Grammalecte  Check-in [a3f82d3151]

    return 0


def createRule (s, nIdLine, sLang, bParagraph, dOptPriority):
    "returns rule as list [option name, regex, bCaseInsensitive, identifier, list of actions]"
    global dJSREGEXES

    sLineId = "#" + str(nIdLine) + ("p" if bParagraph else "s")
    sRuleId = sLineId

    #### GRAPH CALL
    if s.startswith("@@@@"):
        if bParagraph:
            print("Error. Graph call can be made only after the first pass (sentence by sentence)")
            exit()

    if m:
        cWordLimitLeft = m.group('borders_and_case')[0]
        cCaseMode = m.group('borders_and_case')[1]
        cWordLimitRight = m.group('borders_and_case')[2]
        sOption = m.group('option')[1:]  if m.group('option')  else False
        sRuleId =  m.group('ruleid')[1:-1]
        if sRuleId in aRULESET:
            print("# Error. Several rules have the same id: " + sRuleId)
            exit()
        aRULESET.add(sRuleId)
        nPriority = dOptPriority.get(sOption, 4)
        if m.group('priority'):
            nPriority = int(m.group('priority')[1:])
        s = s[m.end(0):]
    else:
        print("# Warning. Rule wrongly shaped at line: " + sLineId)
        exit()

    #### REGEX TRIGGER
    i = s.find(" <<-")
    if i == -1:
        print("# Error: no condition at line " + sLineId)
        return None
    sRegex = s[:i].strip()
    s = s[i+4:]

    # JS groups positioning codes
    m = re.search("@@\\S+", sRegex)
    if m:
        tGroups = jsconv.groupsPositioningCodeToList(sRegex[m.start()+2:])
        sRegex = sRegex[:m.start()].strip()
    # JS regex
    m = re.search("<js>.+</js>i?", sRegex)
    if m:
        dJSREGEXES[sLineId] = m.group(0)
        sRegex = sRegex[:m.start()].strip()
    if "<js>" in sRegex or "</js>" in sRegex:
        print("# Error: JavaScript regex not delimited at line " + sLineId)
        return None

    # quotes ?
    if sRegex.startswith('"') and sRegex.endswith('"'):
        sRegex = sRegex[1:-1]

    ## definitions
    for sDef, sRepl in dDEFINITIONS.items():
        sRegex = sRegex.replace(sDef, sRepl)

    ## count number of groups (must be done before modifying the regex)
    nGroup = countGroupInRegex(sRegex)
    if nGroup > 0:
        if not tGroups:
            print("# Warning: groups positioning code for JavaScript should be defined at line " + sLineId)
        else:
            if nGroup != len(tGroups):
                print("# Error: groups positioning code irrelevant at line " + sLineId)

    ## word limit
    if cWordLimitLeft == '[' and not sRegex.startswith(("^", '’', "'", ",")):
        sRegex = sWORDLIMITLEFT + sRegex
    if cWordLimitRight == ']' and not sRegex.endswith(("$", '’', "'", ",")):
        sRegex = sRegex + sWORDLIMITRIGHT

    ## casing mode
    if cCaseMode == "i":
        bCaseInsensitive = True
        if not sRegex.startswith("(?i)"):
            sRegex = "(?i)" + sRegex
    elif cCaseMode == "s":
        bCaseInsensitive = False
        sRegex = sRegex.replace("(?i)", "")
    elif cCaseMode == "u":
        bCaseInsensitive = False
        sRegex = sRegex.replace("(?i)", "")
        sRegex = uppercase(sRegex, sLang)
    else:
        print("# Unknown case mode [" + cCaseMode + "] at line " + sLineId)

    ## check regex
    try:
        re.compile(sRegex)
    except re.error:
        print("# Regex error at line ", nIdLine)
        print(sRegex)
        return None
    ## groups in non grouping parenthesis
    for _ in re.finditer(r"\(\?:[^)]*\([\[\w -]", sRegex):
        print("# Warning: groups inside non grouping parenthesis in regex at line " + sLineId)

