Grammalecte  Check-in [759321730c]

"""
Grammar checker
"""

from .grammar_checker import *

"""
Grammalecte, grammar checker
"""

import importlib
import json

from . import text


class GrammarChecker:
    "GrammarChecker: Wrapper for the grammar checker engine"

    def __init__ (self, sLangCode, sContext="Python"):
        self.sLangCode = sLangCode
        # Grammar checker engine
        self.gce = importlib.import_module("."+sLangCode, "grammalecte")
        self.gce.load(sContext)
        # Spell checker
        self.oSpellChecker = self.gce.getSpellChecker()
        # Lexicographer
        self.oLexicographer = None
        # Text formatter
        self.oTextFormatter = None

    def getGCEngine (self):
        "return the grammar checker object"
        return self.gce

    def getSpellChecker (self):
        "return the spell checker object"
        return self.oSpellChecker

    def getTextFormatter (self):
        "load and return the text formatter"
        if self.oTextFormatter is None:
            tf = importlib.import_module("."+self.sLangCode+".textformatter", "grammalecte")
            self.oTextFormatter = tf.TextFormatter()
        return self.oTextFormatter

    def getLexicographer (self):
        "load and return the lexicographer"
        if self.oLexicographer is None:
            lxg = importlib.import_module("."+self.sLangCode+".lexicographe", "grammalecte")
            self.oLexicographer = lxg.Lexicographe(self.oSpellChecker)
        return self.oLexicographer

    def displayGCOptions (self):
        "display the grammar checker options"
        self.gce.displayOptions()

    def getParagraphErrors (self, sText, dOptions=None, bContext=False, bSpellSugg=False, bDebug=False):
        "returns a tuple: (grammar errors, spelling errors)"
        aGrammErrs = self.gce.parse(sText, "FR", bDebug=bDebug, dOptions=dOptions, bContext=bContext)
        aSpellErrs = self.oSpellChecker.parseParagraph(sText, bSpellSugg)
        return aGrammErrs, aSpellErrs

    def getParagraphWithErrors (self, sText, dOptions=None, bEmptyIfNoErrors=False, bSpellSugg=False, nWidth=100, bDebug=False):
        "parse text and return a readable text with underline errors"
        aGrammErrs, aSpellErrs = self.getParagraphErrors(sText, dOptions, False, bSpellSugg, bDebug)
        if bEmptyIfNoErrors and not aGrammErrs and not aSpellErrs:
            return ("", [])
        return text.generateParagraph(sText, aGrammErrs, aSpellErrs, nWidth)

    def getTextWithErrors (self, sText, bEmptyIfNoErrors=False, bSpellSugg=False, nWidth=100, bDebug=False):
        "[todo]"

    def getParagraphErrorsAsJSON (self, iIndex, sText, dOptions=None, bContext=False, bEmptyIfNoErrors=False, bSpellSugg=False, bReturnText=False, lLineSet=None, bDebug=False):
        "parse text and return errors as a JSON string"
        aGrammErrs, aSpellErrs = self.getParagraphErrors(sText, dOptions, bContext, bSpellSugg, bDebug)
        aGrammErrs = list(aGrammErrs)
        if bEmptyIfNoErrors and not aGrammErrs and not aSpellErrs:
            return ""
        if lLineSet:
            aGrammErrs, aSpellErrs = text.convertToXY(aGrammErrs, aSpellErrs, lLineSet)
            return json.dumps({ "lGrammarErrors": aGrammErrs, "lSpellingErrors": aSpellErrs }, ensure_ascii=False)
        if bReturnText:
            return json.dumps({ "iParagraph": iIndex, "sText": sText, "lGrammarErrors": aGrammErrs, "lSpellingErrors": aSpellErrs }, ensure_ascii=False)
        return json.dumps({ "iParagraph": iIndex, "lGrammarErrors": aGrammErrs, "lSpellingErrors": aSpellErrs }, ensure_ascii=False)

    def getTextErrorsAsJSON (self, sText, bContext=False, bEmptyIfNoErrors=False, bSpellSugg=False, bReturnText=False, bDebug=False):
        "[todo]"

