Grammalecte  Check-in [82ff352fc9]

Overview
Comment:[core] gc engine: better names for vars and functions
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | core | mtok
Files: files | file ages | folders
SHA3-256: 82ff352fc91b66375eb862e5bec4bfed77c050d5d77b2720a14339567da72b8b
User & Date: olr on 2021-03-10 20:02:04
Other Links: branch diff | manifest | tags
Context
2021-03-12
15:49
[core] function morphx for multi-token check-in: 5b8ec019ea user: olr tags: core, mtok
2021-03-10
20:02
[core] gc engine: better names for vars and functions check-in: 82ff352fc9 user: olr tags: core, mtok
19:54
[core] gc engine: regex for multi-tokens [fr] tests check-in: 3e55498f43 user: olr tags: fr, core, mtok
Changes

Modified gc_core/js/lang_core/gc_engine.js from [acf5e76688] to [f3abee6b18].

297
298
299
300
301
302
303
304

305
306
307
308
309
310
311
312
313
314
315

316
317
318
319
320
321
322
323

324
325
326
327
328
329
330
331
332

333
334
335
336
337
338
339
297
298
299
300
301
302
303

304
305
306
307
308
309
310
311
312
313
314

315
316
317
318
319
320
321
322

323
324
325
326
327
328
329
330
331

332
333
334
335
336
337
338
339







-
+










-
+







-
+








-
+







                }
            }
            else if (!sOption || option(sOption)) {
                for (let [zRegex, bUppercase, sLineId, sRuleId, nPriority, lActions, lGroups, lNegLookBefore] of lRuleGroup) {
                    if (!gc_engine.aIgnoredRules.has(sRuleId)) {
                        while ((m = zRegex.gl_exec2(sText, lGroups, lNegLookBefore)) !== null) {
                            let bCondMemo = null;
                            for (let [sFuncCond, cActionType, sWhat, ...eAct] of lActions) {
                            for (let [sFuncCond, cActionType, sAction, ...eAct] of lActions) {
                                // action in lActions: [ condition, action type, replacement/suggestion/action[, iGroup[, message, URL]] ]
                                try {
                                    bCondMemo = (!sFuncCond || gc_functions[sFuncCond](sText, sText0, m, this.dTokenPos, sCountry, bCondMemo));
                                    if (bCondMemo) {
                                        switch (cActionType) {
                                            case "-":
                                                // grammar error
                                                //console.log("-> error detected in " + sLineId + "\nzRegex: " + zRegex.source);
                                                let nErrorStart = nOffset + m.start[eAct[0]];
                                                if (!this.dError.has(nErrorStart) || nPriority > this.dErrorPriority.get(nErrorStart)) {
                                                    this.dError.set(nErrorStart, this._createErrorFromRegex(sText, sText0, sWhat, nOffset, m, eAct[0], sLineId, sRuleId, bUppercase, eAct[1], eAct[2], bShowRuleId, sOption, bContext));
                                                    this.dError.set(nErrorStart, this._createErrorFromRegex(sText, sText0, sAction, nOffset, m, eAct[0], sLineId, sRuleId, bUppercase, eAct[1], eAct[2], bShowRuleId, sOption, bContext));
                                                    this.dErrorPriority.set(nErrorStart, nPriority);
                                                    this.dSentenceError.set(nErrorStart, this.dError.get(nErrorStart));
                                                }
                                                break;
                                            case "~":
                                                // text processor
                                                //console.log("-> text processor by " + sLineId + "\nzRegex: " + zRegex.source);
                                                sText = this.rewriteText(sText, sWhat, eAct[0], m, bUppercase);
                                                sText = this.rewriteText(sText, sAction, eAct[0], m, bUppercase);
                                                bChange = true;
                                                if (bDebug) {
                                                    console.log("~ " + sText + "  -- " + m[eAct[0]] + "  # " + sLineId);
                                                }
                                                break;
                                            case "=":
                                                // disambiguation
                                                //console.log("-> disambiguation by " + sLineId + "\nzRegex: " + zRegex.source);
                                                gc_functions[sWhat](sText, m, this.dTokenPos);
                                                gc_functions[sAction](sText, m, this.dTokenPos);
                                                if (bDebug) {
                                                    console.log("= " + m[0] + "  # " + sLineId, "\nDA:", this.dTokenPos);
                                                }
                                                break;
                                            case ">":
                                                // we do nothing, this test is just a condition to apply all following actions
                                                break;
387
388
389
390
391
392
393
394

395
396
397
398
399
400
401
387
388
389
390
391
392
393

394
395
396
397
398
399
400
401







-
+







        }
        if (bDebug) {
            console.log("UPDATE:");
            console.log(this.asString());
        }
    }

    * _getMatches (oGraph, oToken, oNode, bKeep=false) {
    * _getNextNodes (oGraph, oToken, oNode, bKeep=false) {
        // generator: return matches where <oToken> “values” match <oNode> arcs
        try {
            let bTokenFound = false;
            // token value
            if (oNode.hasOwnProperty(oToken["sValue"])) {
                yield [" ", oToken["sValue"], oNode[oToken["sValue"]]];
                bTokenFound = true;
575
576
577
578
579
580
581
582