    #### PARSE ACTIONS
    lActions = []
    nAction = 1
    for sAction in s.split(" <<- "):
        t = createAction(sRuleId + "_" + str(nAction), sAction, nGroup)
        nAction += 1
        if t:
            lActions.append(t)
    if not lActions:
        return None

    return [sOption, sRegex, bCaseInsensitive, sLineId, sRuleId, nPriority, lActions, tGroups]


def checkReferenceNumbers (sText, sActionId, nToken):
    "check if token references in <sText> greater than <nToken> (debugging)"
    for x in re.finditer(r"\\(\d+)", sText):
        if int(x.group(1)) > nToken:
            print("# Error in token index at line " + sActionId + " ("+str(nToken)+" tokens only)")
            print(sText)


def checkIfThereIsCode (sText, sActionId):
    "check if there is code in <sText> (debugging)"
    if re.search("[.]\\w+[(]|sugg\\w+[(]|\\([0-9]|\\[[0-9]", sText):
        print("# Warning at line " + sActionId + ":  This message looks like code. Line should probably begin with =")
        print(sText)


def createAction (sIdAction, sAction, nGroup):
    "returns an action to perform as a tuple (condition, action type, action[, iGroup [, message, URL ]])"
    m = re.search(r"([-~=>])(\d*|)>>", sAction)
    if not m:
        print("# No action at line " + sIdAction)
        return None

    #### CONDITION
    sCondition = sAction[:m.start()].strip()
    if sCondition:
        sCondition = prepareFunction(sCondition)
        lFUNCTIONS.append(("_c_"+sIdAction, sCondition))
        checkReferenceNumbers(sCondition, sIdAction, nGroup)
        if ".match" in sCondition:
            print("# Error. JS compatibility. Don't use .match() in condition, use .search()")
        sCondition = "_c_"+sIdAction
    else:
        sCondition = None

    #### iGroup / positioning
    iGroup = int(m.group(2)) if m.group(2) else 0
    if iGroup > nGroup:
        print("# Selected group > group number in regex at line " + sIdAction)

    #### ACTION
    sAction = sAction[m.end():].strip()
    cAction = m.group(1)
    if cAction == "-":
        ## error
        iMsg = sAction.find(" # ")
        if iMsg == -1:
            sMsg = "# Error. Error message not found."
            sURL = ""
            print(sMsg + " Action id: " + sIdAction)
        else:
            sMsg = sAction[iMsg+3:].strip()
            sAction = sAction[:iMsg].strip()
            sURL = ""
            mURL = re.search("[|] *(https?://.*)", sMsg)
            if mURL:
                sURL = mURL.group(1).strip()

        checkIfThereIsCode(sAction, sIdAction)

    if cAction == ">":
        ## no action, break loop if condition is False
        return [sCondition, cAction, ""]

    if not sAction:
        print("# Error in action at line " + sIdAction + ":  This action is empty.")
        return None

    if cAction == "-":
        ## error detected --> suggestion
        if sAction[0:1] == "=":
            lFUNCTIONS.append(("_s_"+sIdAction, sAction[1:]))
            sAction = "=_s_"+sIdAction
        elif sAction.startswith('"') and sAction.endswith('"'):
            sAction = sAction[1:-1]
        if not sMsg:
            print("# Error in action at line " + sIdAction + ":  the message is empty.")
        return [sCondition, cAction, sAction, iGroup, sMsg, sURL]
    if cAction == "~":
        ## text processor
        if sAction[0:1] == "=":
            lFUNCTIONS.append(("_p_"+sIdAction, sAction[1:]))
            sAction = "=_p_"+sIdAction
        elif sAction.startswith('"') and sAction.endswith('"'):
            sAction = sAction[1:-1]
        return [sCondition, cAction, sAction, iGroup]
    if cAction == "=":
        ## disambiguator
        if sAction[0:1] == "=":
            sAction = sAction[1:]
        if "define" in sAction and not re.search(r"define\(dTokenPos, *m\.start.*, \[.*\] *\)", sAction):
            print("# Error in action at line " + sIdAction + ": second argument for define must be a list of strings")
            print(sAction)
        lFUNCTIONS.append(("_d_"+sIdAction, sAction))
        sAction = "_d_"+sIdAction
        return [sCondition, cAction, sAction]
    print("# Unknown action at line " + sIdAction)
    return None


