tegaki-python-0.3.1/ 0000755 0001750 0001750 00000000000 11352066723 014155 5 ustar mathieu mathieu tegaki-python-0.3.1/tests/ 0000755 0001750 0001750 00000000000 11352066723 015317 5 ustar mathieu mathieu tegaki-python-0.3.1/tests/runtests.py 0000755 0001750 0001750 00000003125 11342122457 017560 0 ustar mathieu mathieu #!/usr/bin/env python
# -*- coding: utf-8 -*-
# Copyright (C) 2008 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import glob
import os
import sys
import unittest
import doctest
import tegaki
currdir = os.path.dirname(os.path.abspath(__file__))
parentdir = os.path.join(currdir, "..")
os.chdir(currdir)
sys.path = sys.path + [parentdir]
def gettestnames():
return [name[:-3] for name in glob.glob('test_*.py')]
# Run doctests
for attr in dir(tegaki):
attr = getattr(tegaki, attr)
if type(attr) == type(tegaki):
if hasattr(attr, "__doctest__") and attr.__doctest__:
doctest.testmod(attr)
# Run unittests
suite = unittest.TestSuite()
loader = unittest.TestLoader()
for name in gettestnames():
suite.addTest(loader.loadTestsFromName(name))
testRunner = unittest.TextTestRunner(verbosity=1)
testRunner.run(suite)
tegaki-python-0.3.1/tests/myfile 0000644 0001750 0001750 00000000206 11342122457 016521 0 ustar mathieu mathieu
1000
1000
tegaki-python-0.3.1/tests/test_arrayutils.py 0000644 0001750 0001750 00000005242 11342122457 021126 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import unittest
from tegaki.arrayutils import *
class ArrayTest(unittest.TestCase):
def testArrayFlatten(self):
for arr, expected in (
([[1,2], [3,4]], [1, 2, 3, 4]),
([[]], [])
):
self.assertEquals(array_flatten(arr), expected)
def testArrayReshape(self):
for arr, expected, n in (
([1, 2, 3, 4], [[1,2], [3,4]], 2),
([], [], 2),
([1, 2, 3], [[1, 2]], 2) # expected 4 values
):
self.assertEquals(array_reshape(arr, n), expected)
def testArraySplit(self):
arr = [[1,2], [3,4], [5,6], [7,8], [9, 10], [11, 12]]
expected = [ [[1,2],[3,4]], [[5,6],[7, 8]], [[9,10],[11,12]] ]
self.assertEquals(array_split(arr, 3), expected)
def testArrayMean(self):
arr = [1, 2, 3, 4]
expected = 2.5
self.assertEquals(array_mean(arr), expected)
def testArrayVariance(self):
arr = [1, 2, 3, 4]
expected = 1.25
self.assertEquals(array_variance(arr), expected)
def testArrayMeanVector(self):
arr = [ [1,2], [3,4] ]
expected = [2, 3]
self.assertEquals(array_mean_vector(arr), expected)
def testArrayVarianceVector(self):
arr = [ [1,2], [3,4] ]
expected = [1.0, 1.0]
self.assertEquals(array_variance_vector(arr), expected)
def testArrayAdd(self):
arr1 = [1,2]
arr2 = [3,4]
expected = [4, 6]
self.assertEquals(array_add(arr1, arr2), expected)
def testArrayMul(self):
arr1 = [1,2]
arr2 = [3,4]
expected = [3, 8]
self.assertEquals(array_mul(arr1, arr2), expected)
tegaki-python-0.3.1/tests/test_charcol.py 0000644 0001750 0001750 00000030046 11342122457 020342 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2009 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import unittest
import os
import sys
import StringIO
from tegaki.character import Point, Stroke, Writing, Character
from tegaki.charcol import CharacterCollection
class CharacterCollectionTest(unittest.TestCase):
def setUp(self):
self.currdir = os.path.dirname(os.path.abspath(__file__))
path = os.path.join(self.currdir, "data", "collection", "test.charcol")
self.cc = CharacterCollection()
self.cc.read(path)
f = os.path.join(self.currdir, "data", "character.xml")
self.c = Character()
self.c.read(f)
def testValidate(self):
path = os.path.join(self.currdir, "data", "collection", "test.charcol")
f = open(path)
buf = f.read()
f.close()
invalid = \
"""
防
"""
malformed = \
"""
"""
try:
self.assertTrue(CharacterCollection.validate(buf))
self.assertFalse(CharacterCollection.validate(invalid))
self.assertFalse(CharacterCollection.validate(malformed))
except NotImplementedError:
sys.stderr.write("lxml missing!\n")
pass
def _testReadXML(self, charcol):
self.assertEquals(charcol.get_set_list(), ["一", "三", "二", "四"])
c = {}
for k in ["19968_1", "19968_2", "19968_3", "19977_1", "19977_2",
"20108_1"]:
c[k] = Character()
c[k].read(os.path.join(self.currdir, "data", "collection",
k + ".xml"))
self.assertEquals(charcol.get_characters("一"),
[c["19968_1"], c["19968_2"], c["19968_3"]])
self.assertEquals(charcol.get_characters("三"),
[c["19977_1"], c["19977_2"]])
self.assertEquals(charcol.get_characters("二"),
[c["20108_1"]])
self.assertEquals(charcol.get_characters("四"), [])
self.assertEquals(charcol.get_all_characters(),
[c["19968_1"], c["19968_2"], c["19968_3"],
c["19977_1"], c["19977_2"], c["20108_1"]])
def testReadXMLFile(self):
self._testReadXML(self.cc)
def testToXML(self):
charcol2 = CharacterCollection()
charcol2.read_string(self.cc.to_xml())
self.assertEquals(self.cc.get_set_list(), charcol2.get_set_list())
self.assertEquals(self.cc.get_all_characters(),
charcol2.get_all_characters())
def testWriteGzipString(self):
charcol2 = CharacterCollection()
charcol2.read_string(self.cc.write_string(gzip=True), gzip=True)
self.assertEquals(self.cc.get_set_list(), charcol2.get_set_list())
self.assertEquals(self.cc.get_all_characters(),
charcol2.get_all_characters())
def testWriteBz2String(self):
charcol2 = CharacterCollection()
charcol2.read_string(self.cc.write_string(bz2=True), bz2=True)
self.assertEquals(self.cc.get_set_list(), charcol2.get_set_list())
self.assertEquals(self.cc.get_all_characters(),
charcol2.get_all_characters())
def testAddSame(self):
path = os.path.join(self.currdir, "data", "collection", "test.charcol")
charcol = CharacterCollection()
charcol.read(path)
charcol2 = CharacterCollection()
charcol2.read(path)
charcol3 = charcol.concatenate(charcol2, check_duplicate=True)
self.assertEquals(charcol3.get_set_list(), ["一", "三", "二", "四"])
self.assertEquals(len(charcol3.get_characters("一")), 3)
self.assertEquals(len(charcol3.get_characters("三")), 2)
self.assertEquals(len(charcol3.get_characters("二")), 1)
self.assertEquals(len(charcol3.get_characters("四")), 0)
def testGetChars(self):
all_ = self.cc.get_characters("一")
self.assertEquals(self.cc.get_characters("一", limit=2), all_[0:2])
self.assertEquals(self.cc.get_characters("一", offset=2), all_[2:])
self.assertEquals(self.cc.get_characters("一", limit=1, offset=1),
all_[1:2])
def testAdd(self):
path = os.path.join(self.currdir, "data", "collection", "test.charcol")
charcol = CharacterCollection()
charcol.read(path)
path2 = os.path.join(self.currdir, "data", "collection",
"test2.charcol")
charcol2 = CharacterCollection()
charcol2.read(path2)
charcol3 = charcol + charcol2
self.assertEquals(charcol3.get_set_list(), ["一", "三", "二", "四",
"a", "b", "c", "d"])
self.assertEquals(len(charcol3.get_characters("一")), 3)
self.assertEquals(len(charcol3.get_characters("三")), 2)
self.assertEquals(len(charcol3.get_characters("二")), 1)
self.assertEquals(len(charcol3.get_characters("四")), 0)
self.assertEquals(len(charcol3.get_characters("a")), 3)
self.assertEquals(len(charcol3.get_characters("b")), 2)
self.assertEquals(len(charcol3.get_characters("c")), 1)
self.assertEquals(len(charcol3.get_characters("d")), 0)
def testFromCharDirRecursive(self):
directory = os.path.join(self.currdir, "data")
charcol = CharacterCollection.from_character_directory(directory,
check_duplicate=True)
self.assertEquals(charcol.get_set_list(), ["防", "三", "一", "二"])
self.assertEquals(len(charcol.get_characters("一")), 3)
self.assertEquals(len(charcol.get_characters("三")), 2)
self.assertEquals(len(charcol.get_characters("二")), 1)
self.assertEquals(len(charcol.get_characters("防")), 1)
def testFromCharDirNotRecursive(self):
directory = os.path.join(self.currdir, "data")
charcol = CharacterCollection.from_character_directory(directory,
recursive=False, check_duplicate=True)
self.assertEquals(charcol.get_set_list(), ["防"])
self.assertEquals(len(charcol.get_characters("防")), 1)
def testIncludeChars(self):
self.cc.include_characters_from_text("一三")
self.assertEquals(self.cc.get_set_list(), ["一", "三"])
def testExcludeChars(self):
self.cc.exclude_characters_from_text("三")
self.assertEquals(self.cc.get_set_list(), ["一", "二"])
def testProxy(self):
char = self.cc.get_all_characters()[0]
writing = char.get_writing()
writing.normalize()
strokes = writing.get_strokes(full=True)
stroke = strokes[0]
stroke.smooth()
p = stroke[0]
p.x = 10
char2 = self.cc.get_all_characters()[0]
self.assertEquals(char, char2)
def testNoProxy(self):
self.cc.WRITE_BACK = False
char = self.cc.get_all_characters()[0]
writing = char.get_writing()
writing.normalize()
strokes = writing.get_strokes(full=True)
stroke = strokes[0]
stroke.smooth()
p = stroke[0]
p.x = 10
char2 = self.cc.get_all_characters()[0]
self.assertNotEqual(char, char2)
# manually update the object
self.cc.update_character_object(char)
char2 = self.cc.get_all_characters()[0]
self.assertEquals(char, char2)
def testAddSet(self):
self.cc.add_set("toto")
self.assertEquals(self.cc.get_set_list()[-1], "toto")
def testRemoveSet(self):
before = self.cc.get_set_list()
self.cc.remove_set(before[-1])
after = self.cc.get_set_list()
self.assertEquals(len(before)-1, len(after))
self.assertEquals(before[0:-1], after)
def testGetNSets(self):
self.assertEquals(len(self.cc.get_set_list()), self.cc.get_n_sets())
self.assertEquals(4, self.cc.get_n_sets())
def testGetTotalNCharacters(self):
self.assertEquals(len(self.cc.get_all_characters()),
self.cc.get_total_n_characters())
self.assertEquals(6, self.cc.get_total_n_characters())
def testGetNCharacters(self):
for set_name in self.cc.get_set_list():
self.assertEquals(len(self.cc.get_characters(set_name)),
self.cc.get_n_characters(set_name))
self.assertEquals(self.cc.get_n_characters("一"), 3)
self.assertEquals(self.cc.get_n_characters("三"), 2)
self.assertEquals(self.cc.get_n_characters("二"), 1)
def testSetCharacters(self):
before = self.cc.get_characters("一")[0:2]
self.cc.set_characters("一", before)
after = self.cc.get_characters("一")
self.assertEquals(before, after)
def testAppendCharacter(self):
len_before = len(self.cc.get_characters("一"))
self.cc.append_character("一", self.c)
len_after = len(self.cc.get_characters("一"))
self.assertEquals(len_before + 1, len_after)
def testInsertCharacter(self):
before = self.cc.get_characters("一")[0]
len_before = len(self.cc.get_characters("一"))
self.cc.insert_character("一", 0, self.c)
after = self.cc.get_characters("一")[0]
self.assertNotEqual(before, after)
len_after = len(self.cc.get_characters("一"))
self.assertEqual(len_before+1, len_after)
def testReplaceCharacter(self):
before = self.cc.get_characters("一")[0]
len_before = len(self.cc.get_characters("一"))
self.cc.replace_character("一", 0, self.c)
after = self.cc.get_characters("一")[0]
self.assertNotEqual(before, after)
len_after = len(self.cc.get_characters("一"))
self.assertEqual(len_before, len_after)
def testRemoveCharacter(self):
before = self.cc.get_characters("一")[0]
len_before = len(self.cc.get_characters("一"))
self.cc.remove_character("一", 0)
after = self.cc.get_characters("一")[0]
self.assertNotEqual(before, after)
len_after = len(self.cc.get_characters("一"))
self.assertEqual(len_before-1, len_after)
def testRemoveLastCharacter(self):
before = self.cc.get_characters("一")[-1]
len_before = len(self.cc.get_characters("一"))
self.cc.remove_last_character("一")
after = self.cc.get_characters("一")[-1]
self.assertNotEqual(before, after)
len_after = len(self.cc.get_characters("一"))
self.assertEqual(len_before-1, len_after)
def testRemoveSamples(self):
self.cc.remove_samples(keep_at_most=2)
self.assertEquals(self.cc.get_n_characters("一"), 2)
self.assertEquals(self.cc.get_n_characters("三"), 2)
self.assertEquals(self.cc.get_n_characters("二"), 1)
self.cc.remove_samples(keep_at_most=1)
self.assertEquals(self.cc.get_n_characters("一"), 1)
self.assertEquals(self.cc.get_n_characters("三"), 1)
self.assertEquals(self.cc.get_n_characters("二"), 1)
def testRemoveEmptySets(self):
self.cc.remove_empty_sets()
self.assertEquals(self.cc.get_set_list(), ["一", "三", "二"])
tegaki-python-0.3.1/tests/minjson.py 0000644 0001750 0001750 00000034716 11342122457 017355 0 ustar mathieu mathieu ##############################################################################
#
# Copyright (c) 2006 Zope Corporation and Contributors.
# All Rights Reserved.
#
# This software is subject to the provisions of the Zope Public License,
# Version 2.1 (ZPL). A copy of the ZPL should accompany this distribution.
# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
# FOR A PARTICULAR PURPOSE.
#
##############################################################################
# minjson.py
# reads minimal javascript objects.
# str's objects and fixes the text to write javascript.
#UNICODE USAGE: Minjson tries hard to accommodate naive usage in a
#"Do what I mean" manner. Real applications should handle unicode separately.
# The "right" way to use minjson in an application is to provide minjson a
# python unicode string for reading and accept a unicode output from minjson's
# writing. That way, the assumptions for unicode are yours and not minjson's.
# That said, the minjson code has some (optional) unicode handling that you
# may look at as a model for the unicode handling your application may need.
# Thanks to Patrick Logan for starting the json-py project and making so many
# good test cases.
# Additional thanks to Balazs Ree for replacing the writing module.
# Jim Washington 6 Dec 2006.
# 2006-12-06 Thanks to Koen van de Sande, now handles the case where someone
# might want e.g., a literal "\n" in text not a new-line.
# 2005-12-30 writing now traverses the object tree instead of relying on
# str() or unicode()
# 2005-10-10 on reading, looks for \\uxxxx and replaces with u'\uxxxx'
# 2005-10-09 now tries hard to make all strings unicode when reading.
# 2005-10-07 got rid of eval() completely, makes object as found by the
# tokenizer.
# 2005-09-06 imported parsing constants from tokenize; they changed a bit from
# python2.3 to 2.4
# 2005-08-22 replaced the read sanity code
# 2005-08-21 Search for exploits on eval() yielded more default bad operators.
# 2005-08-18 Added optional code from Koen van de Sande to escape
# outgoing unicode chars above 128
from re import compile, sub, search, DOTALL
from token import ENDMARKER, NAME, NUMBER, STRING, OP, ERRORTOKEN
from tokenize import tokenize, TokenError, NL
#Usually, utf-8 will work, set this to utf-16 if you dare.
emergencyEncoding = 'utf-8'
class ReadException(Exception):
pass
class WriteException(Exception):
pass
#################################
# read JSON object #
#################################
slashstarcomment = compile(r'/\*.*?\*/',DOTALL)
doubleslashcomment = compile(r'//.*\n')
unichrRE = compile(r"\\u[0-9a-fA-F]{4,4}")
def unichrReplace(match):
return unichr(int(match.group()[2:],16))
escapeStrs = (('\n',r'\n'),('\b',r'\b'),
('\f',r'\f'),('\t',r'\t'),('\r',r'\r'), ('"',r'\"')
)
class DictToken:
__slots__=[]
pass
class ListToken:
__slots__=[]
pass
class ColonToken:
__slots__=[]
pass
class CommaToken:
__slots__=[]
pass
class JSONReader(object):
"""raise SyntaxError if it is not JSON, and make the object available"""
def __init__(self,data):
self.stop = False
#make an iterator of data so that next() works in tokenize.
self._data = iter([data])
self.lastOp = None
self.objects = []
self.tokenize()
def tokenize(self):
try:
tokenize(self._data.next,self.readTokens)
except TokenError:
raise SyntaxError
def resolveList(self):
#check for empty list
if isinstance(self.objects[-1],ListToken):
self.objects[-1] = []
return
theList = []
commaCount = 0
try:
item = self.objects.pop()
except IndexError:
raise SyntaxError
while not isinstance(item,ListToken):
if isinstance(item,CommaToken):
commaCount += 1
else:
theList.append(item)
try:
item = self.objects.pop()
except IndexError:
raise SyntaxError
if not commaCount == (len(theList) -1):
raise SyntaxError
theList.reverse()
item = theList
self.objects.append(item)
def resolveDict(self):
theList = []
#check for empty dict
if isinstance(self.objects[-1], DictToken):
self.objects[-1] = {}
return
#not empty; must have at least three values
try:
#value (we're going backwards!)
value = self.objects.pop()
except IndexError:
raise SyntaxError
try:
#colon
colon = self.objects.pop()
if not isinstance(colon, ColonToken):
raise SyntaxError
except IndexError:
raise SyntaxError
try:
#key
key = self.objects.pop()
if not isinstance(key,basestring):
raise SyntaxError
except IndexError:
raise SyntaxError
#salt the while
comma = value
while not isinstance(comma,DictToken):
# store the value
theList.append((key,value))
#do it again...
try:
#might be a comma
comma = self.objects.pop()
except IndexError:
raise SyntaxError
if isinstance(comma,CommaToken):
#if it's a comma, get the values
try:
value = self.objects.pop()
except IndexError:
#print self.objects
raise SyntaxError
try:
colon = self.objects.pop()
if not isinstance(colon, ColonToken):
raise SyntaxError
except IndexError:
raise SyntaxError
try:
key = self.objects.pop()
if not isinstance(key,basestring):
raise SyntaxError
except IndexError:
raise SyntaxError
theDict = {}
for k in theList:
theDict[k[0]] = k[1]
self.objects.append(theDict)
def readTokens(self,type, token, (srow, scol), (erow, ecol), line):
# UPPERCASE consts from tokens.py or tokenize.py
if type == OP:
if token not in "[{}],:-":
raise SyntaxError
else:
self.lastOp = token
if token == '[':
self.objects.append(ListToken())
elif token == '{':
self.objects.append(DictToken())
elif token == ']':
self.resolveList()
elif token == '}':
self.resolveDict()
elif token == ':':
self.objects.append(ColonToken())
elif token == ',':
self.objects.append(CommaToken())
elif type == STRING:
tok = token[1:-1]
parts = tok.split("\\\\")
for k in escapeStrs:
if k[1] in tok:
parts = [part.replace(k[1],k[0]) for part in parts]
self.objects.append("\\".join(parts))
elif type == NUMBER:
if self.lastOp == '-':
factor = -1
else:
factor = 1
try:
self.objects.append(factor * int(token))
except ValueError:
self.objects.append(factor * float(token))
elif type == NAME:
try:
self.objects.append({'true':True,
'false':False,'null':None}[token])
except KeyError:
raise SyntaxError
elif type == ENDMARKER:
pass
elif type == NL:
pass
elif type == ERRORTOKEN:
if ecol == len(line):
#it's a char at the end of the line. (mostly) harmless.
pass
else:
raise SyntaxError
else:
raise SyntaxError
def output(self):
try:
assert len(self.objects) == 1
except AssertionError:
raise SyntaxError
return self.objects[0]
def safeRead(aString, encoding=None):
"""read the js, first sanitizing a bit and removing any c-style comments
If the input is a unicode string, great. That's preferred. If the input
is a byte string, strings in the object will be produced as unicode anyway.
