: Optional[ast.Node]): if node is None: return None raise NotImplementedError @_compile.register def _select(self, node: ast.Select): self.stack.append(self.table) # Compile the FROM clause. c_from_expr = self._compile_from(node.from_clause) # Compile the targets. c_targets = self._compile_targets(node.targets) # Bind the WHERE expression to the execution environment. c_where = self._compile(node.where_clause) # Check that the FROM clause does not contain aggregates. This # should never trigger if the compilation environment does not # contain any aggregate. if c_where is not None and is_aggregate(c_where): raise CompilationError('aggregates are not allowed in WHERE clause') # Combine FROM and WHERE clauses if c_from_expr is not None: c_where = c_from_expr if c_where is None else EvalAnd([c_from_expr, c_where]) # Process the GROUP-BY clause. new_targets, group_indexes, having_index = self._compile_group_by(node.group_by, c_targets) c_targets.extend(new_targets) # Process the ORDER-BY clause. new_targets, order_spec = self._compile_order_by(node.order_by, c_targets) c_targets.extend(new_targets) # If this is an aggregate query (it groups, see list of indexes), check that # the set of non-aggregates match exactly the group indexes. This should # always be the case at this point, because we have added all the necessary # targets to the list of group-by expressions and should have resolved all # the indexes. if group_indexes is not None: non_aggregate_indexes = {index for index, c_target in enumerate(c_targets) if not c_target.is_aggregate} if non_aggregate_indexes != set(group_indexes): missing_names = ['"{}"'.format(c_targets[index].name) for index in non_aggregate_indexes - set(group_indexes)] raise CompilationError( 'all non-aggregates must be covered by GROUP-BY clause in aggregate query: ' 'the following targets are missing: {}'.format(','.join(missing_names))) query = EvalQuery(self.table, c_targets, c_where, group_indexes, having_index, order_spec, node.limit, node.distinct) pivots = self._compile_pivot_by(node.pivot_by, c_targets, group_indexes) if pivots: return EvalPivot(query, pivots) self.stack.pop() return query def _compile_from(self, node): if node is None: return None # Subquery. if isinstance(node, ast.Select): self.table = SubqueryTable(self._compile(node)) return None # Table reference. if isinstance(node, ast.Table): self.table = self.context.tables.get(node.name) if self.table is None: raise CompilationError(f'table "{node.name}" does not exist', node) return None # FROM expression. if isinstance(node, ast.From): c_expression = self._compile(node.expression) # Check that the FROM clause does not contain aggregates. if c_expression is not None and is_aggregate(c_expression): raise CompilationError('aggregates are not allowed in FROM clause') if node.open and node.close and node.open > node.close: raise CompilationError('CLOSE date must follow OPEN date') # Apply OPEN, CLOSE, and CLEAR clauses. self.table = self.table.update(open=node.open, close=node.close, clear=node.clear) return c_expression raise NotImplementedError def _compile_targets(self, targets): """Compile the targets and check for their validity. Process wildcard. Args: targets: A list of target expressions from the parser. Returns: A list of compiled target expressions with resolved names. """ # Bind the targets expressions to the execution context. if isinstance(targets, ast.Asterisk): # Insert the full list of available columns. targets = [ast.Target(ast.Column(name), None) for name in self.table.wildcard_columns] # Compile targets. c_targets = [] for target in targets: c_expr = self._compile(target.expression) name = get_target_name(target) c_targets.append(EvalTarget(c_expr, name, is_aggregate(c_expr))) columns, aggregates = get_columns_and_aggregates(c_expr) # Check for mixed aggregates and non-aggregates. if columns and aggregates: raise CompilationError('mixed aggregates and non-aggregates are not allowed') # Check for aggregates of aggregates. for aggregate in aggregates: for child in aggregate.childnodes(): if is_aggregate(child): raise CompilationError('aggregates of aggregates are not allowed') return c_targets def _compile_order_by(self, order_by, c_targets): """Process an order-by clause. Args: order_by: A OrderBy instance as provided by the parser. c_targets: A list of compiled target expressions. Returns: A tuple of new_targets: A list of new compiled target nodes. order_spec: A list of (integer indexes, sort order) tuples. """ if not order_by: return [], None # Compile order-by expressions and resolve them to their targets if # possible. A ORDER-BY column may be one of the following: # # * A reference to a target by name. # * A reference to a target by index (starting at one). # * A new expression, aggregate or not. # # References by name are converted to indexes. New expressions are # inserted into the list of targets as invisible targets. new_targets = c_targets[:] c_target_expressions = [c_target.c_expr for c_target in c_targets] targets_name_map = {target.name: idx for idx, target in enumerate(c_targets) if target.name is not None} # Only targets appearing in the SELECT targets list can be # referenced by index. These are guaranteed to have a valid name. n_targets = len(targets_name_map) order_spec = [] for spec in order_by: column = spec.column descending = spec.ordering index = None # Process target references by index. if isinstance(column, int): index = column - 1 if not 0 <= index < n_targets: raise CompilationError(f'invalid ORDER-BY column index {column}') else: # Process target references by name. These will be parsed as # simple Column expressions. If they refer to a target name, we # resolve them. if isinstance(column, ast.Column): name = column.name index = targets_name_map.get(name, None) # Otherwise we compile the expression and add it to the list of # targets to evaluate and index into that new target. if index is None: c_expr = self._compile(column) # Attempt to reconcile the expression with one of the existing # target expressions. try: index = c_target_expressions.index(c_expr) except ValueError: # Add the new target. 'None' for the target name implies it # should be invisible, not to be rendered. index = len(new_targets) new_targets.append(EvalTarget(c_expr, None, is_aggregate(c_expr))) c_target_expressions.append(c_expr) assert index is not None, "Internal error, could not index order-by reference." order_spec.append((index, descending)) return new_targets[len(c_targets):], order_spec def _compile_pivot_by(self, pivot_by, targets, group_indexes): """Compiles a PIVOT BY clause. Resolve and validate columns references in the PIVOT BY clause. The PIVOT BY clause accepts two name od index references to columns in the SELECT targets list. The second columns should be a GROUP BY column so that the values of the pivot column are unique. """ if pivot_by is None: return None indexes = [] names = {target.name: index for index, target in enumerate(targets)} for column in pivot_by.columns: # Process target references by index. if isinstance(column, int): index = column - 1 if not 0 <= index < len(targets): raise CompilationError(f'invalid PIVOT BY column index {column}') indexes.append(index) continue # Process target references by name. if isinstance(column, ast.Column): index = names.get(column.name, None) if index is None: raise CompilationError(f'PIVOT BY column {column!r} is not in the targets list') indexes.append(index) continue # Not reached. raise RuntimeError # Sanity checks. if indexes[0] == indexes[1]: raise CompilationError('the two PIVOT BY columns cannot be the same column') if indexes[1] not in group_indexes: raise CompilationError('the second PIVOT BY column must be a GROUP BY column') return indexes def _compile_group_by(self, group_by, c_targets): """Process a group-by clause. Args: group_by: A GroupBy instance as provided by the parser. c_targets: A list of compiled target expressions. Returns: A tuple of new_targets: A list of new compiled target nodes. group_indexes: If the query is an aggregate query, a list of integer indexes to be used for processing grouping. Note that this list may be empty (in the case of targets with only aggregates). On the other hand, if this is not an aggregated query, this is set to None. So do distinguish the empty list vs. None. """ new_targets = c_targets[:] c_target_expressions = [c_target.c_expr for c_target in c_targets] group_indexes = [] having_index = None if group_by: assert group_by.columns, "Internal error with GROUP-BY parsing" # Compile group-by expressions and resolve them to their targets if # possible. A GROUP-BY column may be one of the following: # # * A reference to a target by name. # * A reference to a target by index (starting at one). # * A new, non-aggregate expression. # # References by name are converted to indexes. New expressions are # inserted into the list of targets as invisible targets. targets_name_map = {target.name: index for index, target in enumerate(c_targets)} for column in group_by.columns: index = None # Process target references by index. if isinstance(column, int): index = column - 1 if not 0 <= index < len(c_targets): raise CompilationError(f'invalid GROUP-BY column index {column}') else: # Process target references by name. These will be parsed as # simple Column expressions. If they refer to a target name, we # resolve them. if isinstance(column, ast.Column): name = column.name index = targets_name_map.get(name, None) # Otherwise we compile the expression and add it to the list of # targets to evaluate and index into that new target. if index is None: c_expr = self._compile(column) # Check if the new expression is an aggregate. aggregate = is_aggregate(c_expr) if aggregate: raise CompilationError(f'GROUP-BY expressions may not be aggregates: "{column}"') # Attempt to reconcile the expression with one of the existing # target expressions. try: index = c_target_expressions.index(c_expr) except ValueError: # Add the new target. 'None' for the target name implies it # should be invisible, not to be rendered. index = len(new_targets) new_targets.append(EvalTarget(c_expr, None, aggregate)) c_target_expressions.append(c_expr) assert index is not None, "Internal error, could not index group-by reference." group_indexes.append(index) # Check that the group-by column references a non-aggregate. c_expr = new_targets[index].c_expr if is_aggregate(c_expr): raise CompilationError(f'GROUP-BY expressions may not reference aggregates: "{column}"') # Check that the group-by column has a supported hashable type. if not issubclass(c_expr.dtype, collections.abc.Hashable): raise CompilationError(f'GROUP-BY a non-hashable type is not supported: "{column}"') # Compile HAVING clause. if group_by.having is not None: c_expr = self._compile(group_by.having) if not is_aggregate(c_expr): raise CompilationError('the HAVING clause must be an aggregate expression') having_index = len(new_targets) new_targets.append(EvalTarget(c_expr, None, True)) c_target_expressions.append(c_expr) else: # If it does not have a GROUP-BY clause... aggregate_bools = [c_target.is_aggregate for c_target in c_targets] if any(aggregate_bools): # If the query is an aggregate query, check that all the targets are # aggregates. if all(aggregate_bools): # FIXME: shold we really be checking for the empty # list or is checking for a false value enough? assert group_indexes == [] elif SUPPORT_IMPLICIT_GROUPBY: # If some of the targets aren't aggregates, automatically infer # that they are to be implicit group by targets. This makes for # a much more convenient syntax for our lightweight SQL, where # grouping is optional. group_indexes = [ index for index, c_target in enumerate(c_targets) if not c_target.is_aggregate] else: raise CompilationError('aggregate query without a GROUP-BY should have only aggregates') else: # This is not an aggregate query; don't set group_indexes to # anything useful, we won't need it. group_indexes = None return new_targets[len(c_targets):], group_indexes, having_index @_compile.register def _column(self, node: ast.Column): column = self.table.columns.get(node.name) if column is not None: return column raise CompilationError(f'column "{node.name}" not found in table "{self.table.name}"', node) @_compile.register def _or(self, node: ast.Or): return EvalOr([self._compile(arg) for arg in node.args]) @_compile.register def _and(self, node: ast.And): return EvalAnd([self._compile(arg) for arg in node.args]) _OPERATORS = { '<': ast.Less, '<=': ast.LessEq, '>': ast.Greater, '>=': ast.GreaterEq, '=': ast.Equal, '!=': ast.NotEqual, '~': ast.Match, '!~': ast.NotMatch, '?~': ast.Matches, } # dispatching on an Union is supported only starting with Python 3.11 @_compile.register(ast.All) @_compile.register(ast.Any) def _all(self, node): right = self._compile(node.right) if isinstance(right, EvalQuery): if len(right.columns) != 1: raise CompilationError('subquery has too many columns', node.right) right = EvalConstantSubquery1D(right) right_dtype = typing.get_origin(right.dtype) or right.dtype if right_dtype not in {list, set}: raise CompilationError(f'not a list or set but {right_dtype}', node.right) args = typing.get_args(right.dtype) if args: assert len(args) == 1 right_element_dtype = args[0] else: right_element_dtype = object left = self._compile(node.left) # lookup operator implementaton and check typing op = self._OPERATORS[node.op] for func in OPERATORS[op]: if func.__intypes__ == [right_element_dtype, left.dtype]: break else: raise CompilationError( f'operator "{op.__name__.lower()}(' f'{left.dtype.__name__}, {right_element_dtype.__name__})" not supported', node) # need to instantiate the operaotr implementation to get to the underlying function operator = func(None, None).operator cls = EvalAll if type(node) is ast.All else EvalAny return cls(operator, left, right) @_compile.register def _function(self, node: ast.Function): operands = [self._compile(operand) for operand in node.operands] # ``row(*)`` is parsed like a function call but does something special if node.fname == 'row' and len(operands) == 1 and operands[0].dtype == types.Asterisk: return EvalRow() # ``coalesce()`` is parsed like a function call but it does # not really fit our model for function evaluation, therefore # it gets special threatment here. if node.fname == 'coalesce': for operand in operands: if operand.dtype != operands[0].dtype: dtypes = ', '.join(operand.dtype.__name__ for operand in operands) raise CompilationError(f'coalesce() function arguments must have uniform type, found: {dtypes}', node) return EvalCoalesce(operands) function = types.function_lookup(FUNCTIONS, node.fname, operands) if function is None: sig = '{}({})'.format(node.fname, ', '.join(f'{operand.dtype.__name__.lower()}' for operand in operands)) raise CompilationError(f'no function matches "{sig}" name and argument types', node) # Replace ``meta(key)`` with ``meta[key]``. if node.fname == 'meta': key = node.operands[0] node = ast.Function('getitem', [ast.Column('meta', parseinfo=node.parseinfo), key]) return self._compile(node) # Replace ``entry_meta(key)`` with ``entry.meta[key]``. if node.fname == 'entry_meta': key = node.operands[0] node = ast.Function('getitem', [ast.Attribute(ast.Column('entry', parseinfo=node.parseinfo), 'meta'), key]) return self._compile(node) # Replace ``any_meta(key)`` with ``getitem(meta, key, entry.meta[key])``. if node.fname == 'any_meta': key = node.operands[0] node = ast.Function('getitem', [ast.Column('meta', parseinfo=node.parseinfo), key, ast.Function('getitem', [ ast.Attribute(ast.Column('entry', parseinfo=node.parseinfo), 'meta'), key])]) return self._compile(node) # Replace ``has_account(regexp)`` with ``('(?i)' + regexp) ~? any (accounts)``. if node.fname == 'has_account': node = ast.Any(ast.Add(ast.Constant('(?i)'), node.operands[0]), '?~', ast.Column('accounts')) return self._compile(node) function = function(self.context, operands) # Constants folding. if all(isinstance(operand, EvalConstant) for operand in operands) and function.pure: return EvalConstant(function(None), function.dtype) return function @_compile.register def _subscript(self, node: ast.Subscript): operand = self._compile(node.operand) if issubclass(operand.dtype, dict): return EvalGetItem(operand, node.key) raise CompilationError('column type is not subscriptable', node) @_compile.register def _attribute(self, node: ast.Attribute): operand = self._compile(node.operand) dtype = types.ALIASES.get(operand.dtype, operand.dtype) if issubclass(dtype, types.Structure): getter = dtype.columns.get(node.name) if getter is None: raise CompilationError(f'structured type has no attribute "{node.name}"', node) return EvalGetter(operand, getter, getter.dtype) raise CompilationError('column type is not structured', node) @_compile.register def _unaryop(self, node: ast.UnaryOp): operand = self._compile(node.operand) function = types.function_lookup(OPERATORS, type(node), [operand]) if function is None: raise CompilationError( f'operator "{type(node).__name__.lower()}({types.name(operand.dtype)})" not supported', node) function = function(operand) # Constants folding. if isinstance(operand, EvalConstant): return EvalConstant(function(None), function.dtype) return function @_compile.register def _between(self, node: ast.Between): operand = self._compile(node.operand) lower = self._compile(node.lower) upper = self._compile(node.upper) intypes = [operand.dtype, lower.dtype, upper.dtype] for candidate in OPERATORS[type(node)]: if candidate.__intypes__ == intypes: func = candidate(operand, lower, upper) return func raise CompilationError( f'operator "{types.name(operand.dtype)} BETWEEN {types.name(lower.dtype)} ' f'AND {types.name(upper.dtype)}" not supported', node) @_compile.register(ast.In) @_compile.register(ast.NotIn) def _inop(self, node: Union[ast.In, ast.NotIn]): left = self._compile(node.left) right = self._compile(node.right) if isinstance(right, EvalQuery): if len(right.columns) != 1: raise CompilationError('subquery has too many columns', node.right) right = EvalConstantSubquery1D(right) op = OPERATORS[type(node)][0] return op(left, right) @_compile.register def _binaryop(self, node: ast.BinaryOp): left = self._compile(node.left) right = self._compile(node.right) candidates = OPERATORS[type(node)] while True: intypes = [left.dtype, right.dtype] for op in candidates: if op.__intypes__ == intypes: function = op(left, right) # Constants folding. if isinstance(left, EvalConstant) and isinstance(right, EvalConstant): return EvalConstant(function(None), function.dtype) return function # Implement type inference when one of the operands is not strongly typed. if left.dtype is object and right.dtype is not object: target = right.dtype if target is int: # The Beancount parser does not emit int typed # values, thus casting to int is only going to # loose information. Promote to decimal. target = Decimal name = types.MAP.get(target) if name is None: break left = types.function_lookup(FUNCTIONS, name, [left])(self.context, [left]) continue if right.dtype is object and left.dtype is not object: target = left.dtype if target is int: # The Beancount parser does not emit int typed # values, thus casting to int is only going to # loose information. Promote to decimal. target = Decimal name = types.MAP.get(target) if name is None: break right = types.function_lookup(FUNCTIONS, name, [right])(self.context, [right]) continue # Failure. break raise CompilationError( f'operator "{type(node).__name__.lower()}(' f'{types.name(left.dtype)}, {types.name(right.dtype)})" not supported', node) @_compile.register def _constant(self, node: ast.Constant): return EvalConstant(node.value) @_compile.register def _placeholder(self, node: ast.Placeholder): return EvalConstant(self.parameters[node.name]) @_compile.register def _asterisk(self, node: ast.Asterisk): return EvalConstant(None, dtype=types.Asterisk) @_compile.register def _balances(self, node: ast.Balances): return self._compile(transform_balances(node)) @_compile.register def _journal(self, node: ast.Journal): return self._compile(transform_journal(node)) @_compile.register def _print(self, node: ast.Print): self.table = self.context.tables.get('entries') expr = self._compile_from(node.from_clause) targets = [EvalTarget(EvalRow(), 'ROW(*)', False)] return EvalQuery(self.table, targets, expr, None, None, None, None, False) def transform_journal(journal): """Translate a Journal entry into an uncompiled Select statement. Args: journal: An instance of a Journal object. Returns: An instance of an uncompiled Select object. """ cooked_select = parser.parse(""" SELECT date, flag, MAXWIDTH(payee, 48), MAXWIDTH(narration, 80), account, {summary_func}(position), {summary_func}(balance) {where} """.format(where=('WHERE account ~ "{}"'.format(journal.account) if journal.account else ''), summary_func=journal.summary_func or '')) return ast.Select(cooked_select.targets, journal.from_clause, cooked_select.where_clause, None, None, None, None, None) def transform_balances(balances): """Translate a Balances entry into an uncompiled Select statement. Args: balances: An instance of a Balance object. Returns: An instance of an uncompiled Select object. """ ## FIXME: Change the aggregation rules to allow GROUP-BY not to include the ## non-aggregate ORDER-BY columns, so we could just GROUP-BY accounts here ## instead of having to include the sort-key. I think it should be fine if ## the first or last sort-order value gets used, because it would simplify ## the input statement. cooked_select = parser.parse(""" SELECT account, SUM({}(position)) GROUP BY account, ACCOUNT_SORTKEY(account) ORDER BY ACCOUNT_SORTKEY(account) """.format(balances.summary_func or "")) return ast.Select(cooked_select.targets, balances.from_clause, balances.where_clause, cooked_select.group_by, cooked_select.order_by, None, None, None) def get_target_name(target): """Compute the target name. This uses the same algorithm used by SQLite. If the target has an AS clause assigning it a name, that will be the name used. If the target refers directly to a column, then the target name is the column name. Otherwise use the expression text. """ if target.name is not None: return target.name if isinstance(target.expression, ast.Column): return target.expression.name return target.expression.text.strip() def get_columns_and_aggregates(node): """Find the columns and aggregate nodes below this tree. All nodes under aggregate nodes are ignored. Args: node: An instance of EvalNode. Returns: A pair of (columns, aggregates), both of which are lists of EvalNode instances. columns: The list of all columns accessed not under an aggregate node. aggregates: The list of all aggregate nodes. """ columns = [] aggregates = [] _get_columns_and_aggregates(node, columns, aggregates) return columns, aggregates def _get_columns_and_aggregates(node, columns, aggregates): """Walk down a tree of nodes and fetch the column accessors and aggregates. This function ignores all nodes under aggregate nodes. Args: node: An instance of EvalNode. columns: An accumulator for columns found so far. aggregate: An accumulator for aggregate notes found so far. """ if isinstance(node, EvalAggregator): aggregates.append(node) elif isinstance(node, EvalColumn): columns.append(node) else: for child in node.childnodes(): _get_columns_and_aggregates(child, columns, aggregates) def is_aggregate(node): """Return true if the node is an aggregate. Args: node: An instance of EvalNode. Returns: A boolean. """ # Note: We could be a tiny bit more efficient here, but it doesn't matter # much. Performance of the query compilation matters very little overall. _, aggregates = get_columns_and_aggregates(node) return bool(aggregates) def compile(context, statement, parameters=None): return Compiler(context).compile(statement, parameters) beanquery-master/beanquery/cursor.py000066400000000000000000000076351474576771600202670ustar00rootroot00000000000000from operator import attrgetter from typing import Sequence from . import types from . import parser from . import compiler from . import query_execute class Column(Sequence): __module__ = 'beanquery' def __init__(self, name, datatype): self._name = name self._type = datatype _vars = tuple(attrgetter(name) for name in 'name type_code display_size internal_size precision scale null_ok'.split()) def __eq__(self, other): if isinstance(other, type(self)): return tuple(self) == tuple(other) if isinstance(other, tuple): # Used in tests. return (self._name, self._type) == other return NotImplemented def __repr__(self): return f'{self.__module__}.{self.__class__.__name__}({self._name!r}, {types.name(self._type)})' def __len__(self): return 7 def __getitem__(self, key): if isinstance(key, slice): return tuple(getter(self) for getter in self._vars(key)) return self._vars[key](self) @property def name(self): return self._name @property def datatype(self): # Extension to the DB-API. return self._type @property def type_code(self): # The DB-API specification is vague on this point, but other # database connection libraries expose this as an int. It does # not make much sense to keep a mapping between int type code # and actual types, thus just return the hash of the type # object. return hash(self._type) @property def display_size(self): return None @property def internal_size(self): return None @property def precision(self): return None @property def scale(self): return None @property def null_ok(self): return None class Cursor: def __init__(self, connection): self._context = connection self._description = None self._rows = None self._pos = 0 self.arraysize = 1 @property def connection(self): return self._context def execute(self, query, params=None): if not isinstance(query, parser.ast.Node): query = parser.parse(query) query = compiler.compile(self._context, query, params) description, rows = query_execute.execute_query(query) self._description = description self._rows = rows self._pos = 0 return self def executemany(self, query, params=None): query = parser.parse(query) for p in params: self.execute(query, p) @property def description(self): return self._description @property def rowcount(self): return len(self._rows) if self._rows is not None else -1 @property def rownumber(self): return self._pos def fetchone(self): # This implementation pops items from the front of the results # rows list and is thus not efficient, especially for large # results sets. if self._rows is None or not len(self._rows): return None self._pos += 1 return self._rows.pop(0) def fetchmany(self, size=None): if self._rows is None: return [] n = size if size is not None else self.arraysize rows = self._rows[:n] self._rows = self._rows[n:] self._pos += len(rows) return rows def fetchall(self): if self._rows is None: return [] rows = self._rows self._rows = [] self._pos += len(rows) return rows def close(self): # Required by the DB-API. pass def setinputsizes(self, sizes): # Required by the DB-API. pass def setoutputsize(self, size, column=None): # Required by the DB-API. pass def __iter__(self): return iter(self._rows if self._rows is not None else []) beanquery-master/beanquery/cursor_test.py000066400000000000000000000050641474576771600213200ustar00rootroot00000000000000import unittest import sqlite3 import beanquery from beanquery.sources import test class APITests: def test_description(self): curs = self.conn.cursor() self.assertIsNone(curs.description) curs.execute(f'SELECT x FROM {self.table} WHERE x = 0') self.assertEqual([c[0] for c in curs.description], ['x']) column = curs.description[0] self.assertEqual(len(column), 7) def test_cursor_not_initialized(self): curs = self.conn.cursor() self.assertIsNone(curs.fetchone()) self.assertEqual(curs.fetchmany(), []) self.assertEqual(curs.fetchall(), []) def test_cursor_fetchone(self): curs = self.conn.cursor() curs.execute(f'SELECT x FROM {self.table} WHERE x < 2') row = curs.fetchone() self.assertEqual(row, (0, )) row = curs.fetchone() self.assertEqual(row, (1, )) row = curs.fetchone() self.assertIsNone(row) def test_cursor_fetchall(self): curs = self.conn.cursor() curs.execute(f'SELECT x FROM {self.table} WHERE x < 2') rows = curs.fetchall() self.assertEqual(rows, [(0, ), (1, )]) rows = curs.fetchall() self.assertEqual(rows, []) def test_cursor_fethmany(self): curs = self.conn.cursor() curs.execute(f'SELECT x FROM {self.table} WHERE x < 2') rows = curs.fetchmany() self.assertEqual(rows, [(0, )]) rows = curs.fetchmany() self.assertEqual(rows, [(1, )]) rows = curs.fetchmany() self.assertEqual(rows, []) def test_cursor_iterator(self): curs = self.conn.cursor() o = object() row = next(iter(curs), o) self.assertIs(row, o) curs = self.conn.cursor() curs.execute(f'SELECT x FROM {self.table} WHERE x < 2') iterator = iter(curs) row = next(iterator) self.assertEqual(row, (0, )) row = next(iterator) self.assertEqual(row, (1, )) row = next(iterator, o) self.assertIs(row, o) class TestSQLite(APITests, unittest.TestCase): @classmethod def setUpClass(cls): cls.table = 'test' cls.conn = sqlite3.connect(':memory:') curs = cls.conn.cursor() curs.execute('CREATE TABLE test (x int)') curs.executemany('INSERT INTO test VALUES (?)', [(i, ) for i in range(16)]) class TestBeanquery(APITests, unittest.TestCase): @classmethod def setUpClass(cls): cls.table = '#test' cls.conn = beanquery.Connection() cls.conn.tables['test'] = test.Table(16) beanquery-master/beanquery/errors.py000066400000000000000000000031051474576771600202520ustar00rootroot00000000000000""" Exceptions hierarchy defined by the DB-API: Exception Warning Error InterfaceError DatabaseError DataError OperationalError IntegrityError InternalError ProgrammingError NotSupportedError """ class Warning(Exception): """Exception raised for important warnings.""" __module__ = 'beanquery' class Error(Exception): """Base exception for all errors.""" __module__ = 'beanquery' class InterfaceError(Error): """An error related to the database interface rather than the database itself.""" __module__ = 'beanquery' class DatabaseError(Error): """Exception raised for errors that are related to the database.""" __module__ = 'beanquery' class DataError(DatabaseError): """An error caused by problems with the processed data.""" __module__ = 'beanquery' class OperationalError(DatabaseError): """An error related to the database's operation.""" __module__ = 'beanquery' class IntegrityError(DatabaseError): """An error caused when the relational integrity of the database is affected.""" __module__ = 'beanquery' class InternalError(DatabaseError): """An error generated when the database encounters an internal error.""" __module__ = 'beanquery' class ProgrammingError(DatabaseError): """Exception raised for programming errors.""" __module__ = 'beanquery' class NotSupportedError(DatabaseError): """A method or database API was used which is not supported by the database.""" __module__ = 'beanquery' beanquery-master/beanquery/hashable.py000066400000000000000000000045021474576771600205070ustar00rootroot00000000000000import decimal import pickle import textwrap from . import types # Hashable types. Checking ``issubclass(T, types.Hashable)`` does not # work because named tuples pass that test and beancount uses many named # tuples that have dictionary members making them effectively not # hashable. FUNDAMENTAL = frozenset({ bool, bytes, complex, decimal.Decimal, float, int, str, # These are hashable only if the contained objects are hashable. frozenset, tuple, }) # Function reducing non-hashable types to something hashable. REDUCERS = {} def register(datatype, func): """Register reduce function for non-hashable type. The reduce function maps an non-hashable object into an hashable representation. This representation does not need to capture all the object facets, but it should retrurn someting unique enough to avoid too many hashing collisions. """ REDUCERS[datatype] = func def make(columns): """Build an hashable tuple subclass.""" # When all columns are hashable, pass the input tuple through as is. if all(column.datatype in FUNDAMENTAL for column in columns): return lambda x: x datatypes = ', '.join(types.name(column.datatype) for column in columns) # Code generation inspired by standard library ``dataclasses.py``. parts = [] locals = {} for i, column in enumerate(columns): if column.datatype in FUNDAMENTAL: parts.append(f'self[{i}]') elif column.datatype is dict: parts.append(f'*self[{i}].keys(), *self[{i}].values()') elif column.datatype is set: parts.append(f'*self[{i}]') else: func = REDUCERS.get(column.datatype, pickle.dumps) fname = f'func{i}' locals[fname] = func parts.append(f'{fname}(self[{i}])') objs = ', '.join(parts) names = ', '.join(locals.keys()) code = textwrap.dedent(f''' def create({names}): def __hash__(self): return hash(({objs})) return __hash__ ''') clsname = f'Hashable[{datatypes}]' ns = {} exec(code, globals(), ns) func = ns['create'](**locals) func.__qualname__ = f'{clsname}.{func.__name__}' members = dict(tuple.