583
584
585
586
587
588
589
590
591
592

593
594
595
596
597
598
599
600
601


602
603
604

605
606
607
608
609
610
611
612



613
614
615
616

617
618
619


620
621
622

623
624
625


626
627
628
629

630
631
632


633
634
635

636
637
638
639
640
641
642
575
576
577
578
579
580
581

582
583
584
585
586
587
588
589
590
591

592
593
594
595
596
597
598
599


600
601
602
603

604
605
606
607
608
609



610
611
612
613
614
615

616
617


618
619
620
621

622
623


624
625
626
627
628

629
630


631
632
633
634

635
636
637
638
639
640
641
642







-
+









-
+







-
-
+
+


-
+





-
-
-
+
+
+



-
+

-
-
+
+


-
+

-
-
+
+



-
+

-
-
+
+


-
+







            }
            if (!bTokenFound && bKeep) {
                yield [null, "", -1];
            }
            // JUMP
            // Warning! Recurssion!
            if (oNode.hasOwnProperty("<>")) {
                yield* this._getMatches(oGraph, oToken, oGraph[oNode["<>"]], bKeep=true);
                yield* this._getNextNodes(oGraph, oToken, oGraph[oNode["<>"]], bKeep=true);
            }
        }
        catch (e) {
            console.error(e);
        }
    }

    parseGraph (oGraph, sCountry="${country_default}", dOptions=null, bShowRuleId=false, bDebug=false, bContext=false) {
        // parse graph with tokens from the text and execute actions encountered
        let lPointer = [];
        let lPointers = [];
        let bTagAndRewrite = false;
        try {
            for (let [iToken, oToken] of this.lTokens.entries()) {
                if (bDebug) {
                    console.log("TOKEN: " + oToken["sValue"]);
                }
                // check arcs for each existing pointer
                let lNextPointer = [];
                for (let oPointer of lPointer) {
                let lNextPointers = [];
                for (let oPointer of lPointers) {
                    if (oPointer["nMultiEnd"] != -1) {
                        if (oToken["i"] <= oPointer["nMultiEnd"]) {
                            lNextPointer.push(oPointer);
                            lNextPointers.push(oPointer);
                        }
                        if (oToken["i"] != oPointer["nMultiEnd"]) {
                            continue;
                        }
                    }
                    for (let [cActionType, sMatch, iNode] of this._getMatches(oGraph, oToken, oGraph[oPointer["iNode"]])) {
                        if (cActionType === null) {
                            lNextPointer.push(oPointer);
                    for (let [cNodeType, sMatch, iNode] of this._getNextNodes(oGraph, oToken, oGraph[oPointer["iNode"]])) {
                        if (cNodeType === null) {
                            lNextPointers.push(oPointer);
                            continue;
                        }
                        if (bDebug) {
                            console.log("  MATCH: " + cActionType + sMatch);
                            console.log("  MATCH: " + cNodeType + sMatch);
                        }
                        let nMultiEnd = (cActionType != "&") ? -1 : dToken["nMultiStartTo"];
                        lNextPointer.push({ "iToken1": oPointer["iToken1"], "iNode": iNode, "nMultiEnd": nMultiEnd });
                        let nMultiEnd = (cNodeType != "&") ? -1 : dToken["nMultiStartTo"];
                        lNextPointers.push({ "iToken1": oPointer["iToken1"], "iNode": iNode, "nMultiEnd": nMultiEnd });
                    }
                }
                lPointer = lNextPointer;
                lPointers = lNextPointers;
                // check arcs of first nodes
                for (let [cActionType, sMatch, iNode] of this._getMatches(oGraph, oToken, oGraph[0])) {
                    if (cActionType === null) {
                for (let [cNodeType, sMatch, iNode] of this._getNextNodes(oGraph, oToken, oGraph[0])) {
                    if (cNodeType === null) {
                        continue;
                    }
                    if (bDebug) {
                        console.log("  MATCH: " + cActionType + sMatch);
                        console.log("  MATCH: " + cNodeType + sMatch);
                    }
                    let nMultiEnd = (cActionType != "&") ? -1 : dToken["nMultiStartTo"];
                    lPointer.push({ "iToken1": iToken, "iNode": iNode, "nMultiEnd": nMultiEnd });
                    let nMultiEnd = (cNodeType != "&") ? -1 : dToken["nMultiStartTo"];
                    lPointers.push({ "iToken1": iToken, "iNode": iNode, "nMultiEnd": nMultiEnd });
                }
                // check if there is rules to check for each pointer
                for (let oPointer of lPointer) {
                for (let oPointer of lPointers) {
                    if (oPointer["nMultiEnd"] != -1) {
                        if (oToken["i"] < oPointer["nMultiEnd"]) {
                            continue;
                        }
                        if (oToken["i"] == oPointer["nMultiEnd"]) {
                            oPointer["nMultiEnd"] = -1;
                        }
667
668
669
670
671
672
673
674

