コード例 #1
0
        public IEnumerable <AutoGlosserNote> Gloss(string inputText)
        {
            var words = parser.BreakIntoWords(inputText)
                        .ToList();

            var glosses = new List <AutoGlosserNote>();

            for (int i = 0; i < words.Count; i++)
            {
                var word            = words[i];
                var greedySelection = words.Skip(i).Greedy(wordInfos =>
                {
                    var entireExpression = string.Concat(wordInfos.Select(w => w.RawWord));
                    var l = dictLookup.Lookup(entireExpression);
                    if (l == null)
                    {
                        return(false);
                    }

                    var entireReading = kana.ToHiragana(string.Concat(wordInfos.Select(w => w.Reading)));
                    if (l.Any(e => e.ReadingEntries.Any(r => entireReading == kana.ToHiragana(r.Reading))))
                    {
                        return(true);
                    }

                    return(false);
                }).ToList();
                var lookup = dictLookup.Lookup(word.DictionaryForm ?? word.RawWord)?.ToList();

                if (word.RawWord.All(c => ".!??!⁉、…。.".IndexOf(c) != -1))
                {
                    // skip punctuation
                    continue;
                }

                if (greedySelection.Count > 1)
                {
                    var greedyLookup = dictLookup.Lookup(string.Concat(greedySelection.Select(w => w.RawWord))).Materialize();
                    glosses.Add(new AutoGlosserNote(
                                    string.Join(" ", greedySelection.Select(w => w.RawWord)),
                                    OrderSenses(FilterOutInapplicableSenses(greedyLookup, greedySelection)).Select(FormatSense)));

                    i += greedySelection.Count - 1; // -1 because iteration will result in one extra increase
                    continue;
                }
                else if (lookup != null)
                {
                    glosses.Add(new AutoGlosserNote(word.RawWord, OrderSenses(FilterOutInapplicableSenses(lookup, word)).Select(FormatSense)));
                }
                else
                {
                    glosses.Add(new AutoGlosserNote(word.RawWord, new string[0]));
                }
            }

            return(glosses);
        }
コード例 #2
0
 public Task <Option <RichFormatting> > Answer(Request request, CancellationToken token)
 {
     return(DictUtils.Lookup(
                request,
                t => jdict.Lookup(t),
                r => GreedyLookup(r),
                kana,
                Render));
 }
コード例 #3
0
        public IEnumerable <IEnumerable <WordInfo> > BreakIntoSentences(string input)
        {
            if (string.IsNullOrWhiteSpace(input))
            {
                yield break;
            }

            var entries = analyzer.ParseToEntries(input)
                          .Where(a => a.IsRegular);

            var list          = new List <WordInfo>();
            int previousIndex = 0;
            int currentIndex  = 0;

            foreach (var entry in entries)
            {
                previousIndex = currentIndex;
                currentIndex  = input.IndexOf(entry.SurfaceForm, currentIndex, StringComparison.Ordinal);
                var newlines = input.SubstringFromTo(previousIndex, currentIndex).ReplaceLineEndings("\n")
                               .Count(c => c == '\n');
                for (int i = 0; i < newlines; i++)
                {
                    yield return(list);

                    list = new List <WordInfo>();
                }
                list.Add(Map(entry));
            }

            if (list.Count != 0)
            {
                yield return(list);
            }

            WordInfo Map(IEntry word)
            {
                var reading = word.Reading ??
                              lookup.Lookup(word.DictionaryForm ?? word.SurfaceForm)?.FirstOrDefault()
                              ?.ReadingEntries.First().Reading;

                return(new WordInfo(
                           word.SurfaceForm,
                           word.PartOfSpeech,
                           word.DictionaryForm,
                           word.GetPartOfSpeechInfo().Contains(PartOfSpeechInfo.Pronoun)
                        ? Option.Some(EdictPartOfSpeech.pn)
                        : word.Type,
                           reading,
                           word is UnidicEntry unidicEntry
                        ? unidicEntry.DictionaryFormReading
                        : null));
            };
        }
コード例 #4
0
 public void LookupKana()
 {
     {
         var entries = jmdict.Lookup("みなみ");
         Assert.True(entries.Any(e => e.Senses.Any(s => s.Glosses.Contains("south"))));
         Assert.True(entries.Any(e => e.Senses.Any(s => s.PartOfSpeechInfo.Contains(EdictPartOfSpeech.n))));
     }
 }
コード例 #5
0
    public Task <Option <RichFormatting> > Answer(Request request, CancellationToken token)
    {
        var rich = new RichFormatting();

        var jmDictEntries = jdict.Lookup(request.NotInflected ?? request.QueryText)
                            ?? Enumerable.Empty <JMDictEntry>();

        if (request.Word.DictionaryFormReading != null)
        {
            var normalizedReading = kana.ToKatakana(request.Word.DictionaryFormReading);

