public void MeCabIpaDicNodeMethodExample()
{
var sentence = Example1;
var tagger = MeCabTagger.Create();
var enumerableSet = tagger.ParseToNodes(sentence);
using var enumerator = enumerableSet.GetEnumerator();
enumerator.MoveNext(); // Feature: BOS/EOS,*,*,*,*,*,*,*,*
enumerator.MoveNext();
MeCabNode node = enumerator.Current !;
Trace.WriteLine($"Alpha: {node.Alpha}");
Trace.WriteLine($"Beta: {node.Beta}");
Trace.WriteLine($"`MeCabNode` BNext: {node.BNext}");
Trace.WriteLine($"BPos: {node.BPos}");
Trace.WriteLine($"CharType: {node.CharType}");
Trace.WriteLine($"Cost: {node.Cost}");
Trace.WriteLine($"`MeCabNode` ENext: {node.ENext}");
Trace.WriteLine($"EPos: {node.EPos}");
Trace.WriteLine($"Feature: {node.Feature}");
Trace.WriteLine($"GetConjugatedForm() Feature[4]: {node.GetConjugatedForm()}"); // CTYPE
Trace.WriteLine($"GetInflection() Feature[5]: {node.GetInflection()}"); // CFORM
Trace.WriteLine($"GetOriginalForm() Feature[6]: {node.GetOriginalForm()}"); // BASIC
Trace.WriteLine($"GetPartsOfSpeech() Feature[0]: {node.GetPartsOfSpeech()}");
Trace.WriteLine($"GetPartsOfSpeechSection1() Feature[1]: {node.GetPartsOfSpeechSection1()}");
Trace.WriteLine($"GetPartsOfSpeechSection2() Feature[2]: {node.GetPartsOfSpeechSection2()}");
Trace.WriteLine($"GetPartsOfSpeechSection3() Feature[3]: {node.GetPartsOfSpeechSection3()}");
Trace.WriteLine($"GetPronunciation() Feature[8]: {node.GetPronounciation()}");
Trace.WriteLine($"GetReading() Feature[7]: {node.GetReading()}");
Trace.WriteLine($"IsBest: {node.IsBest}");
Trace.WriteLine($"LCAttr: {node.LCAttr}");
Trace.WriteLine($"Length: {node.Length}");
Trace.WriteLine($"`MeCabNode` Next: {node.Next}");
Trace.WriteLine($"PosId: {node.PosId}");
Trace.WriteLine($"`MeCabNode` Prev: {node.Prev}");
Trace.WriteLine($"Prob: {node.Prob}");
Trace.WriteLine($"RCAttr: {node.RCAttr}");
Trace.WriteLine($"RLength: {node.RLength}");
Trace.WriteLine($"Stat: {node.Stat}");
Trace.WriteLine($"Surface: {node.Surface}");
Trace.WriteLine($"WCost: {node.WCost}");
}
/// <summary>
/// The Generator
/// </summary>
/// <param name="nodeEnumerable">Include IpaDic info</param>
/// <returns></returns>
/// <exception cref="InvalidOperationException"></exception>
public static IEnumerable <VeWord> Words(IEnumerable <MeCabNode> nodeEnumerable)
{
var tokenArray = new List <MeCabNode>(nodeEnumerable).ToArray()[1..^ 1];
List <VeWord> wordList = new();
MeCabNode? previous = null;
// FIXME: Not stable
for (var i = 0; i < tokenArray.Length; i++)
{
var finalSlot = wordList.Count - 1;
MeCabNode current = tokenArray[i];
MeCabNode following;
PartOfSpeech partOfSpeech;
var grammar = Grammar.Unassigned;
var eatNext = false;
var eatLemma = true;
var attachToPrevious = false;
var alsoAttachToLemma = false;
var updatePos = false;
switch (current.GetPartsOfSpeech())
{
case "名詞":
{
partOfSpeech = PartOfSpeech.詞;
if (current.GetPartsOfSpeechSection1().Equals(NoData))
{
break;
}
switch (current.GetPartsOfSpeechSection1())
{
case "固有名詞":
partOfSpeech = PartOfSpeech.固有名詞;
break;
case "代名詞":
partOfSpeech = PartOfSpeech.代名詞;
break;
case "副詞可能":
case "サ変接続":
case "形容動詞語幹":
case "ナイ形容詞語幹":
// Refers to line 213 of Ve.
if (current.GetPartsOfSpeechSection2().Equals(NoData))
{
break;
}
// protects against array overshooting.
if (i == tokenArray.Length - 1)
{
break;
}
following = tokenArray[i + 1];
switch (following.GetConjugatedForm()) // [CTYPE]
{
case "サ変・スル":
partOfSpeech = PartOfSpeech.動詞;
eatNext = true;
break;
case "特殊・ダ":
partOfSpeech = PartOfSpeech.形容詞;
if (following.GetPartsOfSpeechSection1().Equals("体言接続"))
{
eatNext = true;
eatLemma = false;
}
break;
case "特殊・ナイ":
partOfSpeech = PartOfSpeech.形容詞;
eatNext = true;
break;
default:
if (following.GetPartsOfSpeech().Equals("助詞") &&
following.Surface.Equals("に"))
{
// Ve script redundantly (I think) also has eat_next = false here.
partOfSpeech = PartOfSpeech.副詞;
}
break;
}
break;
case "非自立":
case "特殊":
// Refers to line 233 of Ve.
if (current.GetPartsOfSpeechSection2().Equals(NoData))
{
break;
}
// protects against array overshooting.
if (i == tokenArray.Length - 1)
{
break;
}
following = tokenArray[i + 1];
switch (current.GetPartsOfSpeechSection2())
{
case "副詞可能":
if (following.GetPartsOfSpeech().Equals("助詞") &&
following.Surface.Equals("に"))
{
partOfSpeech = PartOfSpeech.副詞;
// Changed this to false because 'case JOSHI' has 'attach_to_previous = true'.
eatNext = false;
}
break;
case "助動詞語幹":
if (following.GetConjugatedForm().Equals("特殊・ダ"))
{
partOfSpeech = PartOfSpeech.動詞;
grammar = Grammar.Auxiliary;
if (following.GetInflection().Equals("体言接続"))
{
eatNext = true;
}
}
else if (following.GetPartsOfSpeech().Equals("助詞") &&
following.GetPartsOfSpeechSection2().Equals("副詞化"))
{
partOfSpeech = PartOfSpeech.副詞;
eatNext = true;
}
break;
case "形容動詞語幹":
partOfSpeech = PartOfSpeech.形容詞;
if (following.GetConjugatedForm().Equals("特殊・ダ") &&
following.GetInflection().Equals("体言接続") ||
following.GetPartsOfSpeechSection1().Equals("連体化"))
{
eatNext = true;
}
break;
}
break;
case "数":
// TODO: "recurse and find following numbers and add to this word. Except non-numbers
// like 幾"
// Refers to line 261.
partOfSpeech = PartOfSpeech.数;
if (wordList.Count > 0 && wordList[finalSlot].PartOfSpeech.Equals(PartOfSpeech.数))
{
attachToPrevious = true;
alsoAttachToLemma = true;
}
break;
case "接尾":
// Refers to line 267.
switch (current.GetPartsOfSpeechSection2())
{
case "人名":
partOfSpeech = PartOfSpeech.人名接尾;
break;
case "特殊":
if (current.GetOriginalForm().Equals("さ"))
{
updatePos = true;
partOfSpeech = PartOfSpeech.詞