675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694

695
696
697
698
699
700
701
702
703
704
705
706
707
708

709
710
711

712
713
714
715
716

717
718

719
720
721
722
723
724
725
726
727
728
729
730
731
732
733

734
735

736
737
738
739

740
741

742
743
744
745
746
747
748
749
750
751
752
753
754
755
756

757
758
759
760
761
762
763
667
668
669
670
671
672
673

674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693

694
695
696
697
698
699
700
701
702
703
704
705
706
707

708
709
710

711
712
713
714
715

716
717

718
719
720
721
722
723
724
725
726
727
728
729
730
731
732

733
734

735
736
737
738

739
740

741
742
743
744
745
746
747
748
749
750
751
752
753
754
755

756
757
758
759
760
761
762
763







-
+



















-
+













-
+


-
+




-
+

-
+














-
+

-
+



-
+

-
+














-
+







        for (let [sLineId, nextNodeKey] of Object.entries(oNode)) {
            let bCondMemo = null;
            for (let sRuleId of oGraph[nextNodeKey]) {
                try {
                    if (bDebug) {
                        console.log("   >TRY: " + sRuleId + " " + sLineId);
                    }
                    let [_, sOption, sFuncCond, cActionType, sWhat, ...eAct] = gc_rules_graph.dRule[sRuleId];
                    let [_, sOption, sFuncCond, cActionType, sAction, ...eAct] = gc_rules_graph.dRule[sRuleId];
                    // Suggestion    [ option, condition, "-", replacement/suggestion/action, iTokenStart, iTokenEnd, cStartLimit, cEndLimit, bCaseSvty, nPriority, sMessage, iURL ]
                    // TextProcessor [ option, condition, "~", replacement/suggestion/action, iTokenStart, iTokenEnd, bCaseSvty ]
                    // Disambiguator [ option, condition, "=", replacement/suggestion/action ]
                    // Tag           [ option, condition, "/", replacement/suggestion/action, iTokenStart, iTokenEnd ]
                    // Immunity      [ option, condition, "!", "",                            iTokenStart, iTokenEnd ]
                    // Immunity      [ option, condition, "&", "",                            iTokenStart, iTokenEnd ]
                    // Test          [ option, condition, ">", "" ]
                    if (!sOption || dOptions.gl_get(sOption, false)) {
                        bCondMemo = !sFuncCond || gc_functions[sFuncCond](this.lTokens, nTokenOffset, nLastToken, sCountry, bCondMemo, this.dTags, this.sSentence, this.sSentence0);
                        if (bCondMemo) {
                            if (cActionType == "-") {
                                // grammar error
                                let [iTokenStart, iTokenEnd, cStartLimit, cEndLimit, bCaseSvty, nPriority, sMessage, iURL] = eAct;
                                let nTokenErrorStart = (iTokenStart > 0) ? nTokenOffset + iTokenStart : nLastToken + iTokenStart;
                                if (!this.lTokens[nTokenErrorStart].hasOwnProperty("sImmunity") || (this.lTokens[nTokenErrorStart]["sImmunity"] != "*" && !this.lTokens[nTokenErrorStart]["sImmunity"].includes(sOption))) {
                                    let nTokenErrorEnd = (iTokenEnd > 0) ? nTokenOffset + iTokenEnd : nLastToken + iTokenEnd;
                                    let nErrorStart = this.nOffsetWithinParagraph + ((cStartLimit == "<") ? this.lTokens[nTokenErrorStart]["nStart"] : this.lTokens[nTokenErrorStart]["nEnd"]);
                                    let nErrorEnd = this.nOffsetWithinParagraph + ((cEndLimit == ">") ? this.lTokens[nTokenErrorEnd]["nEnd"] : this.lTokens[nTokenErrorEnd]["nStart"]);
                                    if (!this.dError.has(nErrorStart) || nPriority > this.dErrorPriority.gl_get(nErrorStart, -1)) {
                                        this.dError.set(nErrorStart, this._createErrorFromTokens(sWhat, nTokenOffset, nLastToken, nTokenErrorStart, nErrorStart, nErrorEnd, sLineId, sRuleId, bCaseSvty,
                                        this.dError.set(nErrorStart, this._createErrorFromTokens(sAction, nTokenOffset, nLastToken, nTokenErrorStart, nErrorStart, nErrorEnd, sLineId, sRuleId, bCaseSvty,
                                                                                                 sMessage, gc_rules_graph.dURL[iURL], bShowRuleId, sOption, bContext));
                                        this.dErrorPriority.set(nErrorStart, nPriority);
                                        this.dSentenceError.set(nErrorStart, this.dError.get(nErrorStart));
                                        if (bDebug) {
                                            console.log("    NEW_ERROR: ",  this.dError.get(nErrorStart));
                                        }
                                    }
                                }
                            }
                            else if (cActionType == "~") {
                                // text processor
                                let nTokenStart = (eAct[0] > 0) ? nTokenOffset + eAct[0] : nLastToken + eAct[0];
                                let nTokenEnd = (eAct[1] > 0) ? nTokenOffset + eAct[1] : nLastToken + eAct[1];
                                this._tagAndPrepareTokenForRewriting(sWhat, nTokenStart, nTokenEnd, nTokenOffset, nLastToken, eAct[2], bDebug);
                                this._tagAndPrepareTokenForRewriting(sAction, nTokenStart, nTokenEnd, nTokenOffset, nLastToken, eAct[2], bDebug);
                                bChange = true;
                                if (bDebug) {
                                    console.log(`    TEXT_PROCESSOR: [${this.lTokens[nTokenStart]["sValue"]}:${this.lTokens[nTokenEnd]["sValue"]}]  > ${sWhat}`);
                                    console.log(`    TEXT_PROCESSOR: [${this.lTokens[nTokenStart]["sValue"]}:${this.lTokens[nTokenEnd]["sValue"]}]  > ${sAction}`);
                                }
                            }
                            else if (cActionType == "=") {
                                // disambiguation
                                gc_functions[sWhat](this.lTokens, nTokenOffset, nLastToken);
                                gc_functions[sAction](this.lTokens, nTokenOffset, nLastToken);
                                if (bDebug) {
                                    console.log(`    DISAMBIGUATOR: (${sWhat})  [${this.lTokens[nTokenOffset+1]["sValue"]}:${this.lTokens[nLastToken]["sValue"]}]`);
                                    console.log(`    DISAMBIGUATOR: (${sAction})  [${this.lTokens[nTokenOffset+1]["sValue"]}:${this.lTokens[nLastToken]["sValue"]}]`);
                                }
                            }
                            else if (cActionType == ">") {
                                // we do nothing, this test is just a condition to apply all following actions
                                if (bDebug) {
                                    console.log("    COND_OK");
                                }
                            }
                            else if (cActionType == "/") {
                                // Tag
                                let nTokenStart = (eAct[0] > 0) ? nTokenOffset + eAct[0] : nLastToken + eAct[0];
                                let nTokenEnd = (eAct[1] > 0) ? nTokenOffset + eAct[1] : nLastToken + eAct[1];
                                for (let i = nTokenStart; i <= nTokenEnd; i++) {
                                    if (this.lTokens[i].hasOwnProperty("aTags")) {
                                        this.lTokens[i]["aTags"].add(...sWhat.split("|"))
                                        this.lTokens[i]["aTags"].add(...sAction.split("|"))
                                    } else {
                                        this.lTokens[i]["aTags"] = new Set(sWhat.split("|"));
                                        this.lTokens[i]["aTags"] = new Set(sAction.split("|"));
                                    }
                                }
                                if (bDebug) {
                                    console.log(`    TAG:  ${sWhat} > [${this.lTokens[nTokenStart]["sValue"]}:${this.lTokens[nTokenEnd]["sValue"]}]`);
                                    console.log(`    TAG:  ${sAction} > [${this.lTokens[nTokenStart]["sValue"]}:${this.lTokens[nTokenEnd]["sValue"]}]`);
                                }
                                for (let sTag of sWhat.split("|")) {
                                for (let sTag of sAction.split("|")) {
                                    if (!this.dTags.has(sTag)) {
                                        this.dTags.set(sTag, [nTokenStart, nTokenEnd]);
                                    } else {
                                        this.dTags.set(sTag, [Math.min(nTokenStart, this.dTags.get(sTag)[0]), Math.max(nTokenEnd, this.dTags.get(sTag)[1])]);
                                    }
                                }
                            }
                            else if (cActionType == "!") {
                                // immunity
                                if (bDebug) {
                                    console.log("    IMMUNITY: " + sLineId + " / " + sRuleId);
                                }
                                let nTokenStart = (eAct[0] > 0) ? nTokenOffset + eAct[0] : nLastToken + eAct[0];
                                let nTokenEnd = (eAct[1] > 0) ? nTokenOffset + eAct[1] : nLastToken + eAct[1];
                                let sImmunity = sWhat || "*";
                                let sImmunity = sAction || "*";
                                if (nTokenEnd - nTokenStart == 0) {
                                    this.lTokens[nTokenStart]["sImmunity"] = sImmunity;
                                    let nErrorStart = this.nOffsetWithinParagraph + this.lTokens[nTokenStart]["nStart"];
                                    if (this.dError.has(nErrorStart)) {
                                        this.dError.delete(nErrorStart);
                                    }
                                } else {
774
775
776
777
778
779
780
781

