︙ | | | ︙ | |
67
68
69
70
71
72
73
74
75
76
77
78
79
80
|
_sAppContext = sContext
_dOptions = dict(gc_options.getOptions(sContext)) # duplication necessary, to be able to reset to default
_oTokenizer = _oSpellChecker.getTokenizer()
_oSpellChecker.activateStorage()
except:
traceback.print_exc()
def _getRules (bParagraph):
try:
if not bParagraph:
return _rules.lSentenceRules
return _rules.lParagraphRules
except:
|
>
>
>
>
>
>
>
|
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
|
_sAppContext = sContext
_dOptions = dict(gc_options.getOptions(sContext)) # duplication necessary, to be able to reset to default
_oTokenizer = _oSpellChecker.getTokenizer()
_oSpellChecker.activateStorage()
except:
traceback.print_exc()
def getSpellChecker ():
"return the spellchecker object"
return _oSpellChecker
#### Rules
def _getRules (bParagraph):
try:
if not bParagraph:
return _rules.lSentenceRules
return _rules.lParagraphRules
except:
|
︙ | | | ︙ | |
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
|
try:
aRule[0] = re.compile(aRule[0])
except:
echo("Bad regular expression in # " + str(aRule[2]))
aRule[0] = "(?i)<Grammalecte>"
#### Rules and options
def ignoreRule (sRuleId):
"disable rule <sRuleId>"
_aIgnoredRules.add(sRuleId)
def resetIgnoreRules ():
"clear all ignored rules"
|
<
<
|
105
106
107
108
109
110
111
112
113
114
115
116
117
118
|
try:
aRule[0] = re.compile(aRule[0])
except:
echo("Bad regular expression in # " + str(aRule[2]))
aRule[0] = "(?i)<Grammalecte>"
def ignoreRule (sRuleId):
"disable rule <sRuleId>"
_aIgnoredRules.add(sRuleId)
def resetIgnoreRules ():
"clear all ignored rules"
|
︙ | | | ︙ | |
136
137
138
139
140
141
142
143
144
145
146
147
148
149
|
def displayRules (sFilter=None):
"display the name of rules, with the filter <sFilter>"
echo("List of rules. Filter: << " + str(sFilter) + " >>")
for sOption, sLineId, sRuleId in listRegexRules(sFilter):
echo("{:<10} {:<10} {}".format(sOption, sLineId, sRuleId))
def setOption (sOpt, bVal):
"set option <sOpt> with <bVal> if it exists"
if sOpt in _dOptions:
_dOptions[sOpt] = bVal
|
>
>
|
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
|
def displayRules (sFilter=None):
"display the name of rules, with the filter <sFilter>"
echo("List of rules. Filter: << " + str(sFilter) + " >>")
for sOption, sLineId, sRuleId in listRegexRules(sFilter):
echo("{:<10} {:<10} {}".format(sOption, sLineId, sRuleId))
#### Options
def setOption (sOpt, bVal):
"set option <sOpt> with <bVal> if it exists"
if sOpt in _dOptions:
_dOptions[sOpt] = bVal
|
︙ | | | ︙ | |
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
|
def resetOptions ():
"set options to default values"
global _dOptions
_dOptions = dict(gc_options.getOptions(_sAppContext))
def getSpellChecker ():
"return the spellchecker object"
return _oSpellChecker
#### Parsing
_zEndOfSentence = re.compile(r'([.?!:;…][ .?!… »”")]*|.$)')
_zBeginOfParagraph = re.compile(r"^\W*")
_zEndOfParagraph = re.compile(r"\W*$")
|
<
<
<
<
<
|
184
185
186
187
188
189
190
191
192
193
194
195
196
197
|
def resetOptions ():
"set options to default values"
global _dOptions
_dOptions = dict(gc_options.getOptions(_sAppContext))
#### Parsing
_zEndOfSentence = re.compile(r'([.?!:;…][ .?!… »”")]*|.$)')
_zBeginOfParagraph = re.compile(r"^\W*")
_zEndOfParagraph = re.compile(r"\W*$")
|
︙ | | | ︙ | |
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
|
#for nPos, dToken in self.dTokenPos.items():
# s += "{}\t{}\n".format(nPos, dToken)
return s
def parse (self, sCountry="${country_default}", bDebug=False, dOptions=None, bContext=False):
"analyses the paragraph sText and returns list of errors"
#sText = unicodedata.normalize("NFC", sText)
dOpt = _dOptions if not dOptions else dOptions
bShowRuleId = option('idrule')
# parse paragraph
try:
self.parseText(self.sText, self.sText0, True, 0, sCountry, dOpt, bShowRuleId, bDebug, bContext)
except:
raise
|
|
|
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
|
#for nPos, dToken in self.dTokenPos.items():
# s += "{}\t{}\n".format(nPos, dToken)
return s
def parse (self, sCountry="${country_default}", bDebug=False, dOptions=None, bContext=False):
"analyses the paragraph sText and returns list of errors"
#sText = unicodedata.normalize("NFC", sText)
dOpt = dOptions or _dOptions
bShowRuleId = option('idrule')
# parse paragraph
try:
self.parseText(self.sText, self.sText0, True, 0, sCountry, dOpt, bShowRuleId, bDebug, bContext)
except:
raise
|
︙ | | | ︙ | |
337
338
339
340
341
342
343
344
345
346
347
348
349
350
