Grammalecte  Diff

Annotate Patch Unified Diff

Differences From Artifact [94bb3a1779]:

To Artifact [841b4146b0]:

︙︙
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444



445
446
447
448
449
450
451

        // préfixe élidé
        let m = /^([ldmtsnjcç]|lorsqu|presqu|jusqu|puisqu|quoiqu|quelqu|qu)[’'‘`ʼ]([a-zA-Zà-öÀ-Ö0-9_ø-ÿØ-ßĀ-ʯfi-st-]+)/i.exec(sWord);
        if (m) {
            sPrefix = m[1] + "’";
            sWord = m[2];
        }
        // mots composés
        m = /^([a-zA-Zà-öÀ-Ö0-9_ø-ÿØ-ßĀ-ʯfi-st-]+)(-(?:(?:les?|la)-(?:moi|toi|lui|[nv]ous|leur)|t-(?:il|elle|on)|y|en|[mts]’(?:y|en)|les?|l[aà]|[mt]oi|leur|lui|je|tu|ils?|elles?|on|[nv]ous|ce))$/i.exec(sWord);
        if (m) {
            sWord = m[1];
            sSuffix = m[2];
        }
        // split word in 3 parts: prefix, root, suffix
        return [sPrefix, sWord, sSuffix];
    },

    analyze: function (sWord) {
        // return meaning of <sWord> if found else an empty string
        sWord = sWord.toLowerCase();
        if (this.dValues.has(sWord)) {
            return this.dValues.get(sWord);
        }
        return "";
    },

    readableMorph: function (sMorph) {



        let sRes = "";
        sMorph = sMorph.replace(/:V([0-3][ea_])[itpqnmr_eaxz]+/, ":V$1");
        let m;
        while ((m = this._zTag.exec(sMorph)) !== null) {
            if (this.dTag.has(m[0])) {
                sRes += this.dTag.get(m[0])[0];
            } else {








|


















>
>
>








419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454

        // préfixe élidé
        let m = /^([ldmtsnjcç]|lorsqu|presqu|jusqu|puisqu|quoiqu|quelqu|qu)[’'‘`ʼ]([a-zA-Zà-öÀ-Ö0-9_ø-ÿØ-ßĀ-ʯfi-st-]+)/i.exec(sWord);
        if (m) {
            sPrefix = m[1] + "’";
            sWord = m[2];
        }
        // mots composés
        m = /^([a-zA-Zà-öÀ-Ö0-9_ø-ÿØ-ßĀ-ʯfi-st]+)(-(?:(?:les?|la)-(?:moi|toi|lui|[nv]ous|leur)|t-(?:il|elle|on)|y|en|[mts]’(?:y|en)|les?|l[aà]|[mt]oi|leur|lui|je|tu|ils?|elles?|on|[nv]ous|ce))$/i.exec(sWord);
        if (m) {
            sWord = m[1];
            sSuffix = m[2];
        }
        // split word in 3 parts: prefix, root, suffix
        return [sPrefix, sWord, sSuffix];
    },

    analyze: function (sWord) {
        // return meaning of <sWord> if found else an empty string
        sWord = sWord.toLowerCase();
        if (this.dValues.has(sWord)) {
            return this.dValues.get(sWord);
        }
        return "";
    },

    readableMorph: function (sMorph) {
        if (!sMorph) {
            return " mot inconnu";
        }
        let sRes = "";
        sMorph = sMorph.replace(/:V([0-3][ea_])[itpqnmr_eaxz]+/, ":V$1");
        let m;
        while ((m = this._zTag.exec(sMorph)) !== null) {
            if (this.dTag.has(m[0])) {
                sRes += this.dTag.get(m[0])[0];
            } else {
︙︙
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546