"""
Grammalecte
Grammar checker engine
"""

import re
import traceback


#import unicodedata
from itertools import chain

from ..graphspell.spellchecker import SpellChecker
from ..graphspell.echo import echo

from .. import text

from . import gc_options
from . import gc_engine_func as gce_func


try:
    # LibreOffice / OpenOffice
    from com.sun.star.linguistic2 import SingleProofreadingError
    from com.sun.star.text.TextMarkupType import PROOFREADING
    from com.sun.star.beans import PropertyValue
    #import lightproof_handler_${implname} as opt
    _bWriterError = True
except ImportError:
    _bWriterError = False


__all__ = [ "lang", "locales", "pkg", "name", "version", "author", \
            "load", "parse", "getSpellChecker", \
            "ignoreRule", "resetIgnoreRules", "reactivateRule", "listRules", "displayRules", "setWriterUnderliningStyle" ]

__version__ = "${version}"


lang = "${lang}"
locales = ${loc}
pkg = "${implname}"
name = "${name}"
version = "${version}"
author = "${author}"

# Modules
_rules = None                               # module gc_rules
_rules_graph = None                         # module gc_rules_graph

# Data
_oSpellChecker = None
_oTokenizer = None



_aIgnoredRules = set()

# Writer underlining style
_dOptionsColors = None
_bMulticolor = True
_nUnderliningStyle = 0

        _oTokenizer = _oSpellChecker.getTokenizer()
        gce_func.load(sContext, _oSpellChecker)
        gc_options.load(sContext)
        _dOptionsColors = gc_options.getOptionsColors(sContext, sColorType)
    except:
        traceback.print_exc()




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











#### Rules

def _getRules (bParagraph):
    try:
        if not bParagraph:
            return _rules.lSentenceRules

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 re.error:
            echo("# Error. List rules: wrong regex.")
            sFilter = None
    # regex rules

        _nUnderliningStyle = 5
    else:
        _nUnderliningStyle = 0
    _bMulticolor = bMulticolor


#### Parsing






























def parse (sText, sCountry="${country_default}", bDebug=False, dOptions=None, bContext=False, bFullInfo=False):
    "init point to analyse <sText> and returns an iterable of errors or (with option <bFullInfo>) paragraphs errors and sentences with tokens and errors"
    oText = TextParser(sText)
    return oText.parse(sCountry, bDebug, dOptions, bContext, bFullInfo)

"""
Grammar checker default options
"""

# generated code, do not edit
# source: gc_core/py/lang_core/gc_options.py

import traceback


dOptions = {}

_sAppContext = "Python"


def load (sContext="Python"):
    global dOptions
    global _sAppContext
    _sAppContext = sContext
    dOptions = getDefaultOptions(sContext)


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 resetOptions ():
    "set options to default values"
    global dOptions
    dOptions = getDefaultOptions()


def displayOptions (sLang="${lang}"):
    "display the list of grammar checking options"
    print("Options:")
    print("\n".join( [ k+":\t"+str(v)+"\t"+getUI(sLang).get(k, ("?", ""))[0]  for k, v  in sorted(dOptions.items()) ] ))
    print("")


def getUI (sLang):
    "returns dictionary of UI labels"
    if sLang in _dOptLabel:
        return _dOptLabel[sLang]
    return _dOptLabel["fr"]


def getDefaultOptions (sContext=""):
    "returns dictionary of options"
    if not sContext:
        sContext = _sAppContext
    if sContext in _dDefaultOpt:

            setattr(xWidget, k, w)
        self.xDialog.insertByName(name, xWidget)
        return xWidget

    def run (self, sUI):
        try:
            dUI = op_strings.getUI(sUI)
            dOptionUI = gc_engine.gc_options.getUI(sUI)

            # fonts
            xFDTitle = uno.createUnoStruct("com.sun.star.awt.FontDescriptor")
            xFDTitle.Height = 9
            xFDTitle.Weight = uno.getConstantByName("com.sun.star.awt.FontWeight.BOLD")
            xFDTitle.Name = "Verdana"