__dict__) members['__hash__'] = func return type(clsname, (tuple,), members) beanquery-master/beanquery/hashable_test.py000066400000000000000000000026721474576771600215540ustar00rootroot00000000000000import dataclasses import unittest from beanquery import hashable from beanquery.cursor import Column class TestHashable(unittest.TestCase): def test_fundamental(self): columns = (Column('b', bool), Column('i', int), Column('s', str)) wrap = hashable.make(columns) obj = (True, 42, 'universe') self.assertIs(wrap(obj), obj) hash(obj) def test_dict(self): columns = (Column('b', bool), Column('d', dict)) wrap = hashable.make(columns) obja = (True, {'answer': 42}) a = hash(wrap(obja)) objb = (True, {'answer': 42}) b = hash(wrap(objb)) self.assertIsNot(obja, objb) self.assertEqual(a, b) objc = (False, {'answer': 42}) c = hash(wrap(objc)) self.assertNotEqual(a, c) objd = (True, {'answer': 43}) d = hash(wrap(objd)) self.assertNotEqual(a, d) def test_registered(self): @dataclasses.dataclass class Foo: xid: int meta: dict hashable.register(Foo, lambda obj: obj.xid) columns = (Column('b', bool), Column('foo', Foo)) wrap = hashable.make(columns) obja = (True, Foo(1, {'test': 1})) a = hash(wrap(obja)) objb = (True, Foo(1, {'test': 2})) b = hash(wrap(objb)) self.assertEqual(a, b) objc = (True, Foo(2, {'test': 2})) c = hash(wrap(objc)) self.assertNotEqual(a, c) beanquery-master/beanquery/numberify.py000066400000000000000000000203241474576771600207400ustar00rootroot00000000000000"""Code to split table columns containing amounts and inventories into number columns. For example, given a column with this content: ----- amount ------ 101.23 USD 200 JPY 99.23 USD 38.34 USD, 100 JPY We can convert this into two columns and remove the currencies: -amount (USD)- -amount (JPY)- 101.23 200 99.23 38.34 100 The point is that the columns should be typed as numbers to make this importable into a spreadsheet and able to be processed. Notes: * This handles the Amount, Position and Inventory datatypes. There is code to automatically recognize columns containing such types from a table of strings and convert such columns to their corresponding guessed data types. * The per-currency columns are ordered in decreasing order of the number of instances of numbers seen for each currency. So if the most numbers you have in a column are USD, then the USD column renders first. * Cost basis specifications should be unmodified and reported to a dedicated extra column, like this: ----- amount ------ 1 AAPL {21.23 USD} We can convert this into two columns and remove the currencies: -amount (AAPL)- -Cost basis- 1 {21.23 USD} (Eventually we might support the conversion of cost amounts as well, but they may contain other information, such as a label or a date, so for now we don't convert them. I'm not sure there's a good practical use case in doing that yet.) * We may provide some options to break out only some of the currencies into columns, in order to handle the case where an inventory contains a large number of currencies and we want to only operate on a restricted set of operating currencies. * If you provide a DisplayFormatter object to the numberification routine, they quantize each column according to their currency's precision. It is recommended that you do that. """ __copyright__ = "Copyright (C) 2015-2017 Martin Blais" __license__ = "GNU GPLv2" import collections from decimal import Decimal from beancount.core import amount from beancount.core import position from beancount.core import inventory from .cursor import Column def numberify_results(columns, drows, dformat=None): """Number rows containing Amount, Position or Inventory types. Args: result_types: A list of items describing the names and data types of the items in each column. result_rows: A list of ResultRow instances. dformat: An optional DisplayFormatter. If set, quantize the numbers by their currency-specific precision when converting the Amount's, Position's or Inventory'es.. Returns: A pair of modified (result_types, result_rows) with converted datatypes. """ # Build an array of converters. converters = [] for index, column in enumerate(columns): convert_col_fun = CONVERTING_TYPES.get(column.datatype) if convert_col_fun is None: converters.append(IdentityConverter(column.name, column.datatype, index)) else: col_converters = convert_col_fun(column.name, drows, index) converters.extend(col_converters) # Derive the output types from the expected outputs from the converters # themselves. otypes = tuple(Column(c.name, c.dtype) for c in converters) # Convert the input rows by processing them through the converters. orows = [] for drow in drows: orow = [] for converter in converters: orow.append(converter(drow, dformat)) orows.append(orow) return otypes, orows class IdentityConverter: """A converter that simply copies its column.""" def __init__(self, name, dtype, index): self.name = name self.dtype = dtype self.index = index def __call__(self, drow, _): return drow[self.index] class AmountConverter: """A converter that extracts the number of an amount for a specific currency.""" dtype = Decimal def __init__(self, name, index, currency): self.name = name self.index = index self.currency = currency def __call__(self, drow, dformat): vamount = drow[self.index] if vamount and vamount.currency == self.currency: number = vamount.number if dformat: number = dformat.quantize(number, self.currency) else: number = None return number def convert_col_Amount(name, drows, index): """Create converters for a column of type Amount. Args: name: A string, the column name. drows: The table of objects. index: The column number. Returns: A list of Converter instances, one for each of the currency types found. """ currency_map = collections.defaultdict(int) for drow in drows: vamount = drow[index] if vamount and vamount.currency: currency_map[vamount.currency] += 1 return [AmountConverter('{} ({})'.format(name, currency), index, currency) for currency, _ in sorted(currency_map.items(), key=lambda item: (item[1], item[0]), reverse=True)] class PositionConverter: """A converter that extracts the number of a position for a specific currency.""" dtype = Decimal def __init__(self, name, index, currency): self.name = name self.index = index self.currency = currency def __call__(self, drow, dformat): pos = drow[self.index] if pos and pos.units.currency == self.currency: number = pos.units.number if dformat: number = dformat.quantize(pos.units.number, self.currency) else: number = None return number def convert_col_Position(name, drows, index): """Create converters for a column of type Position. Args: name: A string, the column name. drows: The table of objects. index: The column number. Returns: A list of Converter instances, one for each of the currency types found. """ currency_map = collections.defaultdict(int) for drow in drows: pos = drow[index] if pos and pos.units.currency: currency_map[pos.units.currency] += 1 return [PositionConverter('{} ({})'.format(name, currency), index, currency) for currency, _ in sorted(currency_map.items(), key=lambda item: (item[1], item[0]), reverse=True)] class InventoryConverter: """A converter that extracts the number of a inventory for a specific currency. If there are multiple lots we aggregate by currency.""" dtype = Decimal def __init__(self, name, index, currency): self.name = name self.index = index self.currency = currency def __call__(self, drow, dformat): inv = drow[self.index] # FIXME:: get_currency_units() returns ZERO and not None when the value # isn't present. This should be fixed to distinguish between the two. number = inv.get_currency_units(self.currency).number if number and dformat: number = dformat.quantize(number, self.currency) return number or None def convert_col_Inventory(name, drows, index): """Create converters for a column of type Inventory. Args: name: A string, the column name. drows: The table of objects. index: The column number. Returns: A list of Converter instances, one for each of the currency types found. """ currency_map = collections.defaultdict(int) for drow in drows: inv = drow[index] for currency in inv.currencies(): currency_map[currency] += 1 return [InventoryConverter('{} ({})'.format(name, currency), index, currency) for currency, _ in sorted(currency_map.items(), key=lambda item: (item[1], item[0]), reverse=True)] # A mapping of data types to their converter factory. CONVERTING_TYPES = { amount.Amount : convert_col_Amount, position.Position : convert_col_Position, inventory.Inventory : convert_col_Inventory, } beanquery-master/beanquery/numberify_test.py000066400000000000000000000131231474576771600217760ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2015-2017 Martin Blais" __license__ = "GNU GPLv2" import datetime import unittest from decimal import Decimal from beancount.core.number import D from beancount.core.amount import A from beancount.core import amount from beancount.core import position from beancount.core import inventory from beancount.core import display_context from beanquery import numberify from beanquery.cursor import Column class TestNumerifySimple(unittest.TestCase): input_amounts = ["24.17 CAD", "-77.02 CAD", "11.39 CAD", "800.00 USD", "41.17 CAD", "950.00 USD", "110 JPY", "-947.00 USD"] expected_types = (('pos (CAD)', Decimal), ('pos (USD)', Decimal), ('pos (JPY)', Decimal)) expected_rows = [[D('24.17'), None, None], [D('-77.02'), None, None], [D('11.39'), None, None], [None, D('800.00'), None], [D('41.17'), None, None], [None, D('950.00'), None], [None, None, D('110')], [None, D('-947.00'), None]] def test_amount(self): itypes = (Column('pos', amount.Amount), ) irows = [(A(string),) for string in self.input_amounts] atypes, arows = numberify.numberify_results(itypes, irows) self.assertEqual(self.expected_types, atypes) self.assertEqual(self.expected_rows, arows) def test_position(self): itypes = (Column('pos', position.Position), ) irows = [(position.from_string(string),) for string in self.input_amounts] atypes, arows = numberify.numberify_results(itypes, irows) self.assertEqual(self.expected_types, atypes) self.assertEqual(self.expected_rows, arows) def test_inventory(self): itypes = (Column('pos', inventory.Inventory), ) irows = [(inventory.from_string(string),) for string in self.input_amounts] atypes, arows = numberify.numberify_results(itypes, irows) self.assertEqual(self.expected_types, atypes) self.assertEqual(self.expected_rows, arows) class TestNumerifyIdentity(unittest.TestCase): def test_identity(self): itypes = (Column('date', datetime.date), Column('name', str), Column('count', int), ) irows = [[datetime.date(2015, 9, 8), 'Testing', 3]] atypes, arows = numberify.numberify_results(itypes, irows) self.assertEqual(itypes, atypes) self.assertEqual(irows, arows) class TestNumerifyInventory(unittest.TestCase): def test_inventory(self): itypes = (Column('balance', inventory.Inventory), ) irows = [[inventory.from_string('10 HOOL {23.00 USD}')], [inventory.from_string('2.11 USD, 3.44 CAD')], [inventory.from_string('-2 HOOL {24.00 USD}, 5.66 CAD')]] atypes, arows = numberify.numberify_results(itypes, irows) self.assertEqual((('balance (HOOL)', Decimal), ('balance (CAD)', Decimal), ('balance (USD)', Decimal), ), atypes) self.assertEqual([[D('10'), None, None], [None, D('3.44'), D('2.11')], [D('-2'), D('5.66'), None]], arows) class TestNumerifyPrecision(unittest.TestCase): def test_precision(self): # Some display context. dcontext = display_context.DisplayContext() dcontext.update(D('111'), 'JPY') dcontext.update(D('1.111'), 'RGAGX') dcontext.update(D('1.11'), 'USD') dformat = dcontext.build() # Input data. itypes = (Column('number', Decimal), Column('amount', amount.Amount), Column('position', position.Position), Column('inventory', inventory.Inventory)) irows = [[D(amt.split()[0]), A(amt), position.from_string(amt), inventory.from_string(amt)] for amt in ['123.45678909876 JPY', '1.67321232123 RGAGX', '5.67345434543 USD']] # First check with no explicit quantization. atypes, arows = numberify.numberify_results(itypes, irows) erows = [[D('123.45678909876'), None, None, D('123.45678909876'), None, None, D('123.45678909876'), None, None, D('123.45678909876')], [D('1.67321232123'), None, D('1.67321232123'), None, None, D('1.67321232123'), None, None, D('1.67321232123'), None], [D('5.67345434543'), D('5.67345434543'), None, None, D('5.67345434543'), None, None, D('5.67345434543'), None, None]] self.assertEqual(erows, arows) # Then compare with quantization. atypes, arows = numberify.numberify_results(itypes, irows, dformat) erows = [[D('123.45678909876'), None, None, D('123'), None, None, D('123'), None, None, D('123')], [D('1.67321232123'), None, D('1.673'), None, None, D('1.673'), None, None, D('1.673'), None], [D('5.67345434543'), D('5.67'), None, None, D('5.67'), None, None, D('5.67'), None, None]] self.assertEqual(erows, arows) if __name__ == '__main__': unittest.main() beanquery-master/beanquery/parser/000077500000000000000000000000001474576771600176615ustar00rootroot00000000000000beanquery-master/beanquery/parser/__init__.py000066400000000000000000000030211474576771600217660ustar00rootroot00000000000000import datetime import decimal import tatsu from ..errors import ProgrammingError from .parser import BQLParser from . import ast class BQLSemantics: def set_context(self, ctx): self._ctx = ctx def null(self, value): return None def integer(self, value): return int(value) def decimal(self, value): return decimal.Decimal(value) def date(self, value): return datetime.date.fromisoformat(value) def string(self, value): return value[1:-1] def boolean(self, value): return value == 'TRUE' def identifier(self, value): return value.lower() def asterisk(self, value): return ast.Asterisk() def list(self, value): return list(value) def ordering(self, value): return ast.Ordering[value or 'ASC'] def _default(self, value, typename=None): if typename is not None: func = getattr(ast, typename) return func(**{name.rstrip('_'): value for name, value in value.items()}) return value class ParseError(ProgrammingError): def __init__(self, parseinfo): super().__init__('syntax error') self.parseinfo = parseinfo def parse(text): try: return BQLParser().parse(text, semantics=BQLSemantics()) except tatsu.exceptions.ParseError as exc: line = exc.tokenizer.line_info(exc.pos).line parseinfo = tatsu.infos.ParseInfo(exc.tokenizer, exc.item, exc.pos, exc.pos + 1, line, []) raise ParseError(parseinfo) from exc beanquery-master/beanquery/parser/ast.py000066400000000000000000000172261474576771600210320ustar00rootroot00000000000000from __future__ import annotations import dataclasses import datetime import enum import sys import textwrap import typing if typing.TYPE_CHECKING: from typing import Any, Optional, Union def _indent(text): return textwrap.indent(text, ' ') def _fields(node): for field in dataclasses.fields(node): if field.repr: yield field.name, getattr(node, field.name) def tosexp(node): if isinstance(node, Node): return f'({node.__class__.__name__.lower()}\n' + _indent( '\n'.join(f'{name.replace("_", "-")}: {tosexp(value)}' for name, value in _fields(node) if value is not None) + ')') if isinstance(node, list): return '(\n' + _indent('\n'.join(tosexp(i) for i in node)) + ')' if isinstance(node, enum.Enum): return node.name.lower() return repr(node) def walk(node): if isinstance(node, Node): for name, child in _fields(node): yield from walk(child) yield node if isinstance(node, list): for child in node: yield from walk(child) class Node: """Base class for BQL AST nodes.""" __slots__ = () parseinfo = None @property def text(self): if not self.parseinfo: return None text = self.parseinfo.tokenizer.text return text[self.parseinfo.pos:self.parseinfo.endpos] def tosexp(self): return tosexp(self) def walk(self): return walk(self) def node(name, fields): """Manufacture an AST node class.""" return dataclasses.make_dataclass( name, [*fields.split(), ('parseinfo', None, dataclasses.field(default=None, compare=False, repr=False))], bases=(Node,), **({'slots': True} if sys.version_info[:2] >= (3, 10) else {})) # A 'select' query action. # # Attributes: # targets: Either a single 'Asterisk' instance of a list of 'Target' # instances. # from_clause: An instance of 'From', or None if absent. # where_clause: A root expression node, or None if absent. # group_by: An instance of 'GroupBy', or None if absent. # order_by: An instance of 'OrderBy', or None if absent. # pivot_by: An instance of 'PivotBy', or None if absent. # limit: An integer, or None is absent. # distinct: A boolean value (True), or None if absent. Select = node('Select', 'targets from_clause where_clause group_by order_by pivot_by limit distinct') # A select query that produces final balances for accounts. # This is equivalent to # # SELECT account, sum(position) # FROM ... # WHERE ... # GROUP BY account # # Attributes: # summary_func: A method on an inventory to call on the position column. # May be to extract units, value at cost, etc. # from_clause: An instance of 'From', or None if absent. Balances = node('Balances', 'summary_func from_clause where_clause') # A select query that produces a journal of postings. # This is equivalent to # # SELECT date, flag, payee, narration, ... FROM # WHERE account = # # Attributes: # account: A string, the name of the account to restrict to. # summary_func: A method on an inventory to call on the position column. # May be to extract units, value at cost, etc. # from_clause: An instance of 'From', or None if absent. Journal = node('Journal', 'account summary_func from_clause') # A query that will simply print the selected entries in Beancount format. # # Attributes: # from_clause: An instance of 'From', or None if absent. Print = node('Print', 'from_clause') # A parsed SELECT column or target. # # Attributes: # expression: A tree of expression nodes from the parser. # name: A string, the given name of the target (given by "AS "). Target = node('Target', 'expression name') # A placeholder in SELECT * or COUNT(*) constructs. Asterisk = node('Asterisk', '') # A FROM clause. # # Attributes: # expression: A tree of expression nodes from the parser. # close: A CLOSE clause, either None if absent, a boolean if the clause # was present by no date was provided, or a datetime.date instance if # a date was provided. @dataclasses.dataclass(**({'slots': True} if sys.version_info[:2] >= (3, 10) else {})) class From(Node): expression: Optional[Node] = None open: Optional[datetime.date] = None close: Optional[Union[datetime.date, bool]] = None clear: Optional[bool] = None parseinfo: Any = dataclasses.field(default=None, compare=False, repr=False) # A GROUP BY clause. # # Attributes: # columns: A list of group-by expressions, simple Column() or otherwise. # having: An expression tree for the optional HAVING clause, or None. GroupBy = node('GroupBy', 'columns having') # An ORDER BY clause. # # Attributes: # column: order-by expression, simple Column() or otherwise. # ordering: The sort order as an Ordering enum value. OrderBy = node('OrderBy', 'column ordering') class Ordering(enum.IntEnum): # The enum values are chosen in this way to be able to use them # directly as the reverse parameter to the list sort() method. ASC = 0 DESC = 1 def __repr__(self): return f"{self.__class__.__name__}.{self.name}" # An PIVOT BY clause. # # Attributes: # columns: A list of group-by expressions, simple Column() or otherwise. PivotBy = node('PivotBy', 'columns') # A reference to a table. # # Attributes: # name: The table name. Table = node('Table', 'name') # A reference to a column. # # Attributes: # name: A string, the name of the column to access. Column = node('Column', 'name') # A function call. # # Attributes: # fname: A string, the name of the function. # operands: A list of other expressions, the arguments of the function to # evaluate. This is possibly an empty list. Function = node('Function', 'fname operands') Attribute = node('Attribute', 'operand name') Subscript = node('Subscript', 'operand key') # A constant node. # # Attributes: # value: The constant value this represents. Constant = node('Constant', 'value') # A query parameter placeholder. # # Attributes: # name: The placeholder name Placeholder = node('Placeholder', 'name') # Base class for unary operators. # # Attributes: # operand: An expression, the operand of the operator. UnaryOp = node('UnaryOp', 'operand') # Base class for binary operators. # # Attributes: # left: An expression, the left operand. # right: An expression, the right operand. BinaryOp = node('BinaryOp', 'left right') # Base class for boolean operators. BoolOp = node('BoolOp', 'args') # Between Between = node('Between', 'operand lower upper') # Negation operator. class Not(UnaryOp): __slots__ = () class IsNull(UnaryOp): __slots__ = () class IsNotNull(UnaryOp): __slots__ = () # Boolean operators. class And(BoolOp): __slots__ = () class Or(BoolOp): __slots__ = () # Equality and inequality comparison operators. class Equal(BinaryOp): __slots__ = () class NotEqual(BinaryOp): __slots__ = () class Greater(BinaryOp): __slots__ = () class GreaterEq(BinaryOp): __slots__ = () class Less(BinaryOp): __slots__ = () class LessEq(BinaryOp): __slots__ = () # Regular expression match operator. class Match(BinaryOp): __slots__ = () class NotMatch(BinaryOp): __slots__ = () class Matches(BinaryOp): __slots__ = () # Membership operators. class In(BinaryOp): __slots__ = () class NotIn(BinaryOp): __slots__ = () # Arithmetic operators. class Neg(UnaryOp): __slots__ = () class Mul(BinaryOp): __slots__ = () class Div(BinaryOp): __slots__ = () class Mod(BinaryOp): __slots__ = () class Add(BinaryOp): __slots__ = () class Sub(BinaryOp): __slots__ = () Any = node('Any', 'left op right') All = node('All', 'left op right') beanquery-master/beanquery/parser/bql.ebnf000066400000000000000000000125131474576771600212750ustar00rootroot00000000000000@@grammar :: BQL @@parseinfo :: True @@ignorecase :: True @@keyword :: 'AND' 'AS' 'ASC' 'BY' 'DESC' 'DISTINCT' 'FALSE' 'FROM' 'GROUP' 'HAVING' 'IN' 'IS' 'LIMIT' 'NOT' 'OR' 'ORDER' 'PIVOT' 'SELECT' 'TRUE' 'WHERE' @@keyword :: 'BALANCES' 'JOURNAL' 'PRINT' @@comments :: /(\/\*([^*]|[\r\n]|(\*+([^*\/]|[\r\n])))*\*+\/)/ @@eol_comments :: /\;[^\n]*?$/ bql = @:statement [';'] $ ; statement = | select | balances | journal | print ; select::Select = 'SELECT' ['DISTINCT' distinct:`True`] targets:(','.{ target }+ | asterisk) ['FROM' from_clause:(table | subselect | from)] ['WHERE' where_clause:expression] ['GROUP' 'BY' group_by:groupby] ['ORDER' 'BY' order_by:','.{order}+] ['PIVOT' 'BY' pivot_by:pivotby] ['LIMIT' limit:integer] ; subselect = '(' @:select ')' ; from::From = | 'OPEN' ~ 'ON' open:date ['CLOSE' ('ON' close:date | {} close:`True`)] ['CLEAR' clear:`True`] | 'CLOSE' ~ ('ON' close:date | {} close:`True`) ['CLEAR' clear:`True`] | 'CLEAR' ~ clear:`True` | expression:expression ['OPEN' 'ON' open:date] ['CLOSE' ('ON' close:date | {} close:`True`)] ['CLEAR' clear:`True`] ; table::Table = name:/#([a-zA-Z_][a-zA-Z0-9_]*)?/ ; groupby::GroupBy = columns:','.{ (integer | expression) }+ ['HAVING' having:expression] ; order::OrderBy = column:(integer | expression) ordering:ordering ; ordering = ['DESC' | 'ASC'] ; pivotby::PivotBy = columns+:(integer | column) ',' columns+:(integer | column) ; target::Target = expression:expression ['AS' name:identifier] ; expression = | disjunction | conjunction ; disjunction = | or | conjunction ; or::Or::BoolOp = args+:conjunction { 'OR' args+:conjunction }+ ; conjunction = | and | inversion ; and::And::BoolOp = args+:inversion { 'AND' args+:inversion }+ ; inversion = | not | comparison ; not::Not::UnaryOp = 'NOT' operand:inversion ; comparison = | any | all | lt | lte | gt | gte | eq | neq | in | notin | match | notmatch | matches | isnull | isnotnull | between | sum ; any::Any = left:sum op:op 'any' '(' right:expression ')' ; all::All = left:sum op:op 'all' '(' right:expression ')' ; op = | '<' | '<=' | '>' | '>=' | '=' | '!=' | '~' | '!~' | '?~' ; lt::Less::BinaryOp = left:sum '<' right:sum ; lte::LessEq::BinaryOp = left:sum '<=' right:sum ; gt::Greater::BinaryOp = left:sum '>' right:sum ; gte::GreaterEq::BinaryOp = left:sum '>=' right:sum ; eq::Equal::BinaryOp = left:sum '=' right:sum ; neq::NotEqual::BinaryOp = left:sum '!=' right:sum ; in::In::BinaryOp = left:sum 'IN' right:sum ; notin::NotIn::BinaryOp = left:sum 'NOT' 'IN' right:sum ; match::Match::BinaryOp = left:sum '~' right:sum ; notmatch::NotMatch::BinaryOp = left:sum '!~' right:sum ; matches::Matches::BinaryOp = left:sum '?~' right:sum ; isnull::IsNull::UnaryOp = operand:sum 'IS' 'NULL' ; isnotnull::IsNotNull::UnaryOp = operand:sum 'IS' 'NOT' 'NULL' ; between::Between = operand:sum 'BETWEEN' lower:sum 'AND' upper:sum ; sum = | add | sub | term ; add::Add::BinaryOp = left:sum '+' ~ right:term ; sub::Sub::BinaryOp = left:sum '-' ~ right:term ; term = | mul | div | mod | factor ; mul::Mul::BinaryOp = left:term '*' ~ right:factor ; div::Div::BinaryOp = left:term '/' ~ right:factor ; mod::Mod::BinaryOp = left:term '%' ~ right:factor ; factor = | unary | '(' @:expression ')' ; unary = | uplus | uminus | primary ; uplus = '+' @:atom ; uminus::Neg::UnaryOp = '-' operand:factor ; primary = | attribute | subscript | atom ; attribute::Attribute = operand:primary '.' name:identifier ; subscript::Subscript = operand:primary '[' key:string ']' ; atom = | select | function | constant | column | placeholder ; placeholder::Placeholder = | '%s' name:`` | '%(' name:identifier ')s' ; function::Function = | fname:identifier '(' operands:','.{ expression } ')' | fname:identifier '(' operands+:asterisk ')' ; column::Column = name:identifier ; literal = | date | decimal | integer | string | null | boolean ; constant::Constant = value:(literal | list) ; list = '(' &( literal ',') @:','.{ (literal | ()) }+ ')' ; @name identifier = /[a-zA-Z_][a-zA-Z0-9_]*/ ; asterisk = '*' ; string = /(\"[^\"]*\"|\'[^\']*\')/ ; boolean = 'TRUE' | 'FALSE' ; null = 'NULL' ; integer = /\d+/ ; decimal = /([0-9]+\.[0-9]*|[0-9]*\.[0-9]+)/ ; date = /(\d{4}-\d{2}-\d{2})/ ; balances::Balances = 'BALANCES' ['AT' summary_func:identifier] ['FROM' from_clause:from] ['WHERE' where_clause:expression] ; journal::Journal = 'JOURNAL' [account:string] ['AT' summary_func:identifier] ['FROM' from_clause:from] ; print::Print = 'PRINT' ['FROM' from_clause:from] ; beanquery-master/beanquery/parser/parser.py000066400000000000000000001035451474576771600215370ustar00rootroot00000000000000#!/usr/bin/env python3 # WARNING: CAVEAT UTILITOR # # This file was automatically generated by TatSu. # # https://pypi.python.org/pypi/tatsu/ # # Any changes you make to it will be overwritten the next time # the file is generated. # ruff: noqa: C405, COM812, I001, F401, PLR1702, PLC2801, SIM117 from __future__ import annotations import sys from pathlib import Path from tatsu.buffering import Buffer from tatsu.parsing import Parser from tatsu.parsing import tatsumasu from tatsu.parsing import leftrec, nomemo, isname from tatsu.infos import ParserConfig from tatsu.util import re, generic_main KEYWORDS: set[str] = { 'AND', 'AS', 'ASC', 'BY', 'DESC', 'DISTINCT', 'FALSE', 'FROM', 'GROUP', 'HAVING', 'IN', 'IS', 'LIMIT', 'NOT', 'OR', 'ORDER', 'PIVOT', 'SELECT', 'TRUE', 'WHERE', 'BALANCES', 'JOURNAL', 'PRINT', } class BQLBuffer(Buffer): def __init__(self, text, /, config: ParserConfig | None = None, **settings): config = ParserConfig.new( config, owner=self, whitespace=None, nameguard=None, ignorecase=True, namechars='', parseinfo=True, comments='(\\/\\*([^*]|[\\r\\n]|(\\*+([^*\\/]|[\\r\\n])))*\\*+\\/)', eol_comments='\\;[^\\n]*?$', keywords=KEYWORDS, start='bql', ) config = config.replace(**settings) super().__init__(text, config=config) class BQLParser(Parser): def __init__(self, /, config: ParserConfig | None = None, **settings): config = ParserConfig.new( config, owner=self, whitespace=None, nameguard=None, ignorecase=True, namechars='', parseinfo=True, comments='(\\/\\*([^*]|[\\r\\n]|(\\*+([^*\\/]|[\\r\\n])))*\\*+\\/)', eol_comments='\\;[^\\n]*?$', keywords=KEYWORDS, start='bql', ) config = config.replace(**settings) super().__init__(config=config) @tatsumasu() def _bql_(self): self._statement_() self.name_last_node('@') with self._optional(): self._token(';') self._check_eof() @tatsumasu() def _statement_(self): with self._choice(): with self._option(): self._select_() with self._option(): self._balances_() with self._option(): self._journal_() with self._option(): self._print_() self._error( 'expecting one of: ' "'BALANCES' 'JOURNAL' 'PRINT' 'SELECT'" ' ' ) @tatsumasu('Select') def _select_(self): self._token('SELECT') with self._optional(): self._token('DISTINCT') self._constant(True) self.name_last_node('distinct') self._define(['distinct'], []) with self._group(): with self._choice(): with self._option(): def sep0(): self._token(',') def block1(): self._target_() self._positive_gather(block1, sep0) with self._option(): self._asterisk_() self._error( 'expecting one of: ' ' ' ) self.name_last_node('targets') with self._optional(): self._token('FROM') with self._group(): with self._choice(): with self._option(): self._table_() with self._option(): self._subselect_() with self._option(): self._from_() self._error( 'expecting one of: ' ' ' ) self.name_last_node('from_clause') self._define(['from_clause'], []) with self._optional(): self._token('WHERE') self._expression_() self.name_last_node('where_clause') self._define(['where_clause'], []) with self._optional(): self._token('GROUP') self._token('BY') self._groupby_() self.name_last_node('group_by') self._define(['group_by'], []) with self._optional(): self._token('ORDER') self._token('BY') def sep2(): self._token(',') def block3(): self._order_() self._positive_gather(block3, sep2) self.name_last_node('order_by') self._define(['order_by'], []) with self._optional(): self._token('PIVOT') self._token('BY') self._pivotby_() self.name_last_node('pivot_by') self._define(['pivot_by'], []) with self._optional(): self._token('LIMIT') self._integer_() self.name_last_node('limit') self._define(['limit'], []) self._define(['distinct', 'from_clause', 'group_by', 'limit', 'order_by', 'pivot_by', 'targets', 'where_clause'], []) @tatsumasu() def _subselect_(self): self._token('(') self._select_() self.name_last_node('@') self._token(')') @tatsumasu('From') @nomemo def _from_(self): with self._choice(): with self._option(): self._token('OPEN') self._cut() self._token('ON') self._date_() self.name_last_node('open') with self._optional(): self._token('CLOSE') with self._group(): with self._choice(): with self._option(): self._token('ON') self._date_() self.name_last_node('close') self._define(['close'], []) with self._option(): self._empty_closure() self._constant(True) self.name_last_node('close') self._define(['close'], []) self._error( 'expecting one of: ' "'ON'" ) self._define(['close'], []) with self._optional(): self._token('CLEAR') self._constant(True) self.name_last_node('clear') self._define(['clear'], []) self._define(['clear', 'close', 'open'], []) with self._option(): self._token('CLOSE') self._cut() with self._group(): with self._choice(): with self._option(): self._token('ON') self._date_() self.name_last_node('close') self._define(['close'], []) with self._option(): self._empty_closure() self._constant(True) self.name_last_node('close') self._define(['close'], []) self._error( 'expecting one of: ' "'ON'" ) with self._optional(): self._token('CLEAR') self._constant(True) self.name_last_node('clear') self._define(['clear'], []) self._define(['clear', 'close'], []) with self._option(): self._token('CLEAR') self._cut() self._constant(True) self.name_last_node('clear') self._define(['clear'], []) with self._option(): self._expression_() self.name_last_node('expression') with self._optional(): self._token('OPEN') self._token('ON') self._date_() self.name_last_node('open') self._define(['open'], []) with self._optional(): self._token('CLOSE') with self._group(): with self._choice(): with self._option(): self._token('ON') self._date_() self.name_last_node('close') self._define(['close'], []) with self._option(): self._empty_closure() self._constant(True) self.name_last_node('close') self._define(['close'], []) self._error( 'expecting one of: ' "'ON'" ) self._define(['close'], []) with self._optional(): self._token('CLEAR') self._constant(True) self.name_last_node('clear') self._define(['clear'], []) self._define(['clear', 'close', 'expression', 'open'], []) self._error( 'expecting one of: ' "'CLEAR' 'CLOSE' 'OPEN' " ' ' ) @tatsumasu('Table') def _table_(self): self._pattern('#([a-zA-Z_][a-zA-Z0-9_]*)?') self.name_last_node('name') @tatsumasu('GroupBy') def _groupby_(self): def sep0(): self._token(',') def block1(): with self._group(): with self._choice(): with self._option(): self._integer_() with self._option(): self._expression_() self._error( 'expecting one of: ' ' ' ) self._positive_gather(block1, sep0) self.name_last_node('columns') with self._optional(): self._token('HAVING') self._expression_() self.name_last_node('having') self._define(['having'], []) self._define(['columns', 'having'], []) @tatsumasu('OrderBy') def _order_(self): with self._group(): with self._choice(): with self._option(): self._integer_() with self._option(): self._expression_() self._error( 'expecting one of: ' ' ' ) self.name_last_node('column') self._ordering_() self.name_last_node('ordering') self._define(['column', 'ordering'], []) @tatsumasu() def _ordering_(self): with self._optional(): with self._choice(): with self._option(): self._token('DESC') with self._option(): self._token('ASC') self._error( 'expecting one of: ' "'ASC' 'DESC'" ) @tatsumasu('PivotBy') def _pivotby_(self): with self._group(): with self._choice(): with self._option(): self._integer_() with self._option(): self._column_() self._error( 'expecting one of: ' ' ' ) self.add_last_node_to_name('columns') self._token(',') with self._group(): with self._choice(): with self._option(): self._integer_() with self._option(): self._column_() self._error( 'expecting one of: ' ' ' ) self.add_last_node_to_name('columns') self._define( [], ['columns'], ) @tatsumasu('Target') def _target_(self): self._expression_() self.name_last_node('expression') with self._optional(): self._token('AS') self._identifier_() self.name_last_node('name') self._define(['name'], []) self._define(['expression', 'name'], []) @tatsumasu() @nomemo def _expression_(self): with self._choice(): with self._option(): self._disjunction_() with self._option(): self._conjunction_() self._error( 'expecting one of: ' ' ' ' ' ) @tatsumasu() @nomemo def _disjunction_(self): with self._choice(): with self._option(): self._or_() with self._option(): self._conjunction_() self._error( 'expecting one of: ' ' ' ) @tatsumasu('Or') @nomemo def _or_(self): self._conjunction_() self.add_last_node_to_name('args') def block0(): self._token('OR') self._conjunction_() self.add_last_node_to_name('args') self._define( [], ['args'], ) self._positive_closure(block0) self._define( [], ['args'], ) @tatsumasu() @nomemo def _conjunction_(self): with self._choice(): with self._option(): self._and_() with self._option(): self._inversion_() self._error( 'expecting one of: ' ' ' ) @tatsumasu('And') @nomemo def _and_(self): self._inversion_() self.add_last_node_to_name('args') def block0(): self._token('AND') self._inversion_() self.add_last_node_to_name('args') self._define( [], ['args'], ) self._positive_closure(block0) self._define( [], ['args'], ) @tatsumasu() @nomemo def _inversion_(self): with self._choice(): with self._option(): self._not_() with self._option(): self._comparison_() self._error( 'expecting one of: ' "'NOT' " ' ' ' ' ' ' ) @tatsumasu('Not') def _not_(self): self._token('NOT') self._inversion_() self.name_last_node('operand') self._define(['operand'], []) @tatsumasu() @nomemo def _comparison_(self): with self._choice(): with self._option(): self._any_() with self._option(): self._all_() with self._option(): self._lt_() with self._option(): self._lte_() with self._option(): self._gt_() with self._option(): self._gte_() with self._option(): self._eq_() with self._option(): self._neq_() with self._option(): self._in_() with self._option(): self._notin_() with self._option(): self._match_() with self._option(): self._notmatch_() with self._option(): self._matches_() with self._option(): self._isnull_() with self._option(): self._isnotnull_() with self._option(): self._between_() with self._option(): self._sum_() self._error( 'expecting one of: ' ' ' ' ' ' ' ' ' ) @tatsumasu('Any') @nomemo def _any_(self): self._sum_() self.name_last_node('left') self._op_() self.name_last_node('op') self._token('any') self._token('(') self._expression_() self.name_last_node('right') self._token(')') self._define(['left', 'op', 'right'], []) @tatsumasu('All') def _all_(self): self._sum_() self.name_last_node('left') self._op_() self.name_last_node('op') self._token('all') self._token('(') self._expression_() self.name_last_node('right') self._token(')') self._define(['left', 'op', 'right'], []) @tatsumasu() def _op_(self): with self._choice(): with self._option(): self._token('<') with self._option(): self._token('<=') with self._option(): self._token('>') with self._option(): self._token('>=') with self._option(): self._token('=') with self._option(): self._token('!=') with self._option(): self._token('~') with self._option(): self._token('!~') with self._option(): self._token('?~') self._error( 'expecting one of: ' "'!=' '!~' '<' '<=' '=' '>' '>=' '?~' '~'" ) @tatsumasu('Less') def _lt_(self): self._sum_() self.name_last_node('left') self._token('<') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('LessEq') def _lte_(self): self._sum_() self.name_last_node('left') self._token('<=') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('Greater') def _gt_(self): self._sum_() self.name_last_node('left') self._token('>') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('GreaterEq') def _gte_(self): self._sum_() self.name_last_node('left') self._token('>=') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('Equal') def _eq_(self): self._sum_() self.name_last_node('left') self._token('=') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('NotEqual') def _neq_(self): self._sum_() self.name_last_node('left') self._token('!=') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('In') def _in_(self): self._sum_() self.name_last_node('left') self._token('IN') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('NotIn') def _notin_(self): self._sum_() self.name_last_node('left') self._token('NOT') self._token('IN') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('Match') def _match_(self): self._sum_() self.name_last_node('left') self._token('~') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('NotMatch') def _notmatch_(self): self._sum_() self.name_last_node('left') self._token('!~') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('Matches') def _matches_(self): self._sum_() self.name_last_node('left') self._token('?~') self._sum_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('IsNull') def _isnull_(self): self._sum_() self.name_last_node('operand') self._token('IS') self._token('NULL') self._define(['operand'], []) @tatsumasu('IsNotNull') def _isnotnull_(self): self._sum_() self.name_last_node('operand') self._token('IS') self._token('NOT') self._token('NULL') self._define(['operand'], []) @tatsumasu('Between') def _between_(self): self._sum_() self.name_last_node('operand') self._token('BETWEEN') self._sum_() self.name_last_node('lower') self._token('AND') self._sum_() self.name_last_node('upper') self._define(['lower', 'operand', 'upper'], []) @tatsumasu() @leftrec def _sum_(self): with self._choice(): with self._option(): self._add_() with self._option(): self._sub_() with self._option(): self._term_() self._error( 'expecting one of: ' ' ' ' ' ) @tatsumasu('Add') @nomemo def _add_(self): self._sum_() self.name_last_node('left') self._token('+') self._cut() self._term_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('Sub') @nomemo def _sub_(self): self._sum_() self.name_last_node('left') self._token('-') self._cut() self._term_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu() @leftrec def _term_(self): with self._choice(): with self._option(): self._mul_() with self._option(): self._div_() with self._option(): self._mod_() with self._option(): self._factor_() self._error( 'expecting one of: ' "'(' " '' ) @tatsumasu('Mul') @nomemo def _mul_(self): self._term_() self.name_last_node('left') self._token('*') self._cut() self._factor_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('Div') @nomemo def _div_(self): self._term_() self.name_last_node('left') self._token('/') self._cut() self._factor_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu('Mod') @nomemo def _mod_(self): self._term_() self.name_last_node('left') self._token('%') self._cut() self._factor_() self.name_last_node('right') self._define(['left', 'right'], []) @tatsumasu() @nomemo def _factor_(self): with self._choice(): with self._option(): self._unary_() with self._option(): self._token('(') self._expression_() self.name_last_node('@') self._token(')') self._error( 'expecting one of: ' "'(' " ) @tatsumasu() @nomemo def _unary_(self): with self._choice(): with self._option(): self._uplus_() with self._option(): self._uminus_() with self._option(): self._primary_() self._error( 'expecting one of: ' "'+' '-' " ' ' ) @tatsumasu() def _uplus_(self): self._token('+') self._atom_() self.name_last_node('@') @tatsumasu('Neg') def _uminus_(self): self._token('-') self._factor_() self.name_last_node('operand') self._define(['operand'], []) @tatsumasu() @leftrec def _primary_(self): with self._choice(): with self._option(): self._attribute_() with self._option(): self._subscript_() with self._option(): self._atom_() self._error( 'expecting one of: ' "'SELECT' " ' ' ' ' ) @tatsumasu('Attribute') @nomemo def _attribute_(self): self._primary_() self.name_last_node('operand') self._token('.') self._identifier_() self.name_last_node('name') self._define(['name', 'operand'], []) @tatsumasu('Subscript') @nomemo def _subscript_(self): self._primary_() self.name_last_node('operand') self._token('[') self._string_() self.name_last_node('key') self._token(']') self._define(['key', 'operand'], []) @tatsumasu() def _atom_(self): with self._choice(): with self._option(): self._select_() with self._option(): self._function_() with self._option(): self._constant_() with self._option(): self._column_() with self._option(): self._placeholder_() self._error( 'expecting one of: ' "'%(' '%s' 'SELECT' " ' ' ' ' ' ' ) @tatsumasu('Placeholder') def _placeholder_(self): with self._choice(): with self._option(): self._token('%s') self._constant('') self.name_last_node('name') self._define(['name'], []) with self._option(): self._token('%(') self._identifier_() self.name_last_node('name') self._token(')s') self._define(['name'], []) self._error( 'expecting one of: ' "'%(' '%s'" ) @tatsumasu('Function') def _function_(self): with self._choice(): with self._option(): self._identifier_() self.name_last_node('fname') self._token('(') def sep0(): self._token(',') def block1(): self._expression_() self._gather(block1, sep0) self.name_last_node('operands') self._token(')') self._define(['fname', 'operands'], []) with self._option(): self._identifier_() self.name_last_node('fname') self._token('(') self._asterisk_() self.add_last_node_to_name('operands') self._token(')') self._define( ['fname'], ['operands'], ) self._error( 'expecting one of: ' ' [a-zA-Z_][a-zA-Z0-9_]*' ) @tatsumasu('Column') def _column_(self): self._identifier_() self.name_last_node('name') @tatsumasu() def _literal_(self): with self._choice(): with self._option(): self._date_() with self._option(): self._decimal_() with self._option(): self._integer_() with self._option(): self._string_() with self._option(): self._null_() with self._option(): self._boolean_() self._error( 'expecting one of: ' "'FALSE' 'NULL' 'TRUE'" '([0-9]+\\.