Grammalecte  Diff

"""
Grammalecte
Grammar checker engine
"""

import re
import sys
import os
import traceback
#import unicodedata
from itertools import chain

# functions
_createRegexError = None


#### Initialization

def load (sContext="Python"):
    "initialization of the grammar checker"
    global _oSpellChecker
    global _sAppContext
    global _dOptions
    global _oTokenizer
    global _createRegexError

    try:
        _oSpellChecker = SpellChecker("${lang}", "${dic_main_filename_py}", "${dic_extended_filename_py}", "${dic_community_filename_py}", "${dic_personal_filename_py}")
        _sAppContext = sContext
        _dOptions = dict(gc_options.getOptions(sContext))   # duplication necessary, to be able to reset to default
        _oTokenizer = _oSpellChecker.getTokenizer()
        _oSpellChecker.activateStorage()
        _createRegexError = _createRegexWriterError  if _bWriterError  else _createRegexDictError

    sMessage = globals()[sMsg[1:]](s, m)  if sMsg[0:1] == "="  else  m.expand(sMsg)
    xErr.aShortComment = sMessage   # sMessage.split("|")[0]     # in context menu
    xErr.aFullComment = sMessage   # sMessage.split("|")[-1]    # in dialog
    if bShowRuleId:
        xErr.aShortComment += "  # " + sLineId + " # " + sRuleId
    # URL
    if sURL:
        xProperty = PropertyValue()
        xProperty.Name = "FullCommentURL"
        xProperty.Value = sURL
        xErr.aProperties = (xProperty,)
    else:
        xErr.aProperties = ()
    return xErr


def _createRegexDictError (s, sx, sRepl, nOffset, m, iGroup, sLineId, sRuleId, bUppercase, sMsg, sURL, bShowRuleId, sOption, bContext):
    "error as a dictionary"

    else:
        sNew = m.expand(sRepl)
        sNew = sNew + " " * (nLen-len(sNew))
    return sSentence[0:m.start(iGroup)] + sNew + sSentence[m.end(iGroup):]


def ignoreRule (sRuleId):
    "disable rule <sRuleId>"
    _aIgnoredRules.add(sRuleId)


def resetIgnoreRules ():
    "clear all ignored rules"
    _aIgnoredRules.clear()


def reactivateRule (sRuleId):
    "(re)activate rule <sRuleId>"
    _aIgnoredRules.discard(sRuleId)


def listRules (sFilter=None):
    "generator: returns typle (sOption, sLineId, sRuleId)"
    if sFilter:
        try:
            zFilter = re.compile(sFilter)
        except:
            echo("# Error. List rules: wrong regex.")
            sFilter = None
    for sOption, lRuleGroup in chain(_getRules(True), _getRules(False)):
        if sOption != "@@@@":
            for _, _, sLineId, sRuleId, _, _ in lRuleGroup:
                if not sFilter or zFilter.search(sRuleId):
                    yield (sOption, sLineId, sRuleId)


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 listRules(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


def setOptions (dOpt):
    "update the dictionary of options with <dOpt>"
    for sKey, bVal in dOpt.items():
        if sKey in _dOptions:
            _dOptions[sKey] = bVal


def getOptions ():
    "return the dictionary of current options"
    return _dOptions


def getDefaultOptions ():
    "return the dictionary of default options"
    return dict(gc_options.getOptions(_sAppContext))


def getOptionsLabels (sLang):
    "return options labels"
    return gc_options.getUI(sLang)


def displayOptions (sLang):
    "display the list of grammar checking options"
    echo("List of options")
    echo("\n".join( [ k+":\t"+str(v)+"\t"+gc_options.getUI(sLang).get(k, ("?", ""))[0]  for k, v  in sorted(_dOptions.items()) ] ))
    echo("")


def resetOptions ():
    "set options to default values"
    global _dOptions
    _dOptions = dict(gc_options.getOptions(_sAppContext))


def getSpellChecker ():
    "return the spellchecker object"
    return _oSpellChecker


def _getPath ():
    return os.path.join(os.path.dirname(sys.modules[__name__].__file__), __name__ + ".py")



