Grammalecte  Diff

"""
Spellchecker.

Useful to check several dictionaries at once.

To avoid iterating over a pile of dictionaries, it is assumed that 3 are enough:
- the main dictionary, bundled with the package
- the extended dictionary
- the community dictionary, added by an organization
- the personal dictionary, created by the user for its own convenience
"""

import importlib
import traceback

from . import ibdawg
from . import tokenizer


dDefaultDictionaries = {
    "fr": "fr-allvars.bdic",
    "en": "en.bdic"
}


class SpellChecker ():
    "SpellChecker: wrapper for the IBDAWG class"

    def __init__ (self, sLangCode, sfMainDic="", sfExtendedDic="", sfCommunityDic="", sfPersonalDic=""):
        "returns True if the main dictionary is loaded"
        self.sLangCode = sLangCode
        if not sfMainDic:
            sfMainDic = dDefaultDictionaries.get(sLangCode, "")
        self.oMainDic = self._loadDictionary(sfMainDic, True)

        except Exception as e:
            if bNecessary:
                raise Exception(str(e), "Error: <" + str(source) + "> not loaded.")
            print("Error: <" + str(source) + "> not loaded.")
            traceback.print_exc()
            return None

    def _loadTokenizer (self):
        self.oTokenizer = tokenizer.Tokenizer(self.sLangCode)

    def getTokenizer (self):
        "load and return the tokenizer object"
        if not self.oTokenizer:
            self._loadTokenizer()
        return self.oTokenizer

    def setMainDictionary (self, source):
        "returns True if the dictionary is loaded"
        self.oMainDic = self._loadDictionary(source, True)
        return bool(self.oMainDic)

    def setExtendedDictionary (self, source, bActivate=True):
        "returns True if the dictionary is loaded"
        self.oExtendedDic = self._loadDictionary(source)
        self.bExtendedDic = False  if not bActivate  else bool(self.oExtendedDic)
        return bool(self.oExtendedDic)

    def setCommunityDictionary (self, source, bActivate=True):
        "returns True if the dictionary is loaded"
        self.oCommunityDic = self._loadDictionary(source)
        self.bCommunityDic = False  if not bActivate  else bool(self.oCommunityDic)
        return bool(self.oCommunityDic)

    def setPersonalDictionary (self, source, bActivate=True):
        "returns True if the dictionary is loaded"
        self.oPersonalDic = self._loadDictionary(source)
        self.bPersonalDic = False  if not bActivate  else bool(self.oPersonalDic)
        return bool(self.oPersonalDic)

    def activateExtendedDictionary (self):
        "activate extended dictionary (if available)"
        self.bExtendedDic = bool(self.oExtendedDic)

    def activateCommunityDictionary (self):
        "activate community dictionary (if available)"
        self.bCommunityDic = bool(self.oCommunityDic)

    def activatePersonalDictionary (self):
        "activate personal dictionary (if available)"
        self.bPersonalDic = bool(self.oPersonalDic)

    def deactivateExtendedDictionary (self):
        "deactivate extended dictionary"
        self.bExtendedDic = False

    def deactivateCommunityDictionary (self):
        "deactivate community dictionary"
        self.bCommunityDic = False

    def deactivatePersonalDictionary (self):
        "deactivate personal dictionary"
        self.bPersonalDic = False


    # Default suggestions

    def loadSuggestions (self, sLangCode):
        "load default suggestion module for <sLangCode>"
        try:
            suggest = importlib.import_module("."+sLangCode, "graphspell")
        except ImportError:
            print("No suggestion module for language <"+sLangCode+">")
            return
        self.dDefaultSugg = suggest.dSugg


    # Storage

    def activateStorage (self):
        "store all lemmas and morphologies retrieved from the word graph"
        self.bStorage = True

    def deactivateStorage (self):
        "stop storing all lemmas and morphologies retrieved from the word graph"
        self.bStorage = False

    def clearStorage (self):
        "clear all stored data"
        self._dLemmas.clear()
        self._dMorphologies.clear()


    # parse text functions

    def parseParagraph (self, sText, bSpellSugg=False):
        "return a list of tokens where token value doesn’t exist in the word graph"
        if not self.oTokenizer:
            self._loadTokenizer()
        aSpellErrs = []
        for dToken in self.oTokenizer.genTokens(sText):
            if dToken['sType'] == "WORD" and not self.isValidToken(dToken['sValue']):
                if bSpellSugg:
                    dToken['aSuggestions'] = []
                    for lSugg in self.suggest(dToken['sValue']):
                        dToken['aSuggestions'].extend(lSugg)
                aSpellErrs.append(dToken)
        return aSpellErrs

    def countWordsOccurrences (self, sText, bByLemma=False, bOnlyUnknownWords=False, dWord={}):
        """count word occurrences.
           <dWord> can be used to cumulate count from several texts."""
        if not self.oTokenizer:
            self._loadTokenizer()
        for dToken in self.oTokenizer.genTokens(sText):
            if dToken['sType'] == "WORD":
                if bOnlyUnknownWords:
                    if not self.isValidToken(dToken['sValue']):
                        dWord[dToken['sValue']] = dWord.get(dToken['sValue'], 0) + 1
                else:
                    if not bByLemma:

    def isValid (self, sWord):
        "checks if sWord is valid (different casing tested if the first letter is a capital)"
        if self.oMainDic.isValid(sWord):
            return True
        if self.bExtendedDic and self.oExtendedDic.isValid(sWord):
            return True
        if self.bCommunityDic and self.oCommunityDic.isValid(sWord):
            return True
        if self.bPersonalDic and self.oPersonalDic.isValid(sWord):
            return True
        return False

    def lookup (self, sWord):
        "checks if sWord is in dictionary as is (strict verification)"
        if self.oMainDic.lookup(sWord):
            return True
        if self.bExtendedDic and self.oExtendedDic.lookup(sWord):
            return True
        if self.bCommunityDic and self.oCommunityDic.lookup(sWord):
            return True
        if self.bPersonalDic and self.oPersonalDic.lookup(sWord):
            return True
        return False

    def getMorph (self, sWord):
        "retrieves morphologies list, different casing allowed"

            yield from self.oExtendedDic.select(sFlexPattern, sTagsPattern)
        if self.bCommunityDic:
            yield from self.oCommunityDic.select(sFlexPattern, sTagsPattern)
        if self.bPersonalDic:
            yield from self.oPersonalDic.select(sFlexPattern, sTagsPattern)

    def drawPath (self, sWord):
        "draw the path taken by <sWord> within the word graph: display matching nodes and their arcs"
        self.oMainDic.drawPath(sWord)
        if self.bExtendedDic:
            print("-----")
            self.oExtendedDic.drawPath(sWord)
        if self.bCommunityDic:
            print("-----")
            self.oCommunityDic.drawPath(sWord)