def _calcRulesStats (lRules):
    "count rules and actions"
    d = {'=':0, '~': 0, '-': 0, '>': 0}
    for aRule in lRules:

def printBookmark (nLevel, sComment, nLine):
    "print bookmark within the rules file"
    print("  {:>6}:  {}".format(nLine, "  " * nLevel + sComment))


def make (spLang, sLang, bUseCache=False):
    "compile rules, returns a dictionary of values"
    # for clarity purpose, don’t create any file here

    dCacheVars = None

    if os.path.isfile("_build/data_cache.json"):
        print("> data cache found")
        sJSON = open("_build/data_cache.json", "r", encoding="utf-8").read()
        dCacheVars = json.loads(sJSON)
        sBuildDate = time.strftime("%Y-%m-%d %H:%M:%S", time.gmtime(dCacheVars.get("fBuildTime", 0)))
        if bUseCache:
            print("> use cache (no rebuild asked)")
            print("  build made at: " + sBuildDate)
            return dCacheVars

    print("> read rules file...")
    try:
        sFileContent = open(spLang + "/rules.grx", 'r', encoding="utf-8").read()
    except OSError:
        print("Error. Rules file in project [" + sLang + "] not found.")
        exit()


    xHasher = hashlib.new("sha3_512")
    xHasher.update(sFileContent.encode("utf-8"))
    sFileHash = xHasher.hexdigest()

    if dCacheVars and sFileHash == dCacheVars.get("sFileHash", ""):

        print("> cache hash identical to file hash, use cache")
        print("  build made at: " + sBuildDate)
        return dCacheVars

    # removing comments, zeroing empty lines, creating definitions, storing tests, merging rule lines
    print("  parsing rules...")
    fBuildTime = time.time()

            pass
        elif sLine.startswith("DEF:"):
            # definition
            m = re.match("DEF: +([a-zA-Z_][a-zA-Z_0-9]*) +(.+)$", sLine.strip())
            if m:
                dDEFINITIONS["{"+m.group(1)+"}"] = m.group(2)
            else:
                print("Error in definition: ", end="")
                print(sLine.strip())
        elif sLine.startswith("DECL:"):
            # declensions
            m = re.match(r"DECL: +(\+\w+) (.+)$", sLine.strip())
            if m:
                dDECLENSIONS[m.group(1)] = m.group(2).strip().split()
            else:

        elif sFuncName.startswith("_s_"): # suggestion
            sParams = "sSentence, m"
        elif sFuncName.startswith("_p_"): # preprocessor
            sParams = "sSentence, m"
        elif sFuncName.startswith("_d_"): # disambiguator
            sParams = "sSentence, m, dTokenPos"
        else:
            print("# Unknown function type in [" + sFuncName + "]")
            continue
        # Python
        sPyCallables += "def {} ({}):\n".format(sFuncName, sParams)
        sPyCallables += "    return " + sReturn + "\n"
        # JavaScript
        sJSCallables += "    {}: function ({})".format(sFuncName, sParams) + " {\n"
        sJSCallables += "        return " + jsconv.py2js(sReturn) + ";\n"
        sJSCallables += "    },\n"

    displayStats(lParagraphRules, lSentenceRules)

    dVars = {
        "fBuildTime": fBuildTime,

        self.dDef = dDef
        self.dDecl = dDecl
        self.dOptPriority = dOptPriority
        self.dAntiPatterns = {}
        self.dActions = {}
        self.dFuncName = {}
        self.dFunctions = {}
        self.dURL = {}

    def _genTokenLines (self, sTokenLine):
        "tokenize a string and return a list of lines of tokens"
        lTokenLines = []
        for sTokBlock in sTokenLine.split():
            # replace merger characters by spaces
            if "␣" in sTokBlock:

        if cAction == "-":
            ## error
            iMsg = sAction.find(" # ")
            if iMsg == -1:
                sMsg = "# Error. Error message not found."
                sURL = ""
                print("\n" + sMsg + " at: ", sLineId, sActionId)
            else:
                sMsg = sAction[iMsg+3:].strip()
                sAction = sAction[:iMsg].strip()
                sURL = ""
                mURL = re.search("[|] *(https?://.*)", sMsg)
                if mURL:
                    sURL = mURL.group(1).strip()