"""
# get rid of trailing null. Konqueror appends this.
CHR0 = chr(0)
while aString.endswith(CHR0):
aString = aString[:-1]
# strip leading and trailing whitespace
aString = aString.strip()
# zap /* ... */ comments
aString = slashstarcomment.sub('',aString)
# zap // comments
aString = doubleslashcomment.sub('',aString)
# detect and handle \\u unicode characters. Note: This has the side effect
# of converting the entire string to unicode. This is probably OK.
unicodechars = unichrRE.search(aString)
if unicodechars:
aString = unichrRE.sub(unichrReplace, aString)
#if it's already unicode, we won't try to decode it
if isinstance(aString, unicode):
s = aString
else:
if encoding:
# note: no "try" here. the encoding provided must work for the
# incoming byte string. UnicodeDecode error will be raised
# in that case. Often, it will be best not to provide the encoding
# and allow the default
s = unicode(aString, encoding)
#print "decoded %s from %s" % (s,encoding)
else:
# let's try to decode to unicode in system default encoding
try:
s = unicode(aString)
#import sys
#print "decoded %s from %s" % (s,sys.getdefaultencoding())
except UnicodeDecodeError:
# last choice: handle as emergencyEncoding
enc = emergencyEncoding
s = unicode(aString, enc)
#print "%s decoded from %s" % (s, enc)
# parse and get the object.
try:
data = JSONReader(s).output()
except SyntaxError:
raise ReadException, 'Unacceptable JSON expression: %s' % aString
return data
read = safeRead
#################################
# write object as JSON #
#################################
import re, codecs
from cStringIO import StringIO
### Codec error handler
def jsonreplace_handler(exc):
'''Error handler for json
If encoding fails, \\uxxxx must be emitted. This
is similar to the "backshashreplace" handler, only
that we never emit \\xnn since this is not legal
according to the JSON syntax specs.
'''
if isinstance(exc, UnicodeEncodeError):
part = exc.object[exc.start]
# repr(part) will convert u'\unnnn' to u'u\\nnnn'
return u'\\u%04x' % ord(part), exc.start+1
else:
raise exc
# register the error handler
codecs.register_error('jsonreplace', jsonreplace_handler)
### Writer
def write(input, encoding='utf-8', outputEncoding=None):
writer = JsonWriter(input_encoding=encoding, output_encoding=outputEncoding)
writer.write(input)
return writer.getvalue()
re_strmangle = re.compile('"|\b|\f|\n|\r|\t|\\\\')
def func_strmangle(match):
return {
'"': '\\"',
'\b': '\\b',
'\f': '\\f',
'\n': '\\n',
'\r': '\\r',
'\t': '\\t',
'\\': '\\\\',
}[match.group(0)]
def strmangle(text):
return re_strmangle.sub(func_strmangle, text)
class JsonStream(object):
def __init__(self):
self.buf = []
def write(self, text):
self.buf.append(text)
def getvalue(self):
return ''.join(self.buf)
class JsonWriter(object):
def __init__(self, stream=None, input_encoding='utf-8', output_encoding=None):
'''
- stream is optional, if specified must also give output_encoding
- The input strings can be unicode or in input_encoding
- output_encoding is optional, if omitted, result will be unicode
'''
if stream is not None:
if output_encoding is None:
raise WriteException, 'If a stream is given, output encoding must also be provided'
else:
stream = JsonStream()
self.stream = stream
self.input_encoding = input_encoding
self.output_encoding = output_encoding
def write(self, obj):
if isinstance(obj, (list, tuple)):
self.stream.write('[')
first = True
for elem in obj:
if first:
first = False
else:
self.stream.write(',')
self.write(elem)
self.stream.write(']'),
elif isinstance(obj, dict):
self.stream.write('{')
first = True
for key, value in obj.iteritems():
if first:
first = False
else:
self.stream.write(',')
self.write(key)
self.stream.write(':')
self.write(value)
self.stream.write('}')
elif obj is True:
self.stream.write('true')
elif obj is False:
self.stream.write('false')
elif obj is None:
self.stream.write('null')
elif not isinstance(obj, basestring):
# if we are not baseobj, convert to it
try:
obj = str(obj)
except Exception, exc:
raise WriteException, 'Cannot write object (%s: %s)' % (exc.__class__, exc)
self.stream.write(obj)
else:
# convert to unicode first
if not isinstance(obj, unicode):
try:
obj = unicode(obj, self.input_encoding)
except (UnicodeDecodeError, UnicodeTranslateError):
obj = unicode(obj, 'utf-8', 'replace')
# do the mangling
obj = strmangle(obj)
# make the encoding
if self.output_encoding is not None:
obj = obj.encode(self.output_encoding, 'jsonreplace')
self.stream.write('"')
self.stream.write(obj)
self.stream.write('"')
def getvalue(self):
return self.stream.getvalue()
tegaki-python-0.3.1/tests/test_arrayutils.pyc 0000644 0001750 0001750 00000006634 11342122457 021277 0 ustar mathieu mathieu
gJc @ s3 d d k Z d d k Td e i f d YZ d S( iN( t *t ArrayTestc B sY e Z d Z d Z d Z d Z d Z d Z d Z d Z d Z
RS( c C sd x] d d g d d g g d d d d g f g g g f f D]" \ } } | i t | | q: Wd S( Ni i i i ( t assertEqualst
array_flatten( t selft arrt expected( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArrayFlatten s
' c C s x d d d d g d d g d d g g d f g g d f d d d g d d g g d f f D]( \ } } } | i t | | | q[ Wd S( Ni i i i ( R t
array_reshape( R R R t n( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArrayReshape% s * " c C s d d g d d g d d g d d g d d
g d d g g } d d g d d g g d d g d d g g d d
g d d g g g } | i t | d | d S(
Ni i i i i i i i i i
i i ( R t array_split( R R R ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArraySplit/ s <Ec C s2 d d d d g } d } | i t | | d S( Ni i i i g @( R t
array_mean( R R R ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt
testArrayMean5 s c C s2 d d d d g } d } | i t | | d S( Ni i i i g ?( R t array_variance( R R R ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArrayVariance; s c C s> d d g d d g g } d d g } | i t | | d S( Ni i i i ( R t array_mean_vector( R R R ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArrayMeanVectorA s c C s> d d g d d g g } d d g } | i t | | d S( Ni i i i g ?( R t array_variance_vector( R R R ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArrayVarianceVectorG s c C sA d d g } d d g } d d g } | i t | | | d S( Ni i i i i ( R t array_add( R t arr1t arr2R ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArrayAddM s c C sA d d g } d d g } d d g } | i t | | | d S( Ni i i i i ( R t array_mul( R R R R ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyt testArrayMulT s ( t __name__t
__module__R R
R R R R R R R ( ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pyR s
( t unittestt tegaki.arrayutilst TestCaseR ( ( ( sI /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_arrayutils.pys s
tegaki-python-0.3.1/tests/mynewfile.xml.bz2 0000644 0001750 0001750 00000000205 11342122457 020525 0 ustar mathieu mathieu BZh91AY&SY~ ߀ PG . u T#@ JPѴBz#L!mhFC2g d?d@dAdBdCdDdEdFdGdHdIdJdKdLdMdNdOdPdQdRdSdTdUdVdWdXdYdZd[d\d]d^d_d`dadbdcdddedfdgdhdidjdkdldmdndodpdqdrdsdtdudvdwdxdydzd{d|d}d~dddddddddddddddddddddddddddddddddddddddddddddddddddddddddgy ddddddddddddddddddg ddddddddddddddddddddddddddddddddddddddddddddddddddddddd ddddddg= ddd d
d d d
ddddddddddddddd d d d d d!d"d#d$d%d&d'd(d)d*d+d,d-d.d/d0d1d2d3d4d5d6d7d8d9d:d;d<d=d>d?d@dAdBdCdDdEdFdGdHdIdJdKdLdMdNdOdPdQdRdSdTdUdVdWdXdYgS dZd[d\d]d^d_d`dadbdcdddedfdgdhdidjdkdldmdndodpdqdrdsdtdudvdwdxdydzd{d|d}d~ddddddddddddddddddddddddddddddddddddddddddddddddddddgY g | _ d S( Ni i i i i i i i i i i i i i i i i i i i% i" i( i, i/ i2 i6 i9 i< i@ iC iF iJ iM iT iZ i^ id ih in iu ix i| i i i i i
i i i i i i i i i i i i iP i i i i i i i i i i i i i i i i i i i i i i i i i i i i i" i) i, i0 i3 i6 i= i@ iD iG iJ i iN iQ iT iX i[ i i i iW ia ik ir i i i i i i i i: i^ ie ih il ir iy i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i& ib i| i i i i i iv io i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i% i" ( i( i" ( i, i" ( i/ i" ( i2 i" ( i6 i" ( i9 i" ( i< i% ( i@ i% ( i@ i( ( i@ i, ( i< i/ ( i< i2 ( i< i6 ( i< i9 ( i9 i< ( i9 i@ ( i6 iC ( i6 iF ( i2 iJ ( i2 iM ( i/ iT ( i, iZ ( i, i^ ( i, id ( i( ih ( i% in ( i" in ( i iu ( i ix ( i i| ( i i ( i i ( i i ( i i ( i i ( i
i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i" i ( i( i ( i, i ( i2 i ( i6 i ( i9 i ( i< i ( i< i ( i@ i ( iC i ( iF i ( iF i ( iJ i ( iM i ( iM i ( iP i ( iP i ( iP i ( iP i ( iP i ( iP i ( iP i ( iP i ( iP i ( iP i ( iP i ( iP i ( iM i ( iM i ( iM i ( iJ i ( iJ i ( iF i ( iF i ( iF i ( iC i ( iC i ( i@ i ( i< i ( i< i ( i9 i ( i6 i ( i2 i ( i2 i ( i/ i ( i, i ( i, i ( i( i" ( i% i" ( i" i) ( i i, ( i i, ( i i0 ( i i3 ( i i6 ( i
i6 ( i i= ( i i= ( i i@ ( i i@ ( i iD ( i iG ( i iG ( i iJ ( i iJ ( i iN ( i iN ( i iQ ( i iT ( i iT ( i iT ( i iX ( i iX ( i iX ( i i[ ( i i[ ( i i[ ( i i ( i i ( i i ( i i ( i i" ( i i% ( i i( ( i i, ( i i, ( i i2 ( i i6 ( i i< ( i i@ ( i iC ( i iF ( i iM ( i iP ( i iT ( i iW ( i iZ ( i ia ( i id ( i ih ( i ik ( i ik ( i ir ( i iu ( i ix ( i i| ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i" ( i i) ( i i, ( i i0 ( i i6 ( i i: ( i i= ( i i@ ( i iD ( i iG ( i iN ( i iQ ( i iT ( i i[ ( i i^ ( i ie ( i ih ( i il ( i ir ( i iy ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i
( i i ( i i ( ir i ( iu i ( ix i ( i| i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i" i ( i& i ( iG i ( iJ i ( iQ i ( iT i ( iX i ( i^ i ( ib i ( ih i ( il i ( ir i ( iy i ( i| i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i ir ( i iu ( i iu ( i" iu ( i& iu ( i: i| ( i= i| ( iD i| ( iX i ( i[ i ( ie i ( ih i ( ir i ( iv i ( i| i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i" ( i i) ( i i, ( i i3 ( i i6 ( i i: ( i i= ( i i@ ( i iG ( i iJ ( i iN ( i iQ ( i iT ( i iX ( i| i[ ( i| i^ ( iy i^ ( iy ib ( iv ib ( ir ib ( io ib ( io ie ( il ie ( ih ie ( ie ie ( ib ie ( i^ ie ( i i% ( i i( ( i i, ( i i/ ( i i2 ( i i6 ( i i< ( i i@ ( i iC ( i iJ ( i iP ( i iW ( i iZ ( i ia ( i ik ( i in ( i iu ( i ix ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i& ( i i) ( i i0 ( i i3 ( i i6 ( i i@ ( i iD ( i iJ ( i iN ( i i[ ( i i^ ( i ie ( i il ( i ir ( i i| ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( t ost patht dirnamet abspatht __file__t currdirt strokes( t self( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt setUp! s ! c C s9 | i | i d | i | i | i i d S( Ns 防( t assertEqualst get_utf8R t get_writingt get_strokes( R t char( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt _testReadXMLu s c
C sG t i i | i d d } x | | d | d d f D] } t | } | o | i | n t t i d } g } | D] \ } } | | q} ~ } x | D] } z- | i
| t | } | i | | Wd t i | Xz* | i
t | }
| i | |
Wd t i | Xq Wq6 Wt } | i
t | i
d S( Nt datas
character.xmls .gzips .bz2s .xmls .xml.gzs .xml.bz2( s .xmls .xml.gzs .xml.bz2( R R t joinR
t NoneR R t mapt tempfilet mkstempt saveR t unlinkt assertRaisest
ValueError( R t file_t fR t filest _[1]t fdR t output_pathst newchart newchar2( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testConstructorAndSave{ s, '
c C sB t i i | i d d } t } | i | | i | d S( NR s
character.xml( R R R R
R t readR ( R t fileR ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testReadXMLFile s
c C sH t i i | i d d } t } | i | d t | i | d S( NR s character.xml.gzipt gzip( R R R R
R R' t TrueR ( R R( R ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testReadXMLGzipFile s c C sH t i i | i d d } t } | i | d t | i | d S( NR s character.xml.bz2t bz2( R R R R
R R' R+ R ( R R( R ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testReadXMLBZ2File s c C sd t i i | i d d } t | } | i } | i t } | i | | i | d S( NR s
character.xml(
R R R R
t openR' t closeR t read_stringR ( R R( R t bufR ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testReadXMLString s
c C sj t i i | i d d } t | } | i } | i t } | i | d t | i
| d S( NR s character.xml.gzipR* ( R R R R
R/ R' R0 R R1 R+ R ( R R( t stringR ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testReadXMLGzipString s
c C sj t i i | i d d } t | } | i } | i t } | i | d t | i
| d S( NR s character.xml.bz2R- ( R R R R
R/ R' R0 R R1 R+ R ( R R( R4 R ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testReadXMLBZ2String s
c C s( t } d | _ d | _ d | _ | S( Ni i i ( R t xt yt timestamp( R t point( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt _getPoint s
c C s& | i } | i | i d d S( Ns# ( R; R t to_xml( R R: ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testPointToXML s c C sJ | i } | i t i | i h d d <d d <d d < d S( Ni u yi u timestampi u x( R; R t minjsonR' t to_json( R R: ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testPointToJSON s c C so t } d | _ d | _ d | _ t } d | _ d | _ d | _ t } | i | | i | | S( Ni i i i i g?( R R7 R8 R9 t pressureR t append_point( R R: t point2t stroke( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt
_getStroke s
c C s, | i } d } | i | | i d S( Ns_
( RE R R< ( R RD t expected( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testStrokeToXML s c C sz | i } h h d d <d d <d d R' R? ( R RD RF ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testStrokeToJSON s / c C s t } d | _ d | _ d | _ t } d | _ d | _ d | _ t } | i | | i | t } | i | | S( Ni i i i i g?( R R7 R8 R9 RA R RB R t
append_stroke( R R: RC RD t writing( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt _getWriting s
c C s, | i } d } | i | | i d S( Ns 1000
1000
( RK R R< ( R RJ RF ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testWritingToXML s c C s | i } h d d <d d <h h d d <d d <d d R' R? ( R RJ RF ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testWritingToJSON) s Oc C s3 | i } t } | i | | i d | S( Nt A( RK R t set_writingt set_utf8( R RJ R ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt
_getCharacter1 s
c C sU | i } t i } | i | t } | i | i | i | | d S( N( RQ t StringIOt writeR R1 t getvalueR ( R R t iot new_char( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testWriteXMLFile: s
c C s | i } h d d <h d d <d d <h h d d <d d <d
d R' R? ( R R RF ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testCharacterToJSONE s 6c C s t } | i d d | i d d | i d d | i d d | i d d | i d d | i d d | i d d | i d d | i } d
d d d
g d d g d d d g g } | i | | d S( Ni i i i i i i i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( i i ( R t move_tot line_toR R ( R RJ R RF ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testNewWritingO s c C s) t } d | _ d | _ d | _ t } d | _ d | _ d | _ t } | i | | i | t } d | _ d | _ d | _ t } d | _ d | _ d | _ t } | i | | i | | i | i d t } | i | | i | | i | i d d S( Ni i i i i i i
i (
R R7 R8 R9 R RB R t get_durationR RI ( R R: RC RD t point3t point4t stroke2RJ ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testDuratione s6
c C si t d d d d } t d d d d } t d d d d } | i | | j | i | | j d S( NR7 i R8 i i ( R t
assertTruet assertFalse( R t p1t p2t p3( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testPointEquality s
c C s? t d d d d } | i | d j | i | d j d S( NR7 i R8 i ( R Rb R Ra ( R Rc ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testPointEqualityNone s c C s8 t d d d d } | i } | i | | j d S( NR7 i R8 i ( R t copyRa ( R Rc Rd ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt
testPointCopy s c C s t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d | i | | j | i | | j d S( NR7 i R8 i i i ( R RB R Ra Rb ( R t s1t s2t s3( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testStrokeEquality s c C sk t } | i t d d d d | i t d d d d | i | d j | i | d j d S( NR7 i R8 i i ( R RB R Rb R Ra ( R Rj ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testStrokeEqualityNone s
c C sd t } | i t d d d d | i t d d d d | i } | i | | j d S( NR7 i R8 i i ( R RB R Rh Ra ( R Rj Rk ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testStrokeCopy s
c C s t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | | i | | | i | | d S( NR7 i R8 i i i ( R RB R R RI R t assertNotEqual( R Rj Rk t w1t w2t w3( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testWritingEquality s:
c C s t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | | i | d j | i | d j d S( NR7 i R8 i i ( R RB R R RI Ra R Rb ( R Rj Rk Rq ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testWritingEqualityNone s
c C s3 t } | i | d j | i | d j d S( N( R Ra R Rb ( R t c( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testCharacterEqualityNone s c C s t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | | i } | i | | j d S( NR7 i R8 i i ( R RB R R RI Rh Ra ( R Rj Rk Rq Rr ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testWritingCopy s
c C s& | i } | i | i d d S( Ni ( RK R t get_n_points( R RJ ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testGetNPoints s c C s t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | | i d | i | i d d g g d S( NR7 i R8 i i i ( i i ( i i ( R RB R R RI t
remove_strokeR R ( R Rj Rk t w( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testRemoveStroke s
c C s+ t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | t } | i t d d d d | i t d d d d | i d | | i | i d
d g d d
g d d g g d S( NR7 i R8 i i i i! i, i ( i i ( i i ( i i! ( i! i, ( i i ( i i ( R RB R R RI t
insert_strokeR R ( R Rj Rk R| Rl ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testInsertStroke! s
!c C s" t } | i t d d d d | i t d d d d t } | i t d d d d | i t d d d d t } | i | | i | t } | i t d d d d | i t d d d d | i d | | i | i d
d g d d
g g d S( NR7 i R8 i i i i! i, i ( i i ( i i ( i i! ( i! i, ( R RB R R RI t replace_strokeR R ( R Rj Rk R| Rl ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testReplaceStroke7 s
c C se t } | i t d d d d | i t d d d d | i | i t | d d S( NR7 i R8 i i i ( R RB R t clearR t len( R Rj ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testClearStrokeK s
c C s t i i | i d d } t | } | i } | i d } d } yF | i t i | | i
t i | | i
t i | Wn# t j
o t i
i d n Xd S( NR s
character.xmls
防
s
防
s lxml missing!
( R R R R
R/ R' R0 Ra R t validateRb t NotImplementedErrort syst stderrRS ( R R R R2 t invalidt malformed( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt testValidateS s
c C s t i i | i d d } t } | i | t t i i | i d d } | i i } | i | i | i
| d S( NR s
character.xmls character.sexp( R R R R
R R' R/ t stripR0 R t to_sexp( R R R t sexp( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyt
testToSexpx s
!
(* t __name__t
__module__R
R R& R) R, R. R3 R5 R6 R; R= R@ RE RG RH RK RL RM RQ RW RX R[ R` Rf Rg Ri Rm Rn Ro Rt Ru Rw Rx Rz R} R R R R R ( ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pyR sP T "
% ( %(
t unittestR R RR R> R t tegaki.characterR R R R t TestCaseR ( ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_character.pys s " tegaki-python-0.3.1/tests/test_mathutils.pyc 0000644 0001750 0001750 00000001672 11342122457 021107 0 ustar mathieu mathieu
gJc @ s3 d d k Z d d k Td e i f d YZ d S( iN( t *t MathTestc B s e Z d Z RS( c C s[ xT d d
d f d d d f f D]4 \ } } } t t | | } | i | | q Wd S(
Ni i
i i i g @i g ( i i
i ( i i
i ( i i ( i i ( t roundt euclidean_distancet assertEquals( t selft v1t v2t expectedt res( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_mathutils.pyt testEuclideanDistance s
( t __name__t
__module__R
( ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_mathutils.pyR s ( t unittestt tegaki.mathutilst TestCaseR ( ( ( sH /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_mathutils.pys s
tegaki-python-0.3.1/tests/charcol.chardb 0000644 0001750 0001750 00000010000 11342122457 020062 0 ustar mathieu mathieu SQLite format 3 @
: : b7!{indexcharacter_setid_indexcharactersCREATE INDEX character_setid_index ON characters(setid)e!!tablecharacterscharactersCREATE TABLE characters(
charid INTEGER PRIMARY KEY,
setid INTEGER REFERENCES character_sets,
utf8 TEXT,
n_strokes INTEGER,
data BLOB, -- gz xml
sha1 TEXT
)x))+tablecharacter_setscharacter_setsCREATE TABLE character_sets(
setid INTEGER PRIMARY KEY,
name TEXT
)
tegaki-python-0.3.1/tests/test_charcol.pyc 0000644 0001750 0001750 00000032211 11342122457 020501 0 ustar mathieu mathieu
Kc @ s d d k Z d d k Z d d k Z d d k Z d d k l Z l Z l Z l Z d d k l
Z
d e i f d YZ d S( iN( t Pointt Stroket Writingt Character( t CharacterCollectiont CharacterCollectionTestc B s
e Z d Z d Z d Z d Z d Z d Z d Z d Z d Z
d Z d
Z d Z
d Z d
Z d Z d Z d Z d Z d Z d Z d Z d Z d Z d Z d Z d Z d Z d Z d Z RS( c C s t i i t i i t | _ t i i | i d d d } t | _ | i i | t i i | i d d } t
| _ | i i | d S( Nt datat
collections test.charcols
character.xml( t ost patht dirnamet abspatht __file__t currdirt joinR t cct readR t c( t selfR t f( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt setUp s ! c C s t i i | i d d d } t | } | i } | i d } d } yF | i t i | | i
t i | | i
t i | Wn# t j
o t i
i d n Xd S( NR R s test.charcols
防
s6
s lxml missing!