782
783
784
785
786
787
788
774
775
776
777
778
779
780

781
782
783
784
785
786
787
788







-
+







                                // multi-tokens
                                let nTokenStart = (eAct[0] > 0) ? nTokenOffset + eAct[0] : nLastToken + eAct[0];
                                let nTokenEnd = (eAct[1] > 0) ? nTokenOffset + eAct[1] : nLastToken + eAct[1];
                                let oMultiToken = {
                                    "nTokenStart": nTokenStart,
                                    "nTokenEnd": nTokenEnd,
                                    "lTokens": this.lTokens.slice(nTokenStart, nTokenEnd+1),
                                    "lMorph": (sWhat) ? sWhat.split("|") : [":HM"]
                                    "lMorph": (sAction) ? sAction.split("|") : [":HM"]
                                }
                                this.lTokens[nTokenStart]["nMultiStartTo"] = nTokenEnd
                                this.lTokens[nTokenEnd]["nMultiEndFrom"] = nTokenStart
                                this.lTokens[nTokenStart]["dMultiToken"] = dMultiToken
                                this.lTokens[nTokenEnd]["dMultiToken"] = dMultiToken
                            }
                            else {
909
910
911
912
913
914
915
916

917
918

919
920
921
922
923
924
925
926
927
928

929
930
931
932
933


934
935

936
937

938
939
940
941
942
943
944
945
946
947
948
949


950
951

952
953
954
955
956
957

958
959

960
961
962
963

964
965
966
967
968
969
970
909
910
911
912
913
914
915

916
917

918
919
920
921
922
923
924
925
926
927

928
929
930
931


932
933
934

935
936

937
938
939
940
941
942
943
944
945
946
947


948
949
950

951
952
953
954
955
956

957
958

959
960
961
962

963
964
965
966
967
968
969
970







-
+

-
+









-
+



-
-
+
+

-
+

-
+










-
-
+
+

-
+





-
+

-
+



-
+







            sNew = sRepl.gl_expand(m);
            sNew = sNew + " ".repeat(ln-sNew.length);
        }
        //console.log(sText+"\nstart: "+m.start[iGroup]+" end:"+m.end[iGroup]);
        return sText.slice(0, m.start[iGroup]) + sNew + sText.slice(m.end[iGroup]);
    }

    _tagAndPrepareTokenForRewriting (sWhat, nTokenRewriteStart, nTokenRewriteEnd, nTokenOffset, nLastToken, bCaseSvty, bDebug) {
    _tagAndPrepareTokenForRewriting (sAction, nTokenRewriteStart, nTokenRewriteEnd, nTokenOffset, nLastToken, bCaseSvty, bDebug) {
        // text processor: rewrite tokens between <nTokenRewriteStart> and <nTokenRewriteEnd> position
        if (sWhat === "*") {
        if (sAction === "*") {
            // purge text
            if (nTokenRewriteEnd - nTokenRewriteStart == 0) {
                this.lTokens[nTokenRewriteStart]["bToRemove"] = true;
            } else {
                for (let i = nTokenRewriteStart;  i <= nTokenRewriteEnd;  i++) {
                    this.lTokens[i]["bToRemove"] = true;
                }
            }
        }
        else if (sWhat === "␣") {
        else if (sAction === "␣") {
            // merge tokens
            this.lTokens[nTokenRewriteStart]["nMergeUntil"] = nTokenRewriteEnd;
        }
        else if (sWhat.startsWith("␣")) {
            sWhat = this._expand(sWhat, nTokenOffset, nLastToken);
        else if (sAction.startsWith("␣")) {
            sAction = this._expand(sAction, nTokenOffset, nLastToken);
            this.lTokens[nTokenRewriteStart]["nMergeUntil"] = nTokenRewriteEnd;
            this.lTokens[nTokenRewriteStart]["sMergedValue"] = sWhat.slice(1);
            this.lTokens[nTokenRewriteStart]["sMergedValue"] = sAction.slice(1);
        }
        else if (sWhat === "_") {
        else if (sAction === "_") {
            // neutralized token
            if (nTokenRewriteEnd - nTokenRewriteStart == 0) {
                this.lTokens[nTokenRewriteStart]["sNewValue"] = "_";
            } else {
                for (let i = nTokenRewriteStart;  i <= nTokenRewriteEnd;  i++) {
                    this.lTokens[i]["sNewValue"] = "_";
                }
            }
        }
        else {
            if (sWhat.startsWith("=")) {
                sWhat = gc_functions[sWhat.slice(1)](this.lTokens, nTokenOffset, nLastToken);
            if (sAction.startsWith("=")) {
                sAction = gc_functions[sAction.slice(1)](this.lTokens, nTokenOffset, nLastToken);
            } else {
                sWhat = this._expand(sWhat, nTokenOffset, nLastToken);
                sAction = this._expand(sAction, nTokenOffset, nLastToken);
            }
            let bUppercase = bCaseSvty && this.lTokens[nTokenRewriteStart]["sValue"].slice(0,1).gl_isUpperCase();
            if (nTokenRewriteEnd - nTokenRewriteStart == 0) {
                // one token
                if (bUppercase) {
                    sWhat = sWhat.gl_toCapitalize();
                    sAction = sAction.gl_toCapitalize();
                }
                this.lTokens[nTokenRewriteStart]["sNewValue"] = sWhat;
                this.lTokens[nTokenRewriteStart]["sNewValue"] = sAction;
            }
            else {
                // several tokens
                let lTokenValue = sWhat.split("|");
                let lTokenValue = sAction.split("|");
                if (lTokenValue.length != (nTokenRewriteEnd - nTokenRewriteStart + 1)) {
                    if (bDebug) {
                        console.log("Error. Text processor: number of replacements != number of tokens.");
                    }
                    return;
                }
                let j = 0;