547
548
549
550
551
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
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
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

                case 'WORD':
                    if (oToken.hasOwnProperty("lMorph")  &&  oToken["lMorph"].length > 0) {
                        // with morphology
                        oToken["aLabels"] = [];
                        for (let sMorph of oToken["lMorph"]) {
                            oToken["aLabels"].push(this.readableMorph(sMorph));
                        }
                        if (oToken.hasOwnProperty("sTags")) {
                            let aTags = [];
                            for (let sTag of oToken["sTags"]) {
                                if (this.dValues.has(sTag)) {
                                    aTags.push(this.dValues.get(sTag))
                                }
                            }
                            if (aTags.length > 0) {
                                oToken["aOtherLabels"] = aTags;
                            }
                        }
                    } else {
                        // no morphology, guessing
                        if (oToken["sValue"].gl_count("-") > 4) {
                            oToken["aLabels"] = ["élément complexe indéterminé"];
                        }
                        else if (m = this._zPartDemForm.exec(oToken["sValue"])) {
                            // mots avec particules démonstratives
                            oToken["aLabels"] = ["mot avec particule démonstrative"];
                        }
                        else if (m = this._zImperatifVerb.exec(oToken["sValue"])) {
                            // formes interrogatives
                            oToken["aLabels"] = ["forme verbale impérative"];
                        }
                        else if (m = this._zInterroVerb.exec(oToken["sValue"])) {
                            // formes interrogatives
                            oToken["aLabels"] = ["forme verbale interrogative"];
                        }
                        else {
                            oToken["aLabels"] = ["mot inconnu du dictionnaire"];
                        }













                    }
                    break;
                default:
                    oToken["aLabels"] = ["token de nature inconnue"];
            }
        } catch (e) {
            console.error(e);
        }
    },

    getInfoForToken: function (oToken) {
        // Token: .sType, .sValue, .nStart, .nEnd
        // return a object {sType, sValue, aLabel}
        let m = null;
        try {
            switch (oToken.sType) {
                case 'PUNC':
                case 'SIGN':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: [this.dValues.gl_get(oToken.sValue, "caractère indéterminé")]
                    };
                    break;
                case 'NUM':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["nombre"]
                    };
                    break;
                case 'LINK':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabel: ["hyperlien"]
                    };
                    break;
                case 'TAG':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["étiquette (hashtag)"]
                    };
                    break;
                case 'HTML':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabel: ["balise HTML"]
                    };
                    break;
                case 'PSEUDOHTML':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["balise pseudo-HTML"]
                    };
                    break;
                case 'HTMLENTITY':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["entité caractère XML/HTML"]
                    };
                    break;
                case 'HOUR':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["heure"]
                    };
                    break;
                case 'WORD_ELIDED':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: [this.dValues.gl_get(oToken.sValue.toLowerCase(), "préfixe élidé inconnu")]
                    };
                    break;
                case 'WORD_ORDINAL':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["nombre ordinal"]
                    };
                    break;
                case 'FOLDERUNIX':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabel: ["dossier UNIX (et dérivés)"]
                    };
                    break;
                case 'FOLDERWIN':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabel: ["dossier Windows"]
                    };
                    break;
                case 'WORD_ACRONYM':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["Sigle ou acronyme"]
                    };
                    break;
                case 'WORD':
                    if (oToken.sValue.gl_count("-") > 4) {
                        return {
                            sType: "COMPLEX",
                            sValue: oToken.sValue,
                            aLabel: ["élément complexe indéterminé"]
                        };
                    } else if (m = this._zPartDemForm.exec(oToken.sValue)) {
                        // mots avec particules démonstratives
                        if (this._aPartDemExceptList.has(m[1].toLowerCase())) {
                            return {
                                sType: "WORD",
                                sValue: oToken.sValue,
                                aLabel: this._getMorph(oToken.sValue)
                            };
                        }
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabel: ["mot avec particule démonstrative"],
                            aSubElem: [
                                { sType: oToken.sType, sValue: m[1], aLabel: this._getMorph(m[1]) },
                                { sType: oToken.sType, sValue: m[2], aLabel: [ this._formatSuffix(m[2]) ] }
                            ]
                        };
                    } else if (m = this._zImperatifVerb.exec(oToken.sValue)) {
                        // formes interrogatives
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabel: ["forme verbale impérative"],
                            aSubElem: [
                                { sType: oToken.sType, sValue: m[1], aLabel: this._getMorph(m[1]) },
                                { sType: oToken.sType, sValue: m[2], aLabel: [ this._formatSuffix(m[2]) ] }
                            ]
                        };
                    } else if (m = this._zInterroVerb.exec(oToken.sValue)) {
                        // formes interrogatives
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabel: ["forme verbale interrogative"],
                            aSubElem: [
                                { sType: oToken.sType, sValue: m[1], aLabel: this._getMorph(m[1]) },
                                { sType: oToken.sType, sValue: m[2], aLabel: [ this._formatSuffix(m[2]) ] }
                            ]
                        };
                    } else if (this.oSpellChecker.isValidToken(oToken.sValue)) {
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabel: this._getMorph(oToken.sValue)
                        };
                    } else {
                        return {
                            sType: "UNKNOWN_WORD",
                            sValue: oToken.sValue,
                            aLabel: ["mot inconnu du dictionnaire"]
                        };
                    }
                    break;
                default:
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabel: ["token inconnu"]
                    }
            }
        } catch (e) {
            console.error(e);
        }
        return null;
    },