( R R R R
t openR t closet
assertTrueR t validatet assertFalset NotImplementedErrort syst stderrt write( R R R t buft invalidt malformed( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testValidate) s
c C sW | i | i d d d d g h } xZ d d d d d d
g D]@ } t | | <| | i t i i | i d d | d
qA W| i | i d | d | d | d g | i | i d | d | d g | i | i d | d
g | i | i d g | i | i | d | d | d | d | d | d
g d S( Ns 一s 三s 二s 四t 19968_1t 19968_2t 19968_3t 19977_1t 19977_2t 20108_1R R s .xml(
t assertEqualst get_set_listR R R R R R
t get_characterst get_all_characters( R t charcolR t k( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt _testReadXMLI s$ "
c C s | i | i d S( N( R. R ( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testReadXMLFile^ s c C sa t } | i | i i | i | i i | i | i | i i | i d S( N( R t read_stringR t to_xmlR( R) R+ ( R t charcol2( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testToXMLa s
c C sm t } | i | i i d t d t | i | i i | i | i | i i | i d S( Nt gzip( R R0 R t write_stringt TrueR( R) R+ ( R R2 ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testWriteGzipStringh s
" c C sm t } | i | i i d t d t | i | i i | i | i | i i | i d S( Nt bz2( R R0 R R5 R6 R( R) R+ ( R R2 ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testWriteBz2Stringo s
" c C s t i i | i d d d } t } | i | t } | i | | i | d t } | i | i d d d d g | i t
| i d d | i t
| i d d
| i t
| i d d | i t
| i d d d S(
NR R s test.charcolt check_duplicates 一s 三s 二s 四i i i i ( R R R R
R R t concatenateR6 R( R) t lenR* ( R R R, R2 t charcol3( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testAddSamev s
" c C s | i i d } | i | i i d d d | d d ! | i | i i d d d | d | i | i i d d d d d | d d ! d S( Ns 一t limiti i t offseti ( R R* R( ( R t all_( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testGetChars s
)&!c
C s t i i | i d d d } t } | i | t i i | i d d d } t } | i | | | } | i | i d d d d d d
d d g | i t | i d d
| i t | i d d | i t | i d d | i t | i d d | i t | i d d
| i t | i d
d | i t | i d d | i t | i d d d S( NR R s test.charcols
test2.charcols 一s 三s 二s 四t at bR t di i i i (
R R R R
R R R( R) R< R* ( R R R, t path2R2 R= ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testAdd s$
c C s t i i | i d } t i | d t } | i | i d d d d g | i t | i
d d | i t | i
d d | i t | i
d d | i t | i
d d d S(
NR R: s 防s 三s 一s 二i i i ( R R R R
R t from_character_directoryR6 R( R) R< R* ( R t directoryR, ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testFromCharDirRecursive s " c C so t i i | i d } t i | d t d t } | i | i d g | i t
| i d d d S( NR t recursiveR: s 防i ( R R R R
R RH t FalseR6 R( R) R< R* ( R RI R, ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testFromCharDirNotRecursive s
c C s3 | i i d | i | i i d d g d S( Ns 一三s 一s 三( R t include_characters_from_textR( R) ( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testIncludeChars s c C s3 | i i d | i | i i d d g d S( Ns 三s 一s 二( R t exclude_characters_from_textR( R) ( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testExcludeChars s c C s | i i d } | i } | i | i d t } | d } | i | d } d | _ | i i d } | i | | d S( Ni t fulli
( R R+ t get_writingt normalizet get_strokesR6 t smootht xR( ( R t chart writingt strokest stroket pt char2( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testProxy s
c C s t | i _ | i i d } | i } | i | i d t } | d } | i | d } d | _ | i i d } | i
| | | i i | | i i d } | i | | d S( Ni RR i
(
RL R t
WRITE_BACKR+ RS RT RU R6 RV RW t assertNotEqualt update_character_objectR( ( R RX RY RZ R[ R\ R] ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testNoProxy s
c C s1 | i i d | i | i i d d d S( Nt totoi( R t add_setR( R) ( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt
testAddSet s c C sm | i i } | i i | d | i i } | i t | d t | | i | d d !| d S( Nii i ( R R) t
remove_setR( R< ( R t beforet after( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt
testRemoveSet s
c C sE | i t | i i | i i | i d | i i d S( Ni ( R( R< R R) t
get_n_sets( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testGetNSets s (c C sE | i t | i i | i i | i d | i i d S( Ni ( R( R< R R+ t get_total_n_characters( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testGetTotalNCharacters s c C s xE | i i D]4 } | i t | i i | | i i | q W| i | i i d d | i | i i d d | i | i i d d d S( Ns 一i s 三i s 二i ( R R) R( R< R* t get_n_characters( R t set_name( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testGetNCharacters s c C sR | i i d d d !} | i i d | | i i d } | i | | d S( Ns 一i i ( R R* t set_charactersR( ( R Rg Rh ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testSetCharacters s c C s^ t | i i d } | i i d | i t | i i d } | i | d | d S( Ns 一i ( R< R R* t append_characterR R( ( R t
len_beforet len_after( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testAppendCharacter s c C s | i i d d } t | i i d } | i i d d | i | i i d d } | i | | t | i i d } | i | d | d S( Ns 一i i ( R R* R< t insert_characterR R` t assertEqual( R Rg Rt Rh Ru ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testInsertCharacter s c C s | i i d d } t | i i d } | i i d d | i | i i d d } | i | | t | i i d } | i | | d S( Ns 一i ( R R* R< t replace_characterR R` Rx ( R Rg Rt Rh Ru ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testReplaceCharacter
s c C s | i i d d } t | i i d } | i i d d | i i d d } | i | | t | i i d } | i | d | d S( Ns 一i i ( R R* R< t remove_characterR` Rx ( R Rg Rt Rh Ru ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testRemoveCharacter s c C s | i i d d } t | i i d } | i i d | i i d d } | i | | t | i i d } | i | d | d S( Ns 一ii ( R R* R< t remove_last_characterR` Rx ( R Rg Rt Rh Ru ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testRemoveLastCharacter# s c C s | i i d d | i | i i d d | i | i i d d | i | i i d d | i i d d | i | i i d d | i | i i d d | i | i i d d d S( Nt keep_at_mosti s 一s 三s 二i ( R t remove_samplesR( Rn ( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testRemoveSamples. s c C s3 | i i | i | i i d d d g d S( Ns 一s 三s 二( R t remove_empty_setsR( R) ( R ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyt testRemoveEmptySets9 s
( t __name__t
__module__R R! R. R/ R3 R7 R9 R> RB RG RJ RM RO RQ R^ Rb Re Ri Rk Rm Rp Rr Rv Ry R{ R} R R R ( ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pyR s:
(
t unittestR R t StringIOt tegaki.characterR R R R t tegaki.charcolR t TestCaseR ( ( ( sF /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/test_charcol.pys s " tegaki-python-0.3.1/tests/test_character.py 0000644 0001750 0001750 00000050760 11342122457 020670 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import unittest
import os
import sys
import StringIO
import minjson
import tempfile
from tegaki.character import Point, Stroke, Writing, Character
class CharacterTest(unittest.TestCase):
def setUp(self):
self.currdir = os.path.dirname(os.path.abspath(__file__))
self.strokes = [[(210, 280), (213, 280), (216, 280), (220, 280), (223,
280), (226, 280), (230, 280), (233, 280), (236, 280), (243, 280), (246, 280),
(253, 280), (260, 280), (263, 280), (270, 283), (276, 283), (283, 286), (286,
286), (293, 290), (296, 290), (300, 290), (303, 290), (306, 290), (310, 290),
(313, 290), (316, 293), (320, 293), (320, 296), (320, 300), (316, 303), (316,
306), (316, 310), (316, 313), (313, 316), (313, 320), (310, 323), (310, 326),
(306, 330), (306, 333), (303, 340), (300, 346), (300, 350), (300, 356), (296,
360), (293, 366), (290, 366), (286, 373), (283, 376), (280, 380), (276, 386),
(273, 386), (273, 390), (270, 390), (270, 396), (266, 396), (263, 400), (260,
403), (256, 406), (253, 406), (253, 410), (250, 410), (250, 413), (246, 413),
(250, 413), (253, 413), (256, 413), (260, 413), (263, 413), (286, 423), (290,
423), (296, 423), (300, 423), (306, 426), (310, 426), (313, 426), (316, 426),
(316, 430), (320, 430), (323, 430), (326, 430), (326, 433), (330, 433), (333,
433), (333, 436), (336, 436), (336, 440), (336, 443), (336, 446), (336, 450),
(336, 453), (336, 456), (336, 460), (336, 463), (336, 466), (336, 470), (336,
476), (333, 480), (333, 483), (333, 486), (330, 490), (330, 496), (326, 496),
(326, 500), (326, 503), (323, 506), (323, 510), (320, 516), (316, 520), (316,
523), (313, 526), (310, 526), (306, 530), (306, 533), (303, 536), (300, 536),
(300, 540), (296, 546), (293, 546), (290, 553), (286, 556), (283, 556), (283,
560), (276, 563), (270, 566), (266, 566), (263, 573), (260, 573), (256, 576),
(253, 576), (250, 580), (250, 583), (246, 583), (243, 586), (240, 586), (240,
590), (236, 590), (233, 593), (230, 596), (226, 596), (220, 596), (220, 600),
(216, 600), (213, 600), (210, 603), (206, 603), (203, 603)], [(200, 276), (200,
280), (200, 283), (200, 286), (200, 290), (203, 293), (203, 296), (203, 300),
(206, 300), (206, 306), (206, 310), (206, 316), (206, 320), (210, 323), (210,
326), (210, 333), (210, 336), (210, 340), (210, 343), (210, 346), (213, 353),
(213, 356), (213, 360), (213, 363), (216, 363), (216, 370), (216, 373), (216,
376), (220, 380), (220, 386), (220, 393), (220, 396), (220, 400), (220, 403),
(220, 410), (220, 416), (220, 420), (220, 423), (220, 426), (220, 433), (220,
436), (220, 443), (220, 446), (220, 450), (220, 453), (220, 460), (220, 466),
(220, 470), (220, 473), (220, 476), (220, 483), (220, 486), (220, 490), (220,
493), (220, 496), (220, 500), (220, 503), (220, 506), (220, 510), (220, 513),
(220, 516), (220, 520), (220, 526), (220, 530), (223, 536), (223, 540), (223,
543), (223, 546), (223, 553), (223, 556), (223, 560), (223, 566), (223, 570),
(223, 573), (223, 576), (223, 580), (223, 583), (223, 590), (223, 593), (223,
596), (223, 603), (223, 606), (223, 613), (223, 616), (223, 620), (223, 626),
(223, 633), (223, 640), (223, 643), (223, 650), (223, 653), (223, 660), (223,
663), (223, 666), (223, 673), (223, 676), (223, 683), (223, 686), (223, 690),
(223, 693), (223, 696), (223, 700), (223, 706), (223, 710), (223, 713), (223,
720), (223, 723), (223, 726), (223, 730), (223, 736), (223, 740), (223, 746),
(223, 750), (223, 753), (223, 756), (223, 760), (223, 763), (223, 766), (223,
773), (223, 776), (223, 780)], [(493, 216), (493, 220), (496, 223), (496, 226),
(496, 230), (500, 233), (500, 236), (500, 240), (500, 243), (503, 246), (503,
250), (506, 253), (506, 256), (506, 260), (510, 263), (510, 266), (510, 270),
(510, 273)], [(370, 283), (373, 283), (376, 283), (380, 283), (386, 283), (390,
283), (400, 283), (403, 283), (413, 283), (423, 283), (426, 283), (436, 283),
(443, 283), (450, 283), (456, 283), (466, 283), (470, 283), (476, 283), (486,
283), (493, 283), (500, 283), (503, 283), (513, 283), (516, 283), (523, 283),
(526, 283), (533, 283), (536, 283), (540, 283), (546, 283), (550, 283), (583,
283), (586, 283), (593, 283), (596, 283), (600, 283), (606, 283), (610, 283),
(616, 283), (620, 283), (626, 283), (633, 283), (636, 283), (643, 283), (646,
283), (650, 283), (653, 283), (656, 283), (663, 283), (666, 283), (670, 283),
(673, 283), (676, 283), (680, 283), (683, 283), (686, 283), (690, 283), (693,
283), (696, 283), (700, 283), (703, 286)], [(530, 370), (536, 373), (540, 373),
(546, 373), (550, 373), (570, 380), (573, 380), (580, 380), (600, 386), (603,
386), (613, 386), (616, 386), (626, 386), (630, 386), (636, 386), (640, 386),
(646, 386), (650, 386), (653, 386), (656, 386), (656, 390), (656, 393), (660,
396), (660, 400), (660, 403), (660, 406), (663, 410), (663, 413), (663, 416),
(663, 420), (663, 423), (663, 426), (663, 430), (663, 433), (663, 436), (663,
440), (663, 446), (663, 450), (663, 456), (663, 460), (663, 463), (663, 470),
(663, 473), (663, 480), (663, 483), (663, 490), (660, 496), (660, 500), (660,
506), (660, 510), (660, 516), (656, 520), (656, 526), (656, 530), (656, 536),
(656, 543), (656, 546), (656, 553), (653, 556), (653, 563), (653, 566), (653,
570), (650, 573), (650, 576), (646, 583), (646, 586), (643, 590), (643, 593),
(640, 596), (640, 600), (636, 603), (636, 606), (633, 606), (633, 610), (630,
610), (626, 610), (623, 610), (623, 613), (620, 613), (616, 613), (613, 613),
(610, 613), (606, 613)], [(490, 293), (490, 296), (490, 300), (490, 303), (490,
306), (490, 310), (490, 316), (490, 320), (493, 323), (493, 330), (493, 336),
(493, 343), (493, 346), (493, 353), (493, 363), (493, 366), (493, 373), (493,
376), (493, 386), (493, 390), (493, 396), (493, 403), (493, 406), (493, 413),
(493, 416), (493, 423), (493, 426), (493, 433), (493, 436), (493, 443), (493,
453), (493, 456), (493, 463), (493, 470), (486, 490), (486, 520), (483, 530),
(483, 530), (483, 540), (480, 543), (480, 550), (480, 553), (476, 560), (476,
563), (476, 566), (476, 576), (476, 580), (473, 586), (473, 590), (460, 603),
(460, 606), (460, 613), (460, 620), (456, 626), (456, 636), (456, 640), (453,
646), (453, 650), (453, 656), (453, 660), (450, 666), (450, 673), (446, 676),
(443, 680), (443, 683), (443, 686), (440, 690), (440, 696), (436, 696), (436,
703), (433, 706), (433, 713), (430, 716), (430, 720), (426, 723), (426, 726),
(423, 730), (420, 736), (420, 740), (420, 743), (420, 746), (416, 746), (416,
750), (416, 753), (413, 756), (413, 760), (410, 760), (410, 763), (406, 763)]]
def _testReadXML(self, char):
self.assertEquals(char.get_utf8(), "防")
self.assertEquals(self.strokes, char.get_writing().get_strokes())
def testConstructorAndSave(self):
file_ = os.path.join(self.currdir, "data", "character.xml")
for f in (file_, file_ + ".gzip", file_ + ".bz2", None):
char = Character(f)
if f:
self._testReadXML(char) # check that it is correctly loaded
files = map(tempfile.mkstemp, (".xml", ".xml.gz", ".xml.bz2"))
output_paths = [path for fd,path in files]
for path in output_paths:
try:
# check that save with a path argument works
char.save(path)
newchar = Character(path)
self.assertEquals(char, newchar)
finally:
os.unlink(path)
try:
# check that save with a path argument works
newchar.save()
newchar2 = Character(path)
self.assertEquals(char, newchar2)
finally:
os.unlink(path)
char = Character()
self.assertRaises(ValueError, char.save)
def testReadXMLFile(self):
file = os.path.join(self.currdir, "data", "character.xml")
char = Character()
char.read(file)
self._testReadXML(char)
def testReadXMLGzipFile(self):
file = os.path.join(self.currdir, "data", "character.xml.gzip")
char = Character()
char.read(file, gzip=True)
self._testReadXML(char)
def testReadXMLBZ2File(self):
file = os.path.join(self.currdir, "data", "character.xml.bz2")
char = Character()
char.read(file, bz2=True)
self._testReadXML(char)
def testReadXMLString(self):
file = os.path.join(self.currdir, "data", "character.xml")
f = open(file)
buf = f.read()
f.close()
char = Character()
char.read_string(buf)
self._testReadXML(char)
def testReadXMLGzipString(self):
file = os.path.join(self.currdir, "data", "character.xml.gzip")
file = open(file)
string = file.read()
file.close()
char = Character()
char.read_string(string, gzip=True)
self._testReadXML(char)
def testReadXMLBZ2String(self):
file = os.path.join(self.currdir, "data", "character.xml.bz2")
file = open(file)
string = file.read()
file.close()
char = Character()
char.read_string(string, bz2=True)
self._testReadXML(char)
def _getPoint(self):
point = Point()
point.x = 1
point.y = 2
point.timestamp = 3
return point
def testPointToXML(self):
point = self._getPoint()
self.assertEquals(point.to_xml(), '')
def testPointToJSON(self):
point = self._getPoint()
self.assertEquals(minjson.read(point.to_json()),
{u'y': 2, u'timestamp': 3, u'x': 1})
def _getStroke(self):
point = Point()
point.x = 1
point.y = 2
point.timestamp = 3
point2 = Point()
point2.x = 4
point2.y = 5
point2.pressure = 0.1
stroke = Stroke()
stroke.append_point(point)
stroke.append_point(point2)
return stroke
def testStrokeToXML(self):
stroke = self._getStroke()
expected = """
"""
self.assertEquals(expected, stroke.to_xml())
def testStrokeToJSON(self):
stroke = self._getStroke()
expected = {u'points': [{u'y': 2, u'timestamp': 3, u'x': 1}, {u'y': 5,
u'pressure': 0, u'x': 4}]}
self.assertEquals(minjson.read(stroke.to_json()), expected)
def _getWriting(self):
point = Point()
point.x = 1
point.y = 2
point.timestamp = 3
point2 = Point()
point2.x = 4
point2.y = 5
point2.pressure = 0.1
stroke = Stroke()
stroke.append_point(point)
stroke.append_point(point2)
writing = Writing()
writing.append_stroke(stroke)
return writing
def testWritingToXML(self):
writing = self._getWriting()
expected = """1000
1000
"""
self.assertEquals(expected, writing.to_xml())
def testWritingToJSON(self):
writing = self._getWriting()
expected = {u'width': 1000, u'height': 1000, u'strokes': [{u'points':
[{u'y': 2, u'timestamp': 3, u'x': 1}, {u'y': 5, u'pressure': 0, u'x': 4}]}]}
self.assertEquals(minjson.read(writing.to_json()), expected)
def _getCharacter(self):
writing = self._getWriting()
char = Character()
char.set_writing(writing)
char.set_utf8("A")
return char
def testWriteXMLFile(self):
char = self._getCharacter()
io = StringIO.StringIO()
char.write(io)
new_char = Character()
new_char.read_string(io.getvalue())
self.assertEquals(char, new_char)
def testCharacterToJSON(self):
char = self._getCharacter()
expected = {u'utf8': u'A', u'writing': {u'width' : 1000,
u'height': 1000, 'strokes': [{u'points': [{u'y':
2, u'timestamp': 3, u'x': 1},
{u'y': 5, u'pressure': 0, u'x': 4}]}]}}
self.assertEquals(minjson.read(char.to_json()), expected)
def testNewWriting(self):
writing = Writing()
writing.move_to(0,0)
writing.line_to(1,1)
writing.line_to(2,2)
writing.line_to(3,3)
writing.move_to(4,4)
writing.line_to(5,5)
writing.move_to(6,6)
writing.line_to(7,7)
writing.line_to(8,8)
strokes = writing.get_strokes()
expected = [ [(0, 0), (1,1), (2,2), (3,3)],
[(4,4), (5,5)],
[(6,6), (7,7), (8,8)] ]
self.assertEquals(strokes, expected)
def testDuration(self):
point = Point()
point.x = 1
point.y = 2
point.timestamp = 0
point2 = Point()
point2.x = 4
point2.y = 5
point2.timestamp = 5
stroke = Stroke()
stroke.append_point(point)