Modified gc_core/py/lang_core/gc_engine.py from [27f95227b1] to [a2c2a9d3d4].

338
339
340
341
342
343
344
345

346
347
348
349
350
351
352
353
354
355
356

357
358
359
360
361

362
363
364
365
366
367
368

369
370
371
372
373
374
375
338
339
340
341
342
343
344

345
346
347
348
349
350
351
352
353
354
355

356
357
358
359
360

361
362
363
364
365
366
367

368
369
370
371
372
373
374
375







-
+










-
+




-
+






-
+







                        sText = self.parseGraph(_rules_graph.dAllGraph[sGraphName], sCountry, dOptions, bShowRuleId, bDebug, bContext)
            elif not sOption or dOptions.get(sOption, False):
                # regex rules
                for zRegex, bUppercase, sLineId, sRuleId, nPriority, lActions in lRuleGroup:
                    if sRuleId not in _aIgnoredRules:
                        for m in zRegex.finditer(sText):
                            bCondMemo = None
                            for sFuncCond, cActionType, sWhat, *eAct in lActions:
                            for sFuncCond, cActionType, sAction, *eAct in lActions:
                                # action in lActions: [ condition, action type, replacement/suggestion/action[, iGroup[, message, URL]] ]
                                try:
                                    bCondMemo = not sFuncCond or getattr(gc_functions, sFuncCond)(sText, sText0, m, self.dTokenPos, sCountry, bCondMemo)
                                    if bCondMemo:
                                        if bDebug:
                                            echo("RULE: " + sLineId)
                                        if cActionType == "-":
                                            # grammar error
                                            nErrorStart = nOffset + m.start(eAct[0])
                                            if nErrorStart not in self.dError or nPriority > self.dErrorPriority.get(nErrorStart, -1):
                                                self.dError[nErrorStart] = self._createErrorFromRegex(sText, sText0, sWhat, nOffset, m, eAct[0], sLineId, sRuleId, bUppercase, eAct[1], eAct[2], bShowRuleId, sOption, bContext)
                                                self.dError[nErrorStart] = self._createErrorFromRegex(sText, sText0, sAction, nOffset, m, eAct[0], sLineId, sRuleId, bUppercase, eAct[1], eAct[2], bShowRuleId, sOption, bContext)
                                                self.dErrorPriority[nErrorStart] = nPriority
                                                self.dSentenceError[nErrorStart] = self.dError[nErrorStart]
                                        elif cActionType == "~":
                                            # text processor
                                            sText = self.rewriteText(sText, sWhat, eAct[0], m, bUppercase)
                                            sText = self.rewriteText(sText, sAction, eAct[0], m, bUppercase)
                                            bChange = True
                                            if bDebug:
                                                echo("~ " + sText + "  -- " + m.group(eAct[0]) + "  # " + sLineId)
                                        elif cActionType == "=":
                                            # disambiguation
                                            if not bParagraph:
                                                getattr(gc_functions, sWhat)(sText, m, self.dTokenPos)
                                                getattr(gc_functions, sAction)(sText, m, self.dTokenPos)
                                                if bDebug:
                                                    echo("= " + m.group(0) + "  # " + sLineId)
                                        elif cActionType == ">":
                                            # we do nothing, this test is just a condition to apply all following actions
                                            pass
                                        else:
                                            echo("# error: unknown action at " + sLineId)
394
395
396
397
398
399
400
401

402
403
404
405
406
407
408
394
395
396
397
398
399
400

401
402
403
404
405
406
407
408







-
+







                dToken["aTags"] = self.dTokenPos[dToken["nStart"]]["aTags"]
        self.lTokens = lNewTokens
        self.dTokenPos = { dToken["nStart"]: dToken  for dToken in self.lTokens  if dToken["sType"] != "INFO" }
        if bDebug:
            echo("UPDATE:")
            echo(self)

    def _getMatches (self, dGraph, dToken, dNode, bKeep=False):
    def _getNextNodes (self, dGraph, dToken, dNode, bKeep=False):
        "generator: return matches where <dToken> “values” match <dNode> arcs"
        bTokenFound = False
        # token value
        if dToken["sValue"] in dNode:
            yield (" ", dToken["sValue"], dNode[dToken["sValue"]])
            bTokenFound = True
        if dToken["sValue"][0:2].istitle(): # we test only 2 first chars, to match words such as "Laissez-les", "Crève-cœur".
552
553
554
555
556
557
558
559

560
561
562
563

564
565
566
567
568
569
570


571
572
573

574
575
576
577
578



579
580
581
582
583
584




585
586
587


588
589
590
591
592



593
594

595
596
597
598
599
600
601
552
553
554
555
556
557
558

559
560
561
562

563
564
565
566
567
568


569
570
571
572

573
574
575



576
577
578
579
580




581
582
583
584
585


586
587
588
589



590
591
592
593

594
595
596
597
598
599
600
601







-
+



-
+





-
-
+
+


-
+


-
-
-
+
+
+


-
-
-
-
+
+
+
+

-
-
+
+


-
-
-
+
+
+

-
+







                        yield ("*", sMeta, dNode["<meta>"][sMeta])
                        bTokenFound = True
        if not bTokenFound and bKeep:
            yield (None, "", -1)
        # JUMP
        # Warning! Recursion!
        if "<>" in dNode:
            yield from self._getMatches(dGraph, dToken, dGraph[dNode["<>"]], bKeep=True)
            yield from self._getNextNodes(dGraph, dToken, dGraph[dNode["<>"]], bKeep=True)