[0-9]*|[0-9]*\\.[0-9]+)' '(\\"[^\\"]*\\"|\\\'[^\\\']*\\\')' '(\\d{4}-\\d{2}-\\d{2}) ' ' \\d+' ) @tatsumasu('Constant') def _constant_(self): with self._group(): with self._choice(): with self._option(): self._literal_() with self._option(): self._list_() self._error( 'expecting one of: ' "'(' 'FALSE' 'NULL' 'TRUE'" '([0-9]+\\.[0-9]*|[0-9]*\\.[0-9]+)' '(\\"[^\\"]*\\"|\\\'[^\\\']*\\\')' '(\\d{4}-\\d{2}-\\d{2}) ' ' ' ' \\d+' ) self.name_last_node('value') @tatsumasu() def _list_(self): self._token('(') with self._if(): with self._group(): self._literal_() self._token(',') def sep0(): self._token(',') def block1(): with self._group(): with self._choice(): with self._option(): self._literal_() with self._option(): self._void() self._error( 'expecting one of: ' '' ) self._positive_gather(block1, sep0) self.name_last_node('@') self._token(')') @tatsumasu() @isname def _identifier_(self): self._pattern('[a-zA-Z_][a-zA-Z0-9_]*') @tatsumasu() def _asterisk_(self): self._token('*') @tatsumasu() def _string_(self): self._pattern('(\\"[^\\"]*\\"|\\\'[^\\\']*\\\')') @tatsumasu() def _boolean_(self): with self._choice(): with self._option(): self._token('TRUE') with self._option(): self._token('FALSE') self._error( 'expecting one of: ' "'FALSE' 'TRUE'" ) @tatsumasu() def _null_(self): self._token('NULL') @tatsumasu() def _integer_(self): self._pattern('\\d+') @tatsumasu() def _decimal_(self): self._pattern('([0-9]+\\.[0-9]*|[0-9]*\\.[0-9]+)') @tatsumasu() def _date_(self): self._pattern('(\\d{4}-\\d{2}-\\d{2})') @tatsumasu('Balances') def _balances_(self): self._token('BALANCES') with self._optional(): self._token('AT') self._identifier_() self.name_last_node('summary_func') self._define(['summary_func'], []) with self._optional(): self._token('FROM') self._from_() self.name_last_node('from_clause') self._define(['from_clause'], []) with self._optional(): self._token('WHERE') self._expression_() self.name_last_node('where_clause') self._define(['where_clause'], []) self._define(['from_clause', 'summary_func', 'where_clause'], []) @tatsumasu('Journal') def _journal_(self): self._token('JOURNAL') with self._optional(): self._string_() self.name_last_node('account') with self._optional(): self._token('AT') self._identifier_() self.name_last_node('summary_func') self._define(['summary_func'], []) with self._optional(): self._token('FROM') self._from_() self.name_last_node('from_clause') self._define(['from_clause'], []) self._define(['account', 'from_clause', 'summary_func'], []) @tatsumasu('Print') def _print_(self): self._token('PRINT') with self._optional(): self._token('FROM') self._from_() self.name_last_node('from_clause') self._define(['from_clause'], []) self._define(['from_clause'], []) def main(filename, **kwargs): if not filename or filename == '-': text = sys.stdin.read() else: text = Path(filename).read_text() parser = BQLParser() return parser.parse( text, filename=filename, **kwargs, ) if __name__ == '__main__': import json from tatsu.util import asjson ast = generic_main(main, BQLParser, name='BQL') data = asjson(ast) print(json.dumps(data, indent=2)) beanquery-master/beanquery/parser_test.py000066400000000000000000000561731474576771600213060ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import datetime import textwrap import unittest from decimal import Decimal as D from beanquery import parser from beanquery.parser import ast def Select(targets, from_clause=None, where_clause=None, **kwargs): defaults = dict(targets=targets, from_clause=from_clause, where_clause=where_clause, group_by=None, order_by=None, pivot_by=None, limit=None, distinct=None) defaults.update(kwargs) return ast.Select(**defaults) class QueryParserTestBase(unittest.TestCase): def parse(self, query): return parser.parse(query.strip()) def assertParse(self, query, expected): self.assertEqual(parser.parse(query), expected) def assertParseTarget(self, query, expected): expr = parser.parse(query) self.assertIsInstance(expr, ast.Select) self.assertEqual(len(expr.targets), 1) self.assertEqual(expr.targets[0].expression, expected) def assertParseFrom(self, query, expected): expr = parser.parse(query) self.assertIsInstance(expr, ast.Select) self.assertEqual(expr.from_clause, expected) class TestParseSelect(QueryParserTestBase): def test_select(self): with self.assertRaises(parser.ParseError): parser.parse("SELECT") with self.assertRaises(parser.ParseError): parser.parse("SELECT ; ") self.assertParse( "SELECT *;", Select(ast.Asterisk())) self.assertParse( "SELECT date;", Select([ ast.Target(ast.Column('date'), None) ])) self.assertParse( "SELECT date, account", Select([ ast.Target(ast.Column('date'), None), ast.Target(ast.Column('account'), None) ])) self.assertParse( "SELECT date as xdate;", Select([ ast.Target(ast.Column('date'), 'xdate') ])) self.assertParse( "SELECT date as x, account, position as y;", Select([ ast.Target(ast.Column('date'), 'x'), ast.Target(ast.Column('account'), None), ast.Target(ast.Column('position'), 'y') ])) def test_literals(self): # null self.assertParseTarget("SELECT NULL;", ast.Constant(None)) # bool self.assertParseTarget("SELECT TRUE;", ast.Constant(True)) self.assertParseTarget("SELECT FALSE;", ast.Constant(False)) # int self.assertParseTarget("SELECT 17;", ast.Constant(17)) # decimal self.assertParseTarget("SELECT 17.345;", ast.Constant(D('17.345'))) self.assertParseTarget("SELECT .345;", ast.Constant(D('.345'))) self.assertParseTarget("SELECT 17.;", ast.Constant(D('17.'))) # string self.assertParseTarget("SELECT 'rainy-day';", ast.Constant('rainy-day')) # date self.assertParseTarget("SELECT 1972-05-28;", ast.Constant(datetime.date(1972, 5, 28))) # not a list self.assertParseTarget("SELECT (1);", ast.Constant(1)) # list self.assertParseTarget("SELECT (1, );", ast.Constant([1])) self.assertParseTarget("SELECT (1, 2);", ast.Constant([1, 2])) self.assertParseTarget("SELECT (1, 2, );", ast.Constant([1, 2])) self.assertParseTarget("SELECT ('x', 'y', 'z');", ast.Constant(['x', 'y', 'z'])) # column self.assertParseTarget("SELECT date;", ast.Column('date')) def test_expressions(self): # comparison operators self.assertParseTarget("SELECT a = 42;", ast.Equal(ast.Column('a'), ast.Constant(42))) self.assertParseTarget("SELECT a != 42;", ast.NotEqual(ast.Column('a'), ast.Constant(42))) self.assertParseTarget("SELECT a > 42;", ast.Greater(ast.Column('a'), ast.Constant(42))) self.assertParseTarget("SELECT a >= 42;", ast.GreaterEq(ast.Column('a'), ast.Constant(42))) self.assertParseTarget("SELECT a < 42;", ast.Less(ast.Column('a'), ast.Constant(42))) self.assertParseTarget("SELECT a <= 42;", ast.LessEq(ast.Column('a'), ast.Constant(42))) self.assertParseTarget("SELECT a ~ 'abc';", ast.Match(ast.Column('a'), ast.Constant('abc'))) self.assertParseTarget("SELECT not a;", ast.Not(ast.Column('a'))) self.assertParseTarget("SELECT a IS NULL;", ast.IsNull(ast.Column('a'))) self.assertParseTarget("SELECT a IS NOT NULL;", ast.IsNotNull(ast.Column('a'))) # bool expressions self.assertParseTarget("SELECT a AND b;", ast.And([ast.Column('a'), ast.Column('b')])) self.assertParseTarget("SELECT a AND b AND c;", ast.And([ast.Column('a'), ast.Column('b'), ast.Column('c')])) self.assertParseTarget("SELECT a OR b;", ast.Or([ast.Column('a'), ast.Column('b')])) self.assertParseTarget("SELECT a OR b OR c;", ast.Or([ast.Column('a'), ast.Column('b'), ast.Column('c')])) self.assertParseTarget("SELECT a AND b OR c;", ast.Or([ast.And([ast.Column('a'), ast.Column('b')]), ast.Column('c')])) self.assertParseTarget("SELECT NOT a;", ast.Not(ast.Column('a'))) # math expressions with identifiers self.assertParseTarget("SELECT a * b;", ast.Mul(ast.Column('a'), ast.Column('b'))) self.assertParseTarget("SELECT a / b;", ast.Div(ast.Column('a'), ast.Column('b'))) self.assertParseTarget("SELECT a + b;", ast.Add(ast.Column('a'), ast.Column('b'))) self.assertParseTarget("SELECT a+b;", ast.Add(ast.Column('a'), ast.Column('b'))) self.assertParseTarget("SELECT a - b;", ast.Sub(ast.Column('a'), ast.Column('b'))) self.assertParseTarget("SELECT a-b;", ast.Sub(ast.Column('a'), ast.Column('b'))) self.assertParseTarget("SELECT +a;", ast.Column('a')) self.assertParseTarget("SELECT -a;", ast.Neg(ast.Column('a'))) # math expressions with numerals self.assertParseTarget("SELECT 2 * 3;", ast.Mul(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT 2 / 3;", ast.Div(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT 2+(3);", ast.Add(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT (2)-3;", ast.Sub(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT 2 + 3;", ast.Add(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT 2+3;", ast.Add(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT 2 - 3;", ast.Sub(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT 2-3;", ast.Sub(ast.Constant(2), ast.Constant(3))) self.assertParseTarget("SELECT +2;", ast.Constant(2)) self.assertParseTarget("SELECT -2;", ast.Neg(ast.Constant(2))) # silly, fails at compile time self.assertParseTarget("SELECT -'abc';", ast.Neg(ast.Constant('abc'))) # functions self.assertParseTarget("SELECT random();", ast.Function('random', [])) self.assertParseTarget("SELECT min(a);", ast.Function('min', [ast.Column('a')])) self.assertParseTarget("SELECT min(a, b);", ast.Function('min', [ast.Column('a'), ast.Column('b')])) self.assertParseTarget("SELECT count(*);", ast.Function('count', [ast.Asterisk()])) def test_non_associative(self): # non associative operators self.assertRaises(parser.ParseError, self.parse, "SELECT 3 > 2 > 1") self.assertRaises(parser.ParseError, self.parse, "SELECT 3 = 2 = 1") def test_complex_expressions(self): self.assertParseTarget( "SELECT NOT a = (b != (42 AND 17));", ast.Not( ast.Equal( ast.Column('a'), ast.NotEqual( ast.Column('b'), ast.And([ ast.Constant(42), ast.Constant(17)]))))) class TestSelectPrecedence(QueryParserTestBase): def test_operators_precedence(self): self.assertParseTarget( "SELECT a AND b OR c AND d;", ast.Or([ast.And([ast.Column('a'), ast.Column('b')]), ast.And([ast.Column('c'), ast.Column('d')])])) self.assertParseTarget( "SELECT a = 2 AND b != 3;", ast.And([ast.Equal(ast.Column('a'), ast.Constant(2)), ast.NotEqual(ast.Column('b'), ast.Constant(3))])) self.assertParseTarget( "SELECT not a AND b;", ast.And([ast.Not(ast.Column('a')), ast.Column('b')])) self.assertParseTarget( "SELECT a + b AND c - d;", ast.And([ast.Add(ast.Column('a'), ast.Column('b')), ast.Sub(ast.Column('c'), ast.Column('d'))])) self.assertParseTarget( "SELECT a * b + c / d - 3;", ast.Sub( ast.Add( ast.Mul( ast.Column(name='a'), ast.Column(name='b')), ast.Div(ast.Column(name='c'), ast.Column(name='d'))), ast.Constant(value=3))) self.assertParseTarget( "SELECT 'orange' IN tags AND 'bananas' IN tags;", ast.And([ ast.In(ast.Constant('orange'), ast.Column('tags')), ast.In(ast.Constant('bananas'), ast.Column('tags'))])) class TestSelectFrom(QueryParserTestBase): def test_select_from(self): expr = ast.Equal(ast.Column('d'), ast.And([ast.Function('max', [ast.Column('e')]), ast.Constant(17)])) with self.assertRaises(parser.ParseError): parser.parse("SELECT a, b FROM;") # simple self.assertParseFrom( "SELECT a, b FROM d = (max(e) and 17);", ast.From(expr, None, None, None)) # open dated self.assertParseFrom( "SELECT a, b FROM d = (max(e) and 17) OPEN ON 2014-01-01;", ast.From(expr, datetime.date(2014, 1, 1), None, None)) # close default self.assertParseFrom( "SELECT a, b FROM d = (max(e) and 17) CLOSE;", ast.From(expr, None, True, None)) # close dated self.assertParseFrom( "SELECT a, b FROM d = (max(e) and 17) CLOSE ON 2014-10-18;", ast.From(expr, None, datetime.date(2014, 10, 18), None)) # close no expression self.assertParseFrom( "SELECT a, b FROM CLOSE;", ast.From(None, None, True, None)) # close no expression dated self.assertParseFrom( "SELECT a, b FROM CLOSE ON 2014-10-18;", ast.From(None, None, datetime.date(2014, 10, 18), None)) # clear default self.assertParseFrom( "SELECT a, b FROM d = (max(e) and 17) CLEAR;", ast.From(expr, None, None, True)) # open close clear self.assertParseFrom( "SELECT a, b FROM d = (max(e) and 17) OPEN ON 2013-10-25 CLOSE ON 2014-10-25 CLEAR;", ast.From(expr, datetime.date(2013, 10, 25), datetime.date(2014, 10, 25), True)) class TestSelectWhere(QueryParserTestBase): def test_where(self): expr = ast.Equal(ast.Column('d'), ast.And([ast.Function('max', [ast.Column('e')]), ast.Constant(17)])) self.assertParse( "SELECT a, b WHERE d = (max(e) and 17);", Select([ ast.Target(ast.Column('a'), None), ast.Target(ast.Column('b'), None) ], None, expr)) with self.assertRaises(parser.ParseError): parser.parse("SELECT a, b WHERE;") class TestSelectFromAndWhere(QueryParserTestBase): def test_from_and_where(self): expr = ast.Equal(ast.Column('d'), ast.And([ast.Function('max', [ast.Column('e')]), ast.Constant(17)])) self.assertParse( "SELECT a, b FROM d = (max(e) and 17) WHERE d = (max(e) and 17);", Select([ ast.Target(ast.Column('a'), None), ast.Target(ast.Column('b'), None) ], ast.From(expr, None, None, None), expr)) class TestSelectFromSelect(QueryParserTestBase): def test_from_select(self): self.assertParse(""" SELECT a, b FROM ( SELECT * FROM date = 2014-05-02 ) WHERE c = 5 LIMIT 100;""", Select([ ast.Target(ast.Column('a'), None), ast.Target(ast.Column('b'), None)], Select( ast.Asterisk(), ast.From( ast.Equal( ast.Column('date'), ast.Constant(datetime.date(2014, 5, 2))), None, None, None)), ast.Equal(ast.Column('c'), ast.Constant(5)), limit=100)) class TestSelectGroupBy(QueryParserTestBase): def test_groupby_one(self): self.assertParse( "SELECT * GROUP BY a;", Select(ast.Asterisk(), group_by=ast.GroupBy([ast.Column('a')], None))) def test_groupby_many(self): self.assertParse( "SELECT * GROUP BY a, b, c;", Select(ast.Asterisk(), group_by=ast.GroupBy([ ast.Column('a'), ast.Column('b'), ast.Column('c')], None))) def test_groupby_expr(self): self.assertParse( "SELECT * GROUP BY length(a) > 0, b;", Select(ast.Asterisk(), group_by=ast.GroupBy([ ast.Greater( ast.Function('length', [ ast.Column('a')]), ast.Constant(0)), ast.Column('b')], None))) def test_groupby_having(self): self.assertParse( "SELECT * GROUP BY a HAVING sum(x) = 0;", Select(ast.Asterisk(), group_by=ast.GroupBy([ast.Column('a')], ast.Equal( ast.Function('sum', [ ast.Column('x')]), ast.Constant(0))))) def test_groupby_numbers(self): self.assertParse( "SELECT * GROUP BY 1;", Select(ast.Asterisk(), group_by=ast.GroupBy([1], None))) self.assertParse( "SELECT * GROUP BY 2, 4, 5;", Select(ast.Asterisk(), group_by=ast.GroupBy([2, 4, 5], None))) def test_groupby_empty(self): with self.assertRaises(parser.ParseError): parser.parse("SELECT * GROUP BY;") class TestSelectOrderBy(QueryParserTestBase): def test_orderby_one(self): self.assertParse( "SELECT * ORDER BY a;", Select(ast.Asterisk(), order_by=[ ast.OrderBy(ast.Column('a'), ast.Ordering.ASC)])) def test_orderby_many(self): self.assertParse( "SELECT * ORDER BY a, b, c;", Select(ast.Asterisk(), order_by=[ ast.OrderBy(ast.Column('a'), ast.Ordering.ASC), ast.OrderBy(ast.Column('b'), ast.Ordering.ASC), ast.OrderBy(ast.Column('c'), ast.Ordering.ASC)])) def test_orderby_asc(self): self.assertParse( "SELECT * ORDER BY a ASC;", Select(ast.Asterisk(), order_by=[ ast.OrderBy(ast.Column('a'), ast.Ordering.ASC)])) def test_orderby_desc(self): self.assertParse( "SELECT * ORDER BY a DESC;", Select(ast.Asterisk(), order_by=[ ast.OrderBy(ast.Column('a'), ast.Ordering.DESC)])) def test_orderby_many_asc_desc(self): self.assertParse( "SELECT * ORDER BY a ASC, b DESC, c;", Select(ast.Asterisk(), order_by=[ ast.OrderBy(ast.Column('a'), ast.Ordering.ASC), ast.OrderBy(ast.Column('b'), ast.Ordering.DESC), ast.OrderBy(ast.Column('c'), ast.Ordering.ASC)])) def test_orderby_empty(self): with self.assertRaises(parser.ParseError): parser.parse("SELECT * ORDER BY;") class TestSelectPivotBy(QueryParserTestBase): def test_pivotby(self): with self.assertRaises(parser.ParseError): parser.parse("SELECT * PIVOT BY;") with self.assertRaises(parser.ParseError): parser.parse("SELECT * PIVOT BY a;") with self.assertRaises(parser.ParseError): parser.parse("SELECT * PIVOT BY a, b, c") self.assertParse( "SELECT * PIVOT BY a, b", Select(ast.Asterisk(), pivot_by=ast.PivotBy([ast.Column('a'), ast.Column('b')]))) self.assertParse( "SELECT * PIVOT BY 1, 2", Select(ast.Asterisk(), pivot_by=ast.PivotBy([1, 2]))) class TestSelectOptions(QueryParserTestBase): def test_distinct(self): self.assertParse( "SELECT DISTINCT x;", Select([ast.Target(ast.Column('x'), None)], distinct=True)) def test_limit_present(self): self.assertParse( "SELECT * LIMIT 45;", Select(ast.Asterisk(), limit=45)) def test_limit_empty(self): with self.assertRaises(parser.ParseError): parser.parse("SELECT * LIMIT;") class TestBalances(QueryParserTestBase): def test_balances_empty(self): self.assertParse( "BALANCES;", ast.Balances(None, None, None)) def test_balances_from(self): self.assertParse( "BALANCES FROM date = 2014-01-01 CLOSE;", ast.Balances( None, ast.From( ast.Equal( ast.Column('date'), ast.Constant(datetime.date(2014, 1, 1))), None, True, None), None)) def test_balances_from_with_transformer(self): self.assertParse( "BALANCES AT units FROM date = 2014-01-01 CLOSE;", ast.Balances('units', ast.From( ast.Equal( ast.Column('date'), ast.Constant(datetime.date(2014, 1, 1))), None, True, None), None)) def test_balances_from_with_transformer_simple(self): self.assertParse( "BALANCES AT units WHERE date = 2014-01-01;", ast.Balances('units', None, ast.Equal( ast.Column('date'), ast.Constant(datetime.date(2014, 1, 1))))) class TestJournal(QueryParserTestBase): def test_journal_empty(self): self.assertParse( "JOURNAL;", ast.Journal(None, None, None)) def test_journal_account(self): self.assertParse( "JOURNAL 'Assets:Checking';", ast.Journal('Assets:Checking', None, None)) def test_journal_summary(self): self.assertParse( "JOURNAL AT cost;", ast.Journal(None, 'cost', None)) def test_journal_account_and_summary(self): self.assertParse( "JOURNAL 'Assets:Foo' AT cost;", ast.Journal('Assets:Foo', 'cost', None)) def test_journal_from(self): self.assertParse( "JOURNAL FROM date = 2014-01-01 CLOSE;", ast.Journal(None, None, ast.From( ast.Equal( ast.Column('date'), ast.Constant(datetime.date(2014, 1, 1)) ), None, True, None))) class TestPrint(QueryParserTestBase): def test_print_empty(self): self.assertParse( "PRINT;", ast.Print(None)) def test_print_from(self): self.assertParse( "PRINT FROM date = 2014-01-01 CLOSE;", ast.Print( ast.From( ast.Equal( ast.Column('date'), ast.Constant(datetime.date(2014, 1, 1)) ), None, True, None))) class TestComments(QueryParserTestBase): def test_comments(self): self.assertParse( """SELECT first, /* comment */ second""", Select([ ast.Target(ast.Column('first'), None), ast.Target(ast.Column('second'), None) ])) self.assertParse( """SELECT first, /* comment */ second;""", Select([ ast.Target(ast.Column('first'), None), ast.Target(ast.Column('second'), None), ])) self.assertParse( """SELECT first, /**/ second;""", Select([ ast.Target(ast.Column('first'), None), ast.Target(ast.Column('second'), None), ])) self.assertParse( """SELECT first, /* /* */ second;""", Select([ ast.Target(ast.Column('first'), None), ast.Target(ast.Column('second'), None), ])) self.assertParse( """SELECT first, /* ; */ second;""", Select([ ast.Target(ast.Column('first'), None), ast.Target(ast.Column('second'), None), ])) class TestRepr(unittest.TestCase): def test_ordering(self): self.assertEqual(repr(ast.Ordering.ASC), 'Ordering.ASC') self.assertEqual(repr(ast.Ordering.DESC), 'Ordering.DESC') def test_ast_node(self): self.assertEqual(repr(ast.Constant(1)), 'Constant(value=1)') self.assertEqual(repr(ast.Not(ast.Constant(False))), 'Not(operand=Constant(value=False))') def test_tosexp(self): sexp = parser.parse('SELECT a + 1 FROM #test WHERE a > 42 ORDER BY b DESC').tosexp() self.assertEqual(sexp, textwrap.dedent('''\ (select targets: ( (target expression: (add left: (column name: 'a') right: (constant value: 1)))) from-clause: (table name: 'test') where-clause: (greater left: (column name: 'a') right: (constant value: 42)) order-by: ( (orderby column: (column name: 'b') ordering: desc)))''')) def test_walk(self): query = parser.parse('SELECT a + 1 FROM #test WHERE a > %(foo)s ORDER BY b DESC') placeholders = [node for node in query.walk() if isinstance(node, ast.Placeholder)] self.assertEqual(placeholders, [ast.Placeholder(name='foo')]) class TestNodeText(unittest.TestCase): def test_text(self): select = parser.parse('SELECT date + 1') self.assertEqual(select.text, 'SELECT date + 1') self.assertEqual(select.targets[0].expression.text, 'date + 1') self.assertEqual(select.targets[0].expression.left.text, 'date') self.assertEqual(select.targets[0].expression.right.text, '1') def test_synthetic(self): node = ast.And([ast.Constant(False), ast.Constant(True)]) self.assertIsNone(node.text) beanquery-master/beanquery/query.py000066400000000000000000000026141474576771600201070ustar00rootroot00000000000000"""A library to run queries. This glues together all the parts of the query engine. """ __copyright__ = "Copyright (C) 2015-2017 Martin Blais" __license__ = "GNU GPLv2" import beanquery import beanquery.numberify def run_query(entries, options, query, *args, numberify=False): """Compile and execute a query, return the result types and rows. Args: entries: A list of entries, as produced by the loader. options: A dict of options, as produced by the loader. query: A string, a single BQL query, optionally containing some new-style (e.g., {}) formatting specifications. args: A tuple of arguments to be formatted in the query. This is just provided as a convenience. numberify: If true, numberify the results before returning them. Returns: A pair of result types and result rows. Raises: ParseError: If the statement cannot be parsed. CompilationError: If the statement cannot be compiled. """ # Execute the query. ctx = beanquery.connect('beancount:', entries=entries, errors=[], options=options) curs = ctx.execute(query.format(*args)) rrows = curs.fetchall() rtypes = curs.description # Numberify the results, if requested. if numberify: dformat = options['dcontext'].build() rtypes, rrows = beanquery.numberify.numberify_results(rtypes, rrows, dformat) return rtypes, rrows beanquery-master/beanquery/query_compile.py000066400000000000000000000426671474576771600216330ustar00rootroot00000000000000"""Interpreter for the query language's AST. This code accepts the abstract syntax tree produced by the query parser, resolves the column and function names, compiles and interpreter and prepares a query to be run against a list of entries. """ from __future__ import annotations __copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import collections import dataclasses import datetime import re import operator from decimal import Decimal from itertools import product from typing import List from dateutil.relativedelta import relativedelta from beanquery.parser import ast from beanquery import query_execute from beanquery import types from beanquery import tables MARKER = object() FUNCTIONS = collections.defaultdict(list) OPERATORS = collections.defaultdict(list) class EvalNode: __slots__ = ('dtype',) def __init__(self, dtype): # The output data type produce by this node. This is intended to be # inferred by the nodes on construction. assert dtype is not None, "Internal erro: Invalid dtype, must be deduced." self.dtype = dtype def __eq__(self, other): if not isinstance(other, type(self)): return NotImplemented return all(getattr(self, attr) == getattr(other, attr) for attr in self.__slots__) def __str__(self): return "{}({})".format(type(self).__name__, ', '.join(repr(getattr(self, child)) for child in self.__slots__)) __repr__ = __str__ def childnodes(self): """Returns the child nodes of this node. Yields: A list of EvalNode instances. """ for attr in self.__slots__: child = getattr(self, attr) if isinstance(child, EvalNode): yield child elif isinstance(child, list): for element in child: if isinstance(element, EvalNode): yield element def __call__(self, context): """Evaluate this node. This is designed to recurse on its children. All subclasses must override and implement this method. Args: context: The evaluation object to which the evaluation need to apply. This is either an entry, a Posting instance, or a particular result set row from a sub-select. This is the provider for the underlying data. Returns: The evaluated value for this sub-expression tree. """ raise NotImplementedError class EvalConstant(EvalNode): __slots__ = ('value',) def __init__(self, value, dtype=None): super().__init__(type(value) if dtype is None else dtype) self.value = value def __call__(self, _): return self.value class EvalUnaryOp(EvalNode): __slots__ = ('operand', 'operator') def __init__(self, operator, operand, dtype): super().__init__(dtype) self.operand = operand self.operator = operator def __call__(self, context): operand = self.operand(context) return self.operator(operand) def __repr__(self): return f'{self.__class__.__name__}({self.operator!r})' class EvalUnaryOpSafe(EvalUnaryOp): def __call__(self, context): operand = self.operand(context) if operand is None: return None return self.operator(operand) class EvalBinaryOp(EvalNode): __slots__ = ('left', 'right', 'operator') def __init__(self, operator, left, right, dtype): super().__init__(dtype) self.operator = operator self.left = left self.right = right def __call__(self, context): left = self.left(context) if left is None: return None right = self.right(context) if right is None: return None return self.operator(left, right) def __repr__(self): return f'{self.__class__.__name__}({self.left!r}, {self.right!r})' class EvalBetween(EvalNode): __slots__ = ('operand', 'lower', 'upper') def __init__(self, operand, lower, upper): super().__init__(bool) self.operand = operand self.lower = lower self.upper = upper def __call__(self, context): operand = self.operand(context) if operand is None: return None lower = self.lower(context) if lower is None: return None upper = self.upper(context) if upper is None: return None return lower <= operand <= upper def unaryop(op, intypes, outtype, nullsafe=False): def decorator(func): class Op(EvalUnaryOp if nullsafe else EvalUnaryOpSafe): __intypes__ = intypes def __init__(self, operand): super().__init__(func, operand, outtype) Op.__name__ = f'{op.__name__}[{intypes[0].__name__}]' OPERATORS[op].append(Op) return func return decorator def binaryop(op, intypes, outtype): def decorator(func): class Op(EvalBinaryOp): __intypes__ = intypes def __init__(self, left, right): super().__init__(func, left, right, outtype) Op.__name__ = f'{op.__name__}[{intypes[0].__name__},{intypes[1].__name__}]' OPERATORS[op].append(Op) return func return decorator def Operator(op, operands): op = types.function_lookup(OPERATORS, op, operands) if op is not None: return op(*operands) raise KeyError unaryop(ast.Not, [types.Any], bool, nullsafe=True)(operator.not_) @unaryop(ast.Neg, [int], int) @unaryop(ast.Neg, [Decimal], Decimal) def neg_(x): return -x @unaryop(ast.IsNull, [types.Any], bool, nullsafe=True) def null(x): return x is None @unaryop(ast.IsNotNull, [types.Any], bool, nullsafe=True) def not_null(x): return x is not None @binaryop(ast.Mul, [Decimal, Decimal], Decimal) @binaryop(ast.Mul, [Decimal, int], Decimal) @binaryop(ast.Mul, [int, Decimal], Decimal) @binaryop(ast.Mul, [int, int], int) def mul_(x, y): return x * y @binaryop(ast.Div, [Decimal, Decimal], Decimal) @binaryop(ast.Div, [Decimal, int], Decimal) @binaryop(ast.Div, [int, Decimal], Decimal) def div_(x, y): if y == 0: return None return x / y @binaryop(ast.Div, [int, int], Decimal) def div_int(x, y): if y == 0: return None return Decimal(x) / y @binaryop(ast.Mod, [int, int], int) @binaryop(ast.Mod, [Decimal, int], Decimal) @binaryop(ast.Mod, [int, Decimal], Decimal) @binaryop(ast.Mod, [Decimal, Decimal], Decimal) def mod_(x, y): if y == 0: return None return x % y @binaryop(ast.Add, [Decimal, Decimal], Decimal) @binaryop(ast.Add, [Decimal, int], Decimal) @binaryop(ast.Add, [int, Decimal], Decimal) @binaryop(ast.Add, [int, int], int) @binaryop(ast.Add, [str, str], str) @binaryop(ast.Add, [datetime.date, relativedelta], datetime.date) @binaryop(ast.Add, [relativedelta, datetime.date], datetime.date) @binaryop(ast.Add, [relativedelta, relativedelta], relativedelta) def add_(x, y): return x + y @binaryop(ast.Sub, [Decimal, Decimal], Decimal) @binaryop(ast.Sub, [Decimal, int], Decimal) @binaryop(ast.Sub, [int, Decimal], Decimal) @binaryop(ast.Sub, [int, int], int) @binaryop(ast.Sub, [datetime.date, relativedelta], datetime.date) @binaryop(ast.Sub, [relativedelta, datetime.date], datetime.date) @binaryop(ast.Sub, [relativedelta, relativedelta], datetime.date) def sub_(x, y): return x - y @binaryop(ast.Add, [datetime.date, int], datetime.date) def add_date_int(x, y): return x + datetime.timedelta(days=y) @binaryop(ast.Add, [int, datetime.date], datetime.date) def add_int_date(x, y): return y + datetime.timedelta(days=x) @binaryop(ast.Sub, [datetime.date, int], datetime.date) def sub_date_int(x, y): return x - datetime.timedelta(days=y) @binaryop(ast.Sub, [datetime.date, datetime.date], int) def sub_date_date(x, y): return (x - y).days @binaryop(ast.Match, [str, str], bool) def match_(x, y): return bool(re.search(y, x, re.IGNORECASE)) @binaryop(ast.NotMatch, [str, str], bool) def not_match_(x, y): return not bool(re.search(y, x, re.IGNORECASE)) @binaryop(ast.Matches, [str, str], bool) def matches_(x, y): return bool(re.search(x, y)) @binaryop(ast.In, [types.Any, set], bool) @binaryop(ast.In, [types.Any, list], bool) @binaryop(ast.In, [types.Any, dict], bool) def in_(x, y): return operator.contains(y, x) @binaryop(ast.NotIn, [types.Any, set], bool) @binaryop(ast.NotIn, [types.Any, list], bool) @binaryop(ast.NotIn, [types.Any, dict], bool) def not_in_(x, y): return not operator.contains(y, x) _comparisons = [ (ast.Equal, operator.eq), (ast.NotEqual, operator.ne), (ast.Greater, operator.gt), (ast.GreaterEq, operator.ge), (ast.Less, operator.lt), (ast.LessEq, operator.le), ] _intypes = [ [int, int], [Decimal, int], [int, Decimal], [Decimal, Decimal], [datetime.date, datetime.date], [str, str], ] for node, op in _comparisons: for intypes in _intypes: binaryop(node, intypes, bool)(op) _comparable = [ # lists of types that can be compared with each other [int, Decimal], [datetime.date], [str], ] for comparable in _comparable: for intypes in product(comparable, repeat=3): class Between(EvalBetween): __intypes__ = list(intypes) OPERATORS[ast.Between].append(Between) class EvalAnd(EvalNode): __slots__ = ('args',) def __init__(self, args): super().__init__(bool) self.args = args def __call__(self, context): for arg in self.args: value = arg(context) if value is None: return None if not value: return False return True class EvalOr(EvalNode): __slots__ = ('args',) def __init__(self, args): super().__init__(bool) self.args = args def __call__(self, context): r = False for arg in self.args: value = arg(context) if value is None: r = None if value: return True return r class EvalCoalesce(EvalNode): __slots__ = ('args',) def __init__(self, args): super().__init__(args[0].dtype) self.args = args def __call__(self, context): for arg in self.args: value = arg(context) if value is not None: return value return None class EvalFunction(EvalNode): __slots__ = ('operands',) # Type constraints on the input arguments. __intypes__ = [] def __init__(self, context, operands, dtype): super().__init__(dtype) self.context = context self.operands = operands class EvalGetItem(EvalNode): __slots__ = ('operand', 'key') def __init__(self, operand, key): super().__init__(object) self.operand = operand self.key = key def __call__(self, context): operand = self.operand(context) if operand is None: return None return operand.get(self.key) class EvalGetter(EvalNode): __slots__ = ('operand', 'getter') def __init__(self, operand, getter, dtype): super().__init__(dtype) self.operand = operand self.getter = getter def __call__(self, context): operand = self.operand(context) if operand is None: return None return self.getter(operand) class EvalAny(EvalNode): __slots__ = ('op', 'left', 'right') def __init__(self, op, left, right): super().__init__(bool) self.op = op self.left = left self.right = right def __call__(self, row): left = self.left(row) if left is None: return None right = self.right(row) if right is None: return None return any(self.op(left, x) for x in right) class EvalAll(EvalNode): __slots__ = ('op', 'left', 'right') def __init__(self, op, left, right): super().__init__(bool) self.op = op self.left = left self.right = right def __call__(self, row): left = self.left(row) if left is None: return None right = self.right(row) if right is None: return None return all(self.op(left, x) for x in right) class EvalRow(EvalNode): __slots__ = () def __init__(self): super().__init__(object) def __call__(self, context): return context class EvalColumn(EvalNode): pass class EvalAggregator(EvalFunction): pure = False def __init__(self, context, operands, dtype=None): super().__init__(context, operands, dtype or operands[0].dtype) self.value = None def allocate(self, allocator): """Allocate handles to store data for a node's aggregate storage. This is called once before beginning aggregations. If you need any kind of per-aggregate storage during the computation phase, get it in this method. Args: allocator: An instance of Allocator, on which you can call allocate() to obtain a handle for a slot to store data on store objects later on. """ self.handle = allocator.allocate() def initialize(self, store): """Initialize this node's aggregate data. Args: store: An object indexable by handles appropriated during allocate(). """ store[self.handle] = self.dtype() self.value = None def update(self, store, context): """Evaluate this node. This is designed to recurse on its children. Args: store: An object indexable by handles appropriated during allocate(). context: The object to which the evaluation need to apply (see __call__). """ # Do nothing by default. def finalize(self, store): """Finalize this node's aggregate data. Args: store: An object indexable by handles appropriated during allocate(). """ self.value = store[self.handle] def __call__(self, context): """Return the value on evaluation. Args: context: The evaluation object to which the evaluation need to apply. Returns: The final aggregated value. """ return self.value class SubqueryTable(tables.Table): def __init__(self, subquery): self.columns = {} self.subquery = subquery for i, target in enumerate(target for target in subquery.c_targets if target.name is not None): column = self.column(i, target.name, target.c_expr.dtype) self.columns[target.name] = column() @staticmethod def column(i, name, dtype): class Column(EvalColumn): def __init__(self): super().__init__(dtype) __call__ = staticmethod(operator.itemgetter(i)) return Column def __iter__(self): columns, rows = query_execute.execute_query(self.subquery) return iter(rows) class EvalConstantSubquery1D(EvalNode): def __init__(self, subquery): self.dtype = List[subquery.columns[0].c_expr.dtype] self.subquery = subquery self.value = MARKER def __call__(self, context): if self.value is MARKER: # Subqueries accessing the current row are not supported yet, # thus the subquery result can be simply cached. columns, rows = query_execute.execute_query(self.subquery) value = [row[0] for row in rows] # Subqueries not returning any row are threates as NULL. self.value = value if value else None return self.value # A compiled target. # # Attributes: # c_expr: A compiled expression tree (an EvalNode root node). # name: The name of the target. If None, this is an invisible # target that gets evaluated but not displayed. # is_aggregate: A boolean, true if 'c_expr' is an aggregate. EvalTarget = collections.namedtuple('EvalTarget', 'c_expr name is_aggregate') # A compiled query, ready for execution. # # Attributes: # c_targets: A list of compiled targets (instancef of EvalTarget). # c_where: An instance of EvalNode, a compiled expression tree, for postings. # group_indexes: A list of integers that describe which target indexes to # group by. All the targets referenced here should be non-aggregates. In fact, # this list of indexes should always cover all non-aggregates in 'c_targets'. # And this list may well include some invisible columns if only specified in # the GROUP BY clause. # order_spec: A list of (integer indexes, sort order) tuples. # This list may refer to either aggregates or non-aggregates. # limit: An optional integer used to cut off the number of result rows returned. # distinct: An optional boolean that requests we should uniquify the result rows. @dataclasses.dataclass class EvalQuery: table: tables.Table c_targets: list c_where: EvalNode group_indexes: list[int] having_index: int order_spec: list[tuple[int, ast.Ordering]] limit: int distinct: bool @property def columns(self): return [t for t in self.c_targets if t.name is not None] # A compiled query with a PIVOT BY clause. # # The PIVOT BY clause causes the structure of the returned table to be # fundamentally alterede, thus it makes sense to model it as a # distinct operation. # # Attributes: # query: The underlying EvalQuery. # pivots: The pivot columns indexes EvalPivot = collections.namedtuple('EvalPivot', 'query pivots') beanquery-master/beanquery/query_compile_test.py000066400000000000000000000713211474576771600226570ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import datetime import unittest from decimal import Decimal as D from beanquery import Connection, CompilationError, ProgrammingError from beanquery import compiler from beanquery import query_compile as qc from beanquery import query_env as qe from beanquery import parser from beanquery import tables from beanquery.parser import ast from beanquery.sources import test class Table: # mock table to be used in tests def __init__(self, name): self.name = name def __eq__(self, other): if not isinstance(other, tables.Table): return NotImplemented return other.name == self.name class Column(qc.EvalColumn): # mock column to be used in tests def __init__(self, name, dtype=str): self.name = name self.dtype = dtype def __eq__(self, other): if not isinstance(other, qc.EvalColumn): return False return getattr(other, 'name', type(other).