<
<
<
<
<
<
<
<
<
<
<




















>
>
>
>
>
>
>
>
>
>
>
>
>












|








|






|






|






|






|






|






|






|






|






|






|






|






|







|







|





|
|
|
|







|
|
|
|







|
|
|
|






|





|







|








512
513
514
515
516
517
518











519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
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
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
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

                case 'WORD':
                    if (oToken.hasOwnProperty("lMorph")  &&  oToken["lMorph"].length > 0) {
                        // with morphology
                        oToken["aLabels"] = [];
                        for (let sMorph of oToken["lMorph"]) {
                            oToken["aLabels"].push(this.readableMorph(sMorph));
                        }











                    } else {
                        // no morphology, guessing
                        if (oToken["sValue"].gl_count("-") > 4) {
                            oToken["aLabels"] = ["élément complexe indéterminé"];
                        }
                        else if (m = this._zPartDemForm.exec(oToken["sValue"])) {
                            // mots avec particules démonstratives
                            oToken["aLabels"] = ["mot avec particule démonstrative"];
                        }
                        else if (m = this._zImperatifVerb.exec(oToken["sValue"])) {
                            // formes interrogatives
                            oToken["aLabels"] = ["forme verbale impérative"];
                        }
                        else if (m = this._zInterroVerb.exec(oToken["sValue"])) {
                            // formes interrogatives
                            oToken["aLabels"] = ["forme verbale interrogative"];
                        }
                        else {
                            oToken["aLabels"] = ["mot inconnu du dictionnaire"];
                        }
                    }
                    if (oToken.hasOwnProperty("lSubTokens")) {
                        for (let oSubToken of oToken["lSubTokens"]) {
                            if (oSubToken["sValue"]) {
                                if (this.dValues.has(oSubToken["sValue"])) {
                                    oSubToken["lMorph"] = [ "" ];
                                    oSubToken["aLabels"] = [ this.dValues.get(oSubToken["sValue"]) ];
                                }
                                else {
                                    oSubToken["aLabels"] = oSubToken["lMorph"].map((sMorph) => this.readableMorph(sMorph));
                                }
                            }
                        }
                    }
                    break;
                default:
                    oToken["aLabels"] = ["token de nature inconnue"];
            }
        } catch (e) {
            console.error(e);
        }
    },