    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 = []
        lPointers = []
        bTagAndRewrite = False
        for iToken, dToken in enumerate(self.lTokens):
            if bDebug:
                echo("TOKEN: " + dToken["sValue"])
            # check arcs for each existing pointer
            lNextPointer = []
            for dPointer in lPointer:
            lNextPointers = []
            for dPointer in lPointers:
                if dPointer["nMultiEnd"] != -1:
                    if dToken["i"] <= dPointer["nMultiEnd"]:
                        lNextPointer.append(dPointer)
                        lNextPointers.append(dPointer)
                    if dToken["i"] != dPointer["nMultiEnd"]:
                        continue
                for cActionType, sMatch, iNode in self._getMatches(dGraph, dToken, dGraph[dPointer["iNode"]]):
                    if cActionType is None:
                        lNextPointer.append(dPointer)
                for cNodeType, sMatch, iNode in self._getNextNodes(dGraph, dToken, dGraph[dPointer["iNode"]]):
                    if cNodeType is None:
                        lNextPointers.append(dPointer)
                        continue
                    if bDebug:
                        echo("  MATCH: " + cActionType + sMatch)
                    nMultiEnd = -1  if cActionType != "&"  else dToken["nMultiStartTo"]
                    lNextPointer.append({ "iToken1": dPointer["iToken1"], "iNode": iNode, "nMultiEnd": nMultiEnd })
            lPointer = lNextPointer
                        echo("  MATCH: " + cNodeType + sMatch)
                    nMultiEnd = -1  if cNodeType != "&"  else dToken["nMultiStartTo"]
                    lNextPointers.append({ "iToken1": dPointer["iToken1"], "iNode": iNode, "nMultiEnd": nMultiEnd })
            lPointers = lNextPointers
            # check arcs of first nodes
            for cActionType, sMatch, iNode in self._getMatches(dGraph, dToken, dGraph[0]):
                if cActionType is None:
            for cNodeType, sMatch, iNode in self._getNextNodes(dGraph, dToken, dGraph[0]):
                if cNodeType is None:
                    continue
                if bDebug:
                    echo("  MATCH: " + cActionType + sMatch)
                nMultiEnd = -1  if cActionType != "&"  else dToken["nMultiStartTo"]
                lPointer.append({ "iToken1": iToken, "iNode": iNode, "nMultiEnd": nMultiEnd })
                    echo("  MATCH: " + cNodeType + sMatch)
                nMultiEnd = -1  if cNodeType != "&"  else dToken["nMultiStartTo"]
                lPointers.append({ "iToken1": iToken, "iNode": iNode, "nMultiEnd": nMultiEnd })
            # check if there is rules to check for each pointer
            for dPointer in lPointer:
            for dPointer in lPointers:
                if dPointer["nMultiEnd"] != -1:
                    if dToken["i"] < dPointer["nMultiEnd"]:
                        continue
                    if dToken["i"] == dPointer["nMultiEnd"]:
                        dPointer["nMultiEnd"] = -1
                if "<rules>" in dGraph[dPointer["iNode"]]:
                    bChange = self._executeActions(dGraph, dGraph[dPointer["iNode"]]["<rules>"], dPointer["iToken1"]-1, iToken, dOptions, sCountry, bShowRuleId, bDebug, bContext)
612
613
614
615
616
617
618
619