__name__) == self.name class TestCompileExpression(unittest.TestCase): @classmethod def setUpClass(cls): cls.context = Connection() cls.context.tables['test'] = test.Table(0) # parser for expressions cls.parser = parser.BQLParser(start='expression') cls.compiler = compiler.Compiler(cls.context) cls.compiler.table = cls.context.tables['test'] def compile(self, expr): expr = self.parser.parse(expr, semantics=parser.BQLSemantics()) return self.compiler.compile(expr) def test_expr_invalid(self): with self.assertRaises(CompilationError): self.compile('''invalid''') def test_expr_column(self): self.assertEqual(self.compile('''x'''), Column('x', int)) def test_expr_function(self): self.assertEqual(self.compile('''sum(x)'''), qe.SumInt(None, [Column('x', int)])) def test_expr_unaryop(self): self.assertEqual(self.compile('''not x'''), qc.Operator(ast.Not, [Column('x', int)])) def test_expr_binaryop(self): self.assertEqual(self.compile('''x = 1'''), qc.Operator(ast.Equal, [Column('x', int), qc.EvalConstant(1)])) def test_expr_constant(self): self.assertEqual(self.compile('''17'''), qc.EvalConstant(D('17'))) def test_expr_function_arity(self): # compile with an incorrect number of arguments. with self.assertRaises(CompilationError): self.compile('''sum(1, 2)''') def test_constants_folding(self): # unary op self.assertEqual(self.compile('''-2'''), qc.EvalConstant(D('-2'))) # binary op self.assertEqual(self.compile('''2 + 2'''), qc.EvalConstant(D('4'))) # funtion self.assertEqual(self.compile('''root('Assets:Cash', 1)'''), qc.EvalConstant('Assets')) class TestCompileAggregateChecks(unittest.TestCase): def test_is_aggregate_derived(self): columns, aggregates = compiler.get_columns_and_aggregates( qc.EvalAnd([ qc.Operator(ast.Equal, [ Column('lineno', int), qc.EvalConstant(42), ]), qc.EvalOr([ qc.Operator(ast.Not, [ qc.Operator(ast.Equal, [ Column('date', datetime.date), qc.EvalConstant(datetime.date(2014, 1, 1)), ]), ]), qc.EvalConstant(False), ]), ])) self.assertEqual((2, 0), (len(columns), len(aggregates))) columns, aggregates = compiler.get_columns_and_aggregates( qc.EvalAnd([ qc.Operator(ast.Equal, [ Column('lineno', int), qc.EvalConstant(42), ]), qc.EvalOr([ qc.Operator(ast.Not, [ qc.Operator(ast.Not, [ qc.Operator(ast.Equal, [ Column('date', datetime.date), qc.EvalConstant(datetime.date(2014, 1, 1)), ]), ]), ]), # Aggregation node deep in the tree. qe.SumInt(None, [qc.EvalConstant(1)]), ]), ])) self.assertEqual((2, 1), (len(columns), len(aggregates))) def test_get_columns_and_aggregates(self): # Simple column. c_query = Column('position') columns, aggregates = compiler.get_columns_and_aggregates(c_query) self.assertEqual((1, 0), (len(columns), len(aggregates))) self.assertFalse(compiler.is_aggregate(c_query)) # Multiple columns. c_query = qc.EvalAnd([Column('position'), Column('date')]) columns, aggregates = compiler.get_columns_and_aggregates(c_query) self.assertEqual((2, 0), (len(columns), len(aggregates))) self.assertFalse(compiler.is_aggregate(c_query)) # Simple aggregate. c_query = qe.SumPosition(None, [Column('position')]) columns, aggregates = compiler.get_columns_and_aggregates(c_query) self.assertEqual((0, 1), (len(columns), len(aggregates))) self.assertTrue(compiler.is_aggregate(c_query)) # Multiple aggregates. c_query = qc.EvalAnd([qe.First(None, [Column('date')]), qe.Last(None, [Column('flag')])]) columns, aggregates = compiler.get_columns_and_aggregates(c_query) self.assertEqual((0, 2), (len(columns), len(aggregates))) self.assertTrue(compiler.is_aggregate(c_query)) # Simple non-aggregate function. c_query = qe.Function('length', [Column('account')]) columns, aggregates = compiler.get_columns_and_aggregates(c_query) self.assertEqual((1, 0), (len(columns), len(aggregates))) self.assertFalse(compiler.is_aggregate(c_query)) # Mix of column and aggregates (this is used to detect this illegal case). c_query = qc.EvalAnd([ qe.Function('length', [Column('account')]), qe.SumPosition(None, [Column('position')]), ]) columns, aggregates = compiler.get_columns_and_aggregates(c_query) self.assertEqual((1, 1), (len(columns), len(aggregates))) self.assertTrue(compiler.is_aggregate(c_query)) class CompileSelectBase(unittest.TestCase): maxDiff = 8192 def setUp(self): self.ctx = Connection() self.ctx.tables['entries'] = qe.EntriesTable(None, None) self.ctx.tables['postings'] = qe.PostingsTable(None, None) def compile(self, query): """Parse one query and compile it. Args: query: An SQL query to be parsed. Returns: The AST. """ c_query = self.ctx.compile(self.ctx.parse(query)) if isinstance(c_query, ast.Select): self.assertSelectInvariants(c_query) return c_query def assertSelectInvariants(self, query): """Assert the invariants on the query. Args: query: An instance of EvalQuery, a compiled query statement. Raises: AssertionError: if the check fails. """ # Check that the group references cover all the simple indexes. if query.group_indexes is not None: non_aggregate_indexes = [index for index, c_target in enumerate(query.c_targets) if not compiler.is_aggregate(c_target.c_expr)] self.assertEqual(set(non_aggregate_indexes), set(query.group_indexes), "Invalid indexes: {}".format(query)) def assertIndexes(self, query, expected_simple_indexes, expected_aggregate_indexes, expected_group_indexes, expected_order_spec): """Check the four lists of indexes for comparison. Args: query: An instance of EvalQuery, a compiled query statement. expected_simple_indexes: The expected visible non-aggregate indexes. expected_aggregate_indexes: The expected visible aggregate indexes. expected_group_indexes: The expected group_indexes. expected_order_spec: The expected order_spec. Raises: AssertionError: if the check fails. """ # Compute the list of _visible_ aggregates and non-aggregates. simple_indexes = [index for index, c_target in enumerate(query.c_targets) if c_target.name and not compiler.is_aggregate(c_target.expression)] aggregate_indexes = [index for index, c_target in enumerate(query.c_targets) if c_target.name and compiler.is_aggregate(c_target.expression)] self.assertEqual(set(expected_simple_indexes), set(simple_indexes)) self.assertEqual(set(expected_aggregate_indexes), set(aggregate_indexes)) self.assertEqual( set(expected_group_indexes) if expected_group_indexes is not None else None, set(query.group_indexes) if query.group_indexes is not None else None) self.assertEqual( set(expected_order_spec) if expected_order_spec is not None else None, set(query.order_spec) if query.order_spec is not None else None) def assertCompile(self, expected, query, debug=False): """Assert parsed and compiled contents from 'query' is 'expected'. Args: expected: An expected AST to compare against the parsed value. query: An SQL query to be parsed. debug: A boolean, if true, print extra debugging information on the console. Raises: AssertionError: If the actual AST does not match the expected one. """ actual = self.compile(query) if debug: print() print() print(actual) print() try: self.assertEqual(expected, actual) return actual except AssertionError: print() print("Expected: {}".format(expected)) print("Actual : {}".format(actual)) raise class TestCompileSelect(CompileSelectBase): def test_compile_from(self): # Test the compilation of from. query = self.compile("SELECT account FROM CLOSE;") self.assertEqual(query.table.close, True) query = self.compile("SELECT account FROM length(payee) != 0;") self.assertTrue(isinstance(query.c_where, qc.EvalNode)) with self.assertRaises(CompilationError): query = self.compile("SELECT account FROM sum(payee) != 0;") def test_compile_from_invalid_dates(self): self.compile(""" SELECT account FROM OPEN ON 2014-03-01 CLOSE ON 2014-03-02; """) self.compile(""" SELECT account FROM OPEN ON 2014-03-02 CLOSE ON 2014-03-02; """) with self.assertRaises(CompilationError): self.compile(""" SELECT account FROM OPEN ON 2014-03-03 CLOSE ON 2014-03-02; """) def test_compile_targets_wildcard(self): # Test the wildcard expansion. query = self.compile("SELECT *;") self.assertTrue(list, type(query.c_targets)) self.assertGreater(len(query.c_targets), 3) self.assertTrue(all(isinstance(target.c_expr, qc.EvalColumn) for target in query.c_targets)) def test_compile_targets_named(self): # Test the wildcard expansion. query = self.compile("SELECT length(account), account as a, date;") self.assertEqual( [qc.EvalTarget(qe.Function('length', [Column('account')]), 'length(account)', False), qc.EvalTarget(Column('account'), 'a', False), qc.EvalTarget(Column('date'), 'date', False)], query.c_targets) def test_compile_mixed_aggregates(self): # Check mixed aggregates and non-aggregates in a target. with self.assertRaises(CompilationError) as assertion: self.compile(""" SELECT length(account) and sum(length(account)); """) self.assertRegex(str(assertion.exception), 'mixed aggregates and non-aggregates') def test_compile_aggregates_of_aggregates(self): # Check mixed aggregates and non-aggregates in a target. with self.assertRaises(CompilationError) as assertion: self.compile(""" SELECT sum(sum(length(account))); """) self.assertRegex(str(assertion.exception), 'aggregates of aggregates') def test_compile_having_non_aggregate(self): with self.assertRaises(CompilationError) as assertion: self.compile(""" SELECT account, sum(number) GROUP BY account HAVING flag; """) self.assertRegex(str(assertion.exception), 'the HAVING clause must be an aggregate') def test_compile_group_by_inventory(self): with self.assertRaises(CompilationError): self.compile(""" SELECT sum(number), balance GROUP BY balance; """) class TestCompileSelectGroupBy(CompileSelectBase): def test_compile_group_by_non_aggregates(self): self.compile(""" SELECT payee GROUP BY payee, length(account); """) with self.assertRaises(CompilationError) as assertion: self.compile(""" SELECT payee GROUP BY payee, last(account); """) self.assertRegex(str(assertion.exception), 'may not be aggregates') def test_compile_group_by_reference_by_name(self): # Valid references to target names. self.compile(""" SELECT payee, last(account) GROUP BY payee; """) self.compile(""" SELECT payee as a, last(account) as len GROUP BY a; """) # References to non-targets have to be valid. self.compile(""" SELECT payee, last(account) as len GROUP BY payee, date; """) with self.assertRaises(CompilationError): self.compile(""" SELECT payee, last(account) as len GROUP BY something; """) def test_compile_group_by_reference_by_number(self): self.compile(""" SELECT date, payee, narration GROUP BY 1, 2, 3; """) with self.assertRaises(CompilationError): self.compile(""" SELECT date, payee, narration GROUP BY 1, 2, 3, 4; """) def test_compile_group_by_reference_an_aggregate(self): # By name. with self.assertRaises(CompilationError): self.compile(""" SELECT payee, last(account) as last GROUP BY last; """) with self.assertRaises(CompilationError): self.compile(""" SELECT account, sum(number) as sum_num GROUP BY account, sum_num; """) # By number. with self.assertRaises(CompilationError): self.compile(""" SELECT payee, last(account) as last GROUP BY 2; """) # Explicit aggregate in group-by clause. with self.assertRaises(CompilationError): self.compile(""" SELECT account, sum(number) GROUP BY account, sum(number); """) def test_compile_group_by_implicit(self): self.compile(""" SELECT payee, last(account); """) self.compile(""" SELECT first(account), last(account); """) def test_compile_group_by_coverage(self): # Non-aggregates. query = self.compile("SELECT account, length(account);") self.assertEqual(None, query.group_indexes) self.assertEqual(None, query.order_spec) # Aggregates only. query = self.compile("SELECT first(account), last(account);") self.assertEqual([], query.group_indexes) # Mixed with non-aggregates in group-by clause. query = self.compile("SELECT account, sum(number) GROUP BY account;") self.assertEqual([0], query.group_indexes) # Mixed with non-aggregates in group-by clause with non-aggregates a # strict subset of the group-by columns. 'account' is a subset of # {'account', 'flag'}. query = self.compile(""" SELECT account, sum(number) GROUP BY account, flag; """) self.assertEqual([0, 2], query.group_indexes) # Non-aggregates not covered by group-by clause. with self.assertRaises(CompilationError): self.compile(""" SELECT account, date, sum(number) GROUP BY account; """) with self.assertRaises(CompilationError): self.compile(""" SELECT payee, last(account) as len GROUP BY date; """) # Non-aggregates not covered by group-by clause, and no aggregates in # the list of targets. with self.assertRaises(CompilationError): self.compile(""" SELECT date, flag, account, number GROUP BY date, flag; """) # All non-aggregates and matching list of aggregates (this is a # pointless list of aggregates, essentially). query = self.compile(""" SELECT date, flag, account GROUP BY date, flag, account; """) self.assertEqual([0, 1, 2], query.group_indexes) def test_compile_group_by_reconcile(self): # Check that no invisible column is created if redundant. query = self.compile(""" SELECT account, length(account), sum(number) GROUP BY account, length(account); """) self.assertEqual([0, 1], query.group_indexes) class TestCompileSelectOrderBy(CompileSelectBase): def test_compile_order_by_simple(self): query = self.compile(""" SELECT account, sum(number) GROUP BY account ORDER BY account; """) self.assertEqual([0], query.group_indexes) self.assertEqual([(0, False)], query.order_spec) def test_compile_order_by_simple_2(self): query = self.compile(""" SELECT account, length(narration) GROUP BY account, 2 ORDER BY 1, 2; """) self.assertEqual([0, 1], query.group_indexes) self.assertEqual([(0, False), (1, False)], query.order_spec) query = self.compile(""" SELECT account, length(narration) as l GROUP BY account, l ORDER BY l; """) self.assertEqual([0, 1], query.group_indexes) self.assertEqual([(1, False)], query.order_spec) def test_compile_order_by_create_non_agg(self): with self.assertRaises(CompilationError): self.compile(""" SELECT account, last(narration) GROUP BY account ORDER BY year(date); """) with self.assertRaises(CompilationError): self.compile(""" SELECT account GROUP BY account ORDER BY year(date); """) query = self.compile(""" SELECT account, year(date) GROUP BY 1, 2 ORDER BY 2; """) self.assertEqual([0, 1], query.group_indexes) self.assertEqual([(1, False)], query.order_spec) # We detect similarity between order-by and targets yet. self.compile(""" SELECT account, year(date) GROUP BY 1, 2 ORDER BY year(date); """) def test_compile_order_by_reconcile(self): # Check that no invisible column is created if redundant. query = self.compile(""" SELECT account, length(account) ORDER BY length(account); """) self.assertEqual([(1, False)], query.order_spec) def test_compile_order_by_reference_invisible(self): # So this is an interesting case: the grouping expression is an # invisible non-aggregate (length(account)) and the ordering expression # refers to the same non-aggregate expression. If they are reconciled to # the same invisible expression, the condition that the grouping # expressions cover all the non-aggregates is fulfilled. Otherwise, it # would fail. In order to support the compilation of this, we must # reconcile the grouping and ordering columns by comparing their values. query = self.compile(""" SELECT count(account) as num, first(account) as first GROUP BY length(account) ORDER BY length(account); """) self.assertEqual([2], query.group_indexes) self.assertEqual([(2, False)], query.order_spec) def test_compile_order_by_aggregate(self): query = self.compile(""" SELECT account, first(narration) GROUP BY account ORDER BY 2; """) self.assertEqual([0], query.group_indexes) self.assertEqual([(1, False)], query.order_spec) query = self.compile(""" SELECT account, first(narration) as f GROUP BY account ORDER BY f; """) self.assertEqual([0], query.group_indexes) self.assertEqual([(1, False)], query.order_spec) query = self.compile(""" SELECT account, first(narration) GROUP BY account ORDER BY sum(number); """) self.assertEqual([0], query.group_indexes) self.assertEqual([(2, False)], query.order_spec) query = self.compile(""" SELECT account GROUP BY account ORDER BY sum(number); """) self.assertEqual([0], query.group_indexes) self.assertEqual([(1, False)], query.order_spec) class TestTranslationJournal(CompileSelectBase): maxDiff = 4096 def test_journal(self): journal = parser.parse("JOURNAL;") select = compiler.transform_journal(journal) self.assertEqual(select, ast.Select([ ast.Target(ast.Column('date'), None), ast.Target(ast.Column('flag'), None), ast.Target(ast.Function('maxwidth', [ ast.Column('payee'), ast.Constant(48)]), None), ast.Target(ast.Function('maxwidth', [ ast.Column('narration'), ast.Constant(80)]), None), ast.Target(ast.Column('account'), None), ast.Target(ast.Column('position'), None), ast.Target(ast.Column('balance'), None), ], None, None, None, None, None, None, None)) def test_journal_with_account(self): journal = parser.parse("JOURNAL 'liabilities';") select = compiler.transform_journal(journal) self.assertEqual(select, ast.Select([ ast.Target(ast.Column('date'), None), ast.Target(ast.Column('flag'), None), ast.Target(ast.Function('maxwidth', [ ast.Column('payee'), ast.Constant(48)]), None), ast.Target(ast.Function('maxwidth', [ ast.Column('narration'), ast.Constant(80)]), None), ast.Target(ast.Column('account'), None), ast.Target(ast.Column('position'), None), ast.Target(ast.Column('balance'), None), ], None, ast.Match(ast.Column('account'), ast.Constant('liabilities')), None, None, None, None, None)) def test_journal_with_account_and_from(self): journal = parser.parse("JOURNAL 'liabilities' FROM year = 2014;") select = compiler.transform_journal(journal) self.assertEqual(select, ast.Select([ ast.Target(ast.Column('date'), None), ast.Target(ast.Column('flag'), None), ast.Target(ast.Function('maxwidth', [ ast.Column('payee'), ast.Constant(48)]), None), ast.Target(ast.Function('maxwidth', [ ast.Column('narration'), ast.Constant(80)]), None), ast.Target(ast.Column('account'), None), ast.Target(ast.Column('position'), None), ast.Target(ast.Column('balance'), None), ], ast.From(ast.Equal(ast.Column('year'), ast.Constant(2014)), None, None, None), ast.Match(ast.Column('account'), ast.Constant('liabilities')), None, None, None, None, None)) def test_journal_with_account_func_and_from(self): journal = parser.parse("JOURNAL 'liabilities' AT cost FROM year = 2014;") select = compiler.transform_journal(journal) self.assertEqual(select, ast.Select([ ast.Target(ast.Column('date'), None), ast.Target(ast.Column('flag'), None), ast.Target(ast.Function('maxwidth', [ ast.Column('payee'), ast.Constant(48)]), None), ast.Target(ast.Function('maxwidth', [ ast.Column('narration'), ast.Constant(80)]), None), ast.Target(ast.Column('account'), None), ast.Target(ast.Function('cost', [ast.Column('position')]), None), ast.Target(ast.Function('cost', [ast.Column('balance')]), None), ], ast.From(ast.Equal(ast.Column('year'), ast.Constant(2014)), None, None, None), ast.Match(ast.Column('account'), ast.Constant('liabilities')), None, None, None, None, None)) class TestTranslationBalance(CompileSelectBase): group_by = ast.GroupBy([ ast.Column('account'), ast.Function('account_sortkey', [ ast.Column(name='account')])], None) order_by = [ast.OrderBy(ast.Function('account_sortkey', [ast.Column('account')]), ast.Ordering.ASC)] def test_balance(self): balance = parser.parse("BALANCES;") select = compiler.transform_balances(balance) self.assertEqual(select, ast.Select([ ast.Target(ast.Column('account'), None), ast.Target(ast.Function('sum', [ ast.Column('position') ]), None), ], None, None, self.group_by, self.order_by, None, None, None)) def test_balance_with_units(self): balance = parser.parse("BALANCES AT cost;") select = compiler.transform_balances(balance) self.assertEqual(select, ast.Select([ ast.Target(ast.Column('account'), None), ast.Target(ast.Function('sum', [ ast.Function('cost', [ ast.Column('position') ]) ]), None) ], None, None, self.group_by, self.order_by, None, None, None)) def test_balance_with_units_and_from(self): balance = parser.parse("BALANCES AT cost FROM year = 2014;") select = compiler.transform_balances(balance) self.assertEqual(select, ast.Select([ ast.Target(ast.Column('account'), None), ast.Target(ast.Function('sum', [ ast.Function('cost', [ ast.Column('position') ]) ]), None), ], ast.From(ast.Equal(ast.Column('year'), ast.Constant(2014)), None, None, None), None, self.group_by, self.order_by, None, None, None)) def test_print(self): self.assertCompile( qc.EvalQuery( Table('entries'), [qc.EvalTarget(qc.EvalRow(), 'ROW(*)', False)], None, None, None, None, None, False), "PRINT;", ) def test_print_from(self): self.assertCompile( qc.EvalQuery( Table('entries'), [qc.EvalTarget(qc.EvalRow(), 'ROW(*)', False)], qc.Operator(ast.Equal, [ Column('year', int), qc.EvalConstant(2014) ]), None, None, None, None, False), "PRINT FROM year = 2014;") class TestCompileParameters(unittest.TestCase): @classmethod def setUpClass(cls): cls.context = Connection() def compile(self, query, params): c = compiler.Compiler(self.context) c.table = self.context.tables.get('') return c.compile(parser.parse(query), params) def test_named_parameters(self): query = self.compile('''SELECT %(x)s + %(y)s''', {'x': 1, 'y': 2}) self.assertEqual(query, qc.EvalQuery( Table(''), [ # addition of constants is optimized away qc.EvalTarget(qc.EvalConstant(3), '%(x)s + %(y)s', False) ], None, None, None, None, None, None)) def test_positional_parameters(self): query = self.compile('''SELECT %s + %s''', (1, 2, )) self.assertEqual(query, qc.EvalQuery( Table(''), [ # addition of constants is optimized away qc.EvalTarget(qc.EvalConstant(3), '%s + %s', False) ], None, None, None, None, None, None)) def test_mixing_parameters(self): with self.assertRaises(ProgrammingError): self.compile('''SELECT %s + %(foo)s''', (1, 2)) def test_missing_parameters_positional(self): with self.assertRaises(ProgrammingError): self.compile('''SELECT %s + %s''', (1, )) def test_missing_parameters_named(self): with self.assertRaises(ProgrammingError): self.compile('''SELECT %(x)s + %(y)s''', {'x': 1}) beanquery-master/beanquery/query_env.py000066400000000000000000001124501474576771600207570ustar00rootroot00000000000000"""Environment object for compiler. This module contains the various column accessors and function evaluators that are made available by the query compiler via their compilation context objects. Define new columns and functions here. """ __copyright__ = "Copyright (C) 2014-2017 Martin Blais" __license__ = "GNU GPLv2" import copy import datetime import decimal import re import textwrap from functools import lru_cache as cache from decimal import Decimal from typing import Set import dateutil.parser from dateutil.relativedelta import relativedelta, weekday from beancount.core.number import ZERO from beancount.core.compare import hash_entry from beancount.core import amount from beancount.core import position from beancount.core import inventory from beancount.core import account from beancount.core import data from beancount.core import getters from beancount.core import convert from beancount.core import prices from beancount.core.account_types import get_account_sign, get_account_sort_key from beancount.ops import summarize from beanquery import query_compile from beanquery import tables from beanquery import types def function(intypes, outtype, pass_context=None, name=None): def decorator(func): class Func(query_compile.EvalFunction): __intypes__ = intypes pure = not pass_context def __init__(self, context, operands): super().__init__(context, operands, outtype) def __call__(self, row): args = [operand(row) for operand in self.operands] for arg in args: if arg is None: return None if pass_context: return func(self.context, *args) return func(*args) Func.__name__ = name if name is not None else func.__name__ Func.__doc__ = func.__doc__ query_compile.FUNCTIONS[Func.__name__].append(Func) return func return decorator def register(name=None): def decorator(cls): if name is not None: cls.__name__ = name query_compile.FUNCTIONS[cls.__name__].append(cls) return cls return decorator @register('getitem') class GetItem2(query_compile.EvalFunction): __intypes__ = [dict, str] def __init__(self, context, operands): super().__init__(context, operands, object) def __call__(self, row): obj, key = self.operands obj = obj(row) if obj is None: return None return obj.get(key(row)) @register('getitem') class GetItem3(query_compile.EvalFunction): __intypes__ = [dict, str, types.Any] def __init__(self, context, operands): super().__init__(context, operands, object) def __call__(self, row): obj, key, default = self.operands obj = obj(row) if obj is None: return None return obj.get(key(row), default(row)) def Function(name, args): func = types.function_lookup(query_compile.FUNCTIONS, name, args) if func is not None: return func(None, args) raise KeyError ## Type casting @function([types.Any], bool, name='bool') def bool_(x): """Convert to bool value.""" return bool(x) @function([int], int, name='int') @function([bool], int, name='int') @function([Decimal], int, name='int') @function([str], int, name='int') @function([object], int, name='int') def int_(x): try: return int(x) except (ValueError, TypeError): return None @function([Decimal], Decimal, name='decimal') @function([int], Decimal, name='decimal') @function([bool], Decimal, name='decimal') @function([str], Decimal, name='decimal') @function([object], Decimal, name='decimal') def decimal_(x): try: return Decimal(x) except (ValueError, TypeError, decimal.InvalidOperation): return None @function([types.Any], str, name='str') def str_(x): if x is True: return 'TRUE' if x is False: return 'FALSE' return str(x) @function([datetime.date], datetime.date, name='date') @function([str], datetime.date, name='date') @function([object], datetime.date, name='date') def date_(x): if isinstance(x, datetime.date): return x if isinstance(x, str): try: return datetime.datetime.strptime(x, '%Y-%m-%d').date() except ValueError: pass return None @function([int, int, int], datetime.date, name='date') def date_from_ymd(year, month, day): """Construct a date with year, month, day arguments.""" try: return datetime.date(year, month, day) except ValueError: return None ## Functions @function([Decimal], Decimal) @function([amount.Amount], amount.Amount) @function([position.Position], position.Position) @function([inventory.Inventory], inventory.Inventory) def neg(x): """Negative value.""" return -x @function([Decimal], Decimal, name='abs') @function([position.Position], position.Position, name='abs') @function([inventory.Inventory], inventory.Inventory, name='abs') def abs_(x): """Absolute value.""" return abs(x) @function([Decimal, Decimal], Decimal) @function([Decimal, int], Decimal) def safediv(x, y): """A division operation that traps division by zero exceptions and outputs zero instead.""" if y == 0: return ZERO return x / y @function([Decimal], Decimal, name='round') @function([Decimal, int], Decimal, name='round') @function([int], int, name='round') @function([int, int], int, name='round') def round_(num, digits=0): """Round the argument.""" return round(num, digits) @function([list], int) @function([set], int) @function([str], int) def length(x): """Compute the length of the argument. This works on sequences.""" return len(x) @function([types.Any], str, name='repr') def repr_(x): """Convert the argument to a string via repr().""" return repr(x) @function([str, int], str) def maxwidth(x, n): """Convert the argument to a substring. This can be used to ensure maximum width. This will insert ellipsis ([...]) if necessary.""" return textwrap.shorten(x, width=n) @function([str, int, int], str) def substr(string, start, end): """Extract a substring of the argument.""" return string[start:end] @function([str, str, int], str) def splitcomp(string, delim, index): """Split a string and extract one of its components.""" return string.split(delim)[index] # Operations on dates. @function([datetime.date], int) def year(x): """Extract the year from a date.""" return x.year @function([datetime.date], int) def month(x): """Extract the month from a date.""" return x.month @function([datetime.date], int) def day(x): """Extract the day from a date.""" return x.day @function([datetime.date], datetime.date) def yearmonth(x): """Extract the year and month from a date.""" return datetime.date(x.year, x.month, 1) @function([datetime.date], str) def quarter(x): """Extract the quarter from a date.""" return '{:04d}-Q{:1d}'.format(x.year, (x.month - 1) // 3 + 1) @function([datetime.date], str, name='weekday') def weekday_(x): """Extract a 3-letter weekday from a date.""" return x.strftime('%a') @function([], datetime.date) def today(): """Today's date""" return datetime.date.today() # Operations on accounts. @function([str], str) @function([str, int], str) def root(acc, n=1): """Get the root name(s) of the account.""" return account.root(n, acc) @function([str], str) def parent(acc): """Get the parent name of the account.""" return account.parent(acc) @function([str], str) def leaf(acc): """Get the name of the leaf subaccount.""" return account.leaf(acc) @function([str, str], str) def grep(pattern, string): """Match a regular expression against a string and return only the matched portion.""" match = re.search(pattern, string) if match: return match.group(0) return None @function([str, str, int], str) def grepn(pattern, string, n): """Match a pattern with subgroups against a string and return the subgroup at the index.""" match = re.search(pattern, string) if match: return match.group(n) return None @function([str, str, str], str) def subst(pattern, repl, string): """Substitute leftmost non-overlapping occurrences of pattern by replacement.""" return re.sub(pattern, repl, string) @function([str], str) def upper(string): """Convert string to uppercase.""" return string.upper() @function([str], str) def lower(string): """Convert string to lowercase.""" return string.lower() # Avoid building a tuple for each function invocation. NONENONE = None, None @function([str], datetime.date, pass_context=True) def open_date(context, acc): """Get the date of the open directive of the account.""" open_entry, _ = context.tables['accounts'].accounts.get(acc, NONENONE) if open_entry is None: return None return open_entry.date @function([str], datetime.date, pass_context=True) def close_date(context, acc): """Get the date of the close directive of the account.""" _, close_entry = context.tables['accounts'].accounts.get(acc, NONENONE) if close_entry is None: return None return close_entry.date @function([str], dict, pass_context=True) @function([str, str], object, pass_context=True) def open_meta(context, account, key=None): """Get the metadata dict of the open directive of the account.""" open_entry, _ = context.tables['accounts'].accounts.get(account, NONENONE) if open_entry is None: return None if key is None: return open_entry.meta return open_entry.meta.get(key) # Stub kept only for function type checking and for generating documentation. @function([str], object) def meta(context, key): """Get some metadata key of the posting.""" raise NotImplementedError # Stub kept only for function type checking and for generating documentation. @function([str], object) def entry_meta(context, key): """Get some metadata key of the transaction.""" raise NotImplementedError # Stub kept only for function type checking and for generating documentation. @function([str], object) def any_meta(context, key): """Get metadata from the posting or its parent transaction if not present.""" raise NotImplementedError @function([str], dict, pass_context=True) @function([str, str], object, pass_context=True) @function([str], dict, pass_context=True, name='commodity_meta') @function([str, str], object, pass_context=True, name='commodity_meta') def currency_meta(context, commodity, key=None): """Get the metadata dict of the commodity directive of the currency.""" entry = context.tables['commodities'].commodities.get(commodity) if entry is None: return None if key is None: return entry.meta return entry.meta.get(key) @function([str], str, pass_context=True) def account_sortkey(context, acc): """Get a string to sort accounts in order taking into account the types.""" account_types = context.tables['accounts'].types index, name = get_account_sort_key(account_types, acc) return '{}-{}'.format(index, name) # Stub kept only for function type checking and for generating documentation. @function([str], bool) def has_account(context, pattern): """True if the transaction has at least one posting matching the regular expression argument.""" raise NotImplementedError # Note: Don't provide this, because polymorphic multiplication on Amount, # Position, Inventory isn't supported yet. # # class AccountSign(query_compile.EvalFunction): # "Produce a +1 / -1 signed value to multiply with to correct balances." # __intypes__ = [str] # # def __init__(self, operands): # super().__init__(operands, Decimal) # # def __call__(self, context): # args = self.eval_args(context) # return Decimal(account_types.get_account_sign(args[0], context.account_types)) # Operation on inventories, positions and amounts. @function([position.Position], amount.Amount, name='units') def position_units(pos): """Get the number of units of a position (stripping cost).""" return convert.get_units(pos) @function([inventory.Inventory], inventory.Inventory, name='units') def inventory_units(inv): """Get the number of units of an inventory (stripping cost).""" return inv.reduce(convert.get_units) @function([position.Position], amount.Amount, name='cost') def position_cost(pos): """Get the cost of a position.""" return convert.get_cost(pos) @function([inventory.Inventory], inventory.Inventory, name='cost') def inventory_cost(inv): """Get the cost of an inventory.""" return inv.reduce(convert.get_cost) @function([amount.Amount, str], amount.Amount, pass_context=True, name='convert') @function([amount.Amount, str, datetime.date], amount.Amount, pass_context=True, name='convert') def convert_amount(context, amount_, currency, date=None): """Coerce an amount to a particular currency.""" price_map = context.tables['prices'].price_map return convert.convert_amount(amount_, currency, price_map, date) @function([position.Position, str], amount.Amount, pass_context=True, name='convert') @function([position.Position, str, datetime.date], amount.Amount, pass_context=True, name='convert') def convert_position(context, pos, currency, date=None): """Coerce an amount to a particular currency.""" price_map = context.tables['prices'].price_map return convert.convert_position(pos, currency, price_map, date) @function([inventory.Inventory, str], inventory.Inventory, pass_context=True, name='convert') @function([inventory.Inventory, str, datetime.date], inventory.Inventory, pass_context=True, name='convert') def convert_inventory(context, inv, currency, date=None): """Coerce an inventory to a particular currency.""" price_map = context.tables['prices'].price_map return inv.reduce(convert.convert_position, currency, price_map, date) @function([position.Position], amount.Amount, pass_context=True, name='value') @function([position.Position, datetime.date], amount.Amount, pass_context=True, name='value') def position_value(context, pos, date=None): """Convert a position to its cost currency at the market value.""" price_map = context.tables['prices'].price_map return convert.get_value(pos, price_map, date) @function([inventory.Inventory], inventory.Inventory, pass_context=True, name='value') @function([inventory.Inventory, datetime.date], inventory.Inventory, pass_context=True, name='value') def inventory_value(context, inv, date=None): """Coerce an inventory to its market value.""" price_map = context.tables['prices'].price_map return inv.reduce(convert.get_value, price_map, date) @function([str, str], Decimal, pass_context=True) @function([str, str, datetime.date], Decimal, pass_context=True, name='getprice') def getprice(context, base, quote, date=None): """Fetch a price.""" price_map = context.tables['prices'].price_map pair = (base.upper(), quote.upper()) _, price = prices.get_price(price_map, pair, date) return price @function([amount.Amount], Decimal) def number(x): """Extract the number from an Amount.""" return x.number @function([amount.Amount], str) @function([amount.Amount], str, name='commodity') def currency(x): """Extract the currency from an Amount.""" return x.currency @function([str, set], str) def findfirst(pattern, values): """Filter a string sequence by regular expression and return the first match.""" if not values: return None for value in sorted(values): if re.match(pattern, value): return value return None @function([set], str) def joinstr(values): """Join a sequence of strings to a single comma-separated string.""" return ','.join(values) @function([str, inventory.Inventory], amount.Amount, name='only') def only_inventory(currency, inventory_): """Get one currency's amount from the inventory.""" return inventory_.get_currency_units(currency) @function([inventory.Inventory], bool, name='empty') def empty_inventory(inventory_): """Determine whether the inventiry is empty.""" return inventory_.is_empty() @function([position.Position, str], position.Position, name='filter_currency') def filter_currency_position(pos, currency): """Filter an inventory to just the specified currency.""" return pos if pos.units.currency == currency else None @function([inventory.Inventory, str], inventory.Inventory, name='filter_currency') def filter_currency_inventory(inv, currency): """Filter an inventory to just the specified currency.""" return inventory.Inventory(pos for pos in inv if pos.units.currency == currency) @function([Decimal, str], Decimal, pass_context=True) @function([amount.Amount, str], amount.Amount, pass_context=True) @function([position.Position, str], position.Position, pass_context=True) @function([inventory.Inventory, str], inventory.Inventory, pass_context=True) def possign(context, x, account): """Correct sign of an Amount based on the usual balance of associated account.""" account_types = context.tables['accounts'].types sign = get_account_sign(account, account_types) return x if sign >= 0 else -x @function([str], datetime.date) @function([str, str], datetime.date) def parse_date(string, frmt=None): """Parse date from string.""" if frmt is None: return dateutil.parser.parse(string).date() return datetime.datetime.strptime(string, frmt).date() @function([datetime.date, datetime.date], int) def date_diff(x, y): """Calculates the difference (in days) between two dates.""" return (x - y).days @function([datetime.date, int], datetime.date) def date_add(x, y): """Adds/subtracts number of days from the given date.""" return x + datetime.timedelta(days=y) @function([str, datetime.date], datetime.date) def date_trunc(field, x): """Truncate a date to the specified precision.""" if field == 'week': return x - relativedelta(weekday=weekday(0, -1)) if field == 'month': return datetime.date(x.year, x.month, 1) if field == 'quarter': return datetime.date(x.year, x.month - (x.month - 1) % 3, 1) if field == 'year': return datetime.date(x.year, 1, 1) if field == 'decade': return datetime.date(x.year - x.year % 10, 1, 1) if field == 'century': return datetime.date(x.year - (x.year - 1) % 100, 1, 1) if field == 'millennium': return datetime.date(x.year - (x.year - 1) % 1000, 1, 1) return None @function([str, datetime.date], int) def date_part(field, x): """Extract the specified field from a date.""" if field == 'weekday' or field == 'dow': return x.weekday() if field == 'isoweekday' or field == 'isodow': return x.isoweekday() if field == 'week': # isocalendar() returns a named tuple only in Python >= 3.9. return x.isocalendar()[1] if field == 'month': return x.month if field == 'quarter': return (x.month - 1) // 3 + 1 if field == 'year': return x.year if field == 'isoyear': # isocalendar() returns a named tuple only in Python >= 3.9. return x.isocalendar()[0] if field == 'decade': return x.year // 10 if field == 'century': return (x.year - 1) // 100 + 1 if field == 'millennium': return (x.year - 1) // 1000 + 1 if field == 'epoch': return int((x - datetime.date(1970, 1, 1)).total_seconds()) return None @function([str], relativedelta) def interval(x): """Construct a relative time interval.""" m = re.fullmatch(r'([-+]?