stroke.append_point(point2)
point3 = Point()
point3.x = 1
point3.y = 2
point3.timestamp = 7
point4 = Point()
point4.x = 4
point4.y = 5
point4.timestamp = 10
stroke2 = Stroke()
stroke2.append_point(point3)
stroke2.append_point(point4)
self.assertEquals(stroke2.get_duration(), 3)
writing = Writing()
writing.append_stroke(stroke)
writing.append_stroke(stroke2)
self.assertEquals(writing.get_duration(), 10)
def testPointEquality(self):
p1 = Point(x=2, y=3)
p2 = Point(x=2, y=3)
p3 = Point(x=2, y=4)
self.assertTrue(p1 == p2)
self.assertFalse(p1 == p3)
def testPointEqualityNone(self):
p1 = Point(x=2, y=3)
self.assertFalse(p1 == None)
self.assertTrue(p1 != None)
def testPointCopy(self):
p1 = Point(x=2, y=3)
p2 = p1.copy()
self.assertTrue(p1 == p2)
def testStrokeEquality(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
s3 = Stroke()
s3.append_point(Point(x=2, y=3))
s3.append_point(Point(x=4, y=5))
self.assertTrue(s1 == s2)
self.assertFalse(s1 == s3)
def testStrokeEqualityNone(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
self.assertFalse(s1 == None)
self.assertTrue(s1 != None)
def testStrokeCopy(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = s1.copy()
self.assertTrue(s1 == s2)
def testWritingEquality(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
w1 = Writing()
w1.append_stroke(s1)
w1.append_stroke(s2)
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
w2 = Writing()
w2.append_stroke(s1)
w2.append_stroke(s2)
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=5))
w3 = Writing()
w3.append_stroke(s1)
w3.append_stroke(s2)
self.assertEquals(w1, w2)
self.assertNotEqual(w1, w3)
def testWritingEqualityNone(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
w1 = Writing()
w1.append_stroke(s1)
w1.append_stroke(s2)
self.assertTrue(w1 != None)
self.assertFalse(w1 == None)
def testCharacterEqualityNone(self):
c = Character()
self.assertTrue(c != None)
self.assertFalse(c == None)
def testWritingCopy(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
w1 = Writing()
w1.append_stroke(s1)
w1.append_stroke(s2)
w2 = w1.copy()
self.assertTrue(w1 == w2)
def testGetNPoints(self):
writing = self._getWriting()
self.assertEquals(writing.get_n_points(), 2)
def testRemoveStroke(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
w = Writing()
w.append_stroke(s1)
w.append_stroke(s2)
w.remove_stroke(1)
self.assertEquals(w.get_strokes(), [[(2,3),(3,4)]])
def testInsertStroke(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
w = Writing()
w.append_stroke(s1)
w.append_stroke(s2)
s3 = Stroke()
s3.append_point(Point(x=22, y=33))
s3.append_point(Point(x=33, y=44))
w.insert_stroke(1, s3)
self.assertEquals(w.get_strokes(), [[(2,3),(3,4)], [(22,33),(33,44)],
[(2,3),(3,4)]])
def testReplaceStroke(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s2 = Stroke()
s2.append_point(Point(x=2, y=3))
s2.append_point(Point(x=3, y=4))
w = Writing()
w.append_stroke(s1)
w.append_stroke(s2)
s3 = Stroke()
s3.append_point(Point(x=22, y=33))
s3.append_point(Point(x=33, y=44))
w.replace_stroke(1, s3)
self.assertEquals(w.get_strokes(), [[(2,3),(3,4)],[(22,33),(33,44)]])
def testClearStroke(self):
s1 = Stroke()
s1.append_point(Point(x=2, y=3))
s1.append_point(Point(x=3, y=4))
s1.clear()
self.assertEquals(len(s1), 0)
def testValidate(self):
path = os.path.join(self.currdir, "data", "character.xml")
f = open(path)
buf = f.read()
f.close()
invalid = \
"""
防
"""
malformed = \
"""
防
"""
try:
self.assertTrue(Character.validate(buf))
self.assertFalse(Character.validate(invalid))
self.assertFalse(Character.validate(malformed))
except NotImplementedError:
sys.stderr.write("lxml missing!\n")
pass
def testToSexp(self):
f = os.path.join(self.currdir, "data", "character.xml")
char = Character()
char.read(f)
f = open(os.path.join(self.currdir, "data", "character.sexp"))
sexp = f.read().strip()
f.close()
self.assertEquals(char.to_sexp(), sexp)
tegaki-python-0.3.1/tests/data/ 0000755 0001750 0001750 00000000000 11352066723 016230 5 ustar mathieu mathieu tegaki-python-0.3.1/tests/data/character.xml 0000755 0001750 0001750 00000041766 11342122457 020723 0 ustar mathieu mathieu
防
tegaki-python-0.3.1/tests/data/character.xml.bz2 0000644 0001750 0001750 00000002553 11342122457 021403 0 ustar mathieu mathieu BZh91AY&SY%q y_TP /` @ ` _p `p* @x qAU* @z LQ@ 4041h0
4h4zѵ hɦH I Q^ԍݶ|Vfsd+9V,QJP@ eUVV Uj,Y*gV2̪$U-]Fj ZM"ЉԵ aXQZ-[XMkX[]i6Uт$U[hʕr[%'5YHTHQJ
U"U()lҵY\imj!hc0 ZjdҨhdŵh(VJH hڵsEPkZֵ)JҔMQ S ~/ 7y\TYJ$DI,&
Dv#$ A$"1df
Z2
M.Ik}ގ$8ĢTޫ}̉{fff!2w{- mtx+C]5zܟ9t<49} m sk/9
0KZ uP$ʽlݜCbCh+^* +}i9f$ Qn[ޓY$#m/3$ܙ2܋$шDI $ Io{ZI$uv$Hf9bl0E n5먂fomC"janS֗.Owndq$#Ice)U-Yotazo[EɨgA%+d&xBP3 nI%&Ϭׯ%'7=[M{$9ʧeHDL >ޖMz$DMVNd#x <]i$l$@]ۆd> 8Ɋ' !n 6
Q0\giuxwuicoxOVM4:8Yho3"l6[in9Oh#dfHm gw}wCjm@m=5< Oؚ 6%/weeK
Z&goCmkWy}T9&nYqSNRU`*516;}YhKM gMYc_Oj
_ڋ[ѣQFe6)LlK3Lѩff&#c*RS53)&4fSLRRK&L̕)26
4%%Fj1IISi4QRX5&$" ]ckmZV5+kϕL68\. 7fH.Ґ : +--%4$;BLH4H ~.p 7JJ tegaki-python-0.3.1/tests/data/collection/ 0000755 0001750 0001750 00000000000 11352066723 020363 5 ustar mathieu mathieu tegaki-python-0.3.1/tests/data/collection/19968_2.xml 0000755 0001750 0001750 00000003244 11342122457 022030 0 ustar mathieu mathieu
一
tegaki-python-0.3.1/tests/data/collection/19968_3.xml 0000755 0001750 0001750 00000005406 11342122457 022033 0 ustar mathieu mathieu
一
tegaki-python-0.3.1/tests/data/collection/19968_1.xml 0000755 0001750 0001750 00000003016 11342122457 022024 0 ustar mathieu mathieu
一
800
900
tegaki-python-0.3.1/tests/data/collection/19977_2.xml 0000755 0001750 0001750 00000011502 11342122457 022024 0 ustar mathieu mathieu
三
tegaki-python-0.3.1/tests/data/collection/test2.charcol 0000644 0001750 0001750 00000005740 11342122457 022763 0 ustar mathieu mathieu
a
800
900
a
a
b
b
c
tegaki-python-0.3.1/tests/data/collection/test.charcol 0000644 0001750 0001750 00000046003 11342122457 022676 0 ustar mathieu mathieu
一
800
900
一
一
三
三
二
tegaki-python-0.3.1/tests/data/collection/20108_1.xml 0000755 0001750 0001750 00000005547 11342122457 022011 0 ustar mathieu mathieu
二
tegaki-python-0.3.1/tests/data/collection/19977_1.xml 0000755 0001750 0001750 00000012726 11342122457 022034 0 ustar mathieu mathieu
三
tegaki-python-0.3.1/tests/data/character.xml.gzip 0000644 0001750 0001750 00000003413 11342122457 021653 0 ustar mathieu mathieu `H Qr\5EY7~*'c1l1T `W;VJΜ _ɸNHݭիwz7w/w,g7w?}˯ή~}? ~<.o ~y> ~Qf@7IMD3fYXB7 a"!!%rXA^q-"ؙ$/a&[0 <1<͇ $$v(Z=J$@c[h$GU|8t1p g 2"I
- NACN0,w@nY` a^ 0 洁5L6aN~ aNg'd9EN6HK%~\o%b,.6b0}ͶJ?w@4 Sσ= Dp@1G J 3!S@30l,fa5 ,tٳHU & yjj$ z JHScXL0 :@ &`7H'`0G`0pi-.qdf^62icf @w0 ty_.
@+,OfGd_Ђ\L %'l}6&;}q16wFyp0 {#ŀILb]f=$z wH`GQT0 | <0 e E4M,P,j*T JaP
Xrsfjf Y0F yVryeVXw2@EzpnS]}d'@ fT4j p5C:y'@!{^0Ķ0@րaf fصP &QnbB>:LDm
kQ,7
'5>9
iGm
kkXGm
L$FM N2`d OS 5c C s6 t d } x | i | o | d } q W| i } t i d | } t i d | } t i | } | o t i t | } n t | t
o
| } nV | o t
| | } n< y t
| } Wn( t j
o t } t
| | } n Xy t
| i } Wn t j
o t d | n X| S( s read the js, first sanitizing a bit and removing any c-style comments
If the input is a unicode string, great. That's preferred. If the input
is a byte string, strings in the object will be produced as unicode anyway.
i it s Unacceptable JSON expression: %s( t chrt endswitht stript slashstarcommentR t doubleslashcommentt unichrRER R R* t unicodet UnicodeDecodeErrort emergencyEncodingR R] R) R
( t aStringt encodingt CHR0t unicodecharst st encR% ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyt safeRead s0
N( t StringIOc C sI t | t o/ | i | i } d t | | i d f Sn | d S( s Error handler for json
If encoding fails, \uxxxx must be emitted. This
is similar to the "backshashreplace" handler, only
that we never emit \xnn since this is not legal
according to the JSON syntax specs.
u \u%04xi N( R* t UnicodeEncodeErrort objectt startt ord( t excRZ ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyt jsonreplace_handler= s t jsonreplacec C s, t d | d | } | i | | i S( Nt input_encodingt output_encoding( t
JsonWritert writet getvalue( t inputRj t outputEncodingt writer( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR{ Q s
s "|| |
|
| |\\c C sP h d d <d d <d d <d d <d d
<d d <d
d <| i d S( Ns \"R s \bs s \fs s \ns
s \rs
s \ts s \\s \i ( R ( R ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyt func_strmangleX s c C s t i t | S( N( t re_strmangleR R ( t text( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyt strmanglec s t
JsonStreamc B s# e Z d Z d Z d Z RS( c C s
g | _ d S( N( t buf( R$ ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR& h s c C s | i i | d S( N( R R- ( R$ R ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR{ k s c C s d i | i S( NR_ ( RI R ( R$ ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR| n s ( R R R& R{ R| ( ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR f s Rz c B s, e Z d d d d Z d Z d Z RS( s utf-8c C sS | d j o | d j o
t d q4 n
t } | | _ | | _ | | _ d S( s
- stream is optional, if specified must also give output_encoding
- The input strings can be unicode or in input_encoding
- output_encoding is optional, if omitted, result will be unicode
s; If a stream is given, output encoding must also be providedN( R! R R t streamRx Ry ( R$ R Rx Ry ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR& s s
c C s t | t t f ol | i i d t } x< | D]4 } | o
t } n | i i d | i | q3 W| i i d f n%t | t o | i i d t } xe | i D]W \ } } | o
t } n | i i d | i | | i i d | i | q W| i i d n| t j o | i i d nb| t j o | i i d nA| d j o | i i d n t | t
pR y t | } Wn+ t j
o } t
d
| i | f n X| i i | n t | t pH y t | | i } WqAt t f j
o t | d d } qAXn t | } | i d j o | i | i d
} n | i i d | i i | | i i d d S( NR< RA R> R= R@ R? RC RD RE s Cannot write object (%s: %s)s utf-8RH Rw R ( R* t listt tupleR R{ RL R t dictt iteritemsR! R4 t strt ExceptionR t __class__Rf Rx Rg t UnicodeTranslateErrorR Ry t encode( R$ t objt firstt elemR7 R5 Ru ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR{ sZ
c C s
| i i S( N( R R| ( R$ ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyR| s N( R R R! R& R{ R| ( ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pyRz q s 4( s
s \n( s s \b( s s \f( s s \t( s
s \r( R s \"( ( ( ( (, t reR R R R RO R R R R R R R
R R Rh R R
R Rc Rd Re R RG R R R R Rr R R! Ro t readt codecst cStringIORp Rv t register_errorR{ R R R R Rz ( ( ( sA /home/mathieu/Desktop/projects/hwr/tegaki-python/tests/minjson.pys 3 s8 ". 1 tegaki-python-0.3.1/tests/test_mathutils.py 0000644 0001750 0001750 00000002301 11342122457 020732 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import unittest
from tegaki.mathutils import *
class MathTest(unittest.TestCase):
def testEuclideanDistance(self):
for v1, v2, expected in ( ( (2, 10, 12), (3, 10, 7), 5.0 ),
( (5, 5), (5, 5), 0.0)
):
res = round(euclidean_distance(v1, v2))
self.assertEquals(res, expected)
tegaki-python-0.3.1/AUTHORS 0000644 0001750 0001750 00000000057 11342122456 015222 0 ustar mathieu mathieu Mathieu Blondel tegaki-python-0.3.1/COPYING 0000644 0001750 0001750 00000043262 11342122457 015213 0 ustar mathieu mathieu GNU GENERAL PUBLIC LICENSE
Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc.
59 Temple Place, Suite 330, Boston, MA 02111-1307 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 Library 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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 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 Library General
Public License instead of this License.
tegaki-python-0.3.1/PKG-INFO 0000644 0001750 0001750 00000000361 11352066723 015252 0 ustar mathieu mathieu Metadata-Version: 1.0
Name: tegaki-python
Version: 0.3.1
Summary: Tegaki Python library
Home-page: http://www.tegaki.org
Author: Mathieu Blondel
Author-email: mathieu ÂT mblondel DÔT org
License: GPL
Description: UNKNOWN
Platform: UNKNOWN
tegaki-python-0.3.1/tegaki/ 0000755 0001750 0001750 00000000000 11352066723 015421 5 ustar mathieu mathieu tegaki-python-0.3.1/tegaki/charcol.py 0000644 0001750 0001750 00000103252 11342122456 017404 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2009 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import sqlite3
import base64
import tempfile
import re
import os
from tegaki.dictutils import SortedDict
from tegaki.character import _XmlBase, Point, Stroke, Writing, Character
def _dict_factory(cursor, row):
d = {}
for idx, col in enumerate(cursor.description):
d[col[0]] = d[idx] = row[idx]
return d
class ObjectProxy(object):
"""
An object that forwards all attribute and method calls to another object.
Object proxies are used to automatically reflect back in the db
changes that are made to objects. For example:
>>> char = charcol.get_all_characters()[0]
>>> char.set_utf8(newvalue) # will be automatically changed in the db
"""
WRITE_METHODS = []
READ_METHODS = []
WRITE_ATTRIBUTES = []
def __init__(self, charpool, obj, charobj=None):
self._charpool = charpool
# the object to redirect attributes and method calls to
self._obj = obj
# the original character object
self._charobj = obj if charobj is None else charobj
def __getattr__(self, attr_):
attr = getattr(self._obj, attr_)
write = False
if attr_ in self.WRITE_METHODS:
write = True
elif not attr_ in self.READ_METHODS:
return attr
def wrapper(*args, **kw):
if write: self._charpool.add_char(self._charobj)
return _apply_proxy(self._charpool, attr(*args, **kw),
self._charobj)
return wrapper
def __setattr__(self, attr, value):
if attr in self.WRITE_ATTRIBUTES:
self._charpool.add_char(self._charobj)
setattr(self._obj, attr, value)
self.__dict__[attr] = value
def __eq__(self, othr):
if othr.__class__.__name__.endswith("Proxy"):
othr = othr._obj
return self._obj == othr
def __ne__(self, othr):
return not(self == othr)
class PointProxy(ObjectProxy):
"""
Proxy to Point.
"""
WRITE_METHODS = ["resize", "move_rel", "copy_from"]
WRITE_ATTRIBUTES = Point.KEYS
def __getitem__(self, x):
return self._obj[x]
class StrokeProxy(ObjectProxy):
"""
Proxy to Stroke.
"""
WRITE_METHODS = ["append_point", "insert", "smooth", "clear",
"downsample", "downsample_threshold",
"upsample", "upsample_threshod"]
READ_METHODS = []
def __getitem__(self, i):
return _apply_proxy(self._charpool, self._obj[i], self._charobj)
def __len__(self):
return len(self._obj)
class WritingProxy(ObjectProxy):
"""
Proxy to Writing.
"""
# Note: Some method calls need not be mentioned below
# because they automatically update the db thanks to
# Point and Stroke methods that are being used in their implementation.
WRITE_METHODS = ["clear", "move_to_point", "line_to_point",
"set_width", "set_height", "remove_stroke"]
READ_METHODS = ["get_strokes"]
class CharacterProxy(ObjectProxy):
"""
Proxy to Writing.
"""
WRITE_METHODS = ["set_utf8", "set_unicode", "set_writing",
"read", "read_string"]
READ_METHODS = ["get_writing"]
def __repr__(self):
return "" % (str(self.get_utf8()), id(self))
OBJ_PROXY = {Character: CharacterProxy,
Writing : WritingProxy,
Stroke : StrokeProxy,
Point : PointProxy}
def _apply_proxy(charpool, obj, charobj):
return _apply_proxy_rec(charpool, obj, charobj)
def _apply_proxy_rec(charpool, obj, charobj, reclevel=0):
try:
return OBJ_PROXY[obj.__class__](charpool, obj, charobj)
except KeyError:
if (isinstance(obj, tuple) or isinstance(obj, list)) and reclevel <= 1:
return [_apply_proxy_rec(charpool, ele, charobj, reclevel+1) \
for ele in obj]
else:
return obj
class _CharPool(dict):
"""
Holds characters that need be updated.
We don't want changes to be immediately reflected back to the db
for performance reasons. The _CharPool keeps track of what objects need
be updated.
"""
def __init__(self, cursor):
self._c = cursor
def add_char(self, char):
self[char.charid] = char
def _update_character(self, char):
self._c.execute("""UPDATE characters
SET utf8=?, n_strokes=?, data=?, sha1=?
WHERE charid=?""", (char.get_utf8(), char.get_writing().get_n_strokes(),
_adapt_character(char), char.hash(), char.charid))
def clear_pool_threshold(self, threshold=100):
if len(self) > threshold:
self.clear_pool()
def clear_pool(self):
for charid, char in self.items():
self._update_character(char)
self.clear()
def _convert_character(data):
# converts a BLOB into an object
char = Character()
char.read_string(base64.b64decode(data), gzip=True)
return char
def _adapt_character(char):
# converts an object into a BLOB
return base64.b64encode(char.write_string(gzip=True))
def _gzipbz2(path):
return (True if path.endswith(".gz") or path.endswith(".gzip") else False,
True if path.endswith(".bz2") or path.endswith(".bzip2") else False)
class CharacterCollection(_XmlBase):
"""
A collection of L{Characters}.
A CharacterCollection is composed of sets.
Each set can be composed of zero, one, or more characters.
/!\ Sets do not necessarily contain only characters of the same class
/ utf8 value. Sets may also be used to group characters in other fashions
(e.g. by number of strokes, by handwriting quality, etc...).
Therefore the set name is not guaranteed to contain the utf8 value of
the characters of that set. The utf8 value must be retrieved from each
character individually.
Building character collection objects
=====================================
A character collection can be built from scratch progmatically:
>>> char = Character()
>>> charcol = CharacterCollection()
>>> charcol.add_set("my set")
>>> charcol.append_character("my set", char)
Reading XML files
=================
A character collection can be read from an XML file:
>>> charcol = CharacterCollection()
>>> charcol.read("myfile")
Gzip-compressed and bzip2-compressed XML files can also be read:
>>> charcol = CharacterCollection()
>>> charcol.read("myfilegz", gzip=True)
>>> charcol = Character()
>>> charcol.read("myfilebz", bz2=True)
A similar method read_string exists to read the XML from a string
instead of a file.
For convenience, you can directly load a character collection by passing it
the file to load. In that case, compression is automatically detected based
on file extension (.gz, .bz2).