#### common functions

def option (sOpt):
    "return True if option <sOpt> is active"
    return _dOptions.get(sOpt, False)


def displayInfo (dTokenPos, tWord):
    "for debugging: retrieve info of word"
    if not tWord:
        echo("> nothing to find")

def morph (dTokenPos, tWord, sPattern, bStrict=True, bNoWord=False):
    "analyse a tuple (position, word), return True if sPattern in morphologies (disambiguation on)"
    if not tWord:
        return bNoWord
    lMorph = dTokenPos[tWord[0]]["lMorph"]  if tWord[0] in dTokenPos and "lMorph" in dTokenPos[tWord[0]]  else _oSpellChecker.getMorph(tWord[1])
    if not lMorph:
        return False
    zPattern = re.compile(sPattern)
    if bStrict:
        return all(zPattern.search(s)  for s in lMorph)
    return any(zPattern.search(s)  for s in lMorph)


def morphex (dTokenPos, tWord, sPattern, sNegPattern, bNoWord=False):
    "analyse a tuple (position, word), returns True if not sNegPattern in word morphologies and sPattern in word morphologies (disambiguation on)"
    if not tWord:
        return bNoWord
    lMorph = dTokenPos[tWord[0]]["lMorph"]  if tWord[0] in dTokenPos and "lMorph" in dTokenPos[tWord[0]]  else _oSpellChecker.getMorph(tWord[1])
    if not lMorph:
        return False
    # check negative condition
    zNegPattern = re.compile(sNegPattern)
    if any(zNegPattern.search(s)  for s in lMorph):
        return False
    # search sPattern
    zPattern = re.compile(sPattern)
    return any(zPattern.search(s)  for s in lMorph)


def analyse (sWord, sPattern, bStrict=True):
    "analyse a word, return True if sPattern in morphologies (disambiguation off)"
    lMorph = _oSpellChecker.getMorph(sWord)
    if not lMorph:
        return False
    zPattern = re.compile(sPattern)
    if bStrict:
        return all(zPattern.search(s)  for s in lMorph)
    return any(zPattern.search(s)  for s in lMorph)


def analysex (sWord, sPattern, sNegPattern):
    "analyse a word, returns True if not sNegPattern in word morphologies and sPattern in word morphologies (disambiguation off)"
    lMorph = _oSpellChecker.getMorph(sWord)
    if not lMorph:
        return False
    # check negative condition
    zNegPattern = re.compile(sNegPattern)
    if any(zNegPattern.search(s)  for s in lMorph):
        return False
    # search sPattern
    zPattern = re.compile(sPattern)
    return any(zPattern.search(s)  for s in lMorph)



## functions to get text outside pattern scope

# warning: check compile_rules.py to understand how it works

        return morphex(dTokenPos, (nPos, sWord), sPatternGroup1, sNegPatternGroup1)
    return morph(dTokenPos, (nPos, sWord), sPatternGroup1, False)


#### Disambiguator

def select (dTokenPos, nPos, sWord, sPattern, lDefault=None):
    "Disambiguation: select morphologies of <sWord> matching <sPattern>"
    if not sWord:
        return True
    if nPos not in dTokenPos:
        print("Error. There should be a token at this position: ", nPos)
        return True
    lMorph = _oSpellChecker.getMorph(sWord)
    if not lMorph or len(lMorph) == 1:
        return True
    lSelect = [ sMorph  for sMorph in lMorph  if re.search(sPattern, sMorph) ]
    if lSelect:
        if len(lSelect) != len(lMorph):
            dTokenPos[nPos]["lMorph"] = lSelect
    elif lDefault:
        dTokenPos[nPos]["lMorph"] = lDefault
    return True