[0-9]+)\s+(day|month|year)s?', x) if not m: return None number = int(m.group(1)) unit = m.group(2) if unit == 'day': return relativedelta(days=number) if unit == 'week': return relativedelta(weeks=number) if unit == 'month': return relativedelta(months=number) if unit == 'year': return relativedelta(years=number) if unit == 'decade': return relativedelta(years=number * 10) if unit == 'century': return relativedelta(years=number * 100) if unit == 'millennium': return relativedelta(years=number * 1000) return None @function([relativedelta, datetime.date, datetime.date], datetime.date) def date_bin(stride, source, origin): """Bin a date into the specified stride aligned with the specified origin. As an extension to the the SQL standard ``date_bin()`` function this function also accepts strides containing units of months and years. """ if stride.months or stride.years: if origin + stride <= origin: # FIXME: this should raise and error: stride must be greater than zero return None if source >= origin: d = n = origin while True: n += stride if n >= source: return d d = n else: n = origin while True: n -= stride if n <= source: return n else: seconds = stride.days * 86400 + stride.hours * 3600 + stride.minutes * 60 + stride.seconds if seconds < 0: # FIXME: this should raise and error: stride must be greater than zero return None diff = (source - origin).total_seconds() modulo = diff % seconds delta = diff - modulo result = origin + datetime.timedelta(seconds=delta) if modulo < 0: result -= datetime.timedelta(seconds=seconds) return result @function([str, datetime.date, datetime.date], datetime.date, name='date_bin') def date_bin_str(stride, source, origin): return date_bin(interval(stride), source, origin) def aggregator(intypes, name=None): def decorator(cls): cls.__intypes__ = intypes if name is not None: cls.__name__ = name query_compile.FUNCTIONS[cls.__name__].append(cls) return cls return decorator @aggregator([types.Asterisk], name='count') class Count(query_compile.EvalAggregator): """Count the number of input rows.""" def __init__(self, context, operands): super().__init__(context, operands, int) def update(self, store, context): store[self.handle] += 1 @aggregator([types.Any], name='count') class CountArg(query_compile.EvalAggregator): """Count the number of non-NULL occurrences of the argument.""" def __init__(self, context, operands): super().__init__(context, operands, int) def update(self, store, context): value = self.operands[0](context) if value is not None: store[self.handle] += 1 @aggregator([int], name='sum') class SumInt(query_compile.EvalAggregator): """Calculate the sum of the numerical argument.""" def __init__(self, context, operands): super().__init__(context, operands, operands[0].dtype) def update(self, store, context): value = self.operands[0](context) if value is not None: store[self.handle] += value @aggregator([Decimal], name='sum') class SumDecimal(query_compile.EvalAggregator): """Calculate the sum of the numerical argument.""" def update(self, store, context): value = self.operands[0](context) if value is not None: store[self.handle] += value @aggregator([amount.Amount], name='sum') class SumAmount(query_compile.EvalAggregator): """Calculate the sum of the amount. The result is an Inventory.""" def __init__(self, context, operands): super().__init__(context, operands, inventory.Inventory) def update(self, store, context): value = self.operands[0](context) if value is not None: store[self.handle].add_amount(value) @aggregator([position.Position], name='sum') class SumPosition(query_compile.EvalAggregator): """Calculate the sum of the position. The result is an Inventory.""" def __init__(self, context, operands): super().__init__(context, operands, inventory.Inventory) def update(self, store, context): value = self.operands[0](context) if value is not None: store[self.handle].add_position(value) @aggregator([inventory.Inventory], name='sum') class SumInventory(query_compile.EvalAggregator): """Calculate the sum of the inventories. The result is an Inventory.""" def __init__(self, context, operands): super().__init__(context, operands, inventory.Inventory) def update(self, store, context): value = self.operands[0](context) if value is not None: store[self.handle].add_inventory(value) @aggregator([types.Any], name='first') class First(query_compile.EvalAggregator): """Keep the first of the values seen.""" def initialize(self, store): store[self.handle] = None def update(self, store, context): if store[self.handle] is None: value = self.operands[0](context) store[self.handle] = value @aggregator([types.Any], name='last') class Last(query_compile.EvalAggregator): """Keep the last of the values seen.""" def initialize(self, store): store[self.handle] = None def update(self, store, context): value = self.operands[0](context) store[self.handle] = value @aggregator([types.Any], name='min') class Min(query_compile.EvalAggregator): """Compute the minimum of the values.""" def initialize(self, store): store[self.handle] = None def update(self, store, context): value = self.operands[0](context) if value is not None: cur = store[self.handle] if cur is None or value < cur: store[self.handle] = value @aggregator([types.Any], name='max') class Max(query_compile.EvalAggregator): """Compute the maximum of the values.""" def initialize(self, store): store[self.handle] = None def update(self, store, context): value = self.operands[0](context) if value is not None: cur = store[self.handle] if cur is None or value > cur: store[self.handle] = value class Row: """A dumb container for information used by a row expression.""" def __init__(self): self.rowid = 0 self.balance = inventory.Inventory() # The current transaction of the posting being evaluated. self.entry = None # The current posting being evaluated. self.posting = None def __hash__(self): # The context hash is used in caching column accessor functions. # Instead than hashing the row context content, use the rowid as # hash. return self.rowid class BeanTable(tables.Table): def __init__(self, entries, options, open=None, close=None, clear=None): super().__init__() self.entries = entries self.options = options self.open = open self.close = close self.clear = clear def update(self, **kwargs): table = copy.copy(self) for name, value in kwargs.items(): setattr(table, name, value) return table def prepare(self): """Filter the entries applying the FROM clause qualifiers OPEN, CLOSE, CLEAR.""" entries = self.entries options = self.options # Process the OPEN clause. if self.open is not None: entries, index = summarize.open_opt(entries, self.open, options) # Process the CLOSE clause. if self.close is not None: if isinstance(self.close, datetime.date): entries, index = summarize.close_opt(entries, self.close, options) elif self.close is True: entries, index = summarize.close_opt(entries, None, options) # Process the CLEAR clause. if self.clear is not None: entries, index = summarize.clear_opt(entries, None, options) return entries class ColumnsRegistry(dict): def register(self, dtype, name=None, help=None): def decorator(func): class Col(query_compile.EvalColumn): def __init__(self): super().__init__(dtype) __call__ = staticmethod(func) Col.__name__ = name or func.__name__ Col.__doc__ = help or func.__doc__ self[Col.__name__] = Col() return func return decorator class EntriesTable(BeanTable): name = 'entries' columns = ColumnsRegistry() def __iter__(self): entries = self.prepare() yield from iter(entries) @columns.register(str) def id(entry): """Unique id of a directive.""" return hash_entry(entry) @columns.register(str) def type(entry): """The data type of the directive.""" return type(entry).__name__.lower() @columns.register(str) def filename(entry): """The filename where the directive was parsed from or created.""" return entry.meta["filename"] @columns.register(int) def lineno(entry): """The line number from the file the directive was parsed from.""" return entry.meta["lineno"] @columns.register(datetime.date) def date(entry): """The date of the directive.""" return entry.date @columns.register(int) def year(entry): """The year of the date year of the directive.""" return entry.date.year @columns.register(int) def month(entry): """The year of the date month of the directive.""" return entry.date.month @columns.register(int) def day(entry): """The year of the date day of the directive.""" return entry.date.day @columns.register(str) def flag(entry): """The flag the transaction.""" if not isinstance(entry, data.Transaction): return None return entry.flag @columns.register(str) def payee(entry): """The payee of the transaction.""" if not isinstance(entry, data.Transaction): return None return entry.payee @columns.register(str) def narration(entry): """The narration of the transaction.""" if not isinstance(entry, data.Transaction): return None return entry.narration @columns.register(str) def description(entry): """A combination of the payee + narration of the transaction, if present.""" if not isinstance(entry, data.Transaction): return None return ' | '.join(filter(None, [entry.payee, entry.narration])) @columns.register(set) def tags(entry): """The set of tags of the transaction.""" return getattr(entry, 'tags', None) @columns.register(set) def links(entry): """The set of links of the transaction.""" return getattr(entry, 'links', None) @columns.register(dict) def meta(entry): return entry.meta @columns.register(Set[str]) def accounts(entry): return getters.get_entry_accounts(entry) class PostingsTable(EntriesTable): name = 'postings' columns = ColumnsRegistry() wildcard_columns = 'date flag payee narration position'.split() def __iter__(self): entries = self.prepare() context = Row() for entry in entries: if isinstance(entry, data.Transaction): context.entry = entry for posting in entry.postings: context.rowid += 1 context.posting = posting yield context @columns.register(str, 'type') def type(context): return 'transaction' @columns.register(str, 'id') def id(context): """Unique id of a directive.""" return hash_entry(context.entry) @columns.register(datetime.date) def date(context): """The date of the directive.""" return context.entry.date @columns.register(int) def year(context): """The year of the date year of the directive.""" return context.entry.date.year @columns.register(int) def month(context): """The year of the date month of the directive.""" return context.entry.date.month @columns.register(int) def day(context): """The year of the date day of the directive.""" return context.entry.date.day @columns.register(str) def filename(context): """The ledger where the posting is defined.""" meta = context.posting.meta # Postings for pad transactions have their meta fields set to # None. See https://github.com/beancount/beancount/issues/767 if meta is None: return None return meta["filename"] @columns.register(int) def lineno(context): """The line number in the ledger file where the posting is defined.""" meta = context.posting.meta # Postings for pad transactions have their meta fields set to # None. See https://github.com/beancount/beancount/issues/767 if meta is None: return None return meta["lineno"] @columns.register(str) def location(context): """The filename:lineno location where the posting is defined.""" meta = context.posting.meta # Postings for pad transactions have their meta fields set to # None. See https://github.com/beancount/beancount/issues/767 if meta is None: return None return '{:s}:{:d}:'.format(meta['filename'], meta['lineno']) @columns.register(str) def flag(context): """The flag of the parent transaction for this posting.""" return context.entry.flag @columns.register(str) def payee(context): """The payee of the parent transaction for this posting.""" return context.entry.payee @columns.register(str) def narration(context): """The narration of the parent transaction for this posting.""" return context.entry.narration @columns.register(str) def description(context): "A combination of the payee + narration for the transaction of this posting." return ' | '.join(filter(None, [context.entry.payee, context.entry.narration])) @columns.register(set) def tags(context): "The set of tags of the parent transaction for this posting." return context.entry.tags @columns.register(set) def links(context): """The set of links of the parent transaction for this posting.""" return context.entry.links @columns.register(str) def posting_flag(context): """The flag of the posting itself.""" return context.posting.flag @columns.register(str, 'account') def account_(context): """The account of the posting.""" return context.posting.account @columns.register(set) def other_accounts(context): """The list of other accounts in the transaction, excluding that of this posting.""" return sorted({posting.account for posting in context.entry.postings if posting is not context.posting}) @columns.register(Decimal) def number(context): """The number of units of the posting.""" return context.posting.units.number @columns.register(str) def currency(context): """The currency of the posting.""" return context.posting.units.currency @columns.register(Decimal) def cost_number(context): """The number of cost units of the posting.""" cost = context.posting.cost return cost.number if cost else None @columns.register(str) def cost_currency(context): """The cost currency of the posting.""" cost = context.posting.cost return cost.currency if cost else None @columns.register(datetime.date) def cost_date(context): """The cost currency of the posting.""" cost = context.posting.cost return cost.date if cost else None @columns.register(str) def cost_label(context): """The cost currency of the posting.""" cost = context.posting.cost return cost.label if cost else '' @columns.register(position.Position, 'position') def position_(context): """The position for the posting. These can be summed into inventories.""" posting = context.posting return position.Position(posting.units, posting.cost) @columns.register(amount.Amount) def price(context): """The price attached to the posting.""" return context.posting.price @columns.register(amount.Amount) def weight(context): """The computed weight used for this posting.""" return convert.get_weight(context.posting) @columns.register(inventory.Inventory) @cache(maxsize=1) # noqa: B019 def balance(context): """The balance for the posting. These can be summed into inventories.""" # Caching protects against multiple balance updates per row when # the columns appears more than once in the execurted query. The # rowid in the row context guarantees that otherwise identical # rows do not hit the cache and thus that the balance is correctly # updated. context.balance.add_position(context.posting) return copy.copy(context.balance) @columns.register(dict) def meta(context): return context.posting.meta @columns.register(data.Transaction) def entry(context): return context.entry @columns.register(Set[str]) def accounts(context): return {p.account for p in context.entry.postings} beanquery-master/beanquery/query_env_test.py000066400000000000000000000147631474576771600220260ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import datetime import unittest from decimal import Decimal from beancount.core.number import D from beancount.parser import parser from beancount import loader from beanquery import query_compile as qc from beanquery import query_env as qe from beanquery import query class TestCompileDataTypes(unittest.TestCase): def test_compile_Length(self): c_length = qe.Function('length', [qc.EvalConstant('testing')]) self.assertEqual(int, c_length.dtype) def test_compile_Sum(self): c_sum = qe.SumInt(None, [qc.EvalConstant(17)]) self.assertEqual(int, c_sum.dtype) c_sum = qe.SumDecimal(None, [qc.EvalConstant(D('17.'))]) self.assertEqual(Decimal, c_sum.dtype) def test_compile_Count(self): c_count = qe.Count(None, [qc.EvalConstant(17)]) self.assertEqual(int, c_count.dtype) def test_compile_First(self): c_first = qe.First(None, [qc.EvalConstant(17.)]) self.assertEqual(float, c_first.dtype) def test_compile_Last(self): c_last = qe.Last(None, [qc.EvalConstant(17.)]) self.assertEqual(float, c_last.dtype) class TestEnv(unittest.TestCase): @parser.parse_doc() def test_GrepN(self, entries, _, options_map): """ 2016-11-20 * "prev match in context next" Assets:Banking 1 USD """ rtypes, rrows = query.run_query(entries, options_map, ''' SELECT GREPN("in", narration, 0) as m ''') self.assertEqual([('in',)], rrows) rtypes, rrows = query.run_query(entries, options_map, ''' SELECT GREPN("match (.*) context", narration, 1) as m ''') self.assertEqual([('in',)], rrows) rtypes, rrows = query.run_query(entries, options_map, ''' SELECT GREPN("(.*) in (.*)", narration, 2) as m ''') self.assertEqual([('context next',)], rrows) rtypes, rrows = query.run_query(entries, options_map, ''' SELECT GREPN("ab(at)hing", "abathing", 1) as m ''') self.assertEqual([('at',)], rrows) @parser.parse_doc() def test_Subst(self, entries, _, options_map): """ 2016-11-20 * "I love candy" Assets:Banking -1 USD 2016-11-21 * "Buy thing thing" Assets:Cash -1 USD """ rtypes, rrows = query.run_query(entries, options_map, ''' SELECT SUBST("[Cc]andy", "carrots", narration) as m where date = 2016-11-20 ''') self.assertEqual([('I love carrots',)], rrows) rtypes, rrows = query.run_query(entries, options_map, ''' SELECT SUBST("thing", "t", narration) as m where date = 2016-11-21 ''') self.assertEqual([('Buy t t',)], rrows) rtypes, rrows = query.run_query(entries, options_map, ''' SELECT SUBST("random", "t", narration) as m where date = 2016-11-21 ''') self.assertEqual([('Buy thing thing',)], rrows) rtypes, rrows = query.run_query(entries, options_map, ''' SELECT SUBST("(love)", "\\1 \\1", narration) as m where date = 2016-11-20 ''') self.assertEqual([('I love love candy',)], rrows) rtypes, rrows = query.run_query(entries, options_map, ''' SELECT SUBST("Assets:.*", "Savings", account) as a, str(sum(position)) as p ''') self.assertEqual([('Savings', '(-2 USD)')], rrows) @parser.parse_doc() def test_Date(self, entries, _, options_map): """ 2016-11-20 * "ok" Assets:Banking 1 USD """ rtypes, rrows = query.run_query(entries, options_map, 'SELECT date(2020, 1, 2) as m') self.assertEqual([(datetime.date(2020, 1, 2),)], rrows) rtypes, rrows = query.run_query(entries, options_map, 'SELECT date(year, month, 1) as m') self.assertEqual([(datetime.date(2016, 11, 1),)], rrows) rtypes, rrows = query.run_query(entries, options_map, 'SELECT date(2020, 2, 32) as m') self.assertEqual([(None,)], rrows) rtypes, rrows = query.run_query(entries, options_map, 'SELECT date("2020-01-02") as m') self.assertEqual([(datetime.date(2020, 1, 2),)], rrows) @parser.parse_doc() def test_DateDiffAdjust(self, entries, _, options_map): """ 2016-11-20 * "ok" Assets:Banking -1 STOCK { 5 USD, 2016-10-30 } """ rtypes, rrows = query.run_query(entries, options_map, 'SELECT date_diff(date, cost_date) as m') self.assertEqual([(21,)], rrows) rtypes, rrows = query.run_query(entries, options_map, 'SELECT date_diff(cost_date, date) as m') self.assertEqual([(-21,)], rrows) rtypes, rrows = query.run_query(entries, options_map, 'SELECT date_add(date, 1) as m') self.assertEqual([(datetime.date(2016, 11, 21),)], rrows) rtypes, rrows = query.run_query(entries, options_map, 'SELECT date_add(date, -1) as m') self.assertEqual([(datetime.date(2016, 11, 19),)], rrows) def test_func_meta(self): # use the loader to have the pad transaction inserted entries, _, options = loader.load_string(''' 2019-01-01 open Assets:Main USD 2019-01-01 open Assets:Other USD 2019-01-14 * "Test" entry: 1 both: 3 Assets:Main 100.00 USD post: 2 both: 3 Assets:Other post: 4 2019-01-15 pad Assets:Main Assets:Other 2019-01-16 balance Assets:Main 1000.00 USD ''', dedent=True) rtypes, rrows = query.run_query(entries, options, ''' SELECT entry_meta('post'), meta('post'), any_meta('post'), entry_meta('entry'), meta('entry'), any_meta('entry'), any_meta('both') ''') self.assertEqual([ (None, D(2), D(2), D(1), None, D(1), D(3)), (None, D(4), D(4), D(1), None, D(1), D(3)), # postings from pad directive (None, None, None, None, None, None, None), (None, None, None, None, None, None, None), ], rrows) if __name__ == '__main__': unittest.main() beanquery-master/beanquery/query_execute.py000066400000000000000000000225371474576771600216370ustar00rootroot00000000000000"""Execution of interpreter on data rows. """ __copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import collections import itertools import operator from . import compiler from . import hashable from . import query_compile from .cursor import Column class Unique: def __init__(self, columns): self.wrap = hashable.make(columns) def __call__(self, iterable): wrap = self.wrap seen = set() add = seen.add for obj in iterable: h = wrap(obj) if h not in seen: add(h) yield obj class Allocator: """A helper class to count slot allocations and return unique handles to them. """ def __init__(self): self.size = 0 def allocate(self): """Allocate a new slot to store row aggregation information. Returns: A unique handle used to index into an row-aggregation store (an integer). """ handle = self.size self.size += 1 return handle def create_store(self): """Create a new row-aggregation store suitable to contain all the node allocations. Returns: A store that can accommodate and be indexed by all the allocated slot handles. """ return [None] * self.size class NullType: """An object that compares smaller than anything. An instance of this class is used to replace None in BQL query results in sort keys to obtain sorting semantics similar to SQL where NULL is sortet at the beginning. """ __slots__ = () def __repr__(self): return 'NULL' __str__ = __repr__ def __lt__(self, other): # Make sure that instances of this class compare equal. if isinstance(other, NullType): return False return True def __gt__(self, other): # Make sure that instances of this class compare equal. if isinstance(other, NullType): return True return False NULL = NullType() def nullitemgetter(item, *items): """An itemgetter() that replaces None values with NULL.""" if items: items = (item, *items) def func(obj): r = [] for i in items: value = obj[i] r.append(value if value is not None else NULL) return tuple(r) return func def func(obj): value = obj[item] return value if value is not None else NULL return func def execute_query(query): """Given a compiled select statement, execute the query. Args: query: The query to execute. entries: A list of directives. options: A parser's option_map. Returns: A list of (name, dtype) tuples describing the results set table and a list of ResultRow tuples with the data.item pairs. """ if isinstance(query, query_compile.EvalQuery): return execute_select(query) if isinstance(query, query_compile.EvalPivot): columns, rows = execute_select(query.query) col1, col2 = query.pivots othercols = [i for i in range(len(columns)) if i not in query.pivots] nother = len(othercols) other = lambda x: tuple(x[i] for i in othercols) keys = sorted({row[col2] for row in rows}) # Compute the new column names and dtypes. if nother > 1: it = itertools.product(keys, other(columns)) names = [f'{columns[col1].name}/{columns[col2].name}'] + [f'{key}/{col.name}' for key, col in it] else: names = [f'{columns[col1].name}/{columns[col2].name}'] + [f'{key}' for key in keys] datatypes = [columns[col1].datatype] + [col.datatype for col in other(columns)] * len(keys) columns = tuple(Column(name, datatype) for name, datatype in zip(names, datatypes)) # Populate the pivoted table. pivoted = [] rows.sort(key=operator.itemgetter(col1)) for field1, group in itertools.groupby(rows, key=operator.itemgetter(col1)): outrow = [field1] + [None] * (len(columns) - 1) for row in group: index = keys.index(row[col2]) * nother + 1 outrow[index:index+nother] = other(row) pivoted.append(tuple(outrow)) return columns, pivoted # Not reached. raise RuntimeError def execute_select(query): """Given a compiled select statement, execute the query. Args: query: An instance of a query_compile.Query entries: A list of directives. options_map: A parser's option_map. Returns: A pair of: result_types: A list of (name, data-type) item pairs. result_rows: A list of ResultRow tuples of length and types described by 'result_types'. """ # Figure out the result types that describe what we return. result_types = tuple(Column(target.name, target.c_expr.dtype) for target in query.c_targets if target.name is not None) # Pre-compute lists of the expressions to evaluate. group_indexes = (set(query.group_indexes) if query.group_indexes is not None else query.group_indexes) # Indexes of the columns for result rows and order rows. result_indexes = [index for index, c_target in enumerate(query.c_targets) if c_target.name] order_spec = query.order_spec # Dispatch between the non-aggregated queries and aggregated queries. c_where = query.c_where rows = [] # Precompute a list of expressions to be evaluated. c_target_exprs = [c_target.c_expr for c_target in query.c_targets] if query.group_indexes is None: # This is a non-aggregated query. # Iterate over all the postings once. for context in query.table: if c_where is None or c_where(context): values = [c_expr(context) for c_expr in c_target_exprs] rows.append(values) else: # This is an aggregated query. # Precompute lists of non-aggregate and aggregate expressions to # evaluate. For aggregate targets, we hunt down the aggregate # sub-expressions to evaluate, to avoid recursion during iteration. c_nonaggregate_exprs = [] c_aggregate_exprs = [] for index, c_expr in enumerate(c_target_exprs): if index in group_indexes: c_nonaggregate_exprs.append(c_expr) else: _, aggregate_exprs = compiler.get_columns_and_aggregates(c_expr) c_aggregate_exprs.extend(aggregate_exprs) # Note: it is possible that there are no aggregates to compute here. You could # have all columns be non-aggregates and group-by the entire list of columns. # Pre-allocate handles in aggregation nodes. allocator = Allocator() for c_expr in c_aggregate_exprs: c_expr.allocate(allocator) def create(): # Create a new row in the aggregates store. store = allocator.create_store() for c_expr in c_aggregate_exprs: c_expr.initialize(store) return store context = None aggregates = collections.defaultdict(create) # Iterate over all the postings to evaluate the aggregates. for context in query.table: if c_where is None or c_where(context): # Compute the non-aggregate expressions. key = tuple(c_expr(context) for c_expr in c_nonaggregate_exprs) # Get an appropriate store for the unique key of this row. store = aggregates[key] # Update the aggregate expressions. for c_expr in c_aggregate_exprs: c_expr.update(store, context) # Iterate over all the aggregations. for key, store in aggregates.items(): key_iter = iter(key) values = [] # Finalize the store. for c_expr in c_aggregate_exprs: c_expr.finalize(store) for index, c_expr in enumerate(c_target_exprs): if index in group_indexes: value = next(key_iter) else: value = c_expr(context) values.append(value) # Skip row if HAVING clause expression is false. if query.having_index is not None: if not values[query.having_index]: continue rows.append(values) # Apply ORDER BY. if order_spec is not None: # Process the order-by clauses grouped by their ordering direction. for reverse, spec in itertools.groupby(reversed(order_spec), key=operator.itemgetter(1)): indexes = reversed([i[0] for i in spec]) # The rows may contain None values: nullitemgetter() # replaces these with a special value that compares # smaller than anything else. rows.sort(key=nullitemgetter(*indexes), reverse=reverse) # Extract results set and convert into tuples. rows = (tuple(row[i] for i in result_indexes) for row in rows) # Apply DISTINCT. if query.distinct: unique = Unique(result_types) rows = unique(rows) # Apply LIMIT. if query.limit is not None: rows = itertools.islice(rows, query.limit) return result_types, list(rows) beanquery-master/beanquery/query_execute_test.py000066400000000000000000001707721474576771600227030ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2014-2017 Martin Blais" __license__ = "GNU GPLv2" import datetime import unittest import textwrap from decimal import Decimal from dateutil.relativedelta import relativedelta from beancount.core.amount import from_string as A from beancount.core.number import D from beancount.core import inventory from beancount.core.inventory import from_string as I from beancount.parser import cmptest from beancount import loader import beanquery from beanquery import CompilationError from beanquery import query_compile as qc from beanquery import query_execute as qx from beanquery import tables from beanquery import compiler class QueryBase(cmptest.TestCase): INPUT = "" maxDiff = 8192 def setUp(self): entries, errors, options = loader.load_string(textwrap.dedent(self.INPUT)) self.ctx = beanquery.connect('beancount:', entries=entries, errors=errors, options=options) def compile(self, query): return self.ctx.compile(self.ctx.parse(query)) def check_query(self, input_string, query, expected_types, expected_rows): entries, errors, options = loader.load_string(textwrap.dedent(input_string)) ctx = beanquery.connect('beancount:', entries=entries, errors=errors, options=options) curs = ctx.execute(query) self.assertEqual(tuple(expected_types), curs.description) result_rows = curs.fetchall() self.assertEqual(expected_rows, result_rows) def assertResult(self, query, result, datatype=None): curs = self.ctx.execute(query) self.assertEqual([c.datatype for c in curs.description], [datatype or type(result)]) self.assertEqual(curs.fetchall(), [(result, )]) def assertError(self, query): with self.assertRaises(CompilationError): self.ctx.execute(query) class CommonInputBase(unittest.TestCase): INPUT = textwrap.dedent(""" 2010-01-01 open Assets:Bank:Checking 2010-01-01 open Assets:ForeignBank:Checking 2010-01-01 open Assets:Bank:Savings 2010-01-01 open Expenses:Restaurant 2010-01-01 * "Dinner with Cero" Assets:Bank:Checking 100.00 USD Expenses:Restaurant -100.00 USD 2011-01-01 * "Dinner with Uno" Assets:Bank:Checking 101.00 USD Expenses:Restaurant -101.00 USD 2012-02-02 * "Dinner with Dos" Assets:Bank:Checking 102.00 USD Expenses:Restaurant -102.00 USD 2013-03-03 * "Dinner with Tres" Assets:Bank:Checking 103.00 USD Expenses:Restaurant -103.00 USD 2013-10-10 * "International Transfer" Assets:Bank:Checking -50.00 USD Assets:ForeignBank:Checking -60.00 CAD @ 1.20 USD 2014-04-04 * "Dinner with Quatro" Assets:Bank:Checking 104.00 USD Expenses:Restaurant -104.00 USD """) class TestFundamentals(QueryBase): INPUT = textwrap.dedent(""" 2022-04-05 commodity TEST rate: 42 2022-04-05 open Assets:Tests 2022-04-05 * "Test" Assets:Tests 1.000 TEST int: 1 decimal: 1.2 bool: TRUE str: "str" str3: "3" str4: "4.0" date: 2022-04-05 null: NULL """) def test_type_casting(self): # bool self.assertResult("SELECT bool(TRUE)", True) self.assertResult("SELECT bool(1)", True) self.assertResult("SELECT bool(1.1)", True) self.assertResult("SELECT bool('foo')", True) self.assertResult("SELECT bool(NULL)", None, bool) self.assertResult("SELECT bool(2022-04-05)", True) self.assertResult("SELECT bool(meta('int'))", True) self.assertResult("SELECT bool(meta('decimal'))", True) self.assertResult("SELECT bool(meta('bool'))", True) self.assertResult("SELECT bool(meta('str'))", True) self.assertResult("SELECT bool(meta('date'))", True) self.assertResult("SELECT bool(meta('null'))", None, bool) self.assertResult("SELECT bool(meta('missing'))", None, bool) # int self.assertResult("SELECT int(TRUE)", 1) self.assertResult("SELECT int(1)", 1) self.assertResult("SELECT int(1.2)", 1) self.assertResult("SELECT int('1')", 1) self.assertResult("SELECT int('foo')", None, int) self.assertResult("SELECT int(NULL)", None, int) self.assertError ("SELECT int(2022-04-05)") self.assertResult("SELECT int(meta('int'))", 1) self.assertResult("SELECT int(meta('decimal'))", 1) self.assertResult("SELECT int(meta('bool'))", 1) self.assertResult("SELECT int(meta('str'))", None, int) self.assertResult("SELECT int(meta('str3'))", 3) self.assertResult("SELECT int(meta('str4'))", None, int) self.assertResult("SELECT int(meta('date'))", None, int) self.assertResult("SELECT int(meta('null'))", None, int) self.assertResult("SELECT int(meta('missing'))", None, int) # decimal self.assertResult("SELECT decimal(TRUE)", Decimal(1)) self.assertResult("SELECT decimal(1)", Decimal(1)) self.assertResult("SELECT decimal(1.2)", Decimal('1.2')) self.assertResult("SELECT decimal('1.2')", Decimal('1.2')) self.assertResult("SELECT decimal('foo')", None, Decimal) self.assertResult("SELECT decimal(NULL)", None, Decimal) self.assertError ("SELECT decimal(2022-04-05)") self.assertResult("SELECT decimal(meta('int'))", Decimal(1)) self.assertResult("SELECT decimal(meta('decimal'))", Decimal('1.2')) self.assertResult("SELECT decimal(meta('bool'))", Decimal(1)) self.assertResult("SELECT decimal(meta('str'))", None, Decimal) self.assertResult("SELECT decimal(meta('str3'))", Decimal(3)) self.assertResult("SELECT decimal(meta('str4'))", Decimal('4.0')) self.assertResult("SELECT decimal(meta('date'))", None, Decimal) self.assertResult("SELECT decimal(meta('null'))", None, Decimal) self.assertResult("SELECT decimal(meta('missing'))", None, Decimal) # str self.assertResult("SELECT str(TRUE)", 'TRUE') self.assertResult("SELECT str(FALSE)", 'FALSE') self.assertResult("SELECT str(1)", '1') self.assertResult("SELECT str(1.1)", '1.1') self.assertResult("SELECT str('foo')", 'foo') self.assertResult("SELECT str(NULL)", None, str) self.assertResult("SELECT str(2022-04-05)", '2022-04-05') self.assertResult("SELECT str(meta('int'))", '1') self.assertResult("SELECT str(meta('decimal'))", '1.2') self.assertResult("SELECT str(meta('bool'))", 'TRUE') self.assertResult("SELECT str(meta('str'))", 'str') self.assertResult("SELECT str(meta('date'))", '2022-04-05') self.assertResult("SELECT str(meta('null'))", None, str) self.assertResult("SELECT str(meta('missing'))", None, str) # date self.assertError ("SELECT date(TRUE)") self.assertError ("SELECT date(1)") self.assertError ("SELECT date(1.2)") self.assertResult("SELECT date('1.2')", None, datetime.date) self.assertResult("SELECT date('foo')", None, datetime.date) self.assertResult("SELECT date('2022-04-05')", datetime.date(2022, 4, 5)) self.assertResult("SELECT date(NULL)", None, datetime.date) self.assertResult("SELECT date(2022-04-05)", datetime.date(2022, 4, 5)) self.assertResult("SELECT date(2022, 4, 5)", datetime.date(2022, 4, 5)) self.assertResult("SELECT date(meta('int'))", None, datetime.date) self.assertResult("SELECT date(meta('decimal'))", None, datetime.date) self.assertResult("SELECT date(meta('bool'))", None, datetime.date) self.assertResult("SELECT date(meta('str'))", None, datetime.date) self.assertResult("SELECT date(meta('date'))", datetime.date(2022, 4, 5)) self.assertResult("SELECT date(meta('null'))", None, datetime.date) self.assertResult("SELECT date(meta('missing'))", None, datetime.date) def test_operators(self): # add self.assertResult("SELECT 1 + 1", 2) self.assertResult("SELECT 1.0 + 1", Decimal(2)) self.assertResult("SELECT 1.0 + 2.00", Decimal(3)) self.assertError ("SELECT 1970-01-01 + 2022-04-01") self.assertResult("SELECT 2022-04-01 + 1", datetime.date(2022, 4, 2)) self.assertResult("SELECT 1 + 2022-04-01", datetime.date(2022, 4, 2)) self.assertResult("SELECT 'ab' + 'c'", 'abc') # sub self.assertResult("SELECT 1 - 1", 0) self.assertResult("SELECT 1.0 - 1", Decimal(0)) self.assertResult("SELECT 1.0 - 2.00", Decimal(-1)) self.assertResult("SELECT 2022-04-01 - 1", datetime.date(2022, 3, 31)) self.assertResult("SELECT 2022-04-01 - 2022-03-31", 1) self.assertError ("SELECT 1 - 2022-04-01") # mul self.assertResult("SELECT 2 * 2", 4) self.assertResult("SELECT 2.0 * 2", Decimal(4)) self.assertResult("SELECT 2 * 2.0", Decimal(4)) self.assertResult("SELECT 2.0 * 2.0", Decimal(4)) # div self.assertResult("SELECT 4 / 2", Decimal(2)) self.assertResult("SELECT 4.0 / 2", Decimal(2)) self.assertResult("SELECT 4 / 2.0", Decimal(2)) self.assertResult("SELECT 4.0 / 2.0", Decimal(2)) self.assertResult("SELECT 4 / 0", None, Decimal) self.assertResult("SELECT 4.0 / 0", None, Decimal) self.assertResult("SELECT 4 / 0.0", None, Decimal) # mod self.assertResult("SELECT 3 % 2", 1) self.assertResult("SELECT 3.0 % 2", Decimal(1)) self.assertResult("SELECT 3 % 2.0", Decimal(1)) self.assertResult("SELECT 3.0 % 2.0", Decimal(1)) # match self.assertResult("SELECT 'foobarbaz' ~ 'bar'", True) self.assertResult("SELECT 'foobarbaz' ~ 'quz'", False) # mot match self.assertResult("SELECT 'foobarbaz' !