620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639

640
641
642
643
644
645
646
647
648
649

650
651
652

653
654
655

656
657

658
659
660
661
662
663
664
665
666
667
668

669
670

671
672
673


674
675
676
677
678
679
680
681
682
683
684
685

686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705

706
707
708
709
710
711
712
612
613
614
615
616
617
618

619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638

639
640
641
642
643
644
645
646
647
648

649
650
651

652
653
654

655
656

657
658
659
660
661
662
663
664
665
666
667

668
669

670
671


672
673
674
675
676
677
678
679
680
681
682
683
684

685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704

705
706
707
708
709
710
711
712







-
+



















-
+









-
+


-
+


-
+

-
+










-
+

-
+

-
-
+
+











-
+



















-
+







        bChange = False
        for sLineId, nextNodeKey in dNode.items():
            bCondMemo = None
            for sRuleId in dGraph[nextNodeKey]:
                try:
                    if bDebug:
                        echo("   >TRY: " + sRuleId + " " + sLineId)
                    _, sOption, sFuncCond, cActionType, sWhat, *eAct = _rules_graph.dRule[sRuleId]
                    _, sOption, sFuncCond, cActionType, sAction, *eAct = _rules_graph.dRule[sRuleId]
                    # Suggestion    [ option, condition, "-", replacement/suggestion/action, iTokenStart, iTokenEnd, cStartLimit, cEndLimit, bCaseSvty, nPriority, sMessage, iURL ]
                    # TextProcessor [ option, condition, "~", replacement/suggestion/action, iTokenStart, iTokenEnd, bCaseSvty ]
                    # Disambiguator [ option, condition, "=", replacement/suggestion/action ]
                    # Tag           [ option, condition, "/", replacement/suggestion/action, iTokenStart, iTokenEnd ]
                    # Immunity      [ option, condition, "!", option,                        iTokenStart, iTokenEnd ]
                    # Multi-token   [ option, condition, "&", morphologies,                  iTokenStart, iTokenEnd ]
                    # Test          [ option, condition, ">", "" ]
                    if not sOption or dOptions.get(sOption, False):
                        bCondMemo = not sFuncCond or getattr(gc_functions, sFuncCond)(self.lTokens, nTokenOffset, nLastToken, sCountry, bCondMemo, self.dTags, self.sSentence, self.sSentence0)
                        if bCondMemo:
                            if cActionType == "-":
                                # grammar error
                                iTokenStart, iTokenEnd, cStartLimit, cEndLimit, bCaseSvty, nPriority, sMessage, iURL = eAct
                                nTokenErrorStart = nTokenOffset + iTokenStart  if iTokenStart > 0  else nLastToken + iTokenStart
                                if "sImmunity" not in self.lTokens[nTokenErrorStart] or (self.lTokens[nTokenErrorStart]["sImmunity"] != "*" and sOption not in self.lTokens[nTokenErrorStart]["sImmunity"]):
                                    nTokenErrorEnd = nTokenOffset + iTokenEnd  if iTokenEnd > 0  else nLastToken + iTokenEnd
                                    nErrorStart = self.nOffsetWithinParagraph + (self.lTokens[nTokenErrorStart]["nStart"] if cStartLimit == "<"  else self.lTokens[nTokenErrorStart]["nEnd"])
                                    nErrorEnd = self.nOffsetWithinParagraph + (self.lTokens[nTokenErrorEnd]["nEnd"] if cEndLimit == ">"  else self.lTokens[nTokenErrorEnd]["nStart"])
                                    if nErrorStart not in self.dError or nPriority > self.dErrorPriority.get(nErrorStart, -1):
                                        self.dError[nErrorStart] = self._createErrorFromTokens(sWhat, nTokenOffset, nLastToken, nTokenErrorStart, nErrorStart, nErrorEnd, sLineId, sRuleId, bCaseSvty, \
                                        self.dError[nErrorStart] = self._createErrorFromTokens(sAction, nTokenOffset, nLastToken, nTokenErrorStart, nErrorStart, nErrorEnd, sLineId, sRuleId, bCaseSvty, \
                                                                                               sMessage, _rules_graph.dURL.get(iURL, ""), bShowRuleId, sOption, bContext)
                                        self.dErrorPriority[nErrorStart] = nPriority
                                        self.dSentenceError[nErrorStart] = self.dError[nErrorStart]
                                        if bDebug:
                                            echo("    NEW_ERROR: {}".format(self.dError[nErrorStart]))
                            elif cActionType == "~":
                                # text processor
                                nTokenStart = nTokenOffset + eAct[0]  if eAct[0] > 0  else nLastToken + eAct[0]
                                nTokenEnd = nTokenOffset + eAct[1]  if eAct[1] > 0  else nLastToken + eAct[1]
                                self._tagAndPrepareTokenForRewriting(sWhat, nTokenStart, nTokenEnd, nTokenOffset, nLastToken, eAct[2], bDebug)
                                self._tagAndPrepareTokenForRewriting(sAction, nTokenStart, nTokenEnd, nTokenOffset, nLastToken, eAct[2], bDebug)
                                bChange = True
                                if bDebug:
                                    echo("    TEXT_PROCESSOR: [{}:{}]  > {}".format(self.lTokens[nTokenStart]["sValue"], self.lTokens[nTokenEnd]["sValue"], sWhat))
                                    echo("    TEXT_PROCESSOR: [{}:{}]  > {}".format(self.lTokens[nTokenStart]["sValue"], self.lTokens[nTokenEnd]["sValue"], sAction))
                            elif cActionType == "=":
                                # disambiguation
                                getattr(gc_functions, sWhat)(self.lTokens, nTokenOffset, nLastToken)
                                getattr(gc_functions, sAction)(self.lTokens, nTokenOffset, nLastToken)
                                if bDebug:
                                    echo("    DISAMBIGUATOR: ({})  [{}:{}]".format(sWhat, self.lTokens[nTokenOffset+1]["sValue"], self.lTokens[nLastToken]["sValue"]))
                                    echo("    DISAMBIGUATOR: ({})  [{}:{}]".format(sAction, self.lTokens[nTokenOffset+1]["sValue"], self.lTokens[nLastToken]["sValue"]))
                            elif cActionType == ">":
                                # we do nothing, this test is just a condition to apply all following actions
                                if bDebug:
                                    echo("    COND_OK")
                            elif cActionType == "/":
                                # Tag
                                nTokenStart = nTokenOffset + eAct[0]  if eAct[0] > 0  else nLastToken + eAct[0]
                                nTokenEnd = nTokenOffset + eAct[1]  if eAct[1] > 0  else nLastToken + eAct[1]
                                for i in range(nTokenStart, nTokenEnd+1):
                                    if "aTags" in self.lTokens[i]:
                                        self.lTokens[i]["aTags"].update(sWhat.split("|"))
                                        self.lTokens[i]["aTags"].update(sAction.split("|"))
                                    else:
                                        self.lTokens[i]["aTags"] = set(sWhat.split("|"))
                                        self.lTokens[i]["aTags"] = set(sAction.split("|"))
                                if bDebug:
                                    echo("    TAG: {} >  [{}:{}]".format(sWhat, self.lTokens[nTokenStart]["sValue"], self.lTokens[nTokenEnd]["sValue"]))
                                for sTag in sWhat.split("|"):
                                    echo("    TAG: {} >  [{}:{}]".format(sAction, self.lTokens[nTokenStart]["sValue"], self.lTokens[nTokenEnd]["sValue"]))
                                for sTag in sAction.split("|"):
                                    if sTag not in self.dTags:
                                        self.dTags[sTag] = [nTokenStart, nTokenEnd]
                                    else:
                                        self.dTags[sTag][0] = min(nTokenStart, self.dTags[sTag][0])
                                        self.dTags[sTag][1] = max(nTokenEnd, self.dTags[sTag][1])
                            elif cActionType == "!":
                                # immunity
                                if bDebug:
                                    echo("    IMMUNITY: " + sLineId + " / " + sRuleId)
                                nTokenStart = nTokenOffset + eAct[0]  if eAct[0] > 0  else nLastToken + eAct[0]
                                nTokenEnd = nTokenOffset + eAct[1]  if eAct[1] > 0  else nLastToken + eAct[1]
                                sImmunity = sWhat or "*"
                                sImmunity = sAction or "*"
                                if nTokenEnd - nTokenStart == 0:
                                    self.lTokens[nTokenStart]["sImmunity"] = sImmunity
                                    nErrorStart = self.nOffsetWithinParagraph + self.lTokens[nTokenStart]["nStart"]
                                    if nErrorStart in self.dError:
                                        del self.dError[nErrorStart]
                                else:
                                    for i in range(nTokenStart, nTokenEnd+1):
                                        self.lTokens[i]["sImmunity"] = sImmunity
                                        nErrorStart = self.nOffsetWithinParagraph + self.lTokens[i]["nStart"]
                                        if nErrorStart in self.dError:
                                            del self.dError[nErrorStart]
                            elif cActionType == "&":
                                # multi-tokens
                                nTokenStart = nTokenOffset + eAct[0]  if eAct[0] > 0  else nLastToken + eAct[0]
                                nTokenEnd = nTokenOffset + eAct[1]  if eAct[1] > 0  else nLastToken + eAct[1]
                                dMultiToken = {
                                    "nTokenStart": nTokenStart,
                                    "nTokenEnd": nTokenEnd,
                                    "lTokens": self.lTokens[nTokenStart:nTokenEnd+1],
                                    "lMorph": sWhat.split("|")  if sWhat else  [":HM"]
                                    "lMorph": sAction.split("|")  if sAction else  [":HM"]
                                }
                                self.lTokens[nTokenStart]["nMultiStartTo"] = nTokenEnd
                                self.lTokens[nTokenEnd]["nMultiEndFrom"] = nTokenStart
                                self.lTokens[nTokenStart]["dMultiToken"] = dMultiToken
                                self.lTokens[nTokenEnd]["dMultiToken"] = dMultiToken
                                print(dMultiToken)
                            else:
822
823
824
825
826
827
828
829