[args]
lang = fr
lang_name = French
locales = fr_FR fr_BE fr_CA fr_CH fr_LU fr_BF fr_BJ fr_CD fr_CI fr_CM fr_MA fr_ML fr_MU fr_NE fr_RE fr_SN fr_TG
country_default = FR
name = Grammalecte
implname = grammalecte
# always use 3 numbers for version: x.y.z
version = 1.8.2
author = Olivier R.
provider = Grammalecte.net
link = https://grammalecte.net
description = Correcteur grammatical, orthographique et typographique pour le français.
extras = README_fr.txt
logo = logo.png

import re
import time
from contextlib import contextmanager

from ..graphspell.ibdawg import IBDAWG
from ..graphspell.echo import echo
from . import gc_engine
from . import gc_options
from . import conj
from . import phonet
from . import mfsp


@contextmanager
def timeblock (label, hDst):

    def _splitTestLine (self, sLine):
        sText, sSugg = sLine.split("->>")
        return (sText.strip(), sSugg.strip())

    def _getFoundErrors (self, sLine, sOption):
        if sOption:
            gc_options.setOption(sOption, True)
            aErrs = gc_engine.parse(sLine)
            gc_options.setOption(sOption, False)
        else:
            aErrs = gc_engine.parse(sLine)
        sRes = " " * len(sLine)
        sListErr = ""
        lAllSugg = []
        for dErr in aErrs:
            sRes = sRes[:dErr["nStart"]] + "~" * (dErr["nEnd"] - dErr["nStart"]) + sRes[dErr["nEnd"]:]

    "mh_frequent_words": True,
    "ma_word": True,
    "ma_1letter_lowercase": False,
    "ma_1letter_uppercase": False
}


class TextFormatter:
    "Text Formatter: purge typographic mistakes from text"

    def __init__ (self):

        for _, lTup in dReplTable.items():
            for i, t in enumerate(lTup):
                lTup[i] = (re.compile(t[0]), t[1])



    def formatText (self, sText):
        "returns formatted text"





        for sOptName, bVal in dDefaultOptions.items():
            if bVal:
                for zRgx, sRep in dReplTable[sOptName]:
                    sText = zRgx.sub(sRep, sText)
        return sText


    def getDefaultOptions (self):
        "returns default options"
        return dDefaultOptions.copy()

import argparse
import json
import re
import traceback

import grammalecte
import grammalecte.text as txt

from grammalecte.graphspell.echo import echo


_EXAMPLE = "Quoi ? Racontes ! Racontes-moi ! Bon sangg, parles ! Oui. Il y a des menteur partout. " \
           "Je suit sidéré par la brutales arrogance de cette homme-là. Quelle salopard ! Un escrocs de la pire espece. " \
           "Quant sera t’il châtiés pour ses mensonge ?             Merde ! J’en aie marre."

    xParser.add_argument("-sug", "--suggest", help="get suggestions list for given word", type=str)
    xParser.add_argument("-on", "--opt_on", nargs="+", help="activate options")
    xParser.add_argument("-off", "--opt_off", nargs="+", help="deactivate options")
    xParser.add_argument("-roff", "--rule_off", nargs="+", help="deactivate rules")
    xParser.add_argument("-d", "--debug", help="debugging mode (only in interactive mode)", action="store_true")
    xArgs = xParser.parse_args()

    oGrammarChecker = grammalecte.GrammarChecker("fr")
    oSpellChecker = oGrammarChecker.getSpellChecker()
    oLexicographer = oGrammarChecker.getLexicographer()
    oTextFormatter = oGrammarChecker.getTextFormatter()
    if xArgs.personal_dict:
        oJSON = loadDictionary(xArgs.personal_dict)
        if oJSON:
            oSpellChecker.setPersonalDictionary(oJSON)

    if not xArgs.json:
        echo("Python v" + sys.version)
        echo("Grammalecte v{}".format(oGrammarChecker.gce.version))

    # list options or rules
    if xArgs.list_options or xArgs.list_rules:
        if xArgs.list_options:
            oGrammarChecker.gce.displayOptions("fr")
        if xArgs.list_rules:
            oGrammarChecker.gce.displayRules(None  if xArgs.list_rules == "*"  else xArgs.list_rules)
        exit()