~ 'bar'", False) self.assertResult("SELECT 'foobarbaz' !~ 'quz'", True) # matches self.assertResult("SELECT '[0-9]+' ?~ '123'", True) self.assertResult("SELECT '[0-9]+' ?~ 'ABC'", False) # and self.assertResult("SELECT 1 and FALSE", False) self.assertResult("SELECT 'something' and FALSE", False) self.assertResult("SELECT 1.0 and FALSE", False) self.assertResult("SELECT TRUE and meta('missing')", None, bool) self.assertResult("SELECT TRUE and not meta('missing')", True) # or self.assertResult("SELECT FALSE or 1", True) self.assertResult("SELECT FALSE or 'something'", True) self.assertResult("SELECT FALSE or 1.0", True) self.assertResult("SELECT TRUE or meta('missing')", True) self.assertResult("SELECT FALSE or not meta('missing')", True) # not self.assertResult("SELECT not TRUE", False) self.assertResult("SELECT not meta('missing')", True) # is null self.assertResult("SELECT date IS NULL", False) self.assertResult("SELECT meta('missing') IS NULL", True) self.assertResult("SELECT meta('int') IS NULL", False) # is not null self.assertResult("SELECT date IS NOT NULL", True) self.assertResult("SELECT meta('missing') IS NOT NULL", False) self.assertResult("SELECT meta('int') IS NOT NULL", True) # in self.assertResult("SELECT 'tag' IN tags", False) self.assertResult("SELECT 3 IN (2, 3, 4)", True) self.assertResult("SELECT 'x' IN ('a', 'b', 'c')", False) self.assertResult("SELECT 3 NOT IN (2, 3, 4)", False) self.assertResult("SELECT 1 NOT IN (2, 3, 4)", True) # between self.assertResult("SELECT 2 BETWEEN 1 AND 3", True) self.assertResult("SELECT 1 BETWEEN 2 AND 3", False) self.assertResult("SELECT 2.0 BETWEEN 1.0 AND 3.0", True) self.assertResult("SELECT 1.0 BETWEEN 2 AND 3.0", False) self.assertResult("SELECT 2023-08-02 BETWEEN 2023-08-01 AND 2023-08-03", True) self.assertResult("SELECT 'b' BETWEEN 'a' AND 'c'", True) self.assertError ("SELECT 2023-08-02 BETWEEN 'a' AND 2") def test_operators_type_inference(self): self.assertResult("SELECT 1 + meta('int')", Decimal(2)) self.assertResult("SELECT 1 + meta('str3')", Decimal(4)) self.assertResult("SELECT meta('int') > 0", True) def test_functions(self): # round self.assertResult("SELECT round(1.2)", Decimal(1)) self.assertResult("SELECT round(1.234, 2)", Decimal('1.23')) self.assertResult("SELECT round(12)", 12) self.assertResult("SELECT round(12, -1)", 10) # neg self.assertResult("SELECT neg(1.0)", Decimal('-1.0')) # abs self.assertResult("SELECT abs(-1.0)", Decimal('1.0')) # safediv self.assertResult("SELECT safediv(1.0, 2.0)", Decimal('0.5')) self.assertResult("SELECT safediv(1.0, 0.0)", Decimal('0')) self.assertResult("SELECT safediv(0.0, 0.0)", Decimal('0')) # repr self.assertResult("SELECT repr(1.0)", 'Decimal(\'1.0\')') # lower self.assertResult('''SELECT lower('AaBbCcDdEe')''', 'aabbccddee') # upper self.assertResult('''SELECT upper('AaBbCcDdEe')''', 'AABBCCDDEE') def test_coalesce(self): # coalesce self.assertResult("SELECT COALESCE(str(meta('missing')), '!')", "!") self.assertError ("SELECT COALESCE(meta('missing'), '!')") def test_count(self): # count number of postings self.assertResult("SELECT COUNT(*)", 1) self.assertResult("SELECT COUNT(1)", 1) self.assertResult("SELECT COUNT(NULL)", 0) self.assertResult("SELECT COUNT(meta('missing'))", 0) def test_getitem(self): self.assertResult("SELECT meta['lineno']", 6, object) self.assertResult("SELECT 1 + meta['lineno']", Decimal('7')) self.assertResult("SELECT meta['filename']", '', object) self.assertResult("SELECT meta['missing']", None, object) self.assertError ("SELECT position['foo']") def test_getatrr(self): self.assertResult("SELECT entry.flag", '*') self.assertResult("SELECT int(entry.meta['lineno'])", 5) self.assertError ("SELECT entry.foo") self.assertError ("SELECT position.foo") def test_parse_date(self): self.assertResult('''SELECT parse_date('2016/11/1')''', datetime.date(2016, 11, 1)) self.assertResult('''SELECT parse_date('2016/11/1', '%Y/%d/%m')''', datetime.date(2016, 1, 11)) def test_date_part(self): self.assertResult('''SELECT date_part('weekday', 2024-06-09)''', 6) self.assertResult('''SELECT date_part('dow', 2024-06-09)''', 6) self.assertResult('''SELECT date_part('isoweekday', 2024-06-09)''', 7) self.assertResult('''SELECT date_part('isodow', 2024-06-09)''', 7) self.assertResult('''SELECT date_part('week', 2024-06-09)''', 23) self.assertResult('''SELECT date_part('month', 2024-06-09)''', 6) self.assertResult('''SELECT date_part('quarter', 2024-06-09)''', 2) self.assertResult('''SELECT date_part('year', 2024-06-09)''', 2024) self.assertResult('''SELECT date_part('isoyear', 2024-06-09)''', 2024) self.assertResult('''SELECT date_part('decade', 2024-06-09)''', 202) self.assertResult('''SELECT date_part('century', 2024-06-09)''', 21) self.assertResult('''SELECT date_part('millennium', 2024-06-09)''', 3) self.assertResult('''SELECT date_part('epoch', 2024-06-09)''', 1717891200) self.assertResult('''SELECT date_part('baz', 2024-06-09)''', None, int) def test_date_trunc(self): self.assertResult('''SELECT date_trunc('week', 2016-11-14)''', datetime.date(2016, 11, 14)) # monday self.assertResult('''SELECT date_trunc('week', 2016-11-15)''', datetime.date(2016, 11, 14)) # tuesday self.assertResult('''SELECT date_trunc('week', 2016-11-20)''', datetime.date(2016, 11, 14)) # sunday self.assertResult('''SELECT date_trunc('month', 2016-11-20)''', datetime.date(2016, 11, 1)) self.assertResult('''SELECT date_trunc('quarter', 2016-09-30)''', datetime.date(2016, 7, 1)) self.assertResult('''SELECT date_trunc('quarter', 2016-10-01)''', datetime.date(2016, 10, 1)) self.assertResult('''SELECT date_trunc('quarter', 2016-11-20)''', datetime.date(2016, 10, 1)) self.assertResult('''SELECT date_trunc('year', 2016-11-20)''', datetime.date(2016, 1, 1)) self.assertResult('''SELECT date_trunc('decade', 2016-11-20)''', datetime.date(2010, 1, 1)) self.assertResult('''SELECT date_trunc('decade', 2020-11-20)''', datetime.date(2020, 1, 1)) self.assertResult('''SELECT date_trunc('decade', 2029-11-20)''', datetime.date(2020, 1, 1)) self.assertResult('''SELECT date_trunc('century', 1999-11-20)''', datetime.date(1901, 1, 1)) self.assertResult('''SELECT date_trunc('century', 2000-11-20)''', datetime.date(1901, 1, 1)) self.assertResult('''SELECT date_trunc('century', 2001-11-20)''', datetime.date(2001, 1, 1)) self.assertResult('''SELECT date_trunc('century', 2016-11-20)''', datetime.date(2001, 1, 1)) self.assertResult('''SELECT date_trunc('millennium', 1991-11-20)''', datetime.date(1001, 1, 1)) self.assertResult('''SELECT date_trunc('millennium', 2000-11-20)''', datetime.date(1001, 1, 1)) self.assertResult('''SELECT date_trunc('millennium', 2001-11-20)''', datetime.date(2001, 1, 1)) self.assertResult('''SELECT date_trunc('millennium', 2016-11-20)''', datetime.date(2001, 1, 1)) self.assertResult('''SELECT date_trunc('millennium', 3456-11-20)''', datetime.date(3001, 1, 1)) self.assertResult('''SELECT date_trunc('foo', 2024-05-24)''', None, datetime.date) def test_interval_ops(self): self.assertResult('''SELECT 2016-11-20 + interval("1 months")''', datetime.date(2016, 12, 20)) self.assertResult('''SELECT 2016-11-01 + interval("1 months") - interval("1 days")''', datetime.date(2016, 11, 30)) self.assertResult('''SELECT 2024-02-01 + interval("1 month") + interval("-1 day")''', datetime.date(2024, 2, 29)) self.assertResult('''SELECT 2024-02-05 + interval("-1 days")''', datetime.date(2024, 2, 4)) self.assertResult('''SELECT 2024-02-05 + interval("1 day") + interval("2 days")''', datetime.date(2024, 2, 8)) self.assertResult('''SELECT 2024-05-24 + interval("1 year")''', datetime.date(2025, 5, 24)) self.assertResult('''SELECT 2024-05-24 + interval("-2 years")''', datetime.date(2022, 5, 24)) self.assertResult('''SELECT interval("1 baz")''', None, relativedelta) self.assertResult('''SELECT interval("A days")''', None, relativedelta) def test_date_bin(self): self.assertResult('''SELECT date_bin(interval('1 year'), 2024-11-10, 2024-06-01)''', datetime.date(2024, 6, 1)) self.assertResult('''SELECT date_bin('1 year', 2024-11-10, 2024-06-01)''', datetime.date(2024, 6, 1)) self.assertResult('''SELECT date_bin('1 year', 2024-11-10, 2025-06-01)''', datetime.date(2024, 6, 1)) self.assertResult('''SELECT date_bin('1 month', 2024-11-10, 2024-06-03)''', datetime.date(2024, 11, 3)) self.assertResult('''SELECT date_bin('3 days', 2024-11-10, 2024-11-02)''', datetime.date(2024, 11, 8)) self.assertResult('''SELECT date_bin('3 days', 2024-11-10, 2024-11-14)''', datetime.date(2024, 11, 8)) class TestBeancountFunctions(QueryBase): INPUT = textwrap.dedent(""" 2024-06-23 commodity TEST answer: "42" 2024-06-23 open Assets:Tests key: "value" 2024-06-23 open Expenses:Tests 2024-06-23 price USD 2.0 TEST 2024-06-23 * "Test" Assets:Tests 10.000 TEST Expenses:Tests 2024-06-24 close Expenses:Tests """) def test_root(self): self.assertResult('''SELECT root('Assets:Foo:Bar', 2) FROM #''', 'Assets:Foo') self.assertResult('''SELECT root('Assets:Foo:Bar', 1) FROM #''', 'Assets') self.assertResult('''SELECT root('Assets:Foo:Bar') FROM #''', 'Assets') def test_open_date(self): self.assertResult('''SELECT open_date('Expenses:Tests') FROM #''', datetime.date(2024, 6, 23)) self.assertResult('''SELECT open_date('Expenses:Foo') FROM #''', None, datetime.date) def test_close_date(self): self.assertResult('''SELECT close_date('Expenses:Tests') FROM #''', datetime.date(2024, 6, 24)) self.assertResult('''SELECT close_date('Expenses:Foo') FROM #''', None, datetime.date) self.assertResult('''SELECT close_date('Assets:Tests') FROM #''', None, datetime.date) def test_open_meta(self): self.assertResult('''SELECT open_meta('Assets:Tests') FROM #''', {'filename': '', 'lineno': 4, 'key': 'value'}) self.assertResult('''SELECT open_meta('Assets:Tests', 'key') FROM #''', 'value', object) self.assertResult('''SELECT open_meta('Expenses:Tests', 'baz') FROM #''', None, object) self.assertResult('''SELECT open_meta('Expenses:Foo', 'key') FROM #''', None, object) def test_commodity_meta(self): self.assertResult('''SELECT commodity_meta('TEST', 'answer') FROM #''', '42', object) self.assertResult('''SELECT commodity_meta('TEST') FROM #''', {'filename': '', 'lineno': 2, 'answer': '42'}) self.assertResult('''SELECT commodity_meta('X') FROM #''', None, dict) def test_account_sortkey(self): self.assertResult('''SELECT account_sortkey('Assets:Foo') LIMIT 1''', '0-Assets:Foo') def test_has_account(self): self.assertResult('''SELECT has_account('Assets:Tests') LIMIT 1''', True) self.assertResult('''SELECT has_account('Assets:Foo') LIMIT 1''', False) def test_convert(self): self.assertResult('''SELECT convert(position, 'USD') LIMIT 1''', A('5.0 USD')) self.assertResult('''SELECT convert(sum(position), 'USD')''', I('0.0 USD')) self.assertResult('''SELECT convert(only('TEST', sum(position)), 'USD')''', A('0.0 USD')) class TestBeancountMetadataFunctions(QueryBase): INPUT = """ 2016-11-20 * "Test" name: "The Name" address: "1 Wrong Way" empty: "Not Empty" Assets:Banking 1 USD color: "Green" address: "1 Right Way" empty: """ def test_meta(self): self.assertResult('''SELECT meta('address')''', '1 Right Way', object) self.assertResult('''SELECT meta('baz')''', None, object) self.assertResult('''SELECT meta('color')''', 'Green', object) self.assertResult('''SELECT meta('empty')''', None, object) self.assertResult('''SELECT meta('name')''', None, object) def test_entry_meta(self): self.assertResult('''SELECT entry_meta('address')''', '1 Wrong Way', object) self.assertResult('''SELECT entry_meta('baz')''', None, object) self.assertResult('''SELECT entry_meta('color')''', None, object) self.assertResult('''SELECT entry_meta('empty')''', 'Not Empty', object) self.assertResult('''SELECT entry_meta('name')''', 'The Name', object) def test_any_meta(self): self.assertResult('''SELECT any_meta('address')''', '1 Right Way', object) self.assertResult('''SELECT any_meta('baz')''', None, object) self.assertResult('''SELECT any_meta('color')''', 'Green', object) self.assertResult('''SELECT any_meta('empty')''', None, object) self.assertResult('''SELECT any_meta('name')''', 'The Name', object) class TestFilterEntries(CommonInputBase, QueryBase): def compile(self, query): # use the ``entries`` table as default table for queries c = compiler.Compiler(self.ctx) c.table = self.ctx.tables.get('entries') return c.compile(self.ctx.parse(query)) @staticmethod def filter_entries(query): entries = [] expr = query.c_where for entry in query.table: if expr is None or expr(entry): entries.append(entry) return entries def test_filter_empty_from(self): # Check that no filter outputs the very same thing. filtered_entries = self.filter_entries(self.compile(""" SELECT * ; """)) self.assertEqualEntries(self.INPUT, filtered_entries) def test_filter_by_year(self): filtered_entries = self.filter_entries(self.compile(""" SELECT date, type FROM year(date) = 2012; """)) self.assertEqualEntries(""" 2012-02-02 * "Dinner with Dos" Assets:Bank:Checking 102.00 USD Expenses:Restaurant -102.00 USD """, filtered_entries) def test_filter_by_expr1(self): filtered_entries = self.filter_entries(self.compile(""" SELECT date, type FROM NOT (type = 'transaction' AND (year(date) = 2012 OR year(date) = 2013)); """)) self.assertEqualEntries(""" 2010-01-01 open Assets:Bank:Checking 2010-01-01 open Assets:Bank:Savings 2010-01-01 open Expenses:Restaurant 2010-01-01 open Assets:ForeignBank:Checking 2010-01-01 * "Dinner with Cero" Assets:Bank:Checking 100.00 USD Expenses:Restaurant -100.00 USD 2011-01-01 * "Dinner with Uno" Assets:Bank:Checking 101.00 USD Expenses:Restaurant -101.00 USD 2014-04-04 * "Dinner with Quatro" Assets:Bank:Checking 104.00 USD Expenses:Restaurant -104.00 USD """, filtered_entries) def test_filter_by_expr2(self): filtered_entries = self.filter_entries(self.compile(""" SELECT date, type FROM date < 2012-06-01; """)) self.assertEqualEntries(""" 2010-01-01 open Assets:Bank:Checking 2010-01-01 open Assets:Bank:Savings 2010-01-01 open Expenses:Restaurant 2010-01-01 open Assets:ForeignBank:Checking 2010-01-01 * "Dinner with Cero" Assets:Bank:Checking 100.00 USD Expenses:Restaurant -100.00 USD 2011-01-01 * "Dinner with Uno" Assets:Bank:Checking 101.00 USD Expenses:Restaurant -101.00 USD 2012-02-02 * "Dinner with Dos" Assets:Bank:Checking 102.00 USD Expenses:Restaurant -102.00 USD """, filtered_entries) def test_filter_close_undated(self): filtered_entries = self.filter_entries(self.compile(""" SELECT date, type FROM CLOSE; """)) self.assertEqualEntries(self.INPUT + textwrap.dedent(""" 2014-04-04 C "Conversion for (-50.00 USD, -60.00 CAD)" Equity:Conversions:Current 50.00 USD @ 0 NOTHING Equity:Conversions:Current 60.00 CAD @ 0 NOTHING """), filtered_entries) def test_filter_close_dated(self): filtered_entries = self.filter_entries(self.compile(""" SELECT date, type FROM CLOSE ON 2013-06-01; """)) self.assertEqualEntries(""" 2010-01-01 open Assets:Bank:Checking 2010-01-01 open Assets:ForeignBank:Checking 2010-01-01 open Assets:Bank:Savings 2010-01-01 open Expenses:Restaurant 2010-01-01 * "Dinner with Cero" Assets:Bank:Checking 100.00 USD Expenses:Restaurant -100.00 USD 2011-01-01 * "Dinner with Uno" Assets:Bank:Checking 101.00 USD Expenses:Restaurant -101.00 USD 2012-02-02 * "Dinner with Dos" Assets:Bank:Checking 102.00 USD Expenses:Restaurant -102.00 USD 2013-03-03 * "Dinner with Tres" Assets:Bank:Checking 103.00 USD Expenses:Restaurant -103.00 USD """, filtered_entries) def test_filter_open_dated(self): filtered_entries = self.filter_entries(self.compile(""" SELECT date, type FROM OPEN ON 2013-01-01; """)) self.assertEqualEntries(""" 2010-01-01 open Assets:Bank:Checking 2010-01-01 open Assets:Bank:Savings 2010-01-01 open Expenses:Restaurant 2010-01-01 open Assets:ForeignBank:Checking 2012-12-31 S "Opening balance for 'Assets:Bank:Checking' (Summarization)" Assets:Bank:Checking 303.00 USD Equity:Opening-Balances -303.00 USD 2012-12-31 S "Opening balance for 'Equity:Earnings:Previous' (Summarization)" Equity:Earnings:Previous -303.00 USD Equity:Opening-Balances 303.00 USD 2013-03-03 * "Dinner with Tres" Assets:Bank:Checking 103.00 USD Expenses:Restaurant -103.00 USD 2013-10-10 * "International Transfer" Assets:Bank:Checking -50.00 USD ; -50.00 USD Assets:ForeignBank:Checking -60.00 CAD @ 1.20 USD ; -72.00 USD 2014-04-04 * "Dinner with Quatro" Assets:Bank:Checking 104.00 USD Expenses:Restaurant -104.00 USD """, filtered_entries) def test_filter_clear(self): filtered_entries = self.filter_entries(self.compile(""" SELECT date, type FROM CLEAR; """)) self.assertEqualEntries(self.INPUT + textwrap.dedent(""" 2014-04-04 T "Transfer balance for 'Expenses:Restaurant' (Transfer balance)" Expenses:Restaurant 510.00 USD Equity:Earnings:Current -510.00 USD """), filtered_entries) class TestAllocation(unittest.TestCase): def test_allocator(self): allocator = qx.Allocator() self.assertEqual(0, allocator.allocate()) self.assertEqual(1, allocator.allocate()) self.assertEqual(2, allocator.allocate()) self.assertEqual([None, None, None], allocator.create_store()) class TestExecuteNonAggregatedQuery(QueryBase): INPUT = """ 2010-01-01 open Assets:Bank:Checking 2010-01-01 open Expenses:Restaurant 2010-02-23 * "Bla" Assets:Bank:Checking 100.00 USD Expenses:Restaurant -100.00 USD """ def test_non_aggregate__one(self): self.check_query( self.INPUT, """ SELECT date; """, [('date', datetime.date)], [(datetime.date(2010, 2, 23),), (datetime.date(2010, 2, 23),)]) def test_non_aggregate__many(self): self.check_query( self.INPUT, """ SELECT date, flag, payee, narration; """, [ ('date', datetime.date), ('flag', str), ('payee', str), ('narration', str), ], [ (datetime.date(2010, 2, 23), '*', None, 'Bla'), (datetime.date(2010, 2, 23), '*', None, 'Bla'), ]) def test_non_aggregated_order_by_visible(self): self.check_query( self.INPUT, """ SELECT account, length(account) ORDER BY 2; """, [ ('account', str), ('length(account)', int), ], [ ('Expenses:Restaurant', 19), ('Assets:Bank:Checking', 20), ]) def test_non_aggregated_order_by_invisible(self): self.check_query( self.INPUT, """ SELECT account ORDER BY length(account); """, [ ('account', str), ], [ ('Expenses:Restaurant',), ('Assets:Bank:Checking',), ]) def test_non_aggregated_order_by_none_date(self): self.check_query( self.INPUT, """ SELECT account ORDER BY cost_date; """, [ ('account', str), ], [ ('Assets:Bank:Checking',), ('Expenses:Restaurant',), ]) def test_non_aggregated_order_by_none_str(self): self.check_query( self.INPUT, """ SELECT account ORDER BY posting_flag; """, [ ('account', str), ], [ ('Assets:Bank:Checking',), ('Expenses:Restaurant',), ]) class TestExecuteAggregatedQuery(QueryBase): INPUT = """ 2010-01-01 open Assets:Bank:Checking 2010-01-01 open Expenses:Restaurant 2010-02-23 * "Bla" Assets:Bank:Checking 100.00 USD Expenses:Restaurant -100.00 USD """ def test_aggregated_group_by_all_implicit(self): # There is no group-by, but all columns are aggregations. self.check_query( self.INPUT, """ SELECT first(account), last(account); """, [ ('first(account)', str), ('last(account)', str), ], [ ('Assets:Bank:Checking', 'Expenses:Restaurant'), ]) def test_aggregated_group_by_all_explicit(self): # All columns ('account', 'len') are subject of a group-by. self.check_query( self.INPUT, """ SELECT account, length(account) as len GROUP BY account, len; """, [ ('account', str), ('len', int), ], [ ('Assets:Bank:Checking', 20), ('Expenses:Restaurant', 19), ]) self.check_query( """ 2010-02-21 * "First" Assets:Bank:Checking -1.00 USD Expenses:Restaurant 1.00 USD 2010-02-23 * "Second" Liabilities:Credit-Card -2.00 USD Expenses:Restaurant 2.00 USD """, """ SELECT account, length(account) as len GROUP BY account, len; """, [ ('account', str), ('len', int), ], [ ('Assets:Bank:Checking', 20), ('Expenses:Restaurant', 19), ('Liabilities:Credit-Card', 23), ]) def test_aggregated_group_by_visible(self): # GROUP-BY: 'account' is visible. self.check_query( self.INPUT, """ SELECT account, sum(position) as amount GROUP BY account; """, [ ('account', str), ('amount', inventory.Inventory), ], [ ('Assets:Bank:Checking', inventory.from_string('100.00 USD')), ('Expenses:Restaurant', inventory.from_string('-100.00 USD')), ]) def test_aggregated_group_by_invisible(self): # GROUP-BY: 'account' is invisible. self.check_query( self.INPUT, """ SELECT count(position) GROUP BY account; """, [ ('count(position)', int), ], [ (1,), (1,), ]) def test_aggregated_group_by_visible_order_by_non_aggregate_visible(self): # GROUP-BY: 'account' is visible. # ORDER-BY: 'account' is a non-aggregate and visible. self.check_query( self.INPUT, """ SELECT account, sum(position) as amount GROUP BY account ORDER BY account; """, [ ('account', str), ('amount', inventory.Inventory), ], [ ('Assets:Bank:Checking', inventory.from_string('100.00 USD')), ('Expenses:Restaurant', inventory.from_string('-100.00 USD')), ]) def test_aggregated_group_by_visible_order_by_non_aggregate_invisible(self): # GROUP-BY: 'account' and 'length(account)' are visible. # ORDER-BY: 'length(account)' is a non-aggregate and invisible. self.check_query( self.INPUT, """ SELECT account, sum(position) as amount GROUP BY account, length(account) ORDER BY length(account); """, [ ('account', str), ('amount', inventory.Inventory), ], [ ('Expenses:Restaurant', inventory.from_string('-100.00 USD')), ('Assets:Bank:Checking', inventory.from_string('100.00 USD')), ]) def test_aggregated_group_by_visible_order_by_aggregate_visible(self): # GROUP-BY: 'account' is visible. # ORDER-BY: 'sum(account)' is an aggregate and visible. self.check_query( """ 2010-02-21 * "First" Assets:Bank:Checking -1.00 USD Expenses:Restaurant 1.00 USD 2010-02-23 * "Second" Liabilities:Credit-Card -2.00 USD Expenses:Restaurant 2.00 USD """, """ SELECT account, count(account) as num, sum(number) as sum GROUP BY account ORDER BY sum(number); """, [ ('account', str), ('num', int), ('sum', Decimal), ], [ ('Liabilities:Credit-Card', 1, D('-2.00')), ('Assets:Bank:Checking', 1, D('-1.00')), ('Expenses:Restaurant', 2, D('3.00')), ]) def test_aggregated_group_by_visible_order_by_aggregate_invisible(self): # GROUP-BY: 'account' is visible. # ORDER-BY: 'sum(number)' is an aggregate and invisible. self.check_query( """ 2010-02-21 * "First" Assets:Bank:Checking -1.00 USD Expenses:Restaurant 1.00 USD 2010-02-23 * "Second" Liabilities:Credit-Card -2.00 USD Expenses:Restaurant 2.00 USD """, """ SELECT account, count(account) as num GROUP BY account ORDER BY sum(number); """, [ ('account', str), ('num', int), ], [ ('Liabilities:Credit-Card', 1), ('Assets:Bank:Checking', 1), ('Expenses:Restaurant', 2), ]) def test_aggregated_group_by_invisible_order_by_non_aggregate_visible(self): # GROUP-BY: 'account' is invisible. # ORDER-BY: 'len(account)' is a non-aggregate and visible. self.check_query( self.INPUT, """ SELECT length(account) as len, sum(position) as amount GROUP BY account, len ORDER BY len; """, [ ('len', int), ('amount', inventory.Inventory), ], [ (19, inventory.from_string('-100.00 USD'),), (20, inventory.from_string('100.00 USD'),), ]) def test_aggregated_group_by_invisible_order_by_non_aggregate_invis(self): # GROUP-BY: 'account' is invisible. # ORDER-BY: 'sum(number)' is an aggregate and invisible. self.check_query( """ 2010-02-21 * "First" Assets:Bank:Checking -1.00 USD Expenses:Restaurant 1.00 USD 2010-02-23 * "Second" Liabilities:Credit-Card -2.00 USD Expenses:Restaurant 2.00 USD """, """ SELECT count(account) as num GROUP BY account ORDER BY sum(number); """, [ ('num', int), ], [ (1,), (1,), (2,), ]) def test_aggregated_group_by_invisible_order_by_aggregate_visible(self): # GROUP-BY: 'account' is invisible. # ORDER-BY: 'sum(account)' is an aggregate and visible. self.check_query( """ 2010-02-21 * "First" Assets:Bank:Checking -1.00 USD Expenses:Restaurant 1.00 USD 2010-02-23 * "Second" Liabilities:Credit-Card -2.00 USD Expenses:Restaurant 2.00 USD """, """ SELECT count(account) as num, sum(number) as sum GROUP BY account ORDER BY sum(number); """, [ ('num', int), ('sum', Decimal), ], [ (1, D('-2.00')), (1, D('-1.00')), (2, D('3.00')), ]) def test_aggregated_group_by_invisible_order_by_aggregate_invisible(self): # GROUP-BY: 'account' is invisible. # ORDER-BY: 'sum(number)' is an aggregate and invisible. self.check_query( """ 2010-02-21 * "First" Assets:Bank:Checking -1.00 USD Expenses:Restaurant 1.00 USD 2010-02-23 * "Second" Liabilities:Credit-Card -2.00 USD Expenses:Restaurant 2.00 USD """, """ SELECT count(account) as num GROUP BY account ORDER BY sum(number); """, [ ('num', int), ], [ (1,), (1,), (2,), ]) def test_aggregated_group_by_with_having(self): self.check_query( """ 2010-02-21 * "First" Assets:Bank:Checking -1.00 USD Expenses:Foo 1.00 USD 2010-02-23 * "Second" Liabilities:Credit-Card -2.00 USD Expenses:Bar 2.00 USD """, """ SELECT account, sum(number) GROUP BY account HAVING sum(number) > 0.0 ORDER BY account """, [ ('account', str), ('sum(number)', Decimal), ], [ ('Expenses:Bar', D(2.0)), ('Expenses:Foo', D(1.0)), ]) class TestExecuteOptions(QueryBase): INPUT = """ 2010-02-23 * Assets:AssetA 5.00 USD Assets:AssetD 2.00 USD Assets:AssetB 4.00 USD Assets:AssetC 3.00 USD Assets:AssetE 1.00 USD Equity:Rest -15.00 USD """ def test_order_by_asc_implicit(self): self.check_query( self.INPUT, """ SELECT account, number ORDER BY number; """, [ ('account', str), ('number', Decimal), ], [ ('Equity:Rest', D('-15.00')), ('Assets:AssetE', D('1.00')), ('Assets:AssetD', D('2.00')), ('Assets:AssetC', D('3.00')), ('Assets:AssetB', D('4.00')), ('Assets:AssetA', D('5.00')), ]) def test_order_by_asc_explicit(self): self.check_query( self.INPUT, """ SELECT account, number ORDER BY number ASC; """, [ ('account', str), ('number', Decimal), ], [ ('Equity:Rest', D('-15.00')), ('Assets:AssetE', D('1.00')), ('Assets:AssetD', D('2.00')), ('Assets:AssetC', D('3.00')), ('Assets:AssetB', D('4.00')), ('Assets:AssetA', D('5.00')), ]) def test_order_by_desc(self): self.check_query( self.INPUT, """ SELECT account, number ORDER BY number DESC; """, [ ('account', str), ('number', Decimal), ], [ ('Assets:AssetA', D('5.00')), ('Assets:AssetB', D('4.00')), ('Assets:AssetC', D('3.00')), ('Assets:AssetD', D('2.00')), ('Assets:AssetE', D('1.00')), ('Equity:Rest', D('-15.00')), ]) def test_distinct(self): self.check_query( """ 2010-02-23 * Assets:AssetA 5.00 USD Assets:AssetA 2.00 USD Assets:AssetA 4.00 USD Equity:Rest """, """ SELECT account ; """, [ ('account', str), ], [ ('Assets:AssetA',), ('Assets:AssetA',), ('Assets:AssetA',), ('Equity:Rest',), ]) self.check_query( """ 2010-02-23 * Assets:AssetA 5.00 USD Assets:AssetA 2.00 USD Assets:AssetA 4.00 USD Equity:Rest -5.00 USD Equity:Rest -2.00 USD Equity:Rest -4.00 USD """, """ SELECT DISTINCT account ; """, [ ('account', str), ], [ ('Assets:AssetA',), ('Equity:Rest',), ]) def test_limit(self): self.check_query( self.INPUT, """ SELECT account, number ORDER BY number LIMIT 3; """, [ ('account', str), ('number', Decimal), ], [ ('Equity:Rest', D('-15.00')), ('Assets:AssetE', D('1.00')), ('Assets:AssetD', D('2.00')), ]) class TestOrderBy(QueryBase): data = """ 2022-03-28 * "Test" Assets:Tests 1.00 USD aa: 1 bb: 1 Assets:Tests 2.00 USD aa: 1 bb: 2 Assets:Tests 3.00 USD aa: 2 Assets:Tests 4.00 USD aa: 2 bb: 1 Expenses:Tests """ def test_order_by_asc_asc(self): self.check_query(self.data, """SELECT account, meta('aa') AS a, meta('bb') AS b ORDER BY 2, 3""", [ ('account', str), ('a', object), ('b', object) ], [ ('Expenses:Tests', None, None), ('Assets:Tests', 1, 1), ('Assets:Tests', 1, 2), ('Assets:Tests', 2, None), ('Assets:Tests', 2, 1), ]) def test_order_by_asc_desc(self): self.check_query(self.data, """SELECT account, meta('aa') AS a, meta('bb') AS b ORDER BY 2, 3 DESC""", [ ('account', str), ('a', object), ('b', object) ], [ ('Expenses:Tests', None, None), ('Assets:Tests', 1, 2), ('Assets:Tests', 1, 1), ('Assets:Tests', 2, 1), ('Assets:Tests', 2, None), ]) def test_order_by_desc_asc(self): self.check_query(self.data, """SELECT account, meta('aa') AS a, meta('bb') AS b ORDER BY 2 DESC, 3""", [ ('account', str), ('a', object), ('b', object) ], [ ('Assets:Tests', 2, None), ('Assets:Tests', 2, 1), ('Assets:Tests', 1, 1), ('Assets:Tests', 1, 2), ('Expenses:Tests', None, None), ]) def test_order_by_desc_desc(self): self.check_query(self.data, """SELECT account, meta('aa') AS a, meta('bb') AS b ORDER BY 2 DESC, 3 DESC""", [ ('account', str), ('a', object), ('b', object) ], [ ('Assets:Tests', 2, 1), ('Assets:Tests', 2, None), ('Assets:Tests', 1, 2), ('Assets:Tests', 1, 1), ('Expenses:Tests', None, None), ]) class TestArithmeticFunctions(QueryBase): # You need some transactions in order to eval a simple arithmetic op. # This also properly sets the data type to Decimal. # In v2, revise this so that this works like a regular DB and support integers. def test_add(self): self.check_query( """ 2010-02-23 * Assets:Something 5.00 USD """, """ SELECT number + 3 as result; """, [('result', Decimal)], [(D("8"),)]) def test_sub(self): self.check_query( """ 2010-02-23 * Assets:Something 5.00 USD """, """ SELECT number - 3 as result; """, [('result', Decimal)], [(D("2"),)]) def test_mul(self): self.check_query( """ 2010-02-23 * Assets:Something 5.00 USD """, """ SELECT number * 1.2 as result; """, [('result', Decimal)], [(D("6"),)]) def test_div(self): self.check_query( """ 2010-02-23 * Assets:Something 5.00 USD """, """ SELECT number / 2 as result; """, [('result', Decimal)], [(D("2.50"),)]) self.check_query( """ 2010-02-23 * Assets:Something 5.00 USD """, """ SELECT number / 0 as result; """, [('result', Decimal)], [(None,)]) class TestExecutePivot(QueryBase): def setUp(self): super().setUp() entries, errors, options = loader.load_string(self.data, dedent=True) self.assertFalse(errors) self.ctx = beanquery.connect('beancount:', entries=entries, errors=errors, options=options) def execute(self, query): curs = self.ctx.execute(query) rows = curs.fetchall() columns = [(c.name, c.datatype) for c in curs.description] return columns, rows data = """ 2012-01-01 open Assets:Cash 2012-01-01 open Expenses:Aaa 2012-01-01 open Expenses:Bbb 2012-01-01 * "Test" Expenses:Bbb 1.00 USD Assets:Cash 2012-01-01 * "Test" Expenses:Aaa 2.00 USD Assets:Cash 2012-02-02 * "Test" Expenses:Aaa 3.00 USD Assets:Cash 2013-01-01 * "Test" Expenses:Aaa 4.00 USD Assets:Cash 2014-02-02 * "Test" Expenses:Bbb 5.00 USD Assets:Cash 2014-03-03 * "Test" Expenses:Aaa 6.00 USD Assets:Cash 2013-01-01 * "Test" Expenses:Bbb 7.00 USD Assets:Cash 2015-04-04 * "Test" Expenses:Aaa 8.00 USD Assets:Cash """ def test_pivot_one_column(self): self.assertEqual(self.execute(""" SELECT account, year(date) AS year, sum(cost(position)) AS balance WHERE account ~ 'Expenses' GROUP BY 1, 2 PIVOT BY 1, 2"""), ( [ ('account/year', str), ('2012', inventory.Inventory), ('2013', inventory.Inventory), ('2014', inventory.Inventory), ('2015', inventory.Inventory), ], [ ('Expenses:Aaa', I('5.00 USD'), I('4.00 USD'), I('6.00 USD'), I('8.00 USD')), ('Expenses:Bbb', I('1.00 USD'), I('7.00 USD'), I('5.00 USD'), None), ])) def test_pivot_one_column_by_name(self): self.assertEqual(self.execute(""" SELECT account, year(date) AS year, sum(cost(position)) AS balance WHERE account ~ 'Expenses' GROUP BY 1, 2 PIVOT BY account, year"""), ( [ ('account/year', str), ('2012', inventory.Inventory), ('2013', inventory.Inventory), ('2014', inventory.Inventory), ('2015', inventory.Inventory), ], [ ('Expenses:Aaa', I('5.00 USD'), I('4.00 USD'), I('6.00 USD'), I('8.00 USD')), ('Expenses:Bbb', I('1.00 USD'), I('7.00 USD'), I('5.00 USD'), None), ])) def test_pivot_two_column(self): self.assertEqual(self.execute(""" SELECT account, year(date) AS year, sum(cost(position)) AS balance, last(date) AS updated WHERE account ~ 'Expenses' AND year >= 2014 GROUP BY 1, 2 PIVOT BY 1, 2"""), ( [ ('account/year', str), ('2014/balance', inventory.Inventory), ('2014/updated', datetime.date), ('2015/balance', inventory.Inventory), ('2015/updated', datetime.date), ], [ ('Expenses:Aaa', I('6.00 USD'), datetime.date(2014, 3, 3), I('8.00 USD'), datetime.date(2015, 4, 4)), ('Expenses:Bbb', I('5.00 USD'), datetime.date(2014, 2, 2), None, None), ])) class TestExecuteSubquery(QueryBase): def execute(self, query): curs = self.ctx.execute(query) rows = curs.fetchall() columns = [(c.name, c.datatype) for c in curs.description] return columns, rows def test_subquery(self): self.assertEqual( self.execute("""SELECT a + 2 AS b FROM (SELECT 3 AS a FROM #)"""), ([('b', int)], [(5, )])) class SimpleColumn(qc.EvalColumn): def __init__(self, name, func, dtype): super().__init__(dtype) self.name = name self.func = func def __call__(self, row): return self.func(row) class SimpleTable(tables.Table): columns = { 'column': SimpleColumn('column', lambda row: row, int) } def __init__(self, name, nrows): self.name = name self.nrows = nrows def __iter__(self): yield from range(self.nrows) class TestExecuteTables(QueryBase): def setUp(self): super().setUp() self.ctx.tables['test'] = SimpleTable('test', 16) def execute(self, query): query = self.compile(query) return qx.execute_query(query) def test_null_table(self): self.assertEqual( self.execute("""SELECT 1 + 1 AS test FROM #"""), ((('test', int), ), [(2, )])) def test_simple_table(self): self.assertEqual( self.execute("""SELECT column FROM #test WHERE column < 2"""), ((('column', int), ), [(0, ), (1, )])) def test_simple_table_aggregation(self): self.assertEqual( self.execute("""SELECT sum(column) FROM #test GROUP BY column % 2"""), ((('sum(column)', int), ), [(56, ), (64, )])) class TestInSubquery(QueryBase): data = """ 2000-01-01 open Income:Contributions:One 2000-01-01 open Income:Contributions:Two 2000-01-01 open Income:Contributions:Three 2000-01-01 open Assets:Cash 2024-10-25 * "One" Income:Contributions:One 100.00 USD Assets:Cash 2024-10-26 * "Two" Income:Contributions:Two 10.00 USD Assets:Cash 2024-10-27 * "Three" Income:Contributions:Three 50.00 USD Assets:Cash 2024-10-28 * "Three" Income:Contributions:Three 1.00 USD Assets:Cash """ def test_in_subquery(self): self.check_query(self.data, """ SELECT account, date FROM #postings WHERE root(account, 2) = 'Income:Contributions' AND account IN ( SELECT account FROM #postings GROUP BY account HAVING number(only('USD', sum(position))) > 50.0 ) ORDER BY date """, (('account', str), ('date', datetime.date)), [ ('Income:Contributions:One', datetime.date(2024, 10, 25)), ('Income:Contributions:Three', datetime.date(2024, 10, 27)), ('Income:Contributions:Three', datetime.date(2024, 10, 28)), ] ) class TestArrayOps(QueryBase): data = """ 1970-01-01 open Assets:Cash 2000-01-01 open Expenses:One 2000-01-02 open Expenses:Two 2000-01-03 open Expenses:Three 2025-01-01 * "One" Expenses:One 100.00 USD Assets:Cash 2025-01-02 * "Two" Expenses:Two 10.00 USD Assets:Cash 2025-01-03 * "Three" Expenses:Three 50.00 USD Assets:Cash 2025-01-03 * "Three" Expenses:Three 1.00 USD Assets:Cash """ def test_in_accounts_entries(self): self.check_query(self.data, """ SELECT date FROM #entries WHERE 'Expenses:Two' IN accounts """, (('date', datetime.date),), [ (datetime.date(2000, 1, 2),), (datetime.date(2025, 1, 2),), ] ) def test_in_accounts_transactions(self): self.check_query(self.data, """ SELECT date, narration FROM #transactions WHERE 'Expenses:Two' IN accounts """, (('date', datetime.date), ('narration', str)), [ (datetime.date(2025, 1, 2), 'Two'), ] ) def test_mathes_any_accounts_transactions(self): self.check_query(self.data, """ SELECT date, narration FROM #transactions WHERE ':Two' ?~ ANY(accounts) """, (('date', datetime.date), ('narration', str)), [ (datetime.date(2025, 1, 2), 'Two'), ] ) def test_mathes_all_accounts_transactions(self): self.check_query(self.data, """ SELECT date, narration FROM #transactions WHERE '(?i):two|:cash' ?~ ALL(accounts) """, (('date', datetime.date), ('narration', str)), [ (datetime.date(2025, 1, 2), 'Two'), ] ) class TestArrayOpsSubquery(unittest.TestCase): @classmethod def setUpClass(cls): cls.conn = beanquery.connect('test:test?end=10') def check_query(self, query, expected_types, expected_rows): curs = self.conn.execute(query) self.assertEqual(tuple(expected_types), curs.description) result_rows = curs.fetchall() self.assertEqual(expected_rows, result_rows) def test_in_subquery(self): self.check_query( """ SELECT 1 IN (SELECT x FROM #test) AS y FROM # """, (('y', bool),), [ (True,), ] ) self.check_query( """ SELECT 1 IN (SELECT x FROM #test WHERE x % 2 = 0) AS y FROM # """, (('y', bool),), [ (False,), ] ) def test_any_subquery(self): self.check_query( """ SELECT 1 = ANY (SELECT x FROM #test) AS y FROM # """, (('y', bool),), [ (True,), ] ) self.check_query( """ SELECT 1 = ANY (SELECT x FROM #test WHERE x % 2 = 0) AS y FROM # """, (('y', bool),), [ (False,), ] ) def test_all_subquery(self): self.check_query( """ SELECT 1 = ALL (SELECT x FROM #test) AS y FROM # """, (('y', bool),), [ (False,), ] ) self.check_query( """ SELECT 1 = ALL (SELECT x FROM #test WHERE x = 1) AS y FROM # """, (('y', bool),), [ (True,), ] ) beanquery-master/beanquery/query_render.py000066400000000000000000000452461474576771600214560ustar00rootroot00000000000000"""Rendering of rows. """ __copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import collections import csv import datetime import enum import typing from decimal import Decimal from beancount.core import amount from beancount.core import display_context from beancount.core import inventory from beancount.core import position class Align(enum.Enum): LEFT = 0 RIGHT = 1 class RenderContext: """Hold the query rendering configuration.""" def __init__(self, dcontext, expand=False, listsep=', ', spaced=False, null=' '): self.dcontext = dcontext self.expand = expand self.listsep = listsep self.spaced = spaced self.null = null # Map of data-types to renderer classes. This is populated by # subclassing ColumnRenderer via an __init_subclass__ hook. RENDERERS = {} class ColumnRenderer: """Base class for classes that render column values. The column renderers are responsible to render uniform type values in a way that will align nicely in a column whereby all the values render to the same width. The formatters are instantiated and are feed all the values in the column via the ``update()`` method to accumulate the dimensions it will need to format them later on. The ``prepare()`` method then computes internal status required to format these values in consistent fashion. The ``width()`` method can then be used to retrieve the computer maximum width of the column. Individual values are formatted with the ``format()`` method. Values are assumed to be of the expected type for the formatter. Formatting values outside the set of the values fed via the ``update()`` method is undefined behavior. """ dtype = None align = Align.LEFT def __init__(self, ctx): self.maxwidth = 0 self.prepared = False def __init_subclass__(cls, **kwargs): super().