830
831

832
833
834
835
836
837
838

839
840
841
842


843
844
845


846
847
848
849
850
851
852
853
854


855
856

857
858
859
860
861
862


863
864
865

866
867
868
869
870
871
872
822
823
824
825
826
827
828

829
830

831
832
833
834
835
836
837

838
839
840


841
842
843


844
845
846
847
848
849
850
851
852


853
854
855

856
857
858
859
860


861
862
863
864

865
866
867
868
869
870
871
872







-
+

-
+






-
+


-
-
+
+

-
-
+
+







-
-
+
+

-
+




-
-
+
+


-
+







            if bUppercase and m.group(iGroup)[0:1].isupper():
                sNew = sNew.capitalize()
        else:
            sNew = m.expand(sRepl)
            sNew = sNew + " " * (nLen-len(sNew))
        return sText[0:m.start(iGroup)] + sNew + sText[m.end(iGroup):]

    def _tagAndPrepareTokenForRewriting (self, sWhat, nTokenRewriteStart, nTokenRewriteEnd, nTokenOffset, nLastToken, bCaseSvty, bDebug):
    def _tagAndPrepareTokenForRewriting (self, sAction, nTokenRewriteStart, nTokenRewriteEnd, nTokenOffset, nLastToken, bCaseSvty, bDebug):
        "text processor: rewrite tokens between <nTokenRewriteStart> and <nTokenRewriteEnd> position"
        if sWhat == "*":
        if sAction == "*":
            # purge text
            if nTokenRewriteEnd - nTokenRewriteStart == 0:
                self.lTokens[nTokenRewriteStart]["bToRemove"] = True
            else:
                for i in range(nTokenRewriteStart, nTokenRewriteEnd+1):
                    self.lTokens[i]["bToRemove"] = True
        elif sWhat == "␣":
        elif sAction == "␣":
            # merge tokens
            self.lTokens[nTokenRewriteStart]["nMergeUntil"] = nTokenRewriteEnd
        elif sWhat.startswith("␣"):
            sWhat = self._expand(sWhat, nTokenOffset, nLastToken)
        elif sAction.startswith("␣"):
            sAction = self._expand(sAction, nTokenOffset, nLastToken)
            self.lTokens[nTokenRewriteStart]["nMergeUntil"] = nTokenRewriteEnd
            self.lTokens[nTokenRewriteStart]["sMergedValue"] = sWhat[1:]
        elif sWhat == "_":
            self.lTokens[nTokenRewriteStart]["sMergedValue"] = sAction[1:]
        elif sAction == "_":
            # neutralized token
            if nTokenRewriteEnd - nTokenRewriteStart == 0:
                self.lTokens[nTokenRewriteStart]["sNewValue"] = "_"
            else:
                for i in range(nTokenRewriteStart, nTokenRewriteEnd+1):
                    self.lTokens[i]["sNewValue"] = "_"
        else:
            if sWhat.startswith("="):
                sWhat = getattr(gc_functions, sWhat[1:])(self.lTokens, nTokenOffset, nLastToken)
            if sAction.startswith("="):
                sAction = getattr(gc_functions, sAction[1:])(self.lTokens, nTokenOffset, nLastToken)
            else:
                sWhat = self._expand(sWhat, nTokenOffset, nLastToken)
                sAction = self._expand(sAction, nTokenOffset, nLastToken)
            bUppercase = bCaseSvty and self.lTokens[nTokenRewriteStart]["sValue"][0:1].isupper()
            if nTokenRewriteEnd - nTokenRewriteStart == 0:
                # one token
                if bUppercase:
                    sWhat = sWhat[0:1].upper() + sWhat[1:]
                self.lTokens[nTokenRewriteStart]["sNewValue"] = sWhat
                    sAction = sAction[0:1].upper() + sAction[1:]
                self.lTokens[nTokenRewriteStart]["sNewValue"] = sAction
            else:
                # several tokens
                lTokenValue = sWhat.split("|")
                lTokenValue = sAction.split("|")
                if len(lTokenValue) != (nTokenRewriteEnd - nTokenRewriteStart + 1):
                    if bDebug:
                        echo("Error. Text processor: number of replacements != number of tokens.")
                    return
                for i, sValue in zip(range(nTokenRewriteStart, nTokenRewriteEnd+1), lTokenValue):
                    if not sValue or sValue == "*":
                        self.lTokens[i]["bToRemove"] = True