    iActionBlock = 0
    aRuleName = set()

    for iLine, sLine in lRule:
        sLine = sLine.rstrip()
        if "\t" in sLine:
            # tabulation not allowed
            print("Error. Tabulation at line: ", iLine)
            exit()
        elif sLine.startswith("@@@@GRAPH: "):
            # rules graph call
            m = re.match(r"@@@@GRAPH: *(\w+) *[|] *(\w+)", sLine.strip())
            if m:
                sGraphName = m.group(1)
                sGraphCode = m.group(2)
                if sGraphName in dAllGraph or sGraphCode in dGraphCode:
                    print(f"Error at line {iLine}. Graph name <{sGraphName}> or graph code <{sGraphCode}> already exists.")
                    exit()
                dAllGraph[sGraphName] = []
                dGraphCode[sGraphName] = sGraphCode
            else:
                print("Error. Graph name not found at line", iLine)
                exit()
        elif sLine.startswith("__") and sLine.endswith("__"):
            # new rule group
            m = re.match("__(\\w+)(!\\d|)__", sLine)
            if m:
                sRuleName = m.group(1)
                if sRuleName in aRuleName:
                    print(f"Error at line {iLine}. Rule name <{sRuleName}> already exists.")
                    exit()
                aRuleName.add(sRuleName)
                iActionBlock = 1
                nPriority = int(m.group(2)[1:]) if m.group(2)  else -1
            else:
                print("Syntax error in rule group: ", sLine, " -- line:", iLine)
                exit()
        elif re.match("    \\S", sLine):
            # tokens line
            lTokenLine.append([iLine, sLine.strip()])
        elif sLine.startswith("        ||"):
            # tokens line continuation
            iPrevLine, sPrevLine = lTokenLine[-1]

            if re.search(r"[-=~/!>](?:-?\d\.?(?::\.?-?\d+|)|):?>>", sLine):
                bActionBlock = False
        elif re.match("[  ]*$", sLine):
            # empty line to end merging
            if not lTokenLine:
                continue
            if bActionBlock or not lActions:
                print("Error. No action found at line:", iLine)
                print(bActionBlock, lActions)
                exit()
            if not sGraphName:
                print("Error. All rules must belong to a named graph. Line: ", iLine)
                exit()
            for j, sTokenLine in lTokenLine:
                dAllGraph[sGraphName].append((j, sRuleName, sTokenLine, iActionBlock, list(lActions), nPriority))
            lTokenLine.clear()
            lActions.clear()
            iActionBlock += 1
        else:
            print("Unknown line at:", iLine)
            print(sLine)

    # processing rules
    print("  processing graph rules...")
    initProcessPoolExecutor(len(dAllGraph))
    fStartTimer = time.time()
    # build graph

def getConfig (sLang):
    "load config.ini in <sLang> at gc_lang/<sLang>, returns xConfigParser object"
    xConfig = configparser.ConfigParser()
    xConfig.optionxform = str
    try:
        xConfig.read_file(open("gc_lang/" + sLang + "/config.ini", "r", encoding="utf-8"))
    except FileNotFoundError:
        print("# Error. Can’t read config file [" + sLang + "]")
        exit()
    return xConfig


def createOptionsLabelProperties (dOptLbl):
    "create content for .properties files (LibreOffice)"
    sContent = ""
    for sOpt, tLabel in dOptLbl.items():
        sContent += sOpt + "=" + tLabel[0] + "\n"
        if tLabel[1]:
            sContent += "hlp_" + sOpt + "=" + tLabel[1] + "\n"
    return sContent