__init_subclass__(**kwargs) RENDERERS[cls.dtype] = cls def update(self, value): """Update the rendered with the given value. Args: value: Any object of type ``dtype``. """ def prepare(self): """Prepare to render the column. Returns: Computed column width. """ self.prepared = True return self.maxwidth @property def width(self): if not self.prepared: raise RuntimeError('width property access before calling prepare()') return self.maxwidth def format(self, value): """Format the value. Args: value: Any object of type ``dtype``. Returns: A string or list of strings representing the rendered value. """ raise NotImplementedError class ObjectRenderer(ColumnRenderer): dtype = object def update(self, value): self.maxwidth = max(self.maxwidth, len(self.format(value))) def format(self, value): return str(value) class DictRenderer(ObjectRenderer): dtype = dict class BoolRenderer(ColumnRenderer): dtype = bool def update(self, value): self.maxwidth = max(self.maxwidth, 4 if value else 5) def format(self, value): return ('TRUE' if value else 'FALSE') class StringRenderer(ObjectRenderer): dtype = str class SetRenderer(ColumnRenderer): dtype = set def __init__(self, ctx): super().__init__(ctx) self.sep = ctx.listsep def update(self, value): self.maxwidth = max(self.maxwidth, sum(len(x) + len(self.sep) for x in value) - len(self.sep)) def format(self, value): return self.sep.join(str(x) for x in sorted(value)) class DateRenderer(ColumnRenderer): dtype = datetime.date def update(self, value): self.maxwidth = 10 def format(self, value): return value.strftime('%Y-%m-%d') class IntRenderer(ObjectRenderer): dtype = int align = Align.RIGHT class EnumRenderer(ObjectRenderer): dtype = enum.Enum def format(self, value): return value.name class DecimalRenderer(ColumnRenderer): """Renderer for Decimal numbers. Numbers are left padded to align on the decimal point:: - 123.40 - 5.000 - -67 """ dtype = Decimal def __init__(self, ctx): super().__init__(ctx) # Max number of digits before the decimal point including sign. self.nintegral = 0 # Max number of digits after the decimal point. self.nfractional = 0 def update(self, value): n = value.as_tuple() if n.exponent > 0: # Special case for decimal numbers with positive exponent # and thus represented in scientific notation. self.nintegral = max(self.nintegral, len(str(value))) else: self.nintegral = max(self.nintegral, max(1, len(n.digits) + n.exponent) + n.sign) self.nfractional = max(self.nfractional, -n.exponent) def prepare(self): self.maxwidth = self.nintegral + self.nfractional + (1 if self.nfractional > 0 else 0) return super().prepare() def format(self, value): n = value.as_tuple() if n.exponent > 0: # Special case for decimal numbers with positive exponent # and thus represented in scientific notation. return str(value).rjust(self.nintegral).ljust(self.maxwidth) # Compute the padding required to align the decimal point. left = self.nintegral - (max(1, len(n.digits) + n.exponent) + n.sign) return f'{"":>{left}}{value:<{self.maxwidth - left}}' class AmountRenderer(ColumnRenderer): """Renderer for Amount instances. The numerical part is formatted with the right quantization determined by ``dcontext`` in the rendering context and aligned on the decimal point across rows. Numbers are right padded with spaces to alignt the commodity symbols across rows:: - 1234.00 USD - 42 TEST - 0.0001 ETH - 567.00 USD """ dtype = amount.Amount def __init__(self, ctx): super().__init__(ctx) # Use the display context inferred from the input ledger to # determine the quantization of the column values. self.quantize = ctx.dcontext.quantize # Use column specific display context for formatting. self.dcontext = display_context.DisplayContext() # Maximum width of the commodity symbol. self.curwidth = 0 def update(self, value): # Need to handle None to reuse this in PositionRenderer. if value is not None: number = self.quantize(value.number, value.currency) self.dcontext.update(number, value.currency) self.curwidth = max(self.curwidth, len(value.currency)) def prepare(self): self.func = self.dcontext.build(display_context.Align.DOT) zero = Decimal() for commodity in self.dcontext.ccontexts: if commodity != '__default__': self.maxwidth = max(self.maxwidth, len(self.func(zero, commodity)) + 1 + self.curwidth) return super().prepare() def format(self, value): return f'{self.func(value.number, value.currency)} {value.currency:<{self.curwidth}}' class CostRenderer(ObjectRenderer): dtype = position.Cost def __init__(self, ctx): super().__init__(ctx) self.amount_renderer = AmountRenderer(ctx) self.date_width = 0 self.label_width = 0 def update(self, value): self.amount_renderer.update(value) if value.date is not None: self.date_width = 10 + 2 if value.label is not None: self.label_width = max(self.label_width, len(value.label) + 4) def prepare(self): cost_width = self.amount_renderer.prepare() self.maxwidth = cost_width + self.date_width + self.label_width return super().prepare() def format(self, value): parts = [self.amount_renderer.format(value)] if value.date is not None: parts.append(f'{value.date:%Y-%m-%d}') if value.label is not None: parts.append(f'"{value.label}"') return ', '.join(parts) class PositionRenderer(ColumnRenderer): """Renderer for Position instrnaces. Both the unit numbers and the cost numbers are aligned:: - 5.000 HOOL {500.23 USD } - 123 CA { 1.000 HOOL} - 3.00 USD - 42.000 HOOL - 3.00 AAPL - 3.0 XY """ dtype = position.Position def __init__(self, ctx): super().__init__(ctx) self.units_renderer = AmountRenderer(ctx) self.cost_renderer = AmountRenderer(ctx) def update(self, value): self.units_renderer.update(value.units) self.cost_renderer.update(value.cost) def prepare(self): units_width = self.units_renderer.prepare() cost_width = self.cost_renderer.prepare() self.maxwidth = units_width + cost_width + (3 if cost_width > 0 else 0) return super().prepare() def format(self, value): units = self.units_renderer.format(value.units) if value.cost is None: return units.ljust(self.maxwidth) cost = self.cost_renderer.format(value.cost) return f'{units} {{{cost}}}' class InventoryRenderer(ColumnRenderer): """Renderer for Inventory instances. Inventories renders as a list of position strings. The format used differs whether expansion of list-like values to multiple rows in enabled or not. When row expansion is enabled, positions in each inventory values are sorted alphabetically by commodity symbol and are formatted with the same position formatter, resulting in all commodity strings to be aligned:: - 1234.00 USD 42 TEST - 0.0001 ETH 567.00 USD When row expansion is disabled, the position formatters are unique for each commodity symbol and the values are rendered in a table like structure. The positions appear sorted by frequency of occurence in the column and alphabetically by commodity symbol:: - 1234.00 USD 0.0001 ETH - 567.00 USD 42 TEST The separator between positions is determined by ``listsep`` in the rendering context. """ dtype = inventory.Inventory def __init__(self, ctx): super().__init__(ctx) self.listsep = ctx.listsep # We look this up for each value, it makes sense to cache it # to avoid the attribute lookup in the context. self.expand = ctx.expand # How many times at most we have seen a commodity in an inventory. self.counts = collections.defaultdict(int) # Commodity specific renderers. self.renderers = collections.defaultdict(lambda: PositionRenderer(ctx)) # How many distinct commodity need to be rendered. self.distinct = 0 def update(self, value): for pos in value.get_positions(): # We use the little indexing trick to do not have to # conditionalize this code on whether rows expansion is # enabled or not. self.renderers[self.expand or pos.units.currency].update(pos) counts = collections.Counter(pos.units.currency for pos in value.get_positions()) for key, val in counts.items(): self.counts[key] = max(self.counts[key], val) def prepare(self): if self.expand: self.maxwidth = self.renderers[self.expand].prepare() else: for commodity, renderer in self.renderers.items(): w = renderer.prepare() self.distinct += self.counts[commodity] self.maxwidth += self.counts[commodity] * (w + len(self.listsep)) self.maxwidth -= len(self.listsep) return super().prepare() @staticmethod def positionsortkey(position): # Sort positions combining fields in a more intuitive way than the default. return (position.units.currency, -position.units.number, (position.cost.currency, -position.cost.number, position.cost.date) if position.cost else ()) def format(self, value): # Expanded row format. if self.expand: strings = [] for pos in sorted(value.get_positions(), key=self.positionsortkey): strings.append(self.renderers[self.expand].format(pos)) return strings # Too many distinct commodities to present in tabular format. if self.distinct > 5: strings = [] for pos in sorted(value.get_positions(), key=self.positionsortkey): strings.append(self.renderers[pos.units.currency].format(pos)) return self.listsep.join(strings).ljust(self.maxwidth) # Tabular format with same commodity positions vertically aligned. positions = collections.defaultdict(list) for pos in sorted(value.get_positions(), key=self.positionsortkey): positions[pos.units.currency].append(pos) strings = [] for commodity, renderer in sorted(self.renderers.items()): strings += [renderer.format(pos) for pos in positions[commodity]] strings += [''.ljust(renderer.width)] * (self.counts[commodity] - len(positions[commodity])) return self.listsep.join(strings) def render_rows(rows, renderers, ctx): """Render results set row.""" # Filler for NULL values. null = ctx.null # Spacing row. spacerow = [''] * len(renderers) for row in rows: # Render the row cells. Do not pass missing values to the # renderers but substitute them with the appropriate # placeholder string. cells = [render.format(value) if value is not None else null for render, value in zip(renderers, row)] if not any(isinstance(cell, list) for cell in cells): # No multi line cells. Yield the row. yield cells else: # At least one multi line cell. Ensure that all cells are lists. cells = [cell if isinstance(cell, list) else [cell] for cell in cells] # Compute the maximum number of lines in any cell. nlines = max(len(cell) for cell in cells) # Add placeholder lines to short multi line cells. for cell in cells: if len(cell) < nlines: cell.extend([''] * (nlines - len(cell))) # Yield the obtained rows. yield from zip(*cells) # Add spacing row when needed. if ctx.spaced: yield spacerow def _get_renderer(datatype, ctx): datatype = typing.get_origin(datatype) or datatype for d in datatype.__mro__: # pragma: no branch renderer = RENDERERS.get(d) if renderer: return renderer(ctx) def render_text(columns, rows, dcontext, file, expand=False, boxed=False, spaced=False, listsep=' ', nullvalue='', narrow=True, unicode=False, **kwargs): """Render the result of executing a query in text format. Args: columns: A list of beanquery.Column descrining the table columns. rows: Data to render. dcontext: A DisplayContext object prepared for rendering numbers. file: A file object to render the results to. expand: When true expand columns that render to lists to multiple rows. boxed: When true draw an ascii-art table borders. spaced: When true insert an empty line between rows. listsep: String to use to separate values in list-like column values. nullvalue: String to use to represent NULL values. narrow: When true truncate headers to the maximum column values width. unicode: When true use unicode box drawing characters to draw tables. """ ctx = RenderContext(dcontext, expand=expand, spaced=spaced, listsep=listsep, null=nullvalue) renderers = [_get_renderer(column.datatype, ctx) for column in columns] headers = [column.name for column in columns] alignment = [renderer.align for renderer in renderers] # Prime the renderers. for row in rows: for value, renderer in zip(row, renderers): if value is not None: renderer.update(value) # Compute columns widths. widths = [max(1, narrow or len(header), len(nullvalue), render.prepare()) for header, render in zip(headers, renderers)] # Initialize table style. For unicode box drawing characters, # see https://www.unicode.org/charts/PDF/U2500.pdf if boxed: if unicode: frmt = '\u2502 {} \u2502\n' colsep = ' \u2502 ' lines = [''.rjust(width, '\u2500') for width in widths] top = '\u250C\u2500{}\u2500\u2510\n'.format('\u2500\u252C\u2500'.join(lines)) hline = '\u251C\u2500{}\u2500\u2524\n'.format('\u2500\u253C\u2500'.join(lines)) bottom = '\u2514\u2500{}\u2500\u2518\n'.format('\u2500\u2534\u2500'.join(lines)) else: frmt = '| {} |\n' colsep = ' | ' top = hline = bottom = '+-{}-+\n'.format('-+-'.join(''.rjust(width, '-') for width in widths)) else: frmt = '{}\n' colsep = ' ' top = bottom = '' hline = '{}\n'.format(colsep.join(''.rjust(width, '\u2500' if unicode else '-') for width in widths)) # Header. file.write(top) file.write(frmt.format(colsep.join(header[:width].center(width) for header, width in zip(headers, widths)))) file.write(hline) # Rows. for row in render_rows(rows, renderers, ctx): file.write(frmt.format(colsep.join(x.ljust(w) if a == Align.LEFT else x.rjust(w) for x, w, a in zip(row, widths, alignment)))) # Footer. file.write(bottom) def render_csv(columns, rows, dcontext, file, expand=False, nullvalue='', **kwargs): """Render the result of executing a query in text format. Args: columns: A list of beanquery.Column describing the table columns. rows: Data to render. dcontext: A DisplayContext object prepared for rendering numbers. file: A file object to render the results to. expand: A boolean, if true, expand columns that render to lists on multiple rows. nullvalue: String to use to represent NULL values. """ ctx = RenderContext(dcontext, expand=expand, spaced=False, listsep=',', null=nullvalue) renderers = [_get_renderer(column.datatype, ctx) for column in columns] headers = [column.name for column in columns] # Prime the renderers. for row in rows: for value, renderer in zip(row, renderers): if value is not None: renderer.update(value) # Prepare the renders. [render.prepare() for render in renderers] # Write the CSV file. writer = csv.writer(file) writer.writerow(headers) writer.writerows(render_rows(rows, renderers, ctx)) beanquery-master/beanquery/query_render_test.py000066400000000000000000000430651474576771600225120ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import datetime import enum import io import textwrap import unittest from decimal import Decimal from beancount.core import display_context from beancount.core.amount import Amount, A from beancount.core.inventory import Inventory, from_string as I from beancount.core.number import D from beancount.core.position import Cost, Position, from_string as P from beanquery import query_render from beanquery.cursor import Column class TestColumnRenderer(unittest.TestCase): def test_column_renderer(self): dcontext = display_context.DisplayContext() ctx = query_render.RenderContext(dcontext) renderer = query_render.ColumnRenderer(ctx) with self.assertRaises(RuntimeError): w = renderer.width w = renderer.prepare() self.assertEqual(w, 0) self.assertEqual(renderer.width, 0) with self.assertRaises(NotImplementedError): renderer.format(None) class RendererTestBase(unittest.TestCase): def setUp(self): self.dcontext = display_context.DisplayContext() def render(self, dtype, values, **kwargs): out = io.StringIO() rows = [(x, ) for x in values] query_render.render_text([Column('', dtype)], rows, self.dcontext, out, **kwargs) return out.getvalue().splitlines()[2:] class Foo(enum.Enum): SHORT = 1 LONG = 2 class TestRenderer(RendererTestBase): def test_object(self): self.assertEqual(self.render(object, ["foo", 1, D('1.23'), datetime.date(1970, 1, 1)]), [ 'foo ', '1 ', '1.23 ', '1970-01-01', ]) def test_bool(self): self.assertEqual(self.render(bool, [True, True]), [ 'TRUE', 'TRUE', ]) self.assertEqual(self.render(bool, [False, True]), [ 'FALSE', 'TRUE ', ]) self.assertEqual(self.render(bool, [False, False]), [ 'FALSE', 'FALSE', ]) def test_str(self): self.assertEqual(self.render(str, ['a', 'bb', 'ccc', '']), [ 'a ', 'bb ', 'ccc', ' ', ]) def test_str_null(self): self.assertEqual(self.render(str, ['', None]), [ ' ', ' ', ]) self.assertEqual(self.render(str, ['', None], nullvalue='NULL'), [ ' ', 'NULL', ]) def test_set_str(self): self.assertEqual(self.render(set, [{}, {'aaaa'}, {'bb', 'ccc'}]), [ ' ', 'aaaa ', 'bb ccc', ]) def test_date(self): self.assertEqual(self.render(datetime.date, [datetime.date(2014, 10, 3)]), [ '2014-10-03' ]) def test_int(self): self.assertEqual(self.render(int, [1, 22, 333]), [ ' 1', ' 22', '333', ]) self.assertEqual(self.render(int, [1, -22, 333]), [ ' 1', '-22', '333', ]) self.assertEqual(self.render(int, [1, 22, -333]), [ ' 1', ' 22', '-333', ]) def test_decimal_integral(self): self.assertEqual(self.render(Decimal, [D('1'), D('12'), D('123'), D('1e4')]), [ ' 1', ' 12', ' 123', '1E+4', ]) self.assertEqual(self.render(Decimal, [D('1'), D('-12'), D('123')]), [ ' 1', '-12', '123', ]) self.assertEqual(self.render(Decimal, [D('1'), D('12'), D('-123')]), [ ' 1', ' 12', '-123', ]) self.assertEqual(self.render(Decimal, [D('1'), D('12'), D('-1e3')]), [ ' 1', ' 12', '-1E+3', ]) def test_decimal_fractional(self): self.assertEqual(self.render(Decimal, [D('0.1'), D('1.2'), D('1.23'), D('1.234')]), [ '0.1 ', '1.2 ', '1.23 ', '1.234', ]) self.assertEqual(self.render(Decimal, [D('12'), D('1.2'), D('1.23'), D('12.345')]), [ '12 ', ' 1.2 ', ' 1.23 ', '12.345', ]) self.assertEqual(self.render(Decimal, [D('12'), D('1.2'), D('1.23'), D('-12.345')]), [ ' 12 ', ' 1.2 ', ' 1.23 ', '-12.345', ]) self.assertEqual(self.render(Decimal, [D('12'), D('1.2'), D('1.23'), D('-1.2e3')]), [ ' 12 ', ' 1.2 ', ' 1.23', '-1.2E+3 ', ]) def test_enum(self): self.assertEqual(self.render(Foo, [Foo.SHORT, Foo.LONG]), [ 'SHORT', 'LONG ', ]) class TestAmountRenderer(RendererTestBase): def render(self, values, **kwargs): return super().render(Amount, values, **kwargs) def test_amount(self): self.assertEqual( self.render([A('100.00 USD')]), ['100.00 USD']) def test_quantization_one(self): self.dcontext.update(Decimal('1.0000'), 'ETH') self.assertEqual(self.dcontext.quantize(Decimal('1.0'), 'ETH'), Decimal('1.0000')) self.assertEqual(self.render([A('1 ETH')]), ['1.0000 ETH']) self.assertEqual(self.render([A('0.00001 ETH')]), ['0.0000 ETH']) def test_quantization_many(self): self.dcontext.update(Decimal('1.0000'), 'ETH') self.dcontext.update(Decimal('1.00'), 'USD') self.dcontext.update(Decimal('1'), 'XYZ') self.assertEqual(self.render([A('1.0 ETH')]), ['1.0000 ETH']) self.assertEqual(self.render([A('1.0 USD')]), ['1.00 USD']) self.assertEqual(self.render([A('1.0 XYZ')]), ['1 XYZ']) def test_number_padding(self): # FIXME: The leading space seems like a bug in # DisplayContext. Either it should always be there or it # should be there to support minus signs not encountered in # training or it shoudl not be there at all. self.assertEqual(self.render([A('1 XY'), A('12 XY'), A('123 XY'), A('-1 XY')]), [ ' 1 XY', ' 12 XY', ' 123 XY', ' -1 XY', ]) self.assertEqual(self.render([A('1 XY'), A('12 XY'), A('-12 XY')]), [ ' 1 XY', ' 12 XY', '-12 XY', ]) def test_decimal_alignment(self): self.assertEqual(self.render([A('1.0 AA'), A('1.00 BB'), A('1.000 CC')]), [ '1.0 AA', '1.00 BB', '1.000 CC', ]) def test_currency_padding(self): self.assertEqual(self.render([A('1.00 XY'), A('1.00 XYZ'), A('1.00 XYZK')]), [ '1.00 XY ', '1.00 XYZ ', '1.00 XYZK', ]) def test_many(self): self.assertEqual(self.render([A('0.0001 USD'), A('20.002 HOOL'), A('33 CA'), A('1098.20 AAPL')]), [ ' 0.0001 USD ', ' 20.002 HOOL', ' 33 CA ', '1098.20 AAPL', ]) class TestPositionRenderer(RendererTestBase): def render(self, values, **kwargs): return super().render(Position, values, **kwargs) def setUp(self): super().setUp() # Prime the display context for some known commodities. self.dcontext = display_context.DisplayContext() self.dcontext.update(D('1.00'), 'USD') self.dcontext.update(D('1.00'), 'CAD') self.dcontext.update(D('1.000'), 'HOOL') self.dcontext.update(D('1'), 'CA') self.dcontext.update(D('1.00'), 'AAPL') def test_simple_poitions(self): self.assertEqual(self.render([P('3.0 USD'), P('3.0 CAD'), P('3.0 HOOL'), P('3.0 CA'), P('3.0 AAPL'), P('3.0 XY')]), [ '3.00 USD ', '3.00 CAD ', '3.000 HOOL', '3 CA ', '3.00 AAPL', '3.0 XY ', ]) def test_positions_with_price(self): self.assertEqual(self.render([P('5 HOOL {500.230000 USD}'), P('123.0 CA {1 HOOL}')]), [ ' 5.000 HOOL {500.23 USD }', '123 CA { 1.000 HOOL}', ]) class TestInventoryRenderer(RendererTestBase): def render(self, values, **kwargs): return super().render(Inventory, values, **kwargs) def setUp(self): super().setUp() # Prime the display context for some known commodities. self.dcontext = display_context.DisplayContext() self.dcontext.update(D('1.00'), 'USD') self.dcontext.update(D('1.00'), 'CAD') self.dcontext.update(D('1.000'), 'HOOL') self.dcontext.update(D('1'), 'CA') self.dcontext.update(D('1.00'), 'AAPL') def test_position_sortkey(self): inventory = I('1 AAAAA, 5 SHARE {100 USD}, 5 SHARE {200 USD}, 5 TESTS {666 USD}') self.assertEqual( sorted(inventory, key=query_render.InventoryRenderer.positionsortkey), [ P('1 AAAAA'), P('5 SHARE {200 USD}'), P('5 SHARE {100 USD}'), P('5 TESTS {666 USD}'), ]) def test_inventory(self): self.assertEqual(self.render([I('100 USD')], expand=True),[ '100.00 USD' ]) self.assertEqual(self.render([I('5 HOOL {500.23 USD}')], expand=True), [ '5.000 HOOL {500.23 USD}' ]) self.assertEqual(self.render([I('5 HOOL {500.23 USD}, 12.3456 CAAD')], expand=True), [ '12.3456 CAAD ', ' 5.000 HOOL {500.23 USD}', ]) def test_inventory_tabular(self): self.assertEqual(self.render([I('100 USD')], expand=False, listsep=' & '), [ '100.00 USD' ]) self.assertEqual(self.render([I('5 HOOL {500.23 USD}')], expand=False, listsep=' & '), [ '5.000 HOOL {500.23 USD}' ]) self.assertEqual(self.render([I('5 HOOL {500.23 USD}, 12.3456 CAAD')], expand=False, listsep=' & '), [ '12.3456 CAAD & 5.000 HOOL {500.23 USD}', ]) self.assertEqual(self.render([I('5 HOOL {500.23 USD}, 12.3456 CAAD'), I('55 HOOL {50.23 USD}, 2.3 CAAD')], expand=False, listsep=' & '), [ '12.3456 CAAD & 5.000 HOOL {500.23 USD}', ' 2.3000 CAAD & 55.000 HOOL { 50.23 USD}', ]) self.assertEqual(self.render([I('5 HOOL {500.23 USD}, 1 HOOL {567.89 USD}'), I('55 HOOL {50.23 USD}, 2.3 CAAD')], expand=False, listsep=' & '), [ ' & 5.000 HOOL {500.23 USD} & 1.000 HOOL {567.89 USD}', '2.3 CAAD & 55.000 HOOL { 50.23 USD} & ', ]) def test_inventory_too_many(self): self.assertEqual(self.render([I('10 AA, 2 BB, 3 CC, 4 DD'), I('5 AA, 6 EE, 7 FF')], expand=False, listsep=' & '), [ '10 AA & 2 BB & 3 CC & 4 DD ', ' 5 AA & 6 EE & 7 FF ', ]) class TestCostRenderer(RendererTestBase): def render(self, values, **kwargs): return super().render(Cost, values, **kwargs) def test_cost(self): self.dcontext.update(Decimal('1.0000'), 'ETH') self.assertEqual(self.render([Cost(D('1.0'), 'ETH', None, None), Cost(D('1.0'), 'ETH', datetime.date(2023, 8, 14), None), Cost(D('1.0'), 'ETH', datetime.date(2023, 8, 14), 'label')]), [ '1.0000 ETH ', '1.0000 ETH, 2023-08-14 ', '1.0000 ETH, 2023-08-14, "label"', ]) class TestQueryRenderText(unittest.TestCase): def setUp(self): self.dcontext = display_context.DisplayContext() def render(self, types, rows, **kwargs): types = [Column(*t) for t in types] oss = io.StringIO() query_render.render_text(types, rows, self.dcontext, oss, **kwargs) return oss.getvalue() def test_render_simple(self): self.assertEqual(self.render( [('x', int), ('y', int), ('z', int)], [(1, 2, 3), (4, 5, 6)]), textwrap.dedent( """\ x y z - - - 1 2 3 4 5 6 """)) def test_render_simple_unicode(self): self.assertEqual(self.render( [('x', int), ('y', int), ('z', int)], [(1, 2, 3), (4, 5, 6)], unicode=True), textwrap.dedent( """\ x y z ─ ─ ─ 1 2 3 4 5 6 """)) def test_render_boxed(self): self.assertEqual(self.render( [('x', int), ('y', int), ('z', int)], [(1, 2, 3), (4, 5, 6)], boxed=True), textwrap.dedent( """\ +---+---+---+ | x | y | z | +---+---+---+ | 1 | 2 | 3 | | 4 | 5 | 6 | +---+---+---+ """)) def test_render_boxed_unicode(self): self.assertEqual(self.render( [('x', int), ('y', int), ('z', int)], [(1, 2, 3), (4, 5, 6)], boxed=True, unicode=True), textwrap.dedent( """\ ┌───┬───┬───┐ │ x │ y │ z │ ├───┼───┼───┤ │ 1 │ 2 │ 3 │ │ 4 │ 5 │ 6 │ └───┴───┴───┘ """)) def test_render_header_centering(self): self.assertEqual(self.render( [('x', int), ('y', int), ('z', int)], [(1, 22222, 3), (4, 5, 6)]), textwrap.dedent( """\ x y z - ----- - 1 22222 3 4 5 6 """)) def test_render_header_truncation(self): self.assertEqual(self.render( [('x', int), ('abcdefg', int), ('z', int)], [(1, 222, 3), (4, 5, 6)]), textwrap.dedent( """\ x abc z - --- - 1 222 3 4 5 6 """)) def test_render_missing_values(self): self.assertEqual(self.render( [('xx', int), ('yy', int), ('zz', int)], [(12, None, 34), (None, 56, 78)]), textwrap.dedent( """\ xx yy zz -- -- -- 12 34 56 78 """)) def test_render_missing_values_boxed(self): self.assertEqual(self.render( [('xx', int), ('yy', int), ('zz', int)], [(12, None, 34), (None, 56, 78)], boxed=True), textwrap.dedent( """\ +----+----+----+ | xx | yy | zz | +----+----+----+ | 12 | | 34 | | | 56 | 78 | +----+----+----+ """)) def test_render_expand(self): self.assertEqual(self.render( [('x', int), ('inv', Inventory), ('q', int)], [(11, I('1.00 USD, 1.00 EUR, 1 TESTS'), 2), (33, I('2.00 EUR, 42 TESTS'), 4)], expand=True, boxed=True), textwrap.dedent( """\ +----+-------------+---+ | x | inv | q | +----+-------------+---+ | 11 | 1.00 EUR | 2 | | | 1 TESTS | | | | 1.00 USD | | | 33 | 2.00 EUR | 4 | | | 42 TESTS | | +----+-------------+---+ """)) def test_render_spaced(self): self.assertEqual(self.render( [('x', int)], [(1, ), (22, ), (333, )], spaced=True, boxed=True), textwrap.dedent( """\ +-----+ | x | +-----+ | 1 | | | | 22 | | | | 333 | | | +-----+ """)) class TestQueryRenderCSV(unittest.TestCase): def setUp(self): self.dcontext = display_context.DisplayContext() def render(self, types, rows, **kwargs): oss = io.StringIO() query_render.render_csv(types, rows, self.dcontext, oss, **kwargs) # The csv modules emits DOS-style newlines. return oss.getvalue().replace('\r\n', '\n') def test_render_simple(self): self.assertEqual(self.render( [Column('x', int), Column('y', int), Column('z', int)], [(1, 2, 3), (4, 5, 6)]), textwrap.dedent( """\ x,y,z 1,2,3 4,5,6 """)) def test_render_missing(self): self.assertEqual(self.render( [Column('x', int), Column('y', int), Column('z', int)], [(None, 2, 3), (4, None, 6)]), textwrap.dedent( """\ x,y,z ,2,3 4,,6 """)) def test_render_expand(self): self.assertEqual(self.render( [Column('x', int), Column('inv', Inventory), Column('q', int)], [(11, I('1.00 USD, 1.00 EUR, 1 TESTS'), 2), (33, I('2.00 EUR, 42 TESTS'), 4)], expand=True), "\n".join([ "x,inv,q", "11, 1.00 EUR ,2", ", 1 TESTS,", ", 1.00 USD ,", "33, 2.00 EUR ,4", ",42 TESTS,", "", ])) beanquery-master/beanquery/query_test.py000066400000000000000000000035401474576771600211450ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2015-2017 Martin Blais" __license__ = "GNU GPLv2" import unittest from beancount import loader from beanquery import query class TestSimple(unittest.TestCase): @loader.load_doc() def test_run_query(self, entries, _, options): """ 2022-01-01 open Assets:Checking USD 2022-01-01 open Income:ACME USD 2022-01-01 open Expenses:Taxes:Federal USD 2022-01-01 open Assets:Federal:401k IRAUSD 2022-01-01 open Expenses:Taxes:401k IRAUSD 2022-01-01 open Assets:Vacation VACHR 2022-01-01 open Income:Vacation VACHR 2022-01-01 open Expenses:Vacation VACHR 2022-01-01 open Expenses:Tests USD 2022-01-01 * "ACME" "Salary" Assets:Checking 10.00 USD Income:ACME -11.00 USD Expenses:Taxes:Federal 1.00 USD Assets:Federal:401k -2.00 IRAUSD Expenses:Taxes:401k 2.00 IRAUSD Assets:Vacation 5 VACHR Income:Vacation -5 VACHR 2022-01-02 * "Holidays" Assets:Vacation -1 VACHR Expenses:Vacation 2022-01-03 * "Test" Assets:Checking 3.00 USD Expenses:Tests """ sql_query = r""" SELECT account, sum(position) AS amount WHERE root(account, 1) = '{0}' GROUP BY 1 ORDER BY 2 DESC """ rtypes, rrows = query.run_query(entries, options, sql_query, 'Expenses', numberify=True) columns = [c.name for c in rtypes] self.assertEqual(columns, ['account', 'amount (USD)', 'amount (VACHR)', 'amount (IRAUSD)']) self.assertEqual(len(rrows[0]), 4) if __name__ == '__main__': unittest.main() beanquery-master/beanquery/render/000077500000000000000000000000001474576771600176445ustar00rootroot00000000000000beanquery-master/beanquery/render/beancount.py000066400000000000000000000007161474576771600222000ustar00rootroot00000000000000from beancount.core import display_context from beancount.parser import printer def render(desc, rows, file, *, dcontext, **kwargs): # Create a display context that renders all numbers with their # natural precision, but honors the commas option in the ledger. commas = dcontext.commas dcontext = display_context.DisplayContext() dcontext.set_commas(commas) return printer.print_entries([entry for entry, in rows], dcontext, file=file) beanquery-master/beanquery/render/csv.py000066400000000000000000000002311474576771600210050ustar00rootroot00000000000000from ..query_render import render_csv def render(desc, rows, file, *, dcontext, **kwargs): return render_csv(desc, rows, dcontext, file, **kwargs) beanquery-master/beanquery/render/text.py000066400000000000000000000003271474576771600212040ustar00rootroot00000000000000from ..query_render import render_text def render(desc, rows, file, *, dcontext, **kwargs): if not rows: return print("(empty)", file=file) return render_text(desc, rows, dcontext, file, **kwargs) beanquery-master/beanquery/shell.py000066400000000000000000000672721474576771600200640ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import atexit import cmd import importlib import io import itertools import os import pkgutil import re import shlex import sys import textwrap import traceback import warnings from contextlib import nullcontext, suppress from dataclasses import dataclass, asdict from os import path import click import beancount from beancount.parser import printer from beancount.core import data from beancount.utils import pager from beancount.utils.misc_utils import get_screen_height import beanquery from beanquery import parser from beanquery import query_compile from beanquery import render from beanquery import types from beanquery.numberify import numberify_results try: import readline except ImportError: readline = None HISTORY_FILENAME = '~/.config/beanquery/history' INIT_FILENAME = '~/.config/beanquery/init' class style: ERROR = '\033[31;1m' WARNING = '\033[31;1m' RESET = '\033[0m' ESCAPES = re.compile(r'\033\[[;?0-9]*[a-zA-Z]') @classmethod def strip(cls, x): return cls.ESCAPES.sub('', x) def render_location(text, pos, endpos, lineno, indent, strip, out): length = endpos - pos lines = text.splitlines(True) for line in lines[:lineno]: pos -= len(line) if strip and not line.rstrip(): continue strip = False out.append(indent + line.rstrip().expandtabs()) out.append(indent + lines[lineno].rstrip().expandtabs()) out.append(indent + ' ' * pos + '^' * length) # FIXME: move the error formatting into the exception classes themselves def render_exception(exc, indent='| ', strip=True): if isinstance(exc, (beanquery.CompilationError, beanquery.ParseError)) and exc.parseinfo: out = [str(exc)] pos = exc.parseinfo.pos endpos = exc.parseinfo.endpos lineno = exc.parseinfo.line render_location(exc.parseinfo.tokenizer.text, pos, endpos, lineno, indent, strip, out) return '\n'.join(out) return '\n' + traceback.format_exc() FORMATS = { name: importlib.import_module('beanquery.render.' + name).render for finder, name, ispkg in pkgutil.iter_modules(render.__path__) } @dataclass class Settings: boxed: bool = False expand: bool = False format: str = 'text' narrow: bool = True nullvalue: str = '' numberify: bool = False pager: bool = True spaced: bool = False unicode: bool = False def _parse_bool(self, value): if value in {True, False}: return value norm = value.strip().lower() if norm in {"1", "true", "t", "yes", "y", "on"}: return True if norm in {"0", "false", "f", "no", "n", "off"}: return False raise ValueError(f'"{value}" is not a valid boolean') def _parse_format(self, value): if value not in FORMATS: raise ValueError(f'"{value}" is not a valid format') return value def getstr(self, name): value = getattr(self, name) if isinstance(value, str): return repr(value) if isinstance(value, bool): return 'true' if value else 'false' return str(value) def setstr(self, name, value): vtype = type(getattr(self, name)) parse = getattr(self, f'_parse_{name}', getattr(self, f'_parse_{vtype.__name__}', vtype)) setattr(self, name, parse(value)) def todict(self): return asdict(self) def __iter__(self): return iter(self.todict().keys()) class DispatchingShell(cmd.Cmd): """A usable convenient shell for interpreting commands, with history.""" # Header for parsed commands. doc_header = "Shell utility commands (type help ):" misc_header = "Beancount query commands:" def __init__(self, outfile, interactive, runinit, settings): """Create a shell with history. Args: outfile: An output file object to write communications to. interactive: A boolean, true if this serves an interactive tty. runinit: When true, execute the commands from the user init file. settings: The shell settings. """ super().__init__() self.identchars += '.' self.outfile = outfile self.interactive = interactive self.settings = settings self.color = interactive and os.environ.get('TERM', 'dumb') != 'dumb' self.add_help() if interactive and readline is not None: # Setup completion on ``tab``. This needs to be done differently # depending on whether the ``readline`` backend is GNU readline or # libedit. This is handled automatically on Python 3.13 an # later. For Python 3.12 and earlier, the recommended way to check # for the string ``linedit`` in ``readline.__doc__`` if 'libedit' in readline.__doc__: readline.parse_and_bind("bind ^I rl_complete") else: readline.parse_and_bind("tab: complete") # ``readline`` is used to complete command names, which are # strictly alphanumeric strings, and named query identifiers, # which may contain any ASCII character. To enable completion of # the latter, reduce the set of completion word delimiters to the # shell default. Notably remove ``-`` from the delimiters list # setup by Python. readline.set_completer_delims(" \t\n\"\\'`@$><=;|&{(") # Setup ``readline`` history handling. history_filepath = path.expanduser(HISTORY_FILENAME) os.makedirs(path.dirname(history_filepath), exist_ok=True) with suppress(FileNotFoundError): readline.read_history_file(history_filepath) readline.set_history_length(2048) atexit.register(readline.write_history_file, history_filepath) warnings.showwarning = self.warning if runinit: with suppress(FileNotFoundError): with open(path.expanduser(INIT_FILENAME)) as f: for line in f: self.onecmd(line) def echo(self, message, file=sys.stdout): if not self.color: message = style.strip(message) print(message, file=file) def error(self, message): self.echo(f'{style.ERROR}error:{style.RESET} {message}', file=sys.stderr) def warning(self, message, *args): self.echo(f'{style.WARNING}warning:{style.RESET} {message}', file=sys.stderr) def add_help(self): "Attach help functions for each of the parsed token handlers." for attrname, func in list(self.__class__.__dict__.items()): match = re.match('on_(.*)', attrname) if not match: continue command_name = match.group(1) setattr(self.__class__, f'help_{command_name.lower()}', lambda _, fun=func: print(textwrap.dedent(fun.__doc__).strip(), file=self.outfile)) @property def output(self): """Where to direct command output. When the output stream is connected to the standard output, and we are running interactively, use an indirection that can send the output to a pager when the number of lines emitted is greater than a threshold. Returns: A context manager that returns a file descriptor on enter. """ if self.outfile is sys.stdout: if self.interactive and self.settings.pager: return pager.ConditionalPager(None, minlines=get_screen_height()) return pager.flush_only(sys.stdout) return nullcontext(self.outfile) def cmdloop(self, intro=None): """Override cmdloop to handle keyboard interrupts.""" while True: try: super().cmdloop(intro) break except KeyboardInterrupt: print('\n(interrupted)', file=sys.stderr) except Exception as exc: self.error(render_exception(exc)) def parseline(self, line): cmd, arg, line = super().parseline(line) if not cmd: return cmd, arg, line if cmd.startswith('.'): cmd = cmd[1:] if cmd == 'EOF': line = '.EOF' return cmd, arg, line def onecmd(self, line): cmd, arg, line = self.parseline(line) if not cmd: return if not line.startswith('.'): cmd = cmd.lower() if cmd not in {'clear', 'errors', 'exit', 'help', 'history', 'parse', 'quit', 'run', 'set'}: return self.execute(line) warnings.warn(f'commands without "." prefix are deprecated. use ".