>>> charcol = Character("myfile.xml.gz")
The recommended extension for XML character collection files is .charcol.
Writing XML files
=================
A character collection can be saved to an XML file by using the write()
method.
>>> charcol.write("myfile")
The write method has gzip and bz2 arguments just like read(). In addition,
there is a write_string method which generates a string instead of a file.
For convenience, you can save a character collection with the save() method.
It automatically detects compression based on the file extension.
>>> charcol.save("mynewfile.xml.bz2")
If the CharacterCollection object was passed a file when it was constructed,
the path can ce omitted.
>>> charcol = Character("myfile.gz")
>>> charcol.save()
Using .chardb files
===================
XML files allow to retain human-readability and are ideal for small
character collections. However, they force the whole database to be kept
in memory. For larger collections, it's recommended to use .chardb files
instead. Their loading is faster and the whole collection doesn't
need be kept entirely in memory. However human-readability ist lost.
>>> charcol = CharacterCollection("charcol.chardb")
[...]
>>> charcol.save()
The .chardb extension is required.
"""
#: With WRITE_BACK set to True, proxy objects are returned in place of
#: character, writing, stroke and point objects in order to automatically
#: reflect changes to these objects back to the sqlite db.
#: However, there is probably overhead usigng them.
WRITE_BACK = True
def get_auto_commit(self):
return True if self._con.isolation_level is None else False
def set_auto_commit(self, auto):
self._con.isolation_level = None if auto else ""
#: With AUTO_COMMIT set to true, data is immediately written to disk
AUTO_COMMIT = property(get_auto_commit, set_auto_commit)
DTD = \
"""
"""
def __init__(self, path=":memory:"):
"""
Construct a collection.
@type path: str
@param path: an XML file or a DB file (see also L{bind})
"""
if path is None:
path = ":memory:"
if not path in ("", ":memory:") and not path.endswith(".chardb"):
# this should be an XML character collection
gzip, bz2 = _gzipbz2(path)
self.bind(":memory:")
self.read(path, gzip=gzip, bz2=bz2)
self._path = path # contains the path to the xml file
else:
# this should be either a .chardb, ":memory:" or ""
self.bind(path)
self._path = None
# DB utils
def _e(self, req, *a, **kw):
self._charpool.clear_pool()
#print req, a, kw
return self._c.execute(req, *a, **kw)
def _em(self, req, *a, **kw):
self._charpool.clear_pool()
#print req, a, kw
return self._c.executemany(req, *a, **kw)
def _fo(self):
return self._c.fetchone()
def _fa(self):
return self._c.fetchall()
def _efo(self, req, *a, **kw):
self._e(req, *a, **kw)
return self._fo()
def _efa(self, req, *a, **kw):
self._e(req, *a, **kw)
return self._fa()
def _has_tables(self):
self._e("SELECT count(type) FROM sqlite_master WHERE type = 'table'")
return self._fo()[0] > 0
def _create_tables(self):
self._c.executescript("""
CREATE TABLE character_sets(
setid INTEGER PRIMARY KEY,
name TEXT
);
CREATE TABLE characters(
charid INTEGER PRIMARY KEY,
setid INTEGER REFERENCES character_sets,
utf8 TEXT,
n_strokes INTEGER,
data BLOB, -- gz xml
sha1 TEXT
);
CREATE INDEX character_setid_index ON characters(setid);
""")
def get_character_from_row(self, row):
# charid, setid, utf8, n_strokes, data, sha1
char = _convert_character(row['data'])
char.charid = row['charid']
if self.WRITE_BACK:
return CharacterProxy(self._charpool, char)
else:
return char
def _update_set_ids(self):
self._SETIDS = SortedDict()
for row in self._efa("SELECT * FROM character_sets ORDER BY setid"):
self._SETIDS[row['name'].encode("utf8")] = row['setid']
# Public API
def __repr__(self):
return "" % \
(self.get_total_n_characters(), id(self))
def bind(self, path):
"""
Bind database to a db file.
All changes to the previous binded database will be lost
if you haven't committed changes with commit().
@type path: str
Possible values for path:
":memory:" for fully in memory database
"" for a in memory database that uses
temp files under pressure
"/path/to/file.chardb" for file-based database
"""
self._con = sqlite3.connect(path)
self._con.text_factory = sqlite3.OptimizedUnicode
self._con.row_factory = _dict_factory #sqlite3.Row
self._c = self._con.cursor()
self._charpool = _CharPool(self._c)
if not self._has_tables():
self._create_tables()
self._update_set_ids()
self._dbpath = path
def get_db_filename(self):
"""
Returns the db file which is internally used by the collection.
@rtype: str or None
@return: file path or None if in memory db
"""
return None if self._dbpath in (":memory:", "") else self._dbpath
def commit(self):
"""
Commit changes since last commit.
"""
self._charpool.clear_pool()
self._con.commit()
def save(self, path=None):
"""
Save collection to a file.
@type path: str
@param path: path where to write the file or None if use the path \
that was given to the constructor
If path ends with .chardb, it's saved as binary db file. Otherwise, it
will be saved as XML.
In the latter case, the file extension is used to determine whether the
file must be saved as plain, gzip-compressed or bzip2-compressed XML.
If path is omitted, the path that was given to the CharacterCollection
constructor is used.
"""
if path is None:
if self._path is not None:
# an XML file was provided to constructor
gzip, bz2 = _gzipbz2(self._path)
self.write(self._path, gzip=gzip, bz2=bz2)
else:
if path.endswith(".chardb"):
if self._dbpath != path:
# the collection changed its database name
# FIXME: this can rewritten more efficiently with
# the ATTACH command
if os.path.exists(path):
os.unlink(path)
newcc = CharacterCollection(path)
newcc.merge([self])
newcc.commit()
del newcc
self.bind(path)
else:
gzip, bz2 = _gzipbz2(path)
self.write(path, gzip=gzip, bz2=bz2)
self.commit()
@staticmethod
def from_character_directory(directory,
extensions=["xml", "bz2", "gz"],
recursive=True,
check_duplicate=False):
"""
Creates a character collection from a directory containing
individual character files.
"""
regexp = re.compile("\.(%s)$" % "|".join(extensions))
charcol = CharacterCollection()
for name in os.listdir(directory):
full_path = os.path.join(directory, name)
if os.path.isdir(full_path) and recursive:
charcol += CharacterCollection.from_character_directory(
full_path, extensions)
elif regexp.search(full_path):
char = Character()
gzip = False; bz2 = False
if full_path.endswith(".gz"): gzip = True
if full_path.endswith(".bz2"): bz2 = True
try:
char.read(full_path, gzip=gzip, bz2=bz2)
except ValueError:
continue # ignore malformed XML files
utf8 = char.get_utf8()
if utf8 is None: utf8 = "Unknown"
charcol.add_set(utf8)
if not check_duplicate or \
not char in charcol.get_characters(utf8):
charcol.append_character(utf8, char)
return charcol
def concatenate(self, other, check_duplicate=False):
"""
Merge two charcols together and return a new charcol
@type other: CharacterCollection
"""
new = CharacterCollection()
new.merge([self, other], check_duplicate=check_duplicate)
return new
def merge(self, charcols, check_duplicate=False):
"""
Merge several charcacter collections into the current collection.
@type charcols: list
@param charcols: a list of CharacterCollection to merge
"""
try:
# it's faster to delete the whole index and rewrite it afterwards
self._e("""DROP INDEX character_setid_index;""")
for charcol in charcols:
for set_name in charcol.get_set_list():
self.add_set(set_name)
if check_duplicate:
existing_chars = self.get_characters(set_name)
chars = charcol.get_characters(set_name)
chars = [c for c in chars if not c in existing_chars]
self.append_characters(set_name, chars)
else:
chars = charcol.get_character_rows(set_name)
self.append_character_rows(set_name, chars)
finally:
self._e("""CREATE INDEX character_setid_index
ON characters(setid);""")
def __add__(self, other):
return self.concatenate(other)
def add_set(self, set_name):
"""
Add a new set to collection.
@type set_name: str
"""
self.add_sets([set_name])
def add_sets(self, set_names):
"""
Add new sets to collection.
@type set_names: list of str
"""
set_names = [(set_name,) for set_name in set_names \
if not set_name in self._SETIDS]
self._em("INSERT INTO character_sets(name) VALUES (?)", set_names)
self._update_set_ids()
def remove_set(self, set_name):
"""
Remove set_name from collection.
@type set_name: str
"""
self.remove_sets([set_name])
def remove_sets(self, set_names):
"""
Remove set_name from collection.
@type set_name: str
"""
set_names = [(set_name,) for set_name in set_names]
self._em("DELETE FROM character_sets WHERE name=?", set_names)
self._update_set_ids()
def get_set_list(self):
"""
Return the sets available in collection.
@rtype: list of str
"""
return self._SETIDS.keys()
def get_n_sets(self):
"""
Return the number of sets available in collection.
@rtype: int
"""
return len(self._SETIDS)
def get_characters(self, set_name, limit=-1, offset=0):
"""
Return character belonging to a set.
@type set_name: str
@param set_name: the set characters belong to
@type limit: int
@param limit: the number of characters needed or -1 if all
@type offset: int
@param offset: the offset to start from (0 if from beginning)
@rtype: list of L{Character}
"""
return list(self.get_characters_gen(set_name, limit, offset))
def get_characters_gen(self, set_name, limit=-1, offset=0):
"""
Return a generator to iterate over characters. See L{get_characters).
"""
rows = self.get_character_rows(set_name, limit, offset)
return (self.get_character_from_row(r) for r in rows)
def get_character_rows(self, set_name, limit=-1, offset=0):
i = self._SETIDS[set_name]
self._e("""SELECT * FROM characters
WHERE setid=? ORDER BY charid LIMIT ? OFFSET ?""", (i, int(limit), int(offset)))
return self._fa()
def get_random_characters(self, n):
"""
Return characters at random.
@type n: int
@param n: number of random characters needed.
"""
return list(self.get_random_characters_gen(n))
def get_random_characters_gen(self, n):
"""
Return a generator to iterate over random characters. See \
L{get_random_characters).
"""
self._e("""SELECT DISTINCT * from characters
ORDER BY RANDOM() LIMIT ?""", (int(n),))
return (self.get_character_from_row(r) for r in self._fa())
def get_n_characters(self, set_name):
"""
Return the number of character belonging to a set.
@type set_name: str
@param set_name: the set characters belong to
@rtype int
"""
try:
i = self._SETIDS[set_name]
return self._efo("""SELECT count(charid) FROM characters
WHERE setid=?""", (i,))[0]
except KeyError:
return 0
def get_all_characters(self, limit=-1, offset=0):
"""
Return all characters in collection.
@type limit: int
@param limit: the number of characters needed or -1 if all
@type offset: int
@param offset: the offset to start from (0 if from beginning)
@rtype: list of L{Character}
"""
return list(self.get_all_characters_gen(limit=-1, offset=0))
def get_all_characters_gen(self, limit=-1, offset=0):
"""
Return a generator to iterate over all characters. See \
L{get_all_characters).
"""
self._e("""SELECT * FROM characters
ORDER BY charid LIMIT ? OFFSET ?""", (int(limit), int(offset)))
return (self.get_character_from_row(r) for r in self._fa())
def get_total_n_characters(self):
"""
Return the total number of characters in collection.
@rtype: int
"""
return self._efo("SELECT COUNT(charid) FROM characters")[0]
def get_total_n_strokes(self):
"""
Return the total number of strokes in collection.
@rtype: int
"""
return self._efo("SELECT SUM(n_strokes) FROM characters")[0]
def get_average_n_strokes(self, set_name):
"""
Return the average number of stroke of the characters in that set.
"""
i = self._SETIDS[set_name]
return self._efo("""SELECT AVG(n_strokes) FROM characters
WHERE setid=?""", (i,))[0]
def set_characters(self, set_name, characters):
"""
Set/Replace the characters of a set.
@type set_name: str
@param set_name: the set that needs be updated
@type characters: list of L{Character}
"""
i = self._SETIDS[set_name]
self._e("DELETE FROM characters WHERE setid=?", (i,))
for char in characters:
self.append_character(set_name, char)
def append_character(self, set_name, character):
"""
Append a new character to a set.
@type set_name: str
@param set_name: the set to which the character needs be added
@type character: L{Character}
"""
self.append_characters(set_name, [character])
def append_characters(self, set_name, characters):
rows = [{'utf8':c.get_utf8(),
'n_strokes':c.get_writing().get_n_strokes(),
'data':_adapt_character(c),
'sha1':c.hash()} for c in characters]
self.append_character_rows(set_name, rows)
def append_character_rows(self, set_name, rows):
i = self._SETIDS[set_name]
tupls = [(i, r['utf8'], r['n_strokes'], r['data'], r['sha1']) \
for r in rows]
self._em("""INSERT INTO
characters (setid, utf8, n_strokes, data, sha1)
VALUES (?,?,?,?,?)""", tupls)
def insert_character(self, set_name, i, character):
"""
Insert a new character to a set at a given position.
@type set_name: str
@param set_name: the set to which the character needs be inserted
@type i: int
@param i: position
@type character: L{Character}
"""
chars = self.get_characters(set_name)
chars.insert(i, character)
self.set_characters(set_name, chars)
def remove_character(self, set_name, i):
"""
Remove a character from a set at a given position.
@type set_name: str
@param set_name: the set from which the character needs be removed
@type i: int
@param i: position
"""
setid = self._SETIDS[set_name]
charid = self._efo("""SELECT charid FROM characters
WHERE setid=? ORDER BY charid LIMIT 1 OFFSET ?""", (setid, i))[0]
if charid:
self._e("DELETE FROM characters WHERE charid=?", (charid,))
def remove_last_character(self, set_name):
"""
Remove the last character from a set.
@type set_name: str
@param set_name: the set from which the character needs be removed
"""
setid = self._SETIDS[set_name]
charid = self._efo("""SELECT charid FROM characters
WHERE setid=? ORDER BY charid DESC LIMIT 1""", (setid,))[0]
if charid:
self._e("DELETE FROM characters WHERE charid=?", (charid,))
def update_character_object(self, character):
"""
Update a character.
@type character: L{Character}
character must have been previously retrieved from the collection.
"""
if not hasattr(character, "charid"):
raise ValueError, "The character object needs a charid attribute"
self._e("""UPDATE characters
SET utf8=?, n_strokes=?, data=?, sha1=?
WHERE charid=?""", (character.get_utf8(),
character.get_writing().get_n_strokes(),
_adapt_character(character),
character.hash(),
character.charid))
def replace_character(self, set_name, i, character):
"""
Replace the character at a given position with a new character.
@type set_name: str
@param set_name: the set where the character needs be replaced
@type i: int
@param i: position
@type character: L{Character}
"""
setid = self._SETIDS[set_name]
charid = self._efo("""SELECT charid FROM characters
WHERE setid=? ORDER BY charid LIMIT 1 OFFSET ?""", (setid, i))[0]
if charid:
character.charid = charid
self.update_character_object(character)
def _get_dict_from_text(self, text):
text = text.replace(" ", "").replace("\n", "").replace("\t", "")
dic = {}
for c in text:
dic[c] = 1
return dic
def include_characters_from_text(self, text):
"""
Only keep characters found in a text.
Or put differently, remove all characters but those found in a text.
@type text: str
"""
dic = self._get_dict_from_text(unicode(text, "utf8"))
utf8values = ",".join(["'%s'" % k for k in dic.keys()])
self._e("DELETE FROM characters WHERE utf8 NOT IN(%s)" % utf8values)
self.remove_empty_sets()
def include_characters_from_files(self, text_files):
"""
Only keep characters found in text_files.
@type text_files: list
@param text_files: a list of file paths
"""
buf = ""
for inc_path in text_files:
f = open(inc_path)
buf += f.read()
f.close()
if len(buf) > 0:
self.include_characters_from_text(buf)
def exclude_characters_from_text(self, text):
"""
Exclude characters found in a text.
@type text: str
"""
dic = self._get_dict_from_text(unicode(text, "utf8"))
utf8values = ",".join(["'%s'" % k for k in dic.keys()])
self._e("DELETE FROM characters WHERE utf8 IN(%s)" % utf8values)
self.remove_empty_sets()
def exclude_characters_from_files(self, text_files):
"""
Exclude characters found in text_files.
@type text_files: list
@param text_files: a list of file paths
"""
buf = ""
for exc_path in text_files:
f = open(exc_path)
buf += f.read()
f.close()
if len(buf) > 0:
self.exclude_characters_from_text(buf)
def remove_samples(self, keep_at_most):
"""
Remove samples.
@type keep_at_most: the maximum number of samples to keep.
"""
for set_name in self.get_set_list():
if self.get_n_characters(set_name) > keep_at_most:
setid = self._SETIDS[set_name]
self._e("""DELETE FROM characters
WHERE charid IN(SELECT charid FROM characters
WHERE setid=? ORDER BY charid LIMIT -1 OFFSET ?)""",
(setid, keep_at_most))
def _get_set_char_counts(self):
rows = self._efa("""SELECT setid, COUNT(charid) AS n_chars
FROM characters GROUP BY setid""")
d = {}
for row in rows:
d[row['setid']] = row['n_chars']
return d
def remove_empty_sets(self):
"""
Remove sets that don't include any character.
"""
charcounts = self._get_set_char_counts()
empty_sets = []
for set_name, setid in self._SETIDS.items():
try:
if charcounts[setid] == 0:
empty_sets.append(set_name)
except KeyError:
empty_sets.append(set_name)
self.remove_sets(empty_sets)
def to_xml(self):
"""
Converts collection to XML.
@rtype: str
"""
s = "\n"
s += "\n"
for set_name in self.get_set_list():
s += "\n" % set_name
for character in self.get_characters(set_name):
s += " \n"
utf8 = character.get_utf8()
if utf8:
s += " %s\n" % utf8
for line in character.get_writing().to_xml().split("\n"):
s += " %s\n" % line
s += " \n"
s += "\n"
s += "\n"
return s
# XML processing...
def _start_element(self, name, attrs):
self._tag = name
if self._first_tag:
self._first_tag = False
if self._tag != "character-collection":
raise ValueError, \
"The very first tag should be "
if self._tag == "set":
if not attrs.has_key("name"):
raise ValueError, " should have a name attribute"
self._curr_set_name = attrs["name"].encode("UTF-8")
self.add_set(self._curr_set_name)
if self._tag == "character":
self._curr_char = Character()
self._curr_writing = self._curr_char.get_writing()
self._curr_width = None
self._curr_height = None
self._curr_utf8 = None
if self._tag == "stroke":
self._curr_stroke = Stroke()
elif self._tag == "point":
point = Point()
for key in ("x", "y", "pressure", "xtilt", "ytilt", "timestamp"):
if attrs.has_key(key):
value = attrs[key].encode("UTF-8")
if key in ("pressure", "xtilt", "ytilt"):
value = float(value)
else:
value = int(float(value))
else:
value = None
setattr(point, key, value)
self._curr_stroke.append_point(point)
def _end_element(self, name):
if name == "character-collection":
for s in ["_tag", "_curr_char", "_curr_writing", "_curr_width",
"_curr_height", "_curr_utf8", "_curr_stroke",
"_curr_chars", "_curr_set_name"]:
if s in self.__dict__:
del self.__dict__[s]
if name == "character":
if self._curr_utf8:
self._curr_char.set_utf8(self._curr_utf8)
if self._curr_width:
self._curr_writing.set_width(self._curr_width)
if self._curr_height:
self._curr_writing.set_height(self._curr_height)
self.append_character(self._curr_set_name, self._curr_char)
if name == "stroke":
if len(self._curr_stroke) > 0:
self._curr_writing.append_stroke(self._curr_stroke)
self._stroke = None
self._tag = None
def _char_data(self, data):
if self._tag == "utf8":
self._curr_utf8 = data.encode("UTF-8")
if self._tag == "width":
self._curr_width = int(data)
elif self._tag == "height":
self._curr_height = int(data)
tegaki-python-0.3.1/tegaki/dictutils.py 0000644 0001750 0001750 00000012021 11342122456 017766 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (c) 2005, the Lawrence Journal-World
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# 1. Redistributions of source code must retain the above copyright notice,
# this list of conditions and the following disclaimer.
#
# 2. Redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
#
# 3. Neither the name of Django nor the names of its contributors may be used
# to endorse or promote products derived from this software without
# specific prior written permission.
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE
# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
class SortedDict(dict):
"""
A dictionary that keeps its keys in the order in which they're inserted.