    getInfoForToken: function (oToken) {
        // Token: .sType, .sValue, .nStart, .nEnd
        // return a object {sType, sValue, aLabels}
        let m = null;
        try {
            switch (oToken.sType) {
                case 'PUNC':
                case 'SIGN':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: [this.dValues.gl_get(oToken.sValue, "caractère indéterminé")]
                    };
                    break;
                case 'NUM':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["nombre"]
                    };
                    break;
                case 'LINK':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabels: ["hyperlien"]
                    };
                    break;
                case 'TAG':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["étiquette (hashtag)"]
                    };
                    break;
                case 'HTML':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabels: ["balise HTML"]
                    };
                    break;
                case 'PSEUDOHTML':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["balise pseudo-HTML"]
                    };
                    break;
                case 'HTMLENTITY':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["entité caractère XML/HTML"]
                    };
                    break;
                case 'HOUR':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["heure"]
                    };
                    break;
                case 'WORD_ELIDED':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: [this.dValues.gl_get(oToken.sValue.toLowerCase(), "préfixe élidé inconnu")]
                    };
                    break;
                case 'WORD_ORDINAL':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["nombre ordinal"]
                    };
                    break;
                case 'FOLDERUNIX':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabels: ["dossier UNIX (et dérivés)"]
                    };
                    break;
                case 'FOLDERWIN':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue.slice(0, 40) + "…",
                        aLabels: ["dossier Windows"]
                    };
                    break;
                case 'WORD_ACRONYM':
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["Sigle ou acronyme"]
                    };
                    break;
                case 'WORD':
                    if (oToken.sValue.gl_count("-") > 4) {
                        return {
                            sType: "COMPLEX",
                            sValue: oToken.sValue,
                            aLabels: ["élément complexe indéterminé"]
                        };
                    } else if (m = this._zPartDemForm.exec(oToken.sValue)) {
                        // mots avec particules démonstratives
                        if (this._aPartDemExceptList.has(m[1].toLowerCase())) {
                            return {
                                sType: "WORD",
                                sValue: oToken.sValue,
                                aLabels: this._getMorph(oToken.sValue)
                            };
                        }
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabels: ["mot avec particule démonstrative"],
                            lSubTokens: [
                                { sType: oToken.sType, sValue: m[1], aLabels: this._getMorph(m[1]) },
                                { sType: oToken.sType, sValue: m[2], aLabels: [ this._formatSuffix(m[2]) ] }
                            ]
                        };
                    } else if (m = this._zImperatifVerb.exec(oToken.sValue)) {
                        // formes interrogatives
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabels: ["forme verbale impérative"],
                            lSubTokens: [
                                { sType: oToken.sType, sValue: m[1], aLabels: this._getMorph(m[1]) },
                                { sType: oToken.sType, sValue: m[2], aLabels: [ this._formatSuffix(m[2]) ] }
                            ]
                        };
                    } else if (m = this._zInterroVerb.exec(oToken.sValue)) {
                        // formes interrogatives
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabels: ["forme verbale interrogative"],
                            lSubTokens: [
                                { sType: oToken.sType, sValue: m[1], aLabels: this._getMorph(m[1]) },
                                { sType: oToken.sType, sValue: m[2], aLabels: [ this._formatSuffix(m[2]) ] }
                            ]
                        };
                    } else if (this.oSpellChecker.isValidToken(oToken.sValue)) {
                        return {
                            sType: oToken.sType,
                            sValue: oToken.sValue,
                            aLabels: this._getMorph(oToken.sValue)
                        };
                    } else {
                        return {
                            sType: "UNKNOWN_WORD",
                            sValue: oToken.sValue,
                            aLabels: ["mot inconnu du dictionnaire"]
                        };
                    }
                    break;
                default:
                    return {
                        sType: oToken.sType,
                        sValue: oToken.sValue,
                        aLabels: ["token inconnu"]
                    }
            }
        } catch (e) {
            console.error(e);
        }
        return null;
    },
︙︙
Fossil 2.20 [210e89a059] 2022-11-16 18:46:32
This page was generated in about 0.005s