            jmDictEntries = jmDictEntries
                            .OrderByDescending(entry =>
                                               entry.ReadingEntries
                                               .Select(readingEntry =>
                                                       kana.ToKatakana(readingEntry.Reading))
                                               .Contains(normalizedReading)
                        ? 1
                        : 0);
        }

        foreach (var jmDictEntry in jmDictEntries)
        {
            rich.Paragraphs.Add(new TextParagraph(
                                    new []
            {
                jmDictEntry.KanjiEntries.Select(kanjiEntry => new Text(kanjiEntry.Kanji, emphasis: true))
                .Intersperse(new Text("; ", emphasis: true)),
                jmDictEntry.ReadingEntries.Select(readingEntry => new Text($"{readingEntry.Reading}", emphasis: true))
                .Intersperse(new Text("; ", emphasis: true))
                .Prepend(new Text("【", emphasis: true))
                .Append(new Text("】", emphasis: true)),
                jmDictEntry.Senses.SelectMany((sense, position) =>
                                              sense.Glosses
                                              .Select(gloss => new Text(gloss))
                                              .Intersperse(new Text("/"))
                                              .Prepend(new Text($" ({position+1}) ", fontSize: FontSize.Small))
                                              .Prepend(new Text(" (" + string.Join(",", sense.PartOfSpeechInfo.Select(pos => pos.ToAbbrevation())) + ") ", fontSize: FontSize.Small)))
            }.SelectMany(x => x)));
        }

        if (rich.Paragraphs.Count == 0)
        {
            return(Task.FromResult(Option.None <RichFormatting>()));
        }

        return(Task.FromResult(Option.Some(rich)));
    }
コード例 #6
0
        public Task <Option <RichFormatting> > Answer(Request request, CancellationToken token)
        {
            var rich = new RichFormatting();

            if (!(request.PartOfSpeech == PartOfSpeech.Verb || request.PartOfSpeech == PartOfSpeech.Unknown))
            {
                rich.Paragraphs.Add(
                    new TextParagraph(
                        EnumerableExt.OfSingle(
                            new Text("The program estimates this word is not a verb. The results below may be garbage."))));
            }


            var verb    = request.NotInflected ?? request.Word.RawWord;
            var entries = jdict.Lookup(verb);

            if (entries == null)
            {
                rich.Paragraphs.Add(
                    new TextParagraph(
                        EnumerableExt.OfSingle(
                            new Text("No word found."))));
                return(Task.FromResult(Option.Some(rich)));
            }

            var verbTypes = entries.Select(e =>
            {
                if (!e.ReadingEntries.Any())
                {
                    return(Option.None <EdictType>());
                }
                return(GetEdictVerbType(e));
            })
                            .OfNonNone()
                            .Distinct()
                            .OrderByDescending(e => Option.Some((int)e) == request.Word.Type.Map(t => (int)t) ? 1 : 0)
                            .ToList();

            if (verbTypes.Count == 0)
            {
                rich.Paragraphs.Add(
                    new TextParagraph(
                        EnumerableExt.OfSingle(
                            new Text("No verb found."))));
                return(Task.FromResult(Option.Some(rich)));
            }
            else
            {
                foreach (var verbType in verbTypes)
                {
                    if (verbTypes.Count > 1)
                    {
                        rich.Paragraphs.Add(new TextParagraph(
                                                EnumerableExt.OfSingle(new Text(verb + ": " + LibJpConjSharp.EdictTypeUtils.ToLongString(verbType)))));
                    }
                    rich.Paragraphs.Add(new TextParagraph(new[]
                    {
                        Form("=", CForm.Present, verbType),
                        Form("@=", CForm.Present, verbType, Politeness.Polite),
                        Form("<", CForm.Past, verbType),
                        Form("?", CForm.Potential, verbType),
                        Form("#", CForm.Passive, verbType),
                        Form("->", CForm.Causative, verbType),
                        Form("if", CForm.Condition, verbType),
                        Form("Te", CForm.TeForm, verbType),
                        Form("!", CForm.Imperative, verbType),
                        Form(":D", CForm.Volitional, verbType),
                    }));
                }
            }

            rich.Paragraphs.Add(new TextParagraph(EnumerableExt.OfSingle(new Text(
                                                                             @"= - Present
< - Past
? - Potential
# - Passive
-> - Causative
if - Conditional
Te - Te Form
! - Imperative
:D - Volitional
~ - Negative form of any of those
@ - Polite form
"))));

            return(Task.FromResult(Option.Some(rich)));

            Text Form(string name, CForm form, EdictType type, Politeness politeness = Politeness.Plain)
            {
                return(new Text(
                           $"{name}: {JpConj.Conjugate(verb, type, form, politeness).Replace("|", "")}\n~{name}: {JpConj.Conjugate(verb, type, form, politeness, Polarity.Negative).Replace("|", "")}\n"));
            }
        }