def createDialogOptionsXDL (dVars):
    "create bundled dialog options file .xdl (LibreOffice)"
    sFixedline = '<dlg:fixedline dlg:id="{0}" dlg:tab-index="{1}" dlg:top="{2}" dlg:left="5" dlg:width="{3}" dlg:height="10" dlg:value="&amp;{0}" />\n'
    sCheckbox = '<dlg:checkbox dlg:id="{0}" dlg:tab-index="{1}" dlg:top="{2}" dlg:left="{3}" dlg:width="{4}" dlg:height="10" dlg:value="&amp;{0}" dlg:checked="{5}" {6} />\n'
    iTabIndex = 1
    nPosY = 5
    nWidth = 240
    sContent = ""
    dOpt = dVars["dOptPython"]
    dOptLabel = dVars["dOptLabel"][dVars["lang"]]
    for sGroup, lGroupOptions in dVars["lStructOpt"]:
        sContent += sFixedline.format(sGroup, iTabIndex, nPosY, nWidth)
        iTabIndex += 1
        for lLineOptions in lGroupOptions:
            nElemWidth = nWidth // len(lLineOptions)
            nPosY += 10
            nPosX = 10
            for sOpt in lLineOptions:
                sHelp = 'dlg:help-text="&amp;hlp_%s"'%sOpt  if dOptLabel[sOpt][1]  else ""
                sContent += sCheckbox.format(sOpt, iTabIndex, nPosY, nPosX, nElemWidth, "true" if dOpt[sOpt] else "false", sHelp)

                iTabIndex += 1
                nPosX += nElemWidth
        nPosY += 10
    return sContent


def createOXT (spLang, dVars, dOxt, spLangPack, bInstall):
    "create extension for Writer"
    print("Building extension for Writer")
    spfZip = "_build/" + dVars['name'] + "-"+ dVars['lang'] +"-v" + dVars['version'] + '.oxt'
    hZip = zipfile.ZipFile(spfZip, mode='w', compression=zipfile.ZIP_DEFLATED)

    # Package and parser
    copyGrammalectePyPackageInZipFile(hZip, spLangPack, "pythonpath/")
    hZip.write("grammalecte-cli.py", "pythonpath/grammalecte-cli.py")

    # Extension files

    dVars["xcs_options"] = "\n".join([ '<prop oor:name="'+sOpt+'" oor:type="xs:string"><value></value></prop>' for sOpt in dVars["dOptPython"] ])
    dVars["xcu_label_values"] = "\n".join([ '<value xml:lang="'+sLang+'">' + dVars["dOptLabel"][sLang]["__optiontitle__"] + '</value>'  for sLang in dVars["dOptLabel"] ])
    hZip.writestr("dialog/options_page.xdl", helpers.fileFile("gc_core/py/oxt/options_page.xdl", dVars))
    hZip.writestr("dialog/OptionsDialog.xcs", helpers.fileFile("gc_core/py/oxt/OptionsDialog.xcs", dVars))
    hZip.writestr("dialog/OptionsDialog.xcu", helpers.fileFile("gc_core/py/oxt/OptionsDialog.xcu", dVars))
    hZip.writestr("dialog/" + dVars['lang'] + "_en.default", "")
    for sLangLbl, dOptLbl in dVars['dOptLabel'].items():
        hZip.writestr("dialog/" + dVars['lang'] + "_" + sLangLbl + ".properties", createOptionsLabelProperties(dOptLbl))

    ## ADDONS OXT
    print("+ OXT: ", end="")
    for spfSrc, spfDst in dOxt.items():
        print(spfSrc, end=", ")
        if os.path.isdir(spLang+'/'+spfSrc):
            for sf in os.listdir(spLang+'/'+spfSrc):

        if dVars.get('unopkg', False):
            cmd = '"'+os.path.abspath(dVars.get('unopkg')+'" add -f '+spfZip)
            print(cmd)
            os.system(cmd)
        else:
            print("# Error: path and filename of unopkg not set in config.ini")


def createServerOptions (sLang, dOptData):
    "create file options for Grammalecte server"
    with open("grammalecte-server-options."+sLang+".ini", "w", encoding="utf-8", newline="\n") as hDst:
        hDst.write("# Server options. Lang: " + sLang + "\n\n[gc_options]\n")
        for sSection, lOpt in dOptData["lStructOpt"]:
            hDst.write("\n########## " + dOptData["dOptLabel"][sLang].get(sSection, sSection + "[no label found]")[0] + " ##########\n")
            for lLineOpt in lOpt:
                for sOpt in lLineOpt:
                    hDst.write("# " + dOptData["dOptLabel"][sLang].get(sOpt, "[no label found]")[0] + "\n")
                    hDst.write(sOpt + " = " + ("1" if dOptData["dOptServer"].get(sOpt, None) else "0") + "\n")
        hDst.write("html = 1\n")