"""
def __new__(cls, *args, **kwargs):
instance = super(SortedDict, cls).__new__(cls, *args, **kwargs)
instance.keyOrder = []
return instance
def __init__(self, data=None):
if data is None:
data = {}
super(SortedDict, self).__init__(data)
if isinstance(data, dict):
self.keyOrder = data.keys()
else:
self.keyOrder = []
for key, value in data:
if key not in self.keyOrder:
self.keyOrder.append(key)
def __deepcopy__(self, memo):
from copy import deepcopy
return self.__class__([(key, deepcopy(value, memo))
for key, value in self.iteritems()])
def __setitem__(self, key, value):
super(SortedDict, self).__setitem__(key, value)
if key not in self.keyOrder:
self.keyOrder.append(key)
def __delitem__(self, key):
super(SortedDict, self).__delitem__(key)
self.keyOrder.remove(key)
def __iter__(self):
for k in self.keyOrder:
yield k
def pop(self, k, *args):
result = super(SortedDict, self).pop(k, *args)
try:
self.keyOrder.remove(k)
except ValueError:
# Key wasn't in the dictionary in the first place. No problem.
pass
return result
def popitem(self):
result = super(SortedDict, self).popitem()
self.keyOrder.remove(result[0])
return result
def items(self):
return zip(self.keyOrder, self.values())
def iteritems(self):
for key in self.keyOrder:
yield key, super(SortedDict, self).__getitem__(key)
def keys(self):
return self.keyOrder[:]
def iterkeys(self):
return iter(self.keyOrder)
def values(self):
return map(super(SortedDict, self).__getitem__, self.keyOrder)
def itervalues(self):
for key in self.keyOrder:
yield super(SortedDict, self).__getitem__(key)
def update(self, dict_):
for k, v in dict_.items():
self.__setitem__(k, v)
def setdefault(self, key, default):
if key not in self.keyOrder:
self.keyOrder.append(key)
return super(SortedDict, self).setdefault(key, default)
def value_for_index(self, index):
"""Returns the value of the item at the given zero-based index."""
return self[self.keyOrder[index]]
def insert(self, index, key, value):
"""Inserts the key, value pair before the item with the given index."""
if key in self.keyOrder:
n = self.keyOrder.index(key)
del self.keyOrder[n]
if n < index:
index -= 1
self.keyOrder.insert(index, key)
super(SortedDict, self).__setitem__(key, value)
def copy(self):
"""Returns a copy of this object."""
# This way of initializing the copy means it works for subclasses, too.
obj = self.__class__(self)
obj.keyOrder = self.keyOrder[:]
return obj
def __repr__(self):
"""
Replaces the normal dict.__repr__ with a version that returns the keys
in their sorted order.
"""
return '{%s}' % ', '.join(['%r: %r' % (k, v) for k, v in self.items()])
def clear(self):
super(SortedDict, self).clear()
self.keyOrder = [] tegaki-python-0.3.1/tegaki/__init__.py 0000644 0001750 0001750 00000000240 11352064314 017520 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
VERSION = '0.3.1'
import tegaki.character
import tegaki.charcol
import tegaki.recognizer
import tegaki.trainer
import tegaki.arrayutils tegaki-python-0.3.1/tegaki/character.py 0000644 0001750 0001750 00000110517 11342122456 017727 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008-2009 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import xml.parsers.expat
import cStringIO
import gzip as gzipm
try:
import bz2 as bz2m
except ImportError:
pass
from math import floor, atan, sin, cos, pi
import os
import hashlib
try:
# lxml is used for DTD validation
# for server-side applications, it is recommended to install it
# for desktop applications, it is optional
from lxml import etree
except ImportError:
pass
from tegaki.mathutils import euclidean_distance
from tegaki.dictutils import SortedDict
class Point(dict):
"""
A point in a 2-dimensional space.
"""
#: Attributes that a point can have.
KEYS = ("x", "y", "pressure", "xtilt", "ytilt", "timestamp")
def __init__(self, x=None, y=None,
pressure=None, xtilt=None, ytilt=None,
timestamp=None):
"""
@type x: int
@type y: int
@type pressure: float
@type xtilt: float
@type ytilt: float
@type timestamp: int
@param timestamp: ellapsed time since first point in milliseconds
"""
dict.__init__(self)
self.x = x
self.y = y
self.pressure = pressure
self.xtilt = xtilt
self.ytilt = ytilt
self.timestamp = timestamp
def __getattr__(self, attr):
try:
return self[attr]
except KeyError:
raise AttributeError
def __setattr__(self, attr, value):
try:
self[attr] = value
except KeyError:
raise AttributeError
def get_coordinates(self):
"""
Return (x,y) coordinates.
@rtype: tuple of two int
@return: (x,y) coordinates
"""
return (self.x, self.y)
def resize(self, xrate, yrate):
"""
Scale point.
@type xrate: float
@param xrate: the x scaling factor
@type yrate: float
@param yrate: the y scaling factor
"""
self.x = int(self.x * xrate)
self.y = int(self.y * yrate)
def move_rel(self, dx, dy):
"""
Translate point.
@type dx: int
@param dx: relative distance from x
@type dy: int
@param yrate: relative distance from y
"""
self.x = self.x + dx
self.y = self.y + dy
def to_xml(self):
"""
Converts point to XML.
@rtype: str
"""
attrs = []
for key in self.KEYS:
if self[key] is not None:
attrs.append("%s=\"%s\"" % (key, str(self[key])))
return "" % " ".join(attrs)
def to_json(self):
"""
Converts point to JSON.
@rtype: str
"""
attrs = []
for key in self.KEYS:
if self[key] is not None:
attrs.append("\"%s\" : %d" % (key, int(self[key])))
return "{ %s }" % ", ".join(attrs)
def to_sexp(self):
"""
Converts point to S-expressions.
@rtype: str
"""
return "(%d %d)" % (self.x, self.y)
def __eq__(self, othr):
if not othr.__class__.__name__ in ("Point", "PointProxy"):
return False
for key in self.KEYS:
if self[key] != othr[key]:
return False
return True
def __ne__(self, othr):
return not(self == othr)
def copy_from(self, p):
"""
Replace point with another point.
@type p: L{Point}
@param p: the point to copy from
"""
self.clear()
for k in p.keys():
if p[k] is not None:
self[k] = p[k]
def copy(self):
"""
Return a copy of point.
@rtype: L{Point}
"""
return Point(**self)
def __repr__(self):
return "" % (self.x, self.y, id(self))
class Stroke(list):
"""
A sequence of L{Points}.
"""
def __init__(self):
list.__init__(self)
self._is_smoothed = False
def get_coordinates(self):
"""
Return (x,y) coordinates.
@rtype: a list of tuples
"""
return [(p.x, p.y) for p in self]
def get_duration(self):
"""
Return the time that it took to draw the stroke.
@rtype: int or None
@return: time in millisecons or None if the information is not available
"""
if len(self) > 0:
if self[-1].timestamp is not None and self[0].timestamp is not None:
return self[-1].timestamp - self[0].timestamp
return None
def append_point(self, point):
"""
Append point to stroke.
@type point: L{Point}
"""
self.append(point)
def to_xml(self):
"""
Converts stroke to XML.
@rtype: str
"""
s = "\n"
for point in self:
s += " %s\n" % point.to_xml()
s += ""
return s
def to_json(self):
"""
Converts stroke to JSON.
@rtype: str
"""
s = "{\"points\" : ["
s += ",".join([point.to_json() for point in self])
s += "]}"
return s
def to_sexp(self):
"""
Converts stroke to S-expressions.
@rtype: str
"""
return "(" + "".join([p.to_sexp() for p in self]) + ")"
def __eq__(self, othr):
if not othr.__class__.__name__ in ("Stroke", "StrokeProxy"):
return False
if len(self) != len(othr):
return False
for i in range(len(self)):
if self[i] != othr[i]:
return False
return True
def __ne__(self, othr):
return not(self == othr)
def copy_from(self, s):
"""
Replace stroke with another stroke.
@type s: L{Stroke}
@param s: the stroke to copy from
"""
self.clear()
self._is_smoothed = s.get_is_smoothed()
for p in s:
self.append_point(p.copy())
def copy(self):
"""
Return a copy of stroke.
@rtype: L{Stroke}
"""
c = Stroke()
c.copy_from(self)
return c
def get_is_smoothed(self):
"""
Return whether the stroke has been smoothed already or not.
@rtype: boolean
"""
return self._is_smoothed
def smooth(self):
"""
Visually improve the rendering of stroke by averaging points
with their neighbours.
The method is based on a (simple) moving average algorithm.
Let p = p(0), ..., p(N) be the set points of this stroke,
w = w(-M), ..., w(0), ..., w(M) be a set of weights.
This algorithm aims at replacing p with a set p' such as
p'(i) = (w(-M)*p(i-M) + ... + w(0)*p(i) + ... + w(M)*p(i+M)) / S
and where S = w(-M) + ... + w(0) + ... w(M). End points are not
affected.
"""
if self._is_smoothed:
return
weights = [1, 1, 2, 1, 1] # Weights to be used
times = 3 # Number of times to apply the algorithm
if len(self) < len(weights):
return
offset = int(floor(len(weights) / 2.0))
wsum = sum(weights)
for n in range(times):
s = self.copy()
for i in range(offset, len(self) - offset):
self[i].x = 0
self[i].y = 0
for j in range(len(weights)):
self[i].x += weights[j] * s[i + j - offset].x
self[i].y += weights[j] * s[i + j - offset].y
self[i].x = int(round(self[i].x / wsum))
self[i].y = int(round(self[i].y / wsum))
self._is_smoothed = True
def clear(self):
"""
Remove all points from stroke.
"""
while len(self) != 0:
del self[0]
self._is_smoothed = False
def downsample(self, n):
"""
Downsample by keeping only 1 sample every n samples.
@type n: int
"""
if len(self) == 0:
return
new_s = Stroke()
for i in range(len(self)):
if i % n == 0:
new_s.append_point(self[i])
self.copy_from(new_s)
def downsample_threshold(self, threshold):
"""
Downsample by removing consecutive samples for which
the euclidean distance is inferior to threshold.
@type threshod: int
"""
if len(self) == 0:
return
new_s = Stroke()
new_s.append_point(self[0])
last = 0
for i in range(1, len(self) - 2):
u = [self[last].x, self[last].y]
v = [self[i].x, self[i].y]
if euclidean_distance(u, v) > threshold:
new_s.append_point(self[i])
last = i
new_s.append_point(self[-1])
self.copy_from(new_s)
def upsample(self, n):
"""
'Artificially' increase sampling by adding n linearly spaced points
between consecutive points.
@type n: int
"""
self._upsample(lambda d: n)
def upsample_threshold(self, threshold):
"""
'Artificially' increase sampling, using threshold to determine
how many samples to add between consecutive points.
@type threshold: int
"""
self._upsample(lambda d: int(floor(float(d) / threshold - 1)))
def _upsample(self, func):
"""
'Artificially' increase sampling, using func(distance) to determine how
many samples to add between consecutive points.
"""
if len(self) == 0:
return
new_s = Stroke()
for i in range(len(self)- 1):
x1, y1 = [self[i].x, self[i].y]
x2, y2 = [self[i+1].x, self[i+1].y]
new_s.append_point(self[i])
dx = x2 - x1
dy = y2 - y1
if dx == 0:
alpha = pi / 2
cosalpha = 0.0
sinalpha = 1.0
else:
alpha = atan(float(abs(dy)) / abs(x2 - x1))
cosalpha = cos(alpha)
sinalpha = sin(alpha)
d = euclidean_distance([x1, y1], [x2, y2])
signx = cmp(dx, 0)
signy = cmp(dy, 0)
n = func(d)
for j in range(1, n+1):
dx = cosalpha * 1.0 / (n + 1) * d
dy = sinalpha * 1.0 / (n + 1) * d
new_s.append_point(Point(x=int(x1+j*dx*signx),
y=int(y1+j*dy*signy)))
new_s.append_point(self[-1])
self.copy_from(new_s)
def __repr__(self):
return "" % (len(self), id(self))
class Writing(object):
"""
A sequence of L{Strokes}.
"""
#: Default width and height of the canvas
#: If the canvas used to create the Writing object
#: has a different width or height, then
#: the methods set_width and set_height need to be used
WIDTH = 1000
HEIGHT = 1000
NORMALIZE_PROPORTION = 0.7 # percentage of the drawing area
NORMALIZE_MIN_SIZE = 0.1 # don't nornalize if below that percentage
def __init__(self):
self._width = Writing.WIDTH
self._height = Writing.HEIGHT
self.clear()
def clear(self):
"""
Remove all strokes from writing.
"""
self._strokes = []
def get_duration(self):
"""
Return the time that it took to draw the strokes.
@rtype: int or None
@return: time in millisecons or None if the information is not available
"""
if self.get_n_strokes() > 0:
if self._strokes[0][0].timestamp is not None and \
self._strokes[-1][-1].timestamp is not None:
return self._strokes[-1][-1].timestamp - \
self._strokes[0][0].timestamp
return None
def move_to(self, x, y):
"""
Start a new stroke at (x,y).
@type x: int
@type y: int
"""
# For compatibility
point = Point()
point.x = x
point.y = y
self.move_to_point(point)
def line_to(self, x, y):
"""
Add point with coordinates (x,y) to the current stroke.
@type x: int
@type y: int
"""
# For compatibility
point = Point()
point.x = x
point.y = y
self.line_to_point(point)
def move_to_point(self, point):
"""
Start a new stroke at point.
@type point: L{Point}
"""
stroke = Stroke()
stroke.append_point(point)
self.append_stroke(stroke)
def line_to_point(self, point):
"""
Add point to the current stroke.
@type point: L{Point}
"""
self._strokes[-1].append(point)
def get_n_strokes(self):
"""
Return the number of strokes.
@rtype: int
"""
return len(self._strokes)
def get_n_points(self):
"""
Return the total number of points.
"""
return sum([len(s) for s in self._strokes])
def get_strokes(self, full=False):
"""
Return strokes.
@type full: boolean
@param full: whether to return strokes as objects or as (x,y) pairs
"""
if not full:
# For compatibility
return [[(int(p.x), int(p.y)) for p in s] for s in self._strokes]
else:
return self._strokes
def append_stroke(self, stroke):
"""
Add a new stroke.
@type stroke: L{Stroke}
"""
self._strokes.append(stroke)
def insert_stroke(self, i, stroke):
"""
Insert a stroke at a given position.
@type stroke: L{Stroke}
@type i: int
@param i: position at which to add the stroke (starts at 0)
"""
self._strokes.insert(i, stroke)
def remove_stroke(self, i):
"""
Remove the ith stroke.
@type i: int
@param i: position at which to delete a stroke (starts at 0)
"""
if self.get_n_strokes() - 1 >= i:
del self._strokes[i]
def remove_last_stroke(self):
"""
Remove last stroke.
Equivalent to remove_stroke(n-1) where n is the number of strokes.
"""
if self.get_n_strokes() > 0:
del self._strokes[-1]
def replace_stroke(self, i, stroke):
"""
Replace the ith stroke with a new stroke.
@type i: int
@param i: position at which to replace a stroke (starts at 0)
@type stroke: L{Stroke}
@param stroke: the new stroke
"""
if self.get_n_strokes() - 1 >= i:
self.remove_stroke(i)
self.insert_stroke(i, stroke)
def resize(self, xrate, yrate):
"""
Scale writing.
@type xrate: float
@param xrate: the x scaling factor
@type yrate: float
@param yrate: the y scaling factor
"""
for stroke in self._strokes:
if len(stroke) == 0:
continue
stroke[0].resize(xrate, yrate)
for point in stroke[1:]:
point.resize(xrate, yrate)
def move_rel(self, dx, dy):
"""
Translate writing.
@type dx: int
@param dx: relative distance from current position
@type dy: int
@param yrate: relative distance from current position
"""
for stroke in self._strokes:
if len(stroke) == 0:
continue
stroke[0].move_rel(dx, dy)
for point in stroke[1:]:
point.move_rel(dx, dy)
def size(self):
"""
Return writing size.
@rtype: (x, y, width, height)
@return: (x,y) are the coordinates of the upper-left point
"""
xmin, ymin = 4294967296, 4294967296 # 2^32
xmax, ymax = 0, 0
for stroke in self._strokes:
for point in stroke:
xmin = min(xmin, point.x)
ymin = min(ymin, point.y)
xmax = max(xmax, point.x)
ymax = max(ymax, point.y)
return (xmin, ymin, xmax-xmin, ymax-ymin)
def normalize(self):
"""
Call L{normalize_size} and L{normalize_position} consecutively.
"""
self.normalize_size()
self.normalize_position()
def normalize_position(self):
"""
Translate character so as to have the same amount of space to
each side of the drawing box.
It improves the quality of characters by making them
more centered on the drawing box.
"""
x, y, width, height = self.size()
dx = (self._width - width) / 2 - x
dy = (self._height - height) / 2 - y
self.move_rel(dx, dy)
def normalize_size(self):
"""
Scale character to match a given, fixed size.
This improves the quality of characters which are too big or too small.
"""
# Note: you should call normalize_position() after normalize_size()
x, y, width, height = self.size()
if float(width) / self._width > Writing.NORMALIZE_MIN_SIZE:
xrate = self._width * Writing.NORMALIZE_PROPORTION / width
else:
# Don't normalize if too thin in width
xrate = 1.0
if float(height) / self._height > Writing.NORMALIZE_MIN_SIZE:
yrate = self._height * Writing.NORMALIZE_PROPORTION / height
else:
# Don't normalize if too thin in height
yrate = 1.0
self.resize(xrate, yrate)
def downsample(self, n):
"""
Downsample by keeping only 1 sample every n samples.
@type n: int
"""
for s in self._strokes:
s.downsample(n)
def downsample_threshold(self, threshold):
"""
Downsample by removing consecutive samples for which
the euclidean distance is inferior to threshold.
@type threshod: int
"""
for s in self._strokes:
s.downsample_threshold(threshold)
def upsample(self, n):
"""
'Artificially' increase sampling by adding n linearly spaced points
between consecutive points.
@type n: int
"""
for s in self._strokes:
s.upsample(n)
def upsample_threshold(self, threshold):
"""
'Artificially' increase sampling, using threshold to determine
how many samples to add between consecutive points.
@type threshold: int
"""
for s in self._strokes:
s.upsample_threshold(threshold)
def get_size(self):
"""
Return the size of the drawing box.
@rtype: tuple
Not to be confused with size() which returns the size the writing.
"""
return (self.get_width(), self.get_height())
def set_size(self, w, h):
self.set_width(w)
self.set_height(h)
def get_width(self):
"""
Return the width of the drawing box.
@rtype: int
"""
return self._width
def set_width(self, width):
"""
Set the drawing box width.
This is necessary if the points which are added were not drawn in
1000x1000 drawing box.
"""
self._width = width
def get_height(self):
"""
Return the height of the drawing box.
@rtype: int
"""
return self._height
def set_height(self, height):
"""
Set the drawing box height.
This is necessary if the points which are added were not drawn in
1000x1000 drawing box.
"""
self._height = height
def to_xml(self):
"""
Converts writing to XML.
@rtype: str
"""
s = "%d\n" % self.get_width()
s += "%d\n" % self.get_height()
s += "\n"
for stroke in self._strokes:
for line in stroke.to_xml().split("\n"):
s += " %s\n" % line
s += ""
return s
def to_json(self):
"""
Converts writing to JSON.
@rtype: str
"""
s = "{ \"width\" : %d, " % self.get_width()
s += "\"height\" : %d, " % self.get_height()
s += "\"strokes\" : ["
s += ", ".join([stroke.to_json() for stroke in self._strokes])
s += "]}"
return s
def to_sexp(self):
"""
Converts writing to S-expressions.
@rtype: str
"""
return "((width %d)(height %d)(strokes %s))" % \
(self._width, self._height,
"".join([s.to_sexp() for s in self._strokes]))
def __eq__(self, othr):
if not othr.__class__.__name__ in ("Writing", "WritingProxy"):
return False
if self.get_n_strokes() != othr.get_n_strokes():
return False
if self.get_width() != othr.get_width():
return False
if self.get_height() != othr.get_height():
return False
othr_strokes = othr.get_strokes(full=True)
for i in range(len(self._strokes)):
if self._strokes[i] != othr_strokes[i]:
return False
return True
def __ne__(self, othr):
return not(self == othr)
self.clear()
self._is_smoothed = s.get_is_smoothed()
for p in s:
self.append_point(p.copy())
def copy_from(self, w):
"""
Replace writing with another writing.
@type w: L{Writing}
@param w: the writing to copy from
"""
self.clear()
self.set_width(w.get_width())
self.set_height(w.get_height())
for s in w.get_strokes(True):
self.append_stroke(s.copy())
def copy(self):
"""
Return a copy writing.
@rtype: L{Writing}
"""
c = Writing()
c.copy_from(self)
return c
def smooth(self):
"""
Smooth all strokes. See L{Stroke.smooth}.
"""
for stroke in self._strokes:
stroke.smooth()
def __repr__(self):
return "" % (self.get_n_strokes(),
id(self))
class _XmlBase(object):
"""
Class providing XML functionality to L{Character} and \
L{CharacterCollection}.
"""
@classmethod
def validate(cls, string):
"""
Validate XML against a DTD.
@type string: str
@param string: a string containing XML
DTD must be an attribute of cls.
"""
try:
dtd = etree.DTD(cStringIO.StringIO(cls.DTD))
root = etree.XML(string.strip())
return dtd.validate(root)
except etree.XMLSyntaxError:
return False
except NameError:
# this means that the functionality is not available on that
# system so you have to catch that exception if you want to
# ignore it
raise NotImplementedError
def read(self, file, gzip=False, bz2=False, compresslevel=9):
"""
Read XML from a file.
