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
|
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
340
341
342
|
-
+
-
+
-
+
-
+
|
}
}
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;
|
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
|
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
643
644
645
646
647
648
649
650
651
652
653
|
-
+
-
+
-
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+
+
-
+
-
+
-
-
+
+
-
+
+
-
+
-
+
+
+
+
+
+
+
+
+
|
}
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) {
for (let [cActionType, sMatch, iNode] of this._getMatches(oGraph, oToken, oGraph[oPointer["iNode"]])) {
if (cActionType === null) {
lNextPointer.push(oPointer);
let lNextPointers = [];
for (let oPointer of lPointers) {
if (oPointer["nMultiEnd"] != -1) {
if (oToken["i"] <= oPointer["nMultiEnd"]) {
lNextPointers.push(oPointer);
}
if (oToken["i"] != oPointer["nMultiEnd"]) {
continue;
}
}
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 = (cNodeType != "&") ? -1 : dToken["nMultiStartTo"];
lNextPointer.push({ "iToken1": oPointer["iToken1"], "iNode": iNode });
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 = (cNodeType != "&") ? -1 : dToken["nMultiStartTo"];
lPointer.push({ "iToken1": iToken, "iNode": iNode });
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;
}
}
if (oGraph[oPointer["iNode"]].hasOwnProperty("<rules>")) {
let bChange = this._executeActions(oGraph, oGraph[oPointer["iNode"]]["<rules>"], oPointer["iToken1"]-1, iToken, dOptions, sCountry, bShowRuleId, bDebug, bContext);
if (bChange) {
bTagAndRewrite = true;
}
}
}
|
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
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
|
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
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
|
-
+
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+
|
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 {
for (let i = nTokenStart; i <= nTokenEnd; i++) {
this.lTokens[i]["sImmunity"] = sImmunity;
let nErrorStart = this.nOffsetWithinParagraph + this.lTokens[i]["nStart"];
if (this.dError.has(nErrorStart)) {
this.dError.delete(nErrorStart);
}
}
}
}
else if (cActionType == "&") {
// 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": (sAction) ? sAction.split("|") : [":HM"]
}
this.lTokens[nTokenStart]["nMultiStartTo"] = this.lTokens[nTokenEnd]["i"];
this.lTokens[nTokenEnd]["nMultiEndFrom"] = this.lTokens[nTokenStart]["i"];
this.lTokens[nTokenStart]["oMultiToken"] = oMultiToken;
this.lTokens[nTokenEnd]["oMultiToken"] = oMultiToken;
if (bDebug) {
console.log(`" MULTI-TOKEN: ${sAction} [${this.lTokens[nTokenStart]["sValue"]}:${this.lTokens[nTokenEnd]["sValue"]}]`);
}
}
} else {
|
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
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
|
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
971
972
973
974
975
976
|
-
+
-
+
-
+
-
-
+
+
-
+
-
+
-
-
+
+
-
+
-
+
-
+
-
+
|
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;
|