def createPackageZip (dVars, spLangPack):
    "create server zip"
    spfZip = "_build/" + dVars['name'] + "-"+ dVars['lang'] +"-v" + dVars['version'] + '.zip'
    hZip = zipfile.ZipFile(spfZip, mode='w', compression=zipfile.ZIP_DEFLATED)
    copyGrammalectePyPackageInZipFile(hZip, spLangPack)
    for spf in ["grammalecte-cli.py", "grammalecte-server.py", \
                "README.txt", "LICENSE.txt", "LICENSE.fr.txt"]:
        hZip.write(spf)
    hZip.writestr("setup.py", helpers.fileFile("gc_lang/fr/setup.py", dVars))

    dVars["dic_personal_filename_js"] = ""
    lDict = [ ("main", s)  for s in dVars['dic_filenames'].split(",") ]
    if bCommunityDict:
        lDict.append(("community", dVars['dic_community_filename']))
    if bPersonalDict:
        lDict.append(("personal", dVars['dic_personal_filename']))
    for sType, sFileName in lDict:
        spfPyDic = "graphspell/_dictionaries/" + sFileName + ".bdic"
        spfJSDic = "graphspell-js/_dictionaries/" + sFileName + ".json"
        if not os.path.isfile(spfPyDic) or (bJavaScript and not os.path.isfile(spfJSDic)):
            buildDictionary(dVars, sType, bJavaScript)
        print(spfPyDic)
        file_util.copy_file(spfPyDic, "grammalecte/graphspell/_dictionaries")
        dVars['dic_'+sType+'_filename_py'] = sFileName + '.bdic'
        if bJavaScript:
            print(spfJSDic)

    print("Python: " + sys.version)
    if sys.version < "3.7":
        print("Python 3.7+ required")
        return
    xParser = argparse.ArgumentParser()
    xParser.add_argument("lang", type=str, nargs='+', help="lang project to generate (name of folder in /lang)")
    xParser.add_argument("-uc", "--use_cache", help="use data cache instead of rebuilding rules", action="store_true")

    xParser.add_argument("-b", "--build_data", help="launch build_data.py (part 1 and 2)", action="store_true")
    xParser.add_argument("-bb", "--build_data_before", help="launch build_data.py (only part 1: before dictionary building)", action="store_true")
    xParser.add_argument("-ba", "--build_data_after", help="launch build_data.py (only part 2: before dictionary building)", action="store_true")
    xParser.add_argument("-d", "--dict", help="generate FSA dictionary", action="store_true")
    xParser.add_argument("-t", "--tests", help="run unit tests", action="store_true")
    xParser.add_argument("-p", "--perf", help="run performance tests", action="store_true")
    xParser.add_argument("-pm", "--perf_memo", help="run performance tests and store results in perf_memo.txt", action="store_true")

            if databuild and xArgs.build_data_after:
                databuild.after('gc_lang/'+sLang, dVars, xArgs.javascript)

            # copy dictionaries from Graphspell
            copyGraphspellDictionaries(dVars, xArgs.javascript, xArgs.add_community_dictionary, xArgs.add_personal_dictionary)

            # make





            sVersion = create(sLang, xConfig, xArgs.install, xArgs.javascript, xArgs.use_cache)

            # tests
            if xArgs.tests or xArgs.perf or xArgs.perf_memo:
                print("> Running tests")
                try:
                    tests = importlib.import_module("grammalecte."+sLang+".tests")
                    print(tests.__file__)
                except ImportError:
                    print("# Error. Import failed:" + "grammalecte."+sLang+".tests")
                    traceback.print_exc()
                else:
                    if xArgs.tests:
                        xTestSuite = unittest.TestLoader().loadTestsFromModule(tests)
                        unittest.TextTestRunner().run(xTestSuite)
                    if xArgs.perf or xArgs.perf_memo:
                        hDst = open("./gc_lang/"+sLang+"/perf_memo.txt", "a", encoding="utf-8", newline="\n")  if xArgs.perf_memo  else None