@type file: str or file
@param file: path to file or file object
@type gzip: boolean
@param gzip: whether the file is gzip-compressed or not
@type bz2: boolean
@param bz2: whether the file is bzip2-compressed or not
@type compresslevel: int
@param compresslevel: compression level (see gzip module documentation)
Raises ValueError if incorrect XML.
"""
parser = self._get_parser()
try:
if type(file) == str:
if gzip:
file = gzipm.GzipFile(file, compresslevel=compresslevel)
elif bz2:
try:
file = bz2m.BZ2File(file, compresslevel=compresslevel)
except NameError:
raise NotImplementedError
else:
file = open(file)
parser.ParseFile(file)
file.close()
else:
parser.ParseFile(file)
except (IOError, xml.parsers.expat.ExpatError):
raise ValueError
def read_string(self, string, gzip=False, bz2=False, compresslevel=9):
"""
Read XML from string.
@type string: str
@param string: string containing XML
Other parameters are identical to L{read}.
"""
if gzip:
io = cStringIO.StringIO(string)
io = gzipm.GzipFile(fileobj=io, compresslevel=compresslevel)
string = io.read()
elif bz2:
try:
string = bz2m.decompress(string)
except NameError:
raise NotImplementedError
parser = self._get_parser()
parser.Parse(string)
def write(self, file, gzip=False, bz2=False, compresslevel=9):
"""
Write XML to a file.
@type file: str or file
@param file: path to file or file object
@type gzip: boolean
@param gzip: whether the file need be gzip-compressed or not
@type bz2: boolean
@param bz2: whether the file need be bzip2-compressed or not
@type compresslevel: int
@param compresslevel: compression level (see gzip module documentation)
"""
if type(file) == str:
if gzip:
file = gzipm.GzipFile(file, "w", compresslevel=compresslevel)
elif bz2:
try:
file = bz2m.BZ2File(file, "w", compresslevel=compresslevel)
except NameError:
raise NotImplementedError
else:
file = open(file, "w")
file.write(self.to_xml())
file.close()
else:
file.write(self.to_xml())
def write_string(self, gzip=False, bz2=False, compresslevel=9):
"""
Write XML to string.
@rtype: str
@return: string containing XML
Other parameters are identical to L{write}.
"""
if bz2:
try:
return bz2m.compress(self.to_xml(), compresslevel=compresslevel)
except NameError:
raise NotImplementedError
elif gzip:
io = cStringIO.StringIO()
f = gzipm.GzipFile(fileobj=io, mode="w",
compresslevel=compresslevel)
f.write(self.to_xml())
f.close()
return io.getvalue()
else:
return self.to_xml()
def _get_parser(self):
parser = xml.parsers.expat.ParserCreate(encoding="UTF-8")
parser.StartElementHandler = self._start_element
parser.EndElementHandler = self._end_element
parser.CharacterDataHandler = self._char_data
self._first_tag = True
return parser
class Character(_XmlBase):
"""
A handwritten character.
A Character is composed of meta-data and handwriting data.
Handwriting data are contained in L{Writing} objects.
Building character objects
==========================
A character can be built from scratch progmatically:
>>> s = Stroke()
>>> s.append_point(Point(10, 20))
>>> w = Writing()
>>> w.append_stroke(s)
>>> c = Character()
>>> c.set_writing(writing)
Reading XML files
=================
A character can be read from an XML file:
>>> c = Character()
>>> c.read("myfile")
Gzip-compressed and bzip2-compressed XML files can also be read:
>>> c = Character()
>>> c.read("myfilegz", gzip=True)
>>> c = Character()
>>> c.read("myfilebz", bz2=True)
A similar method read_string exists to read the XML from a string
instead of a file.
For convenience, you can directly load a character by passing it the
file to load. In that case, compression is automatically detected based on
file extension (.gz, .bz2).
>>> c = Character("myfile.xml.gz")
The recommended extension for XML character files is .xml.
Writing XML files
=================
A character can be saved to an XML file by using the write() method.
>>> c.write("myfile")
The write method has gzip and bz2 arguments just like read(). In addition,
there is a write_string method which generates a string instead of a file.
For convenience, you can save a character with the save() method.
It automatically detects compression based on the file extension.
>>> c.save("mynewfile.xml.bz2")
If the Character object was passed a file when it was constructed,
the path can ce omitted.
>>> c = Character("myfile.gz")
>>> c.save()
>>> c = Character()
>>> c.save()
Traceback (most recent call last):
File "", line 1, in
File "tegaki/character.py", line 1238, in save
raise ValueError, "A path must be specified"
ValueError: A path must be specified
"""
DTD = \
"""
"""
def __init__(self, path=None):
"""
Creates a new Character.
@type path: str or None
@param path: path to file to load or None if empty character
The file extension is used to determine whether the file is plain,
gzip-compressed or bzip2-compressed XML.
"""
self._writing = Writing()
self._utf8 = None
self._path = path
if path is not None:
gzip = True if path.endswith(".gz") or path.endswith(".gzip") \
else False
bz2 = True if path.endswith(".bz2") or path.endswith(".bzip2") \
else False
self.read(path, gzip=gzip, bz2=bz2)
def get_utf8(self):
"""
Return the label of the character.
@rtype: str
"""
return self._utf8
def get_unicode(self):
"""
Return the label character.
@rtype: unicode
"""
return unicode(self.get_utf8(), "utf8")
def set_utf8(self, utf8):
"""
Set the label the character.
@type utf8: str
"""
self._utf8 = utf8
def set_unicode(self, uni):
"""
Set the label of the character.
@type uni: unicode
"""
self._utf8 = uni.encode("utf8")
def get_writing(self):
"""
Return the handwriting data of the character.
@rtype: L{Writing}
"""
return self._writing
def set_writing(self, writing):
"""
Set the handwriting data of the character.
@type writing: L{Writing}
"""
self._writing = writing
def hash(self):
"""
Return a sha1 digest for that character.
"""
return hashlib.sha1(self.to_xml()).hexdigest()
def save(self, path=None):
"""
Save character to file.
@type path: str
@param path: path where to write the file or None if use the path \
that was given to the constructor
The file extension is used to determine whether the file is plain,
gzip-compressed or bzip2-compressed XML.
"""
if [path, self._path] == [None, None]:
raise ValueError, "A path must be specified"
elif path is None:
path = self._path
gzip = True if path.endswith(".gz") or path.endswith(".gzip") \
else False
bz2 = True if path.endswith(".bz2") or path.endswith(".bzip2") \
else False
self.write(path, gzip=gzip, bz2=bz2)
def to_xml(self):
"""
Converts character to XML.
@rtype: str
"""
s = "\n"
s += "\n"
if self._utf8:
s += " %s\n" % self._utf8
for line in self._writing.to_xml().split("\n"):
s += " %s\n" % line
s += ""
return s
def to_json(self):
"""
Converts character to JSON.
@rtype: str
"""
s = "{"
attrs = ["\"utf8\" : \"%s\"" % self._utf8,
"\"writing\" : " + self._writing.to_json()]
s += ", ".join(attrs)
s += "}"
return s
def to_sexp(self):
"""
Converts character to S-expressions.
@rtype: str
"""
return "(character (value %s)" % self._utf8 + \
self._writing.to_sexp()[1:-1]
def __eq__(self, char):
if not char.__class__.__name__ in ("Character", "CharacterProxy"):
return False
return self._utf8 == char.get_utf8() and \
self._writing == char.get_writing()
def __ne__(self, othr):
return not(self == othr)
self.clear()
self.set_width(w.get_width())
self.set_height(w.get_height())
for s in w.get_strokes(True):
self.append_stroke(s.copy())
def copy_from(self, c):
"""
Replace character with another character.
@type c: L{Character}
@param c: the character to copy from
"""
self.set_utf8(c.get_utf8())
self.set_writing(c.get_writing().copy())
def copy(self):
"""
Return a copy of character.
@rtype: L{Character}
"""
c = Character()
c.copy_from(self)
return c
def __repr__(self):
return "" % (str(self.get_utf8()), id(self))
# Private...
def _start_element(self, name, attrs):
self._tag = name
if self._first_tag:
self._first_tag = False
if self._tag != "character":
raise ValueError, "The very first tag should be "
if self._tag == "stroke":
self._stroke = Stroke()
elif self._tag == "point":
point = Point()
for key in ("x", "y", "pressure", "xtilt", "ytilt", "timestamp"):
if attrs.has_key(key):
value = attrs[key].encode("UTF-8")
if key in ("pressure", "xtilt", "ytilt"):
value = float(value)
else:
value = int(float(value))
else:
value = None
setattr(point, key, value)
self._stroke.append_point(point)
def _end_element(self, name):
if name == "character":
for s in ["_tag", "_stroke"]:
if s in self.__dict__:
del self.__dict__[s]
if name == "stroke":
if len(self._stroke) > 0:
self._writing.append_stroke(self._stroke)
self._stroke = None
self._tag = None
def _char_data(self, data):
if self._tag == "utf8":
self._utf8 = data.encode("UTF-8")
elif self._tag == "width":
self._writing.set_width(int(data))
elif self._tag == "height":
self._writing.set_height(int(data))
tegaki-python-0.3.1/tegaki/engine.py 0000644 0001750 0001750 00000006642 11342122456 017243 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2009 The Tegaki project contributors
#
# 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.
import os
import platform
from tegaki.dictutils import SortedDict
class Engine(object):
"""
Base class for Recognizer and Trainer.
"""
@classmethod
def _get_search_path(cls, what):
"""
Return a list of search path.
@typ what: str
@param what: "models" or "engines"
"""
libdir = os.path.dirname(os.path.abspath(__file__))
try:
# UNIX
homedir = os.environ['HOME']
homeengines = os.path.join(homedir, ".tegaki", what)
except KeyError:
# Windows
homedir = os.environ['USERPROFILE']
homeengines = os.path.join(homedir, "tegaki", what)
search_path = [# For Unix
"/usr/local/share/tegaki/%s/" % what,
"/usr/share/tegaki/%s/" % what,
# for Maemo
"/media/mmc1/tegaki/%s/" % what,
"/media/mmc2/tegaki/%s/" % what,
# personal directory
homeengines,
# lib dir
os.path.join(libdir, what)]
# For Windows
try:
search_path += [os.path.join(os.environ["APPDATA"], "tegaki",
what),
r"C:\Python25\share\tegaki\%s" % what,
r"C:\Python26\share\tegaki\%s" % what]
except KeyError:
pass
# For OSX
if platform.system() == "Darwin":
search_path += [os.path.join(homedir, "Library",
"Application Support", "tegaki", what),
os.path.join("/Library", "Application Support",
"tegaki", what)]
try:
env = {"engines": "TEGAKI_ENGINE_PATH",
"models" : "TEGAKI_MODEL_PATH"}[what]
if env in os.environ and \
os.environ[env].strip() != "":
search_path += os.environ[env].strip().split(os.path.pathsep)
except KeyError:
pass
return search_path
@classmethod
def read_meta_file(cls, meta_file):
"""
Read a .meta file.
@type meta_file: str
@param meta_file: meta file file to read
@rtype: dict
"""
f = open(meta_file)
ret = SortedDict()
for line in f.readlines():
try:
key, value = [s.strip() for s in line.strip().split("=")]
ret[key] = value
except ValueError:
continue
f.close()
return ret
tegaki-python-0.3.1/tegaki/mathutils.py 0000644 0001750 0001750 00000002362 11342122456 020003 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
from math import sqrt, hypot, atan2, pi
def euclidean_distance(v1, v2):
assert(len(v1) == len(v2))
return sqrt(sum([(v2[i] - v1[i]) ** 2 for i in range(len(v1))]))
def cartesian_to_polar(x, y):
"""
Cartesian to polar coordinates conversion.
r is the distance to the point.
teta is the angle to the point between 0 and 2 pi.
"""
r = hypot(x, y)
teta = atan2(y, x) + pi
return (r, teta) tegaki-python-0.3.1/tegaki/engines/ 0000755 0001750 0001750 00000000000 11352066723 017051 5 ustar mathieu mathieu tegaki-python-0.3.1/tegaki/engines/__init__.py 0000644 0001750 0001750 00000000000 11342122456 021143 0 ustar mathieu mathieu tegaki-python-0.3.1/tegaki/engines/tegakizinnia.py 0000644 0001750 0001750 00000006362 11342122456 022102 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008-2009 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import os
from tegaki.recognizer import Recognizer, RecognizerError
from tegaki.trainer import Trainer, TrainerError
try:
import zinnia
class ZinniaRecognizer(Recognizer):
RECOGNIZER_NAME = "zinnia"
def __init__(self):
Recognizer.__init__(self)
self._recognizer = zinnia.Recognizer()
def open(self, path):
ret = self._recognizer.open(path)
if not ret: raise RecognizerError, "Could not open!"
def recognize(self, writing, n=10):
s = zinnia.Character()
s.set_width(writing.get_width())
s.set_height(writing.get_height())
strokes = writing.get_strokes()
for i in range(len(strokes)):
stroke = strokes[i]
for x, y in stroke:
s.add(i, x, y)
result = self._recognizer.classify(s, n+1)
size = result.size()
return [(result.value(i), result.score(i)) \
for i in range(0, (size - 1))]
RECOGNIZER_CLASS = ZinniaRecognizer
class ZinniaTrainer(Trainer):
TRAINER_NAME = "zinnia"
def __init__(self):
Trainer.__init__(self)
def train(self, charcol, meta, path=None):
self._check_meta(meta)
trainer = zinnia.Trainer()
zinnia_char = zinnia.Character()
for set_name in charcol.get_set_list():
for character in charcol.get_characters(set_name):
if (not zinnia_char.parse(character.to_sexp())):
raise TrainerError, zinnia_char.what()
else:
trainer.add(zinnia_char)
if not path:
if "path" in meta:
path = meta["path"]
else:
path = os.path.join(os.environ['HOME'], ".tegaki", "models",
"zinnia", meta["name"] + ".model")
else:
path = os.path.abspath(path)
if not os.path.exists(os.path.dirname(path)):
os.makedirs(os.path.dirname(path))
meta_file = path.replace(".model", ".meta")
if not meta_file.endswith(".meta"): meta_file += ".meta"
trainer.train(path)
self._write_meta_file(meta, meta_file)
TRAINER_CLASS = ZinniaTrainer
except ImportError:
pass
tegaki-python-0.3.1/tegaki/trainer.py 0000644 0001750 0001750 00000010034 11342122456 017430 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008-2009 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import glob
import os
import imp
from cStringIO import StringIO
from tegaki.engine import Engine
from tegaki.dictutils import SortedDict
class TrainerError(Exception):
"""
Raised when something went wrong in a Trainer.
"""
pass
class Trainer(Engine):
"""
Base Trainer class.
A trainer can train models based on sample data annotated with labels.
"""
def __init__(self):
pass
@classmethod
def get_available_trainers(cls):
"""
Return trainers installed on the system.
@rtype: dict
@return: a dict where keys are trainer names and values \
are trainer classes
"""
if not "available_trainers" in cls.__dict__:
cls._load_available_trainers()
return cls.available_trainers
@classmethod
def _load_available_trainers(cls):
cls.available_trainers = SortedDict()
for directory in cls._get_search_path("engines"):
if not os.path.exists(directory):
continue
for f in glob.glob(os.path.join(directory, "*.py")):
if f.endswith("__init__.py") or f.endswith("setup.py"):
continue
module_name = os.path.basename(f).replace(".py", "")
module_name += "trainer"
module = imp.load_source(module_name, f)
try:
name = module.TRAINER_CLASS.TRAINER_NAME
cls.available_trainers[name] = module.TRAINER_CLASS
except AttributeError:
pass
def set_options(self, options):
"""
Process trainer/model specific options.
@type options: dict
@param options: a dict where keys are option names and values are \
option values
"""
pass
# To be implemented by child class
def train(self, character_collection, meta, path=None):
"""
Train a model.
@type character_collection: L{CharacterCollection}
@param character_collection: collection containing training data
@type meta: dict
@param meta: meta dict obtained with L{Engine.read_meta_file}
@type path: str
@param path: path to the ouput model \
(if None, the personal directory is assumed)
The meta dict needs the following keys:
- name: full name (mandatory)
- shortname: name with less than 3 characters (mandatory)
- language: model language (optional)
"""
raise NotImplementedError
def _check_meta(self, meta):
if not meta.has_key("name") or not meta.has_key("shortname"):
raise TrainerError, "meta must contain a name and a shortname"
def _write_meta_file(self, meta, meta_file):
io = StringIO()
for k,v in meta.items():
io.write("%s = %s\n" % (k,v))
if os.path.exists(meta_file):
f = open(meta_file)
contents = f.read()
f.close()
# don't rewrite the file if same
if io.getvalue() == contents:
return
f = open(meta_file, "w")
f.write(io.getvalue())
f.close()
tegaki-python-0.3.1/tegaki/arrayutils.py 0000644 0001750 0001750 00000016054 11342122456 020173 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
__doctest__ = True
def array_sample(arr, rate):
"""
Sample array.
@type arr: list/tuple/array
@param arr: the list/tuple/array to sample
@type rate: float
@param rate: the rate (between 0 and 1.0) at which to sample
@rtype: list
@return: the sampled list
>>> array_sample([1,2,3,4,5,6], 0.5)
[1, 3, 5]
"""
n = int(round(1 / rate))
return [arr[i] for i in range(0, len(arr), n)]
def array_flatten(l, ltypes=(list, tuple)):
"""
Reduce array of possibly multiple dimensions to one dimension.
@type l: list/tuple/array
@param l: the list/tuple/array to flatten
@rtype: list
@return: the flatten list
>>> array_flatten([[1,2,3], [4,5], [[7,8]]])
[1, 2, 3, 4, 5, 7, 8]
"""
i = 0
while i < len(l):
while isinstance(l[i], ltypes):
if not l[i]:
l.pop(i)
if not len(l):
break
else:
l[i:i+1] = list(l[i])
i += 1
return l
def array_reshape(arr, n):
"""
Reshape one-dimensional array to a list of n-element lists.
@type arr: list/tuple/array
@param arr: the array to reshape
@type n: int
@param n: the number of elements in each list
@rtype: list
@return: the reshaped array
>>> array_reshape([1,2,3,4,5,6,7,8,9], 3)
[[1, 2, 3], [4, 5, 6], [7, 8, 9]]
"""
newarr = []
subarr = []
i = 0
for ele in arr:
subarr.append(ele)
i += 1
if i % n == 0 and i != 0:
newarr.append(subarr)
subarr = []
return newarr
def array_split(seq, p):
"""
Split an array into p arrays of about the same size.
@type seq: list/tuple/array
@param seq: the array to split
@type p: int
@param p: the split size
@rtype: list
@return: the split array
>>> array_split([1,2,3,4,5,6,7], 3)
[[1, 2, 3], [4, 5], [6, 7]]
"""
newseq = []
n = len(seq) / p # min items per subsequence
r = len(seq) % p # remaindered items
b,e = 0, n + min(1, r) # first split
for i in range(p):
newseq.append(seq[b:e])
r = max(0, r-1) # use up remainders
b,e = e, e + n + min(1, r) # min(1,r) is always 0 or 1
return newseq
def array_mean(arr):
"""
Calculate the mean of the elements contained in an array.
@type arr: list/tuple/array
@rtype: float
@return: mean
>>> array_mean([100, 150, 300])
183.33333333333334
"""
return float(sum(arr)) / float(len(arr))
def array_variance(arr, mean=None):
"""
Calculate the variance of the elements contained in an array.
@type arr: list/tuple/array
@rtype: float
@return: variance
>>> array_variance([100, 150, 300])
7222.2222222222226
"""
if mean is None:
mean = array_mean(arr)
var = array_mean([(val - mean) ** 2 for val in arr])
if var == 0.0:
return 1.0
else:
return var
def array_mean_vector(vectors):
"""
Calculate the mean of the vectors, element-wise.
@type arr: list of vectors
@rtype: list of floats
@return: list of means
>>> array_mean_vector([[10,20], [100, 200]])
[55.0, 110.0]
"""
assert(len(vectors) > 0)
n_dimensions = len(vectors[0])
mean_vector = []
for i in range(n_dimensions):
arr = [vector[i] for vector in vectors]
mean_vector.append(array_mean(arr))
return mean_vector
def array_variance_vector(vectors, means=None):
"""
Calculate the variance of the vectors, element-wise.
@type arr: list of vectors
@rtype: list of floats
@return: list of variances
>>> array_variance_vector([[10,20], [100, 200]])
[2025.0, 8100.0]
"""
assert(len(vectors) > 0)
n_dimensions = len(vectors[0])
if means is not None:
assert(n_dimensions == len(means))
else:
means = array_mean_vector(vectors)
variance_vector = []
for i in range(n_dimensions):
arr = [vector[i] for vector in vectors]
variance_vector.append(array_variance(arr, means[i]))
return variance_vector
def array_covariance_matrix(vectors, non_diagonal=False):
"""
Calculate the covariance matrix of vectors.