|
def update (self, sSentence, bDebug=False):
"update <sSentence> and retokenize"
self.sSentence = sSentence
lNewToken = list(_oTokenizer.genTokens(sSentence, True))
for dToken in lNewToken:
if "lMorph" in self.dTokenPos.get(dToken["nStart"], {}):
dToken["lMorph"] = self.dTokenPos[dToken["nStart"]]["lMorph"]
self.lToken = lNewToken
self.dTokenPos = { dToken["nStart"]: dToken for dToken in self.lToken if dToken["sType"] != "INFO" }
if bDebug:
echo("UPDATE:")
echo(self)
def _getNextPointers (self, dToken, dGraph, dPointer, bDebug=False):
|
>
>
|
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
|
def update (self, sSentence, bDebug=False):
"update <sSentence> and retokenize"
self.sSentence = sSentence
lNewToken = list(_oTokenizer.genTokens(sSentence, True))
for dToken in lNewToken:
if "lMorph" in self.dTokenPos.get(dToken["nStart"], {}):
dToken["lMorph"] = self.dTokenPos[dToken["nStart"]]["lMorph"]
if "tags" in self.dTokenPos.get(dToken["nStart"], {}):
dToken["tags"] = self.dTokenPos[dToken["nStart"]]["tags"]
self.lToken = lNewToken
self.dTokenPos = { dToken["nStart"]: dToken for dToken in self.lToken if dToken["sType"] != "INFO" }
if bDebug:
echo("UPDATE:")
echo(self)
def _getNextPointers (self, dToken, dGraph, dPointer, bDebug=False):
|
︙ | | | ︙ | |
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
|
bTokenFound = True
elif "¬" in sMeta:
if dToken["sType"] not in sMeta:
if bDebug:
echo(" MATCH: *" + sMeta)
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<meta>"][sMeta]] }
bTokenFound = True
if "bKeep" in dPointer and not bTokenFound:
yield dPointer
# JUMP
# Warning! Recurssion!
if "<>" in dNode:
dPointer2 = { "iNode1": iNode1, "dNode": dGraph[dNode["<>"]], "bKeep": True }
yield from self._getNextPointers(dToken, dGraph, dPointer2, bDebug)
def parseGraph (self, dGraph, sCountry="${country_default}", dOptions=None, bShowRuleId=False, bDebug=False, bContext=False):
"parse graph with tokens from the text and execute actions encountered"
dOpt = _dOptions if not dOptions else dOptions
lPointer = []
bTagAndRewrite = False
for iToken, dToken in enumerate(self.lToken):
if bDebug:
echo("TOKEN: " + dToken["sValue"])
# check arcs for each existing pointer
lNextPointer = []
for dPointer in lPointer:
lNextPointer.extend(self._getNextPointers(dToken, dGraph, dPointer, bDebug))
lPointer = lNextPointer
# check arcs of first nodes
lPointer.extend(self._getNextPointers(dToken, dGraph, { "iNode1": iToken, "dNode": dGraph[0] }, bDebug))
# check if there is rules to check for each pointer
for dPointer in lPointer:
#if bDebug:
# echo("+", dPointer)
if "<rules>" in dPointer["dNode"]:
bChange = self._executeActions(dGraph, dPointer["dNode"]["<rules>"], dPointer["iNode1"]-1, iToken, dOpt, sCountry, bShowRuleId, bDebug, bContext)
if bChange:
bTagAndRewrite = True
if bTagAndRewrite:
self.rewriteFromTags(bDebug)
if bDebug:
echo(self)
return self.sSentence
|
|
<
|
|
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
|
bTokenFound = True
elif "¬" in sMeta:
if dToken["sType"] not in sMeta:
if bDebug:
echo(" MATCH: *" + sMeta)
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<meta>"][sMeta]] }
bTokenFound = True
if not bTokenFound and "bKeep" in dPointer:
yield dPointer
# JUMP
# Warning! Recurssion!
if "<>" in dNode:
dPointer2 = { "iNode1": iNode1, "dNode": dGraph[dNode["<>"]], "bKeep": True }
yield from self._getNextPointers(dToken, dGraph, dPointer2, bDebug)
def parseGraph (self, dGraph, sCountry="${country_default}", dOptions=None, bShowRuleId=False, bDebug=False, bContext=False):
"parse graph with tokens from the text and execute actions encountered"
lPointer = []
bTagAndRewrite = False
for iToken, dToken in enumerate(self.lToken):
if bDebug:
echo("TOKEN: " + dToken["sValue"])
# check arcs for each existing pointer
lNextPointer = []
for dPointer in lPointer:
lNextPointer.extend(self._getNextPointers(dToken, dGraph, dPointer, bDebug))
lPointer = lNextPointer
# check arcs of first nodes
lPointer.extend(self._getNextPointers(dToken, dGraph, { "iNode1": iToken, "dNode": dGraph[0] }, bDebug))
# check if there is rules to check for each pointer
for dPointer in lPointer:
#if bDebug:
# echo("+", dPointer)
if "<rules>" in dPointer["dNode"]:
bChange = self._executeActions(dGraph, dPointer["dNode"]["<rules>"], dPointer["iNode1"]-1, iToken, dOptions, sCountry, bShowRuleId, bDebug, bContext)
if bChange:
bTagAndRewrite = True
if bTagAndRewrite:
self.rewriteFromTags(bDebug)
if bDebug:
echo(self)
return self.sSentence
|
︙ | | | ︙ | |