def exclude (dTokenPos, nPos, sWord, sPattern, lDefault=None):
    "Disambiguation: exclude morphologies of <sWord> matching <sPattern>"
    if not sWord:
        return True
    if nPos not in dTokenPos:
        print("Error. There should be a token at this position: ", nPos)
        return True
    lMorph = _oSpellChecker.getMorph(sWord)
    if not lMorph or len(lMorph) == 1:
        return True
    lSelect = [ sMorph  for sMorph in lMorph  if not re.search(sPattern, sMorph) ]
    if lSelect:
        if len(lSelect) != len(lMorph):
            dTokenPos[nPos]["lMorph"] = lSelect
    elif lDefault:
        dTokenPos[nPos]["lMorph"] = lDefault
    return True


def define (dTokenPos, nPos, lMorph):
    "Disambiguation: set morphologies of token at <nPos> with <lMorph>"
    if nPos not in dTokenPos:
        print("Error. There should be a token at this position: ", nPos)
        return True
    dTokenPos[nPos]["lMorph"] = lMorph
    return True




#### TOKEN SENTENCE CHECKER

class TokenSentence:
    "Text parser"

    def __init__ (self, sSentence, sSentence0, nOffset):
        self.sSentence = sSentence
        self.sSentence0 = sSentence0
        self.nOffsetWithinParagraph = nOffset
        self.lToken = list(_oTokenizer.genTokens(sSentence, True))
        self.dTokenPos = { dToken["nStart"]: dToken  for dToken in self.lToken }
        self.dTags = {}
        self.dError = {}
        self.createError = self._createWriterError  if _bWriterError  else self._createDictError

    def update (self, sSentence):
        "update <sSentence> and retokenize"
        self.sSentence = sSentence
        self.lToken = list(_oTokenizer.genTokens(sSentence, True))

    def _getNextMatchingNodes (self, dToken, dGraph, dNode, bDebug=False):
        "generator: return nodes where <dToken> “values” match <dNode> arcs"
        # token value
        if dToken["sValue"] in dNode:

                elif dNode["sType"] in sMeta:
                    if bDebug:
                        print("  MATCH: *" + sMeta)
                    yield dGraph[dNode["<meta>"][sMeta]]


    def parse (self, dGraph, dPriority, sCountry="${country_default}", dOptions=None, bShowRuleId=False, bDebug=False, bContext=False):
        "parse tokens from the text and execute actions encountered"
        self.dError = {}
        dPriority = {}  # Key = position; value = priority
        dOpt = _dOptions  if not dOptions  else dOptions
        lPointer = []
        bTagAndRewrite = False
        for dToken in self.lToken:
            if bDebug:

        sMessage = globals()[sMsg[1:]](self.lToken)  if sMsg[0:1] == "="  else self._expand(sMsg, nTokenOffset)
        xErr.aShortComment = sMessage   # sMessage.split("|")[0]     # in context menu
        xErr.aFullComment = sMessage   # sMessage.split("|")[-1]    # in dialog
        if bShowRuleId:
            xErr.aShortComment += "  " + sLineId + " # " + sRuleId
        # URL
        if sURL:
            xProperty = PropertyValue()
            xProperty.Name = "FullCommentURL"
            xProperty.Value = sURL
            xErr.aProperties = (xProperty,)
        else:
            xErr.aProperties = ()
        return xErr

    def _createDictError (self, sSugg, nTokenOffset, iFirstToken, nStart, nEnd, sLineId, sRuleId, bUppercase, sMsg, sURL, bShowRuleId, sOption, bContext):
        "error as a dictionary"
        dErr = {}

                if bDebug:
                    print("Merged token:", dTokenMerger["sValue"])
                bKeepToken = False
            if "nMergeUntil" in dToken:
                if dToken["i"] > nMergeUntil: # this token is not already merged with a previous token
                    dTokenMerger = dToken
                if dToken["nMergeUntil"] > nMergeUntil:
                    nMergeUntil = dToken["nMergeUntil"]
                del dToken["nMergeUntil"]
            elif "bToRemove" in dToken:
                # remove useless token
                self.sSentence = self.sSentence[:dToken["nStart"]] + " " * (dToken["nEnd"] - dToken["nStart"]) + self.sSentence[dToken["nEnd"]:]
                if bDebug:
                    print("removed:", dToken["sValue"])
                bKeepToken = False