@type vectors: list of arrays
@type non_diagonal: boolean
@param non_diagonal: whether to calculate non-diagonal elements of the \
matrix or not
>>> array_covariance_matrix([[10,20], [100, 200]])
[2025.0, 0.0, 0.0, 8100.0]
>>> array_covariance_matrix([[10,20], [100, 200]], non_diagonal=True)
[2025.0, 4050.0, 4050.0, 8100.0]
"""
assert(len(vectors) > 0)
n_dimensions = len(vectors[0])
cov_matrix = []
for i in range(n_dimensions):
for j in range(n_dimensions):
if i == j:
# diagonal value: COV(X,X) = VAR(X)
arr = [vector[i] for vector in vectors]
cov_matrix.append(array_variance(arr))
else:
# non-diagonal value
if non_diagonal:
# COV(X,Y) = E(XY) - E(X)E(Y)
arr_x = [vector[i] for vector in vectors]
arr_y = [vector[j] for vector in vectors]
arr_xy = array_mul(arr_x, arr_y)
mean_xy = array_mean(arr_xy)
mean_x = array_mean(arr_x)
mean_y = array_mean(arr_y)
cov_matrix.append(mean_xy - mean_x * mean_y)
else:
# X and Y indep => COV(X,Y) = 0
cov_matrix.append(0.0)
return cov_matrix
def array_add(arr1, arr2):
"""
Add two arrays element-wise.
>>> array_add([1,2],[3,4])
[4, 6]
"""
assert(len(arr1) == len(arr1))
newarr = []
for i in range(len(arr1)):
newarr.append(arr1[i] + arr2[i])
return newarr
def array_mul(arr1, arr2):
"""
Multiply two arrays element-wise.
>>> array_mul([1,2],[3,4])
[3, 8]
"""
assert(len(arr1) == len(arr1))
newarr = []
for i in range(len(arr1)):
newarr.append(arr1[i] * arr2[i])
return newarr
if __name__ == '__main__':
import doctest
doctest.testmod() tegaki-python-0.3.1/tegaki/recognizer.py 0000644 0001750 0001750 00000016714 11342122456 020146 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
# Copyright (C) 2008-2009 The Tegaki project contributors
#
# 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.
# Contributors to this file:
# - Mathieu Blondel
import glob
import os
import imp
from tegaki.engine import Engine
from tegaki.dictutils import SortedDict
class RecognizerError(Exception):
"""
Raised when something went wrong in a Recognizer.
"""
pass
class Recognizer(Engine):
"""
Base Recognizer class.
A recognizer can recognize handwritten characters based on a model.
The L{open} method should be used to load a model from an
absolute path on the disk.
The L{set_model} method should be used to load a model from its name.
Two models can't have the same name within one recognizer.
However, two models can be named the same if they belong to two different
recognizers.
Recognizers usually have a corresponding L{Trainer}.
"""
def __init__(self):
self._model = None
@classmethod
def get_available_recognizers(cls):
"""
Return recognizers installed on the system.
@rtype: dict
@return: a dict where keys are recognizer names and values \
are recognizer classes
"""
if not "available_recognizers" in cls.__dict__:
cls._load_available_recognizers()
return cls.available_recognizers
@classmethod
def _load_available_recognizers(cls):
cls.available_recognizers = SortedDict()
for directory in cls._get_search_path("engines"):
if not os.path.exists(directory):
continue
for f in glob.glob(os.path.join(directory, "*.py")):
if f.endswith("__init__.py") or f.endswith("setup.py"):
continue
module_name = os.path.basename(f).replace(".py", "")
module_name += "recognizer"
module = imp.load_source(module_name, f)
try:
name = module.RECOGNIZER_CLASS.RECOGNIZER_NAME
cls.available_recognizers[name] = module.RECOGNIZER_CLASS
except AttributeError:
pass
@staticmethod
def get_all_available_models():
"""
Return available models from all recognizers.
@rtype: list
@return: a list of tuples (recognizer_name, model_name, meta_dict)
"""
all_models = []
for r_name, klass in Recognizer.get_available_recognizers().items():
for model_name, meta in klass.get_available_models().items():
all_models.append([r_name, model_name, meta])
return all_models
@classmethod
def get_available_models(cls):
"""
Return available models for the current recognizer.
@rtype; dict
@return: a dict where keys are models names and values are meta dict
"""
if "available_models" in cls.__dict__:
return cls.available_models
else:
name = cls.RECOGNIZER_NAME
cls.available_models = cls._get_available_models(name)
return cls.__dict__["available_models"]
@classmethod
def _get_available_models(cls, recognizer):
available_models = SortedDict()
for directory in cls._get_search_path("models"):
directory = os.path.join(directory, recognizer)
if not os.path.exists(directory):
continue
meta_files = glob.glob(os.path.join(directory, "*.meta"))
for meta_file in meta_files:
meta = Recognizer.read_meta_file(meta_file)
if not meta.has_key("name") or \
not meta.has_key("shortname"):
continue
model_file = meta_file.replace(".meta", ".model")
if meta.has_key("path") and not os.path.exists(meta["path"]):
# skip model if specified path is incorrect
continue
elif not meta.has_key("path") and os.path.exists(model_file):
# if path option is missing, assume the .model file
# is in the same directory
meta["path"] = model_file
available_models[meta["name"]] = meta
return available_models
def open(self, path):
"""
Open a model.
@type path: str
@param path: model path
Raises RecognizerError if could not open.
"""
raise NotImplementedError
def set_options(self, options):
"""
Process recognizer/model specific options.
@type options: dict
@param options: a dict where keys are option names and values are \
option values
"""
pass
def get_model(self):
"""
Return the currently selected model.
@rtype: str
@return: name which identifies model uniquely on the system
"""
return self._model
def set_model(self, model_name):
"""
Set the currently selected model.
@type model_name: str
@param model_name: name which identifies model uniquely on the system
model_name must exist for that recognizer.
"""
if not model_name in self.__class__.get_available_models():
raise RecognizerError, "Model does not exist"
self._model = model_name
meta = self.__class__.get_available_models()[model_name]
self.set_options(meta)
path = meta["path"]
self.open(path)
# To be implemented by child class
def recognize(self, writing, n=10):
"""
Recognizes handwriting.
@type writing: L{Writing}
@param writing: the handwriting to recognize
@type n: int
@param n: the number of candidates to return
@rtype: list
@return: a list of tuple (label, probability/distance)
A model must be loaded with open or set_model() beforehand.
"""
raise NotImplementedError
if __name__ == "__main__":
import sys
from tegaki.character import Character
recognizer = sys.argv[1] # name of recognizer
model = sys.argv[2] # name of model file
char = Character()
char.read(sys.argv[3]) # path of .xml file
writing = char.get_writing()
recognizers = Recognizer.get_available_recognizers()
print "Available recognizers", recognizers
if not recognizer in recognizers:
raise Exception, "Not an available recognizer"
recognizer_klass = recognizers[recognizer]
recognizer = recognizer_klass()
models = recognizer_klass.get_available_models()
print "Available models", models
if not model in models:
raise Exception, "Not an available model"
recognizer.set_model(model)
print recognizer.recognize(writing)
tegaki-python-0.3.1/README 0000644 0001750 0001750 00000001471 11342122457 015034 0 ustar mathieu mathieu tegaki-python
=============
Base Python library for the Tegaki project.
Requirements
------------
Python 2.4 http://www.python.org
Zinnia * 0.05 http://zinnia.sourceforge.net/
[*] The Python bindings are also required.
Install
-------
($ su)
# python setup.py install
Developer
---------
If you intend to contribute to Tegaki, the most convenient way to install it
is to add the following to your ~/.bashrc file.
tegakiroot=/path/to/hwr/
export PYTHONPATH=$tegakiroot/tegaki-python:$tegakiroot/tegaki-pygtk
In order to download the latest source code, please use git.
$ git clone http://www.mblondel.org/code/hwr.git
License
-------
This package is distributed under the terms of the GNU General Public
License. See the COPYING file for more information.
Homepage
--------
http://www.tegaki.org
tegaki-python-0.3.1/ChangeLog 0000644 0001750 0001750 00000023443 11352066723 015735 0 ustar mathieu mathieu commit 5b0b86343c91cbf3d7ee683e82f41f9df75edf4d
Author: Mathieu Blondel
Date: Mon Mar 22 15:32:33 2010 +0900
[all] Set current version to 0.3.1.
commit d232e5c84913508872ab3c161d1dfbf3cbefad6a
Author: Mathieu Blondel
Date: Tue Jan 12 13:43:05 2010 +0900
[tegaki-python] Fixed bug in CharacterCollection.remove_samples.
commit 48884434905f6c8e8b103b1acccaf26dfb957365
Author: Mathieu Blondel
Date: Wed Dec 23 20:25:51 2009 +0900
[tegaki-python] Added get_size() and set_size().
commit dc82b0befb3339f69193be3f68900ace4eff0045
Author: Mathieu Blondel
Date: Fri Dec 11 22:24:28 2009 +0900
[tegaki-python] Added __repr__ for a few objects.
commit 9b396782428e65e2b97b2dfe1b5b8a9362287104
Author: Mathieu Blondel
Date: Thu Dec 10 17:44:16 2009 +0900
[tegaki-python] Fixed pb in CharacterCollection.get_n_characters().
commit 47afc05133c10b29a38d7aec1e1f180d69f4e365
Author: Mathieu Blondel
Date: Wed Dec 9 17:32:18 2009 +0900
[tegaki-python] Updated ToDo list.
commit 62a09950deea0094f16974bf509c3ee03bfca2da
Author: Mathieu Blondel
Date: Wed Dec 9 14:10:39 2009 +0900
[tegaki-python] Various optimizations to CharacterCollection.
commit 3ce946ca19fe4b13a737b5e3ec2834cdab30ee6c
Author: Mathieu Blondel
Date: Fri Dec 4 21:52:28 2009 +0900
[tegaki-python] Added get_total_n_strokes to CharacterCollection.
commit e26b64edc059dd40a33f259546b6dc2da1ace52d
Author: Mathieu Blondel
Date: Thu Dec 3 18:48:04 2009 +0900
[tegaki-python] Added get_random_characters() to CharacterCollection.
commit cb9274dd1e879e80ecb219d268a3f3c9b7de5cb7
Author: Mathieu Blondel
Date: Thu Dec 3 16:36:33 2009 +0900
[tegaki-python] Renamed PROXY option to WRITE_BACK.
commit 50cad5c4779dde302de7fec6fb6af5b763797617
Author: Mathieu Blondel
Date: Thu Dec 3 16:30:42 2009 +0900
[tegaki-python] Added get_average_n_strokes to CharacterCollection.
commit 9d01b64a775f15733a9a62098fe534faa9b33d79
Author: Mathieu Blondel
Date: Thu Dec 3 16:22:11 2009 +0900
[tegaki-python] Sync SortedDict with upstream.
commit a656fed9c7473daa790fc52179bd602800e4ff7e
Author: Mathieu Blondel
Date: Thu Dec 3 16:01:51 2009 +0900
[tegaki-python] Fixed variance.
commit b0106a723d71c2f79f01d9a8f8b6e2bb05f18a13
Author: Mathieu Blondel
Date: Thu Dec 3 12:43:49 2009 +0900
[tegaki-python] Added generator-based methods for get_characters and get_all_characters.
Use these methods when you retrieve a lot of characters.
commit 4257abf1f0c8d5302bc874ce46dae783def9f12d
Author: Mathieu Blondel
Date: Thu Dec 3 12:05:37 2009 +0900
[tegaki-python] Run doctests as well in runtests.py.
commit 167c43ea557d08f4b4e31ccda45750bd231211ee
Author: Mathieu Blondel
Date: Wed Dec 2 20:57:02 2009 +0900
[tegaki-python] More documentation about Character and CharacterCollection.
commit f95f32762ac38cd12b6659593353dd7b5c1a49be
Author: Mathieu Blondel
Date: Wed Dec 2 20:37:26 2009 +0900
[tegaki-python] Added add_sets() method to CharacterCollection.
commit 164d927e1b399e411c7d8519c70845d344b321dc
Author: Mathieu Blondel
Date: Wed Dec 2 20:22:28 2009 +0900
[tegaki-python] Handle XML files CharacterCollection constructor and add a new save() method.
commit 3ad6e525dd61bfa15ae2539042d97ee9759ede64
Author: Mathieu Blondel
Date: Wed Dec 2 18:34:26 2009 +0900
[tegaki-python] Added optional path argument to Character constructor and a new save() method.
commit a092a357343c67826efb1466ad33bc470c3487d7
Author: Mathieu Blondel
Date: Wed Dec 2 16:45:03 2009 +0900
[tegaki-python] Added inline documentation.
commit 58fbebb75d7ee6b5484c2838c0272b950dc1e35a
Author: Mathieu Blondel
Date: Tue Dec 1 17:52:32 2009 +0900
[tegaki-python] Moved read_meta_file to Engine class.
commit 9829c06728ffe4ab00c7f5766090d467e178e9c1
Author: Mathieu Blondel
Date: Tue Dec 1 17:26:17 2009 +0900
[tegaki-python] CharacterCollection now uses sqlite as backend.
commit 04f1db381e24f20f3b406694e092a810186e944a
Author: Mathieu Blondel
Date: Sun Nov 29 21:26:48 2009 +0900
[tegaki-python] Fixed system-wide path on Mac OS X.
commit 23428970b5125e6effc233ff3261e6cb368ff93d
Author: Mathieu Blondel
Date: Sun Nov 1 20:56:48 2009 +0900
[all] Include ChangeLog.
commit ac086c4a5e34df5552483b26cee5ff1f6d0aad2b
Author: Mathieu Blondel
Date: Sun Nov 1 19:34:42 2009 +0900
[tegaki-python] Removed NEWS file.
commit 4ac1c409c8b6e80fe86e6109837685f7fa390556
Author: Mathieu Blondel
Date: Sun Oct 25 20:16:59 2009 +0900
[tegaki-python] Bumped to version 0.3
commit 3014ea2a47ba070169e270b1fd2ff104708bd64b
Author: Mathieu Blondel
Date: Mon Oct 19 17:38:21 2009 +0900
[tegaki-python] Added include_characters_from_files() and exclude_characters_from_files() to CharacterCollection.
commit 5cad9cbf608c32adc6209848a28bb5b02b8cccb7
Author: Mathieu Blondel
Date: Sun Oct 11 05:21:05 2009 +0900
[tegaki-engines] Fixed bug when meta file is a relative path.
commit e055d58ef725ab0d7f1a94cebcc3edcc7983155b
Author: Mathieu Blondel
Date: Sun Sep 27 02:27:43 2009 +0900
[tegaki-python] Added directories for win32 and osx plus refactored some code.
commit 799a160957b39ee1d87cad99c64366dfb368bbf7
Author: Mathieu Blondel
Date: Sun Sep 27 01:30:07 2009 +0900
[all] Distutils on python 2.6 for win32 doesn't like end slashes for directories.
commit 8d081be5fe5dafa2e295cd2c3f8abbaaf2626fde
Author: Mathieu Blondel
Date: Fri Sep 18 01:44:35 2009 +0900
[tegaki-python] Added remove_samples() to CharacterCollection.
commit 2686e5c55b1a801508a1ff9441de83d15d51dbbd
Author: Mathieu Blondel
Date: Thu Sep 17 20:23:57 2009 +0900
[tegaki-python] Added a check_duplicate option to __add__.
commit 2261ed94de25d0202d1e841f1f5c6ce9053035c1
Author: Mathieu Blondel
Date: Thu Sep 17 00:32:28 2009 +0900
[tegaki-lab] Added Jahmm (http://code.google.com/p/jahmm/) support when run with Jython2.5.
commit dee6ba1a576041f0f1c8d34a00e74cfb1685bd2a
Author: Mathieu Blondel
Date: Sat Sep 12 18:07:23 2009 +0900
[tegaki-python] A few new methods + tests.
commit 6d85820de3d29bc6db4893c6ab2fa94ade514019
Author: Mathieu Blondel
Date: Wed Aug 19 10:19:48 2009 +0900
[tegaki-python] The engines didn't get installed.
commit c62e4450127c064d4f11ab708a2e1696d62babf6
Author: Mathieu Blondel
Date: Fri Jul 31 21:30:41 2009 +0900
[tegaki-python] Don't rewrite .meta file if its content didn't change.
commit 59f1a1d79a7ba289abeba9434267d2be2b12468b
Author: Mathieu Blondel
Date: Fri Jul 31 20:44:17 2009 +0900
[tegaki-python] Support recognizer/trainer options.
commit 296311816535680513cbaf29d4d359f0a9049d35
Author: ian johnson
Date: Tue Jul 28 14:18:55 2009 -0400
[tegaki-db] merging with Erik's branch
commit ef55bbf1902c3fcd2252cf327829c4a6c664cc24
Merge: 6d58e08 19e72ea
Author: ian johnson
Date: Tue Jul 28 13:16:47 2009 -0400
Merge branch 'master' of git://github.com/enoex/hwr
commit 19e72eaf1dd3cc26654f8e9e93af4e5b32dff3b8
Author: erik
Date: Mon Jul 27 16:33:32 2009 -0400
Updated layout for most pages
commit a33a4ad2adb308212537b13a76b1f50ff3952636
Author: Mathieu Blondel
Date: Sun Jul 26 16:18:04 2009 +0200
[tegaki-python] Don't load empty strokes.
commit 7e679657da6bf4aac8a215f207aafdc74329ea08
Author: Mathieu Blondel
Date: Sun Jul 26 16:06:37 2009 +0200
[tegaki-python] Don't load modules called setup.py.
commit a10ef239742e7d42c38a9183b3058f723a3cb21b
Author: Mathieu Blondel
Date: Fri Jul 24 08:34:53 2009 +0200
[tegaki-python] Added TEGAKI_ENGINE_PATH and TEGAKI_MODEL_PATH environment variables.
commit 430adef612884ea7b2bc619851eac391932063c7
Author: Mathieu Blondel
Date: Fri Jul 24 05:20:52 2009 +0200
[tegaki-python] Merged recognizers and trainers into one file.
This is more convenient.
commit 05cb3ee3471599829412d8ccc3bc4adabe581dea
Author: Mathieu Blondel
Date: Thu Jul 23 20:38:38 2009 +0900
[tegaki-python] Pluginify Recognizer and Trainer.
commit 072589a2417565b3697ae3abc8e547b06ccde39d
Author: Mathieu Blondel
Date: Sun Jul 19 11:45:53 2009 +0200
[all] Update project URL.
commit f09bf2f5a99907d9484758f214b8f8c9cda3ba9e
Author: Mathieu Blondel
Date: Sun Jul 19 11:37:41 2009 +0200
[tegaki-python] Fixed setup.py.
commit beaaed5713b0ce999a1e13fec2f4b9a2b864f8f9
Author: Mathieu Blondel
Date: Sun Jul 19 07:07:18 2009 +0200
[tegaki-python] Updated README.
commit be3647c410dadf5a647973312728976812e15c64
Author: Mathieu Blondel
Date: Sun Jul 19 05:48:36 2009 +0200
[tegaki-python] Added NEWS file.
commit 10ea23607f9a0879e1c1c459cf83e763286871da
Author: Mathieu Blondel
Date: Sun Jul 19 05:23:37 2009 +0200
[tegaki,tegaki-gtk] Renamed the packages to respectively tegaki-python and tegaki-pygtk to make it clear that they are Python packages.
tegaki-python-0.3.1/setup.py 0000644 0001750 0001750 00000001371 11342122456 015664 0 ustar mathieu mathieu # -*- coding: utf-8 -*-
from distutils.core import setup
import os
def getversion():
currdir = os.path.dirname(os.path.abspath(__file__))
path = os.path.join(currdir, "tegaki", "__init__.py")
import re
regexp = re.compile(r"VERSION = '([^']*)'")
f = open(path)
buf = f.read()
f.close()
return regexp.search(buf).group(1)
# Please run
# python setup.py install
setup(
name = 'tegaki-python',
description = 'Tegaki Python library',
author = 'Mathieu Blondel',
author_email = 'mathieu ÂT mblondel DÔT org',
url = 'http://www.tegaki.org',
version = getversion(),
license='GPL',
packages = ['tegaki', 'tegaki.engines'],
package_dir = {'tegaki':'tegaki', 'tegaki.engines':'tegaki/engines'}
) tegaki-python-0.3.1/MANIFEST.in 0000644 0001750 0001750 00000000244 11342122456 015706 0 ustar mathieu mathieu include AUTHORS COPYING COPYRIGHT README TODO MANIFEST.in ChangeLog
exclude *~ *.orig *.bak *.tmp *.pyc MANIFEST
recursive-include tests *
recursive-include data *
tegaki-python-0.3.1/COPYRIGHT 0000644 0001750 0001750 00000000274 11342122457 015447 0 ustar mathieu mathieu The code included in this package is copyrighted by Mathieu Blondel :
Copyright (C) 2008 Mathieu Blondel. All Rights Reserved.
See the COPYING File for the terms of distribution licence. tegaki-python-0.3.1/TODO 0000644 0001750 0001750 00000000570 11342122456 014642 0 ustar mathieu mathieu - Character transition model (bigram)
- API to retrieve the most suitable recognizer and model available for some task
or given some requirements (language, characters included, stroke-order
free...)
-> This is useful for applications (e.g. games) that want to use Tegaki
as a service without asking the user to explicitly choose a recognizer
or model.