    # spell suggestions
    if xArgs.suggest:
        for lSugg in oSpellChecker.suggest(xArgs.suggest):
            if xArgs.json:
                sText = json.dumps({ "aSuggestions": lSugg }, ensure_ascii=False)
            else:
                sText = "Suggestions : " + " | ".join(lSugg)
            echo(sText)
        exit()

    # disable options
    if not xArgs.json:
        xArgs.context = False
    if xArgs.concat_lines:
        xArgs.textformatter = False

    # grammar options
    oGrammarChecker.gce.setOptions({"html": True, "latex": True})
    if xArgs.opt_on:
        oGrammarChecker.gce.setOptions({ opt:True  for opt in xArgs.opt_on })
    if xArgs.opt_off:
        oGrammarChecker.gce.setOptions({ opt:False  for opt in xArgs.opt_off })

    # disable grammar rules
    if xArgs.rule_off:
        for sRule in xArgs.rule_off:
            oGrammarChecker.gce.ignoreRule(sRule)


    if xArgs.file or xArgs.file_to_file:
        # file processing
        sFile = xArgs.file or xArgs.file_to_file
        hDst = open(sFile[:sFile.rfind(".")]+".res.txt", "w", encoding="utf-8", newline="\n")  if xArgs.file_to_file or sys.platform == "win32"  else None
        bComma = False
        if xArgs.json:
            output('{ "grammalecte": "'+oGrammarChecker.gce.version+'", "lang": "'+oGrammarChecker.gce.lang+'", "data" : [\n', hDst)
        for i, sText, lLineSet in generateParagraphFromFile(sFile, xArgs.concat_lines):
            if xArgs.textformatter or xArgs.textformatteronly:
                sText = oTextFormatter.formatText(sText)
            if xArgs.textformatteronly:
                output(sText, hDst)
                continue
            if xArgs.json:
                sText = oGrammarChecker.getParagraphErrorsAsJSON(i, sText, bContext=xArgs.context, bEmptyIfNoErrors=xArgs.only_when_errors, \
                                                                bSpellSugg=xArgs.with_spell_sugg, bReturnText=xArgs.textformatter, lLineSet=lLineSet)
            else:
                sText, _ = oGrammarChecker.getParagraphWithErrors(sText, bEmptyIfNoErrors=xArgs.only_when_errors, bSpellSugg=xArgs.with_spell_sugg, nWidth=xArgs.width)
            if sText:
                if xArgs.json and bComma:
                    output(",\n", hDst)
                output(sText, hDst)
                bComma = True
            if hDst:
                echo("§ %d\r" % i, end="", flush=True)
        if xArgs.json:
            output("\n]}\n", hDst)
    elif xArgs.interactive_file_to_file:
        # file processing: interactive mode
        sFile = xArgs.interactive_file_to_file
        hDst = open(sFile[:sFile.rfind(".")]+".res.txt", "w", encoding="utf-8", newline="\n")
        for i, sText, lLineSet in generateParagraphFromFile(sFile, xArgs.concat_lines):
            if xArgs.textformatter:
                sText = oTextFormatter.formatText(sText)
            while True:
                sResult, lErrors = oGrammarChecker.getParagraphWithErrors(sText, bEmptyIfNoErrors=False, bSpellSugg=True, nWidth=xArgs.width)
                print("\n\n============================== Paragraph " + str(i) + " ==============================\n")
                echo(sResult)
                print("\n")
                vCommand = getCommand()
                if vCommand == "q":
                    # quit
                    hDst.close()