{cmd}" instead', stacklevel=0) func = getattr(self, 'do_' + cmd, None) if func is not None: return func(arg) self.error(f'unknown command "{cmd}"') def completenames(self, text, *ignored): if text.startswith('.'): dotext = 'do_' + text[1:] return ['.' + a[3:] for a in self.get_names() if a.startswith(dotext)] def do_help(self, arg): """List available commands with "help" or detailed help with "help cmd".""" super().do_help(arg.lower()) def do_history(self, arg): """Print the command-line history.""" if readline is not None: num_entries = readline.get_current_history_length() try: max_entries = int(arg) start = max(0, num_entries - max_entries) except ValueError: start = 0 for index in range(start, num_entries): line = readline.get_history_item(index + 1) print(line, file=self.outfile) def do_clear(self, arg): """Clear the command-line history.""" if readline is not None: readline.clear_history() def do_set(self, arg): """Set shell settings variables.""" if not arg: for name in self.settings: value = self.settings.getstr(name) print(f'{name}: {value}', file=self.outfile) else: components = shlex.split(arg) name = components[0] if len(components) == 1: try: value = self.settings.getstr(name) print(f'{name}: {value}', file=self.outfile) except AttributeError: self.error(f'variable "{name}" does not exist') elif len(components) == 2: value = components[1] try: self.settings.setstr(name, value) except ValueError as ex: self.error(str(ex)) except AttributeError: self.error(f'variable "{name}" does not exist') else: self.error('invalid number of arguments') def complete_set(self, text, _line, _begidx, _endidx): return [name for name in self.settings if name.startswith(text)] def do_format(self, arg): """Set output format to FRMT.""" if not arg: frmt = self.settings.getstr('format') print(f'format: {frmt}', file=self.outfile) else: value, *others = shlex.split(arg) if others: self.error('invalid number of arguments') try: self.settings.setstr('format', value) except ValueError as ex: self.error(str(ex)) def complete_format(self, text, _line, _begidx, _endidx): return [frmt for frmt in FORMATS if frmt.startswith(text)] def do_output(self, arg): """Send output to FILE or stdout if FILE is omitted.""" self.outfile = open(arg, "w") if arg else sys.stdout def do_parse(self, arg): """Run the parser on the following command and print the output.""" print(self.parse(arg).tosexp()) def parse(self, query, **kwargs): raise NotImplementedError def execute(self, query, **kwargs): """Handle statements via our parser instance and dispatch to appropriate methods. Args: query: The string to be parsed. """ statement = self.parse(query, **kwargs) name = type(statement).__name__ method = getattr(self, f'on_{name}') return method(statement) def do_exit(self, arg): """Exit the command interpreter.""" return True do_quit = do_exit def do_EOF(self, arg): """Exit the command interpreter.""" print('exit', file=self.outfile) return self.do_exit(arg) class BQLShell(DispatchingShell): """An interactive shell interpreter for the Beancount query language.""" prompt = 'beanquery> ' def __init__(self, filename, outfile, interactive=False, runinit=False, format='text', numberify=False, errors=True): settings = Settings(format=format, numberify=numberify) super().__init__(outfile, interactive, runinit, settings) self.context = beanquery.connect(None) self.filename = filename self.show_load_errors = errors self.queries = {} self.do_reload() def parse(self, line, default_close_date=None, **kwargs): statement = self.context.parse(line) if (isinstance(statement, parser.ast.Select) and isinstance(statement.from_clause, parser.ast.From) and not statement.from_clause.close): statement.from_clause.close = default_close_date return statement def do_reload(self, arg=None): "Reload the Beancount input file." if not self.filename: return self.context.errors.clear() self.context.options.clear() self.context.attach('beancount:' + self.filename) table = self.context.tables['entries'] self._extract_queries(table.entries) if self.context.errors and self.show_load_errors: printer.print_errors(self.context.errors, file=sys.stderr) if self.interactive: print_statistics(table.entries, table.options, self.context.errors, self.outfile) def _extract_queries(self, entries): self.queries = {} for entry in entries: if isinstance(entry, data.Query): x = self.queries.setdefault(entry.name, entry) if x is not entry: warnings.warn(f'duplicate query name "{entry.name}"', stacklevel=0) def do_errors(self, arg): "Print the errors that occurred during Beancount input file parsing." if self.context.errors: printer.print_errors(self.context.errors) else: print('(no errors)', file=self.outfile) def do_run(self, arg): "Run a named query defined in the Beancount input file." arg = arg.rstrip('; \t') if not arg: # List the available queries. if self.queries: print('\n'.join(name for name in sorted(self.queries))) return if arg == "*": # Execute all. for name, query in sorted(self.queries.items()): print(f'{name}:') self.execute(query.query_string, default_close_date=query.date) print() print() return name, *args = shlex.split(arg) if args: self.error('too many arguments for "run" command') return query = self.queries.get(name) if not query: self.error(f'query "{name}" not found') return self.execute(query.query_string, default_close_date=query.date) def complete_run(self, text, line, begidx, endidx): return [name for name in self.queries if name.startswith(text)] def do_tables(self, arg): """List tables.""" print('\n'.join(name for name in sorted(self.context.tables.keys()) if name), file=self.outfile) def do_describe(self, arg): """Describe table or structured type.""" def describe(obj): return '\n'.join(f' {name} ({types.name(column.dtype)})' for name, column in obj.columns.items()) names = shlex.split(arg) for name in names: table = self.context.tables.get(name) if table: print(f'table {name}:', file=self.outfile) print(describe(table), file=self.outfile) datatype = types.TYPES.get(name) if datatype: print(f'structured type {name}:', file=self.outfile) print(describe(datatype), file=self.outfile) def complete_describe(self, text, line, begidx, endidx): names = itertools.chain(self.context.tables.keys(), types.TYPES.keys()) return [name for name in names if name and name.startswith(text)] def do_explain(self, arg): """Compile and print a compiled statement for debugging.""" p = lambda x: print(x, file=self.outfile) statement = self.context.parse(arg) p('parsed statement') p('----------------') p(textwrap.indent(statement.tosexp(), ' ')) p('') query = self.context.compile(statement) p('compiled query') p('--------------') p(f' {query}') p('') p('targets') p('-------') for target in query.c_targets: name = target.name or '' datatype = types.name(target.c_expr.dtype) if target.is_aggregate: datatype += ', aggregate' p(f' {name}: {datatype}') def on_Print(self, statement): """ Print entries in Beancount format. The general form of a PRINT statement includes an SQL-like FROM selector: PRINT [FROM ...] Where: from_expr: A logical expression that matches on the attributes of the directives. See SELECT command for details (this FROM expression supports all the same expressions including its OPEN, CLOSE and CLEAR operations). """ frmt = self.settings.format try: self.settings.format = 'beancount' self.on_Select(statement) finally: self.settings.format = frmt def on_Select(self, statement): """ Extract data from a query on the postings. The general form of a SELECT statement loosely follows SQL syntax, with some mild and idiomatic extensions: SELECT [DISTINCT] [|*] [FROM [OPEN ON ] [CLOSE [ON ]] [CLEAR]] [WHERE ] [GROUP BY ] [ORDER BY [ASC|DESC]] [LIMIT num] Where: targets: A list of desired output attributes from the postings, and expressions on them. Some of the attributes of the parent transaction directive are made available in this context as well. Simple functions (that return a single value per row) and aggregation functions (that return a single value per group) are available. For the complete list of supported columns and functions, see help on "targets". You can also provide a wildcard here, which will select a reasonable default set of columns for rendering a journal. from_expr: A logical expression that matches on the attributes of the directives (not postings). This allows you to select a subset of transactions, so the accounting equation is respected for balance reports. For the complete list of supported columns and functions, see help on "from". where_expr: A logical expression that matches on the attributes of postings. The available columns are similar to those in the targets clause, without the aggregation functions. OPEN clause: replace all the transactions before the given date by summarizing entries and transfer Income and Expenses balances to Equity. CLOSE clause: Remove all the transactions after the given date and CLEAR: Transfer final Income and Expenses balances to Equity. """ cursor = self.context.execute(statement) desc = cursor.description rows = cursor.fetchall() dcontext = self.context.options['dcontext'] if self.settings.numberify: desc, rows = numberify_results(desc, rows, dcontext.build()) with self.output as out: render = FORMATS.get(self.settings.format) if render is not None: return render(desc, rows, out, dcontext=dcontext, **self.settings.todict()) raise NotImplementedError def on_Journal(self, journal): """ Select a journal of some subset of postings. This command is a convenience and converts into an equivalent Select statement, designed to extract the most sensible list of columns for the register of a list of entries as a table. The general form of a JOURNAL statement loosely follows SQL syntax: JOURNAL [FROM_CLAUSE] See the SELECT query help for more details on the FROM clause. """ return self.on_Select(journal) def on_Balances(self, balance): """ Select balances of some subset of postings. This command is a convenience and converts into an equivalent Select statement, designed to extract the most sensible list of columns for the register of a list of entries as a table. The general form of a JOURNAL statement loosely follows SQL syntax: BALANCE [FROM_CLAUSE] See the SELECT query help for more details on the FROM clause. """ return self.on_Select(balance) def help_targets(self): template = textwrap.dedent(""" The list of comma-separated target expressions may consist of columns, simple functions and aggregate functions. If you use any aggregate function, you must also provide a GROUP-BY clause. Columns ------- {columns} Functions --------- {functions} Aggregate functions ------------------- {aggregates} """) print(template.format(**_describe(self.context.tables['postings'], query_compile.FUNCTIONS)), file=self.outfile) def help_from(self): template = textwrap.dedent(""" A logical expression that consist of columns on directives (mostly transactions) and simple functions. Columns ------- {columns} Functions --------- {functions} """) print(template.format(**_describe(self.context.tables['entries'], query_compile.FUNCTIONS)), file=self.outfile) def help_where(self): template = textwrap.dedent(""" A logical expression that consist of columns on postings and simple functions. Columns ------- {columns} Functions --------- {functions} """) print(template.format(**_describe(self.context.tables['postings'], query_compile.FUNCTIONS)), file=self.outfile) def _describe_columns(columns): out = io.StringIO() wrapper = textwrap.TextWrapper(initial_indent=' ', subsequent_indent=' ', width=80) for name, column in columns.items(): print(f'{name}: {types.name(column.dtype)}', file=out) print(wrapper.fill(re.sub(r'[ \n\t]+', ' ', column.__doc__ or '')), file=out) print(file=out) return out.getvalue().rstrip() def _describe_functions(functions, aggregates=False): entries = [] for name, funcs in functions.items(): if aggregates != issubclass(funcs[0], query_compile.EvalAggregator): continue name = name.lower() for func in funcs: args = ', '.join(types.name(d) for d in func.__intypes__) doc = re.sub(r'[ \n\t]+', ' ', func.__doc__ or '') entries.append((name, doc, args)) entries.sort() out = io.StringIO() wrapper = textwrap.TextWrapper(initial_indent=' ', subsequent_indent=' ', width=80) for key, entries in itertools.groupby(entries, key=lambda x: x[:2]): # noqa: B020 for name, doc, args in entries: print(f'{name}({args})', file=out) print(wrapper.fill(doc), file=out) print(file=out) return out.getvalue().rstrip() def _describe(table, functions): return dict( columns=_describe_columns(table.columns), functions=_describe_functions(functions, aggregates=False), aggregates=_describe_functions(functions, aggregates=True)) def summary_statistics(entries): """Calculate basic summary statistics to output a brief welcome message. Args: entries: A list of directives. Returns: A tuple of three integers, the total number of directives parsed, the total number of transactions and the total number of postings there in. """ num_directives = len(entries) num_transactions = 0 num_postings = 0 for entry in entries: if isinstance(entry, data.Transaction): num_transactions += 1 num_postings += len(entry.postings) return num_directives, num_transactions, num_postings def print_statistics(entries, options, errors, outfile): """Print summary statistics to stdout. Args: entries: A list of directives. options: An options map. as produced by the parser. outfile: A file object to write to. """ num_directives, num_transactions, num_postings = summary_statistics(entries) if 'title' in options: print(f'''Input file: "{options['title']}"''', file=outfile) print(f"Ready with {num_directives} directives " f"({num_postings} postings in {num_transactions} transactions, " f"{len(errors)} validation errors)", file=outfile) @click.command() @click.argument('filename') @click.argument('query', nargs=-1) @click.option('--output', '-o', type=click.File('w'), default='-', help="Output filename.") @click.option('--format', '-f', type=click.Choice(FORMATS.keys()), default='text', help="Output format.") @click.option('--numberify', '-m', is_flag=True, help="Numberify the output, removing the currencies.") @click.option('--no-errors', '-q', is_flag=True, help="Do not report ledger validation errors on load.") @click.version_option('', message=f'beanquery {beanquery.__version__}, beancount {beancount.__version__}') def main(filename, query, numberify, format, output, no_errors): """An interactive interpreter for the Beancount Query Language. Load Beancount ledger FILENAME and run Beancount Query Language QUERY on it, if specified, or drop into the interactive shell. If not explicitly set with the dedicated option, the output format is inferred from the output file name, if specified. """ # Create the shell. interactive = sys.stdin.isatty() and not query shell = BQLShell(filename, output, interactive, True, format, numberify, not no_errors) # Run interactively if we're a TTY and no query is supplied. if interactive: warnings.filterwarnings('always') shell.cmdloop() else: # Run in batch mode (Non-interactive). if query: # We have a query to run. query = ' '.join(query) else: # If we have no query and we're not a TTY, read the BQL command from # standard input. query = sys.stdin.read() shell.onecmd(query) beanquery-master/beanquery/shell_test.py000066400000000000000000000255611474576771600211160ustar00rootroot00000000000000__copyright__ = "Copyright (C) 2014-2016 Martin Blais" __license__ = "GNU GPLv2" import functools import re import sys import textwrap import unittest import click.testing from beancount import loader from beancount.utils import test_utils from beanquery import shell @functools.lru_cache(None) def load(): entries, errors, options = loader.load_string(textwrap.dedent(""" 2022-01-01 open Assets:Checking USD 2022-01-01 open Assets:Federal:401k IRAUSD 2022-01-01 open Assets:Gold GLD 2022-01-01 open Assets:Vacation VACHR 2022-01-01 open Assets:Vanguard:RGAGX RGAGX 2022-01-01 open Expenses:Commissions USD 2022-01-01 open Expenses:Food USD 2022-01-01 open Expenses:Home:Rent USD 2022-01-01 open Expenses:Taxes:401k IRAUSD 2022-01-01 open Expenses:Taxes:Federal USD 2022-01-01 open Expenses:Tests USD 2022-01-01 open Expenses:Vacation VACHR 2022-01-01 open Income:ACME USD 2022-01-01 open Income:Gains USD 2022-01-01 open Income:Vacation VACHR 2022-01-01 * "ACME" "Salary" Assets:Checking 10.00 USD Income:ACME -11.00 USD Expenses:Taxes:Federal 1.00 USD Assets:Federal:401k -2.00 IRAUSD Expenses:Taxes:401k 2.00 IRAUSD Assets:Vacation 5 VACHR Income:Vacation -5 VACHR 2022-01-01 * "Rent" Assets:Checking 42.00 USD Expenses:Home:Rent 42.00 USD 2022-01-02 * "Holidays" Assets:Vacation -1 VACHR Expenses:Vacation 2022-01-03 * "Test 01" Assets:Checking 1.00 USD Expenses:Tests 2022-01-04 * "My Fovorite Plase" "Eating out alone" Assets:Checking 4.00 USD Expenses:Food 2022-01-05 * "Invest" Assets:Checking -359.94 USD Assets:Vanguard:RGAGX 2.086 RGAGX {172.55 USD} 2013-10-23 * "Buy Gold" Assets:Checking -1278.67 USD Assets:Gold 9 GLD {141.08 USD} Expenses:Commissions 8.95 USD 2022-01-07 * "Sell Gold" Assets:Gold -16 GLD {147.01 USD} @ 135.50 USD Assets:Checking 2159.05 USD Expenses:Commissions 8.95 USD Income:Gains 184.16 USD 2022-01-08 * "Sell Gold" Assets:Gold -16 GLD {147.01 USD} @ 135.50 USD Assets:Checking 2159.05 USD Expenses:Commissions 8.95 USD Income:Gains 184.16 USD 2022-02-01 * "ACME" "Salary" Assets:Checking 10.00 USD Income:ACME -11.00 USD Expenses:Taxes:Federal 1.00 USD Assets:Federal:401k -2.00 IRAUSD Expenses:Taxes:401k 2.00 IRAUSD Assets:Vacation 5 VACHR Income:Vacation -5 VACHR 2022-02-01 * "Rent" Assets:Checking 43.00 USD Expenses:Home:Rent 43.00 USD 2022-02-02 * "Test 02" Assets:Checking 2.00 USD Expenses:Tests 2030-01-01 query "taxes" " SELECT date, description, position, balance WHERE account ~ 'Taxes' ORDER BY date DESC LIMIT 20" 2015-01-01 query "home" " SELECT last(date) as latest, account, sum(position) as total WHERE account ~ ':Home:' GROUP BY account" """)) return entries, errors, options def run_shell_command(cmd): """Run a shell command and return its output.""" with test_utils.capture('stdout') as stdout, test_utils.capture('stderr') as stderr: shell_obj = shell.BQLShell(None, sys.stdout) entries, errors, options = load() shell_obj.context.attach('beancount:', entries=entries, errors=errors, options=options) shell_obj._extract_queries(entries) # pylint: disable=protected-access shell_obj.onecmd(cmd) return stdout.getvalue(), stderr.getvalue() def runshell(function): """Decorate a function to run the shell and return the output.""" def wrapper(self): out, err = run_shell_command(function.__doc__) return function(self, out, err) return wrapper class TestUseCases(unittest.TestCase): """Testing all the use cases from the proposal here. I'm hoping to replace reports by these queries instead.""" @runshell def test_print_from(self, out, err): """ PRINT FROM narration ~ 'alone' """ self.assertRegex(out, 'Eating out alone') @runshell def test_accounts(self, out, err): """ SELECT DISTINCT account, open_date(account) ORDER BY account_sortkey(account); """ self.assertRegex(out, 'Assets:Checking *2022-01-01') self.assertRegex(out, 'Income:ACME *2022-01-01') @runshell def test_commodities(self, out, err): """ SELECT DISTINCT currency ORDER BY 1; """ self.assertRegex(out, 'USD') self.assertRegex(out, 'IRAUSD') self.assertRegex(out, 'VACHR') @runshell def test_commodities_cost(self, out, err): """ SELECT DISTINCT cost_currency ORDER BY 1; """ self.assertRegex(out, 'USD') @runshell def test_commodities_pairs(self, out, err): """ SELECT DISTINCT currency, cost_currency ORDER BY 1, 2; """ self.assertRegex(out, 'GLD *USD') @runshell def test_balances(self, out, err): """ BALANCES AT cost; """ self.assertRegex(out, r'Assets:Gold *\d+\.\d+ USD') @runshell def test_balances_with_where(self, out, err): """ JOURNAL 'Assets:Checking'; """ self.assertRegex(out, 'Salary') @runshell def test_balance_sheet(self, out, err): """ BALANCES AT cost FROM OPEN ON 2022-01-02 CLOSE ON 2022-02-01 CLEAR; """ self.assertRegex(out, r'Assets:Gold * \d+\.\d+ USD') @runshell def test_income_statement(self, out, err): """ SELECT account, cost(sum(position)) FROM OPEN ON 2022-01-01 CLOSE ON 2023-01-01 WHERE account ~ '(Income|Expenses):*' GROUP BY account, account_sortkey(account) ORDER BY account_sortkey(account); """ self.assertRegex( out, 'Expenses:Taxes:401k *4.00 IRAUSD') @runshell def test_journal(self, out, err): """ JOURNAL 'Assets:Checking' FROM OPEN ON 2022-02-01 CLOSE ON 2022-03-01; """ self.assertRegex(out, "2022-01-31 +S +Opening balance for 'Assets:Checking'") self.assertRegex(out, "Test 02") @runshell def test_conversions(self, out, err): """ SELECT date, payee, narration, position, balance FROM OPEN ON 2022-01-01 CLOSE ON 2023-01-01 WHERE flag = 'C' """ self.assertRegex(out, "2022-12-31 *Conversion for") @runshell def test_documents(self, out, err): """ SELECT date, account, narration WHERE type = 'Document'; """ ## FIXME: Make this possible, we need an example with document entries. @runshell def test_holdings(self, out, err): """ SELECT account, currency, cost_currency, sum(position) GROUP BY account, currency, cost_currency; """ ## FIXME: Here we need to finally support FLATTEN to make this happen properly. class TestRun(unittest.TestCase): @runshell def test_run_custom__list(self, out, err): """ .run """ self.assertEqual("home taxes", re.sub(r'[] \n\t]+', ' ', out).strip()) @runshell def test_run_custom__query_not_exists(self, out, err): """ .run something """ self.assertEqual('error: query "something" not found', err.strip()) @runshell def test_run_custom__query_id(self, out, err): """ .run taxes """ self.assertRegex(out, 'date +description +position +balance') self.assertRegex(out, r'ACME \| Salary') @runshell def test_run_custom__query_string(self, out, err): """ RUN "taxes" """ self.assertRegex(out, 'date +description +position +balance') self.assertRegex(out, r'ACME \| Salary') @runshell def test_run_custom__all(self, out, err): """ RUN * """ self.assertRegex(out, 'date +description +position +balance') self.assertRegex(out, r'ACME \| Salary') self.assertRegex(out, 'account +total') self.assertRegex(out, 'Expenses:Home:Rent') class TestHelp(unittest.TestCase): def test_help_functions(self): for name in dir(shell.BQLShell): if name.startswith('help_'): run_shell_command('help ' + name[5:]) class ClickTestCase(unittest.TestCase): """Base class for command-line program test cases.""" def main(self, *args): init_filename = shell.INIT_FILENAME history_filename = shell.HISTORY_FILENAME try: shell.INIT_FILENAME = '' shell.HISTORY_FILENAME = '' runner = click.testing.CliRunner() result = runner.invoke(shell.main, args, catch_exceptions=False) self.assertEqual(result.exit_code, 0) return result finally: shell.INIT_FILENAME = init_filename shell.HISTORY_FILENAME = history_filename class TestShell(ClickTestCase): @test_utils.docfile def test_success(self, filename): """ 2013-01-01 open Assets:Account1 2013-01-01 open Assets:Account2 2013-01-01 open Assets:Account3 2013-01-01 open Equity:Unknown 2013-04-05 * Equity:Unknown Assets:Account1 5000 USD 2013-04-05 * Assets:Account1 -3000 USD Assets:Account2 30 BOOG {100 USD} 2013-04-05 * Assets:Account1 -1000 USD Assets:Account3 800 EUR @ 1.25 USD """ result = self.main(filename, "SELECT 1;") self.assertTrue(result.stdout) @test_utils.docfile def test_format_csv(self, filename): """ """ r = self.main(filename, '--format=csv', "SELECT 111 AS one, 222 AS two FROM #") self.assertEqual(r.stdout, textwrap.dedent('''\ one,two 111,222 ''')) @test_utils.docfile def test_format_text(self, filename): """ """ r = self.main(filename, '--format=text', "SELECT 111 AS one, 222 AS two FROM #") self.assertEqual(r.stdout, textwrap.dedent('''\ one two --- --- 111 222 ''')) if __name__ == '__main__': unittest.main() beanquery-master/beanquery/sources/000077500000000000000000000000001474576771600200505ustar00rootroot00000000000000beanquery-master/beanquery/sources/beancount.py000066400000000000000000000140221474576771600223770ustar00rootroot00000000000000import sys import types as _types import typing from urllib.parse import urlparse from beancount import loader from beancount import parser from beancount.core import data from beancount.core import position from beancount.core import prices from beancount.core.getters import get_account_open_close, get_commodity_directives from beanquery import tables from beanquery import types from beanquery import query_compile from beanquery import query_env from beanquery import query_render from beanquery import hashable if sys.version_info >= (3, 10): _UNIONTYPES = {typing.Union, _types.UnionType} else: _UNIONTYPES = {typing.Union} hashable.register(data.Transaction, lambda t: (t.date, t.flag, t.narration)) TABLES = [query_env.EntriesTable, query_env.PostingsTable] def attach(context, dsn, *, entries=None, errors=None, options=None): filename = urlparse(dsn).path if filename: entries, errors, options = loader.load_file(filename) for table in TABLES: context.tables[table.name] = table(entries, options) context.options.update(options) context.errors.extend(errors) class Metadata(dict): pass class MetadataRenderer(query_render.ObjectRenderer): dtype = Metadata def format(self, value): return str({k: v for k, v in value.items() if k not in {'filename', 'lineno'} and not k.startswith('__')}) class GetAttrColumn(query_compile.EvalColumn): def __init__(self, name, dtype): super().__init__(dtype) self.name = name def __call__(self, context): return getattr(context, self.name) def _simplify_typing_annotation(dtype): if typing.get_origin(dtype) in _UNIONTYPES: args = typing.get_args(dtype) if len(args) == 2: if args[0] is type(None): return args[1], True if args[1] is type(None): return args[0], True raise NotImplementedError return typing.get_origin(dtype) or dtype, False def _typed_namedtuple_to_columns(cls, renames=None): columns = {} for name, dtype in typing.get_type_hints(cls).items(): dtype, nullable = _simplify_typing_annotation(dtype) if name == 'meta' and dtype is dict: dtype = Metadata if dtype is frozenset: dtype = set colname = renames.get(name, name) if renames is not None else name columns[colname] = GetAttrColumn(name, dtype) return columns class Position(types.Structure): name = 'position' columns = _typed_namedtuple_to_columns(position.Position) class Cost(types.Structure): name = 'cost' columns = _typed_namedtuple_to_columns(data.Cost) class Amount(types.Structure): name = 'amount' columns = _typed_namedtuple_to_columns(data.Amount) class Transaction(types.Structure): name = 'transaction' columns = _typed_namedtuple_to_columns(data.Transaction) del columns['postings'] def _accounts(row): return {p.account for p in row.postings} _accounts.dtype = typing.Set[str] _accounts.childnodes = () Transaction.columns['accounts'] = _accounts types.ALIASES[position.Position] = Position types.ALIASES[data.Cost] = Cost types.ALIASES[data.Amount] = Amount types.ALIASES[data.Transaction] = Transaction class Open(types.Structure): name = 'open' columns = _typed_namedtuple_to_columns(data.Open) class Close(types.Structure): name = 'close' columns = _typed_namedtuple_to_columns(data.Close) class Table(tables.Table): datatype = None def __init__(self, entries, options): self.entries = entries self.options = options def __init_subclass__(cls): TABLES.append(cls) def __iter__(self): datatype = self.datatype for entry in self.entries: if isinstance(entry, datatype): yield entry @property def wildcard_columns(self): return tuple(col for col in self.columns.keys() if col != 'meta') class TransactionsTable(Table): name = 'transactions' datatype = data.Transaction # There is not a way to inherit the __init_subclass__() and # __iter__() methods while inheriting the columns attribute # definition from the Transaction class. columns = Transaction.columns class PricesTable(Table): name = 'prices' datatype = data.Price columns = _typed_namedtuple_to_columns(datatype) def __init__(self, entries, options): super().__init__(entries, options) self.price_map = prices.build_price_map(entries) class BalancesTable(Table): name = 'balances' datatype = data.Balance columns = _typed_namedtuple_to_columns(datatype, {'diff_amount': 'discrepancy'}) class NotesTable(Table): name = 'notes' datatype = data.Note columns = _typed_namedtuple_to_columns(datatype) class EventsTable(Table): name = 'events' datatype = data.Event columns = _typed_namedtuple_to_columns(datatype) class DocumentsTable(Table): name = 'documents' datatype = data.Document columns = _typed_namedtuple_to_columns(datatype) class GetItemColumn(query_compile.EvalColumn): def __init__(self, key, dtype): super().__init__(dtype) self.key = key def __call__(self, row): return row[self.key] class AccountsTable(tables.Table): name = 'accounts' columns = { 'account': GetItemColumn(0, str), 'open': GetItemColumn(1, Open), 'close': GetItemColumn(2, Close), } def __init__(self, entries, options): self.accounts = get_account_open_close(entries) self.types = parser.options.get_account_types(options) def __iter__(self): return ((name, value[0], value[1]) for name, value in self.accounts.items()) TABLES.append(AccountsTable) class CommoditiesTable(tables.Table): name = 'commodities' columns = _typed_namedtuple_to_columns(data.Commodity, {'currency': 'name'}) def __init__(self, entries, options): self.commodities = get_commodity_directives(entries) def __iter__(self): return iter(self.commodities.values()) TABLES.append(CommoditiesTable) beanquery-master/beanquery/sources/test.py000066400000000000000000000032161474576771600214030ustar00rootroot00000000000000import re from urllib.parse import urlparse, parse_qsl from ..query_compile import EvalColumn from .. import tables class Column(EvalColumn): def __init__(self, datatype, func): super().__init__(datatype) self.func = func def __call__(self, row): return self.func(row) class TableMeta(type): def __new__(mcs, name, bases, dct): columns = {} members = {} for key, value in dct.items(): if isinstance(value, EvalColumn): value.name = key columns[key] = value continue members[key] = value assert 'columns' not in members members['columns'] = columns return super().__new__(mcs, name, bases, members) class Table(tables.Table, metaclass=TableMeta): x = Column(int, lambda row: row) def __init__(self, *args): self.rows = range(*args) def __iter__(self): return iter(self.rows) class MagicColumnsRegistry(dict): def get(self, key, default=None): if re.fullmatch(r'x+', key): return Column(int, lambda x: x * len(key)) return default class MagicTable(tables.Table): columns = MagicColumnsRegistry() def __init__(self, *args): self.rows = range(*args) def __iter__(self): return iter(self.rows) def attach(context, dsn): parts = urlparse(dsn) params = dict(parse_qsl(parts.query, strict_parsing=True)) name = params.get('name', 'test') cls = MagicTable if parts.path == 'magic' else Table context.tables[name] = cls(int(params.get('start', 0)), int(params.get('stop', 11)), int(params.get('step', 1))) beanquery-master/beanquery/tables.py000066400000000000000000000007331474576771600202140ustar00rootroot00000000000000class Table: columns = {} name = '' def __getitem__(self, name): return self.columns[name] @property def wildcard_columns(self): # For backward compatibility. Remove once the postings table # is updated to return all columns upon ``SELECT *`` and the # query compiler is updated not to rely on this property return self.columns.keys() class NullTable(Table): def __iter__(self): return iter([None]) beanquery-master/beanquery/types.py000066400000000000000000000047001474576771600201040ustar00rootroot00000000000000import datetime import decimal import itertools import typing # Only Python >= 3.10 exposes NoneType in the types module. NoneType = type(None) class AnyType: """As written on the tin.""" __slots__ = () __name__ = 'any' def __eq__(self, other): """Compares equal to any other type.""" return isinstance(other, type) # Used in BQL functions signatures for arguments that can have any type. Any = AnyType() # Used for COUNT(*) Asterisk = typing.NewType('*', object) # noqa: PLC0132 Asterisk.__mro__ = Asterisk, # Keep track of the defined structured types to allow introspection. TYPES = {} class Structure: """Base class for structured data types.""" name = None columns = {} def __init_subclass__(cls): if cls.name: TYPES[cls.name] = cls def _bases(t): if t is NoneType: return (object,) bases = t.__mro__ if len(bases) > 1 and bases[-1] is object: # All types that are not ``object`` have more than one class # in their ``__mro__``. BQL uses ``object`` for untypes # values. Do not return ``object`` as base for strict types, # to avoid functions taking untyped onjects to accept all # values. return bases[:-1] return bases def function_lookup(functions, name, operands): """Lookup a BQL function implementation. Args: functions: The functions registry to interrogate. name: The function name. operands: Function operands. Returns: A EvalNode (or subclass) instance or None if the function was not found. """ for signature in itertools.product(*(_bases(operand.dtype) for operand in operands)): for func in functions[name]: if func.__intypes__ == list(signature): return func return None # Map types to their BQL name. Used to find the name of the type cast funtion. MAP = { bool: 'bool', datetime.date: 'date', decimal.Decimal: 'decimal', int: 'int', str: 'str', } # Map between Python types and BQL structured types. Functions and # columns definitions can use Python types. The corresponding BQL # structured type is looked up when compiling the subscrip operator. ALIASES = {} def name(datatype): if datatype is NoneType: return 'NULL' if isinstance(datatype, typing._GenericAlias): return str(datatype).rsplit('.', 1)[-1].lower() return getattr(datatype, 'name', datatype.__name__.lower()) beanquery-master/beanquery/types_test.py000066400000000000000000000011071474576771600211410ustar00rootroot00000000000000import unittest from beanquery import types from beanquery.sources import beancount class TestName(unittest.TestCase): def test_transactions_names(self): table = beancount.TransactionsTable columns = {name: types.name(column.dtype) for name, column in table.columns.items()} self.assertEqual(columns, { 'meta': 'metadata', 'date': 'date', 'flag': 'str', 'payee': 'str', 'narration': 'str', 'tags': 'set', 'links': 'set', 'accounts': 'set[str]', }) beanquery-master/bin/000077500000000000000000000000001474576771600151425ustar00rootroot00000000000000beanquery-master/bin/bean-query000077500000000000000000000002221474576771600171340ustar00rootroot00000000000000#!/usr/bin/env python3 __copyright__ = "Copyright (C) 2013-2017 Martin Blais" __license__ = "GNU GPLv2" from beanquery.shell import main; main() beanquery-master/docs/000077500000000000000000000000001474576771600153225ustar00rootroot00000000000000beanquery-master/docs/conf.py000066400000000000000000000014241474576771600166220ustar00rootroot00000000000000project = 'beanquery' copyright = '2014-2022, beanquery Contributors' author = 'beanquery Contributors' version = '0.1' language = 'en' html_theme = 'furo' html_title = f'{project} {version}' html_logo = 'logo.svg' extensions = [ 'sphinx.ext.autodoc', 'sphinx.ext.napoleon', 'sphinx.ext.intersphinx', 'sphinx.ext.extlinks', 'sphinx.ext.githubpages', ] extlinks = { 'issue': ('https://github.com/beancount/beanquery/issues/%s', '#'), 'pull': ('https://github.com/beancount/beanquery/pull/%s', '#'), } intersphinx_mapping = { 'python': ('https://docs.python.org/3', None), 'beancount': ('https://beancount.github.io/docs/', None), } napoleon_google_docstring = True napoleon_use_param = False autodoc_typehints = 'none' autodoc_member_order = 'bysource' beanquery-master/docs/index.rst000066400000000000000000000004171474576771600171650ustar00rootroot00000000000000beanquery: Customizable lightweight SQL query tool ================================================== beanquery is a customizable and extensible lightweight SQL query tool that works on tabular data, including `Beancount`__ ledger data. __ https://beancount.github.io/ beanquery-master/docs/logo.inkscape000066400000000000000000000131471474576771600200070ustar00rootroot00000000000000 beanquery-master/docs/logo.svg000066400000000000000000000143731474576771600170130ustar00rootroot00000000000000 beanquery-master/etc/000077500000000000000000000000001474576771600151455ustar00rootroot00000000000000beanquery-master/etc/env000077500000000000000000000001121474576771600156550ustar00rootroot00000000000000#!/bin/sh USERPATH=$USERPATH:$PROJDIR/bin PYTHONPATH=$PYTHONPATH:$PROJDIR beanquery-master/pyproject.toml000066400000000000000000000040551474576771600173120ustar00rootroot00000000000000[build-system] requires = ['setuptools'] build-backend = 'setuptools.build_meta' [project] name = 'beanquery' version = '0.2.0.dev0' description = 'Customizable lightweight SQL query tool' license = { file = 'LICENSE' } readme = 'README.rst' authors = [ { name = 'Martin Blais', email = 'blais@furius.ca' }, { name = 'Daniele Nicolodi', email = 'daniele@grinta.net' }, ] maintainers = [ { name = 'Daniele Nicolodi', email = 'daniele@grinta.net' }, ] keywords = [ 'accounting', 'ledger', 'beancount', 'SQL', 'BQL' ] classifiers = [ 'License :: OSI Approved :: GNU General Public License v2 (GPLv2)', 'Programming Language :: Python :: 3 :: Only', 'Programming Language :: Python :: 3.8', 'Programming Language :: Python :: 3.9', 'Programming Language :: Python :: 3.10', 'Programming Language :: Python :: 3.11', 'Programming Language :: Python :: 3.12', 'Programming Language :: Python :: 3.13', 'Programming Language :: SQL', 'Topic :: Office/Business :: Financial :: Accounting', ] requires-python = '>= 3.8' dependencies = [ 'beancount >= 2.3.4', 'click > 7.0', 'python-dateutil >= 2.6.0', 'tatsu-lts', ] [project.optional-dependencies] docs = [ 'furo >= 2024.08.06', 'sphinx ~= 8.1.0', ] [project.scripts] bean-query = 'beanquery.shell:main' [project.urls] homepage = 'https://github.com/beancount/beanquery' issues = 'https://github.com/beancount/beanquery/issues' [tool.setuptools.packages] find = {} [tool.coverage.run] branch = true [tool.coverage.report] exclude_also = [ 'if typing.TYPE_CHECKING:', ] [tool.ruff] line-length = 128 target-version = 'py38' [tool.ruff.lint] select = ['E', 'F', 'W', 'UP', 'B', 'C4', 'PL', 'RUF'] ignore = [ 'B007', 'B905', 'C408', 'E731', 'PLR0911', 'PLR0912', 'PLR0913', 'PLR0915', 'PLR1714', 'PLR2004', 'PLW2901', 'RUF012', 'RUF023', # unsorted-dunder-slots 'UP007', 'UP032', ] exclude = [ 'beanquery/parser/parser.py' ] [tool.ruff.lint.per-file-ignores] 'beanquery/query_env.py' = ['F811'] beanquery-master/requirements.txt000066400000000000000000000001131474576771600176510ustar00rootroot00000000000000beancount >=2.3.4 click >7.0 python-dateutil >=2.6.0 tatsu >=5.7.4, <5.8.0