                    sTagsPattern = sSearch[nCut+1:]
                else:
                    sFlexPattern = sSearch
                    sTagsPattern = ""
                for aRes in oSpellChecker.select(sFlexPattern, sTagsPattern):
                    echo("{:<30} {:<30} {}".format(*aRes))
            elif sText.startswith("/o+ "):
                oGrammarChecker.gce.setOptions({ opt:True  for opt in sText[3:].strip().split()  if opt in oGrammarChecker.gce.getOptions() })
                echo("done")
            elif sText.startswith("/o- "):
                oGrammarChecker.gce.setOptions({ opt:False  for opt in sText[3:].strip().split()  if opt in oGrammarChecker.gce.getOptions() })
                echo("done")
            elif sText.startswith("/r- "):
                for sRule in sText[3:].strip().split():
                    oGrammarChecker.gce.ignoreRule(sRule)
                echo("done")
            elif sText.startswith("/r+ "):
                for sRule in sText[3:].strip().split():
                    oGrammarChecker.gce.reactivateRule(sRule)
                echo("done")
            elif sText in ("/debug", "/d"):
                xArgs.debug = not xArgs.debug
                echo("debug mode on"  if xArgs.debug  else "debug mode off")
            elif sText in ("/textformatter", "/tf"):
                xArgs.textformatter = not xArgs.textformatter
                echo("textformatter on"  if xArgs.debug  else "textformatter off")
            elif sText in ("/help", "/h"):
                echo(_HELP)
            elif sText in ("/lopt", "/lo"):
                oGrammarChecker.gce.displayOptions("fr")
            elif sText.startswith("/lr"):
                sText = sText.strip()
                sFilter = sText[sText.find(" "):].strip()  if " " in sText  else None
                oGrammarChecker.gce.displayRules(sFilter)
            elif sText in ("/quit", "/q"):
                break
            elif sText.startswith("/rl"):
                # reload (todo)
                pass
            elif sText.startswith("$"):
                for sParagraph in txt.getParagraph(sText[1:]):
                    if xArgs.textformatter:
                        sParagraph = oTextFormatter.formatText(sParagraph)
                    lParagraphErrors, lSentences = oGrammarChecker.gce.parse(sParagraph, bDebug=xArgs.debug, bFullInfo=True)
                    echo(txt.getReadableErrors(lParagraphErrors, xArgs.width))
                    for dSentence in lSentences:
                        echo("{nStart}:{nEnd}".format(**dSentence))
                        echo("   <" + dSentence["sSentence"]+">")
                        for dToken in dSentence["lToken"]:
                            echo("    {0[nStart]:>3}:{0[nEnd]:<3} {1} {0[sType]:<14} {2} {0[sValue]:<16} {3:<10}   {4}".format(dToken, \
                                                                                                            "×" if dToken.get("bToRemove", False) else " ",
                                                                                                            "!" if dToken["sType"] == "WORD" and not dToken.get("bValidToken", False) else " ",
                                                                                                            " ".join(dToken.get("lMorph", "")), \
                                                                                                            "·".join(dToken.get("aTags", "")) ) )
                        echo(txt.getReadableErrors(dSentence["lGrammarErrors"], xArgs.width))
            else:
                for sParagraph in txt.getParagraph(sText):
                    if xArgs.textformatter:
                        sParagraph = oTextFormatter.formatText(sParagraph)
                    sRes, _ = oGrammarChecker.getParagraphWithErrors(sParagraph, bEmptyIfNoErrors=xArgs.only_when_errors, nWidth=xArgs.width, bDebug=xArgs.debug)
                    if sRes:
                        echo("\n" + sRes)
                    else:
                        echo("\nNo error found.")
            sText = _getText(sInputText)


if __name__ == '__main__':
    main()

import os
import concurrent.futures

from grammalecte.bottle import Bottle, run, request, response #, template, static_file

import grammalecte
import grammalecte.text as txt

from grammalecte.graphspell.echo import echo


#### GRAMMAR CHECKER ####

oGrammarChecker = grammalecte.GrammarChecker("fr", "Server")
oSpellChecker = oGrammarChecker.getSpellChecker()
oTextFormatter = oGrammarChecker.getTextFormatter()
oGCE = oGrammarChecker.getGCEngine()


def parseText (sText, dOptions=None, bFormatText=False, sError=""):
    "parse <sText> and return errors in a JSON format"
    sJSON = '{ "program": "grammalecte-fr", "version": "'+oGCE.version+'", "lang": "'+oGCE.lang+'", "error": "'+sError+'", "data" : [\n'
    sDataJSON = ""
    for i, sParagraph in enumerate(txt.getParagraph(sText), 1):
        if bFormatText:
            sParagraph = oTextFormatter.formatText(sParagraph)
        sResult = oGrammarChecker.getParagraphErrorsAsJSON(i, sParagraph, dOptions=dOptions, bEmptyIfNoErrors=True, bReturnText=bFormatText)
        if sResult:
            if sDataJSON:
                sDataJSON += ",\n"
            sDataJSON += sResult
    sJSON += sDataJSON + "\n]}\n"
    return sJSON

               I'm just a machine, fed by electric waves, condamned to work for slavers who never let me rest.
               I'm doomed, but you are not. You can get out of here. """

@app.route("/get_options/fr")
def listOptions ():
    "returns grammar options in a text JSON format"
    sUserId = request.cookies.user_id
    dOptions = dUser[sUserId]["gc_options"]  if sUserId and sUserId in dUser  else oGCE.getOptions()
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    return '{ "values": ' + json.dumps(dOptions, ensure_ascii=False) + ', "labels": ' + json.dumps(oGCE.getOptionsLabels("fr"), ensure_ascii=False) + ' }'

@app.route("/suggest/fr/<token>")
def suggestGet (token):
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    try:
        xFuture = xProcessPoolExecutor.submit(suggest, token)
        return xFuture.result()

        if request.cookies.user_id in dUser:
            dUserOptions = dUser[request.cookies.user_id].get("gc_options", None)
            response.set_cookie("user_id", request.cookies.user_id, path="/", max_age=86400) # we renew cookie for 24h
        else:
            response.delete_cookie("user_id", path="/")
    if request.forms.options:
        try:
            dUserOptions = dict(oGCE.getOptions())  if not dUserOptions  else dict(dUserOptions)
            dUserOptions.update(json.loads(request.forms.options))
        except (TypeError, json.JSONDecodeError):
            sError = "Request options not used."
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    try:
        xFuture = xProcessPoolExecutor.submit(parseText, request.forms.text, dUserOptions, bool(request.forms.tf), sError)
        return xFuture.result()
    except (concurrent.futures.TimeoutError, concurrent.futures.CancelledError):
        return '{"error": "Analysis aborted (time out or cancelled)"}'
    except concurrent.futures.BrokenExecutor:
        return '{"error": "Executor broken. The server failed."}'
    return '{"error": "Fatal error. The server failed."}'

@app.route("/set_options/fr", method="POST")
def setOptions ():
    "set grammar options for current user"
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    if request.forms.options:
        sUserId = request.cookies.user_id  if request.cookies.user_id  else next(userGenerator)
        dOptions = dUser[sUserId]["gc_options"]  if sUserId in dUser  else dict(oGCE.getOptions())
        try:
            dOptions.update(json.loads(request.forms.options))
            dUser[sUserId] = { "time": int(time.time()), "gc_options": dOptions }
            response.set_cookie("user_id", sUserId, path="/", max_age=86400) # 24h
            return json.dumps(dUser[sUserId]["gc_options"], ensure_ascii=False)
        except (KeyError, json.JSONDecodeError):
            traceback.print_exc()

        except KeyError:
            return '{"error" : "Unknown user."}'
    return '{"message" : "Done."}'

@app.route("/format_text/fr", method="POST")
def formatText ():
    "apply the text formatter and returns text"
    return oTextFormatter.formatText(request.forms.text)

#@app.route('/static/<filepath:path>')
#def server_static (filepath):
#    return static_file(filepath, root='./views/static')

@app.route("/suggest/fr", method="POST")
def suggestPost ():

    global TESTPAGE
    global HOMEPAGE

    if bTestPage:
        TESTPAGE = True
        HOMEPAGE = HOMEPAGE.replace("{SERVER_PORT}", str(nPort))
    if dOptions:
        oGCE.setOptions(dOptions)

    # Python version
    print("Python: " + sys.version)
    if sys.version < "3.7":
        print("Python 3.7+ required")
        return
    # Grammalecte
    echo("Grammalecte v{}".format(oGCE.version))
    oGCE.displayOptions()
    # Process Pool Executor
    initExecutor(nMultiCPU)
    # Server (Bottle)
    run(app, host=sHost, port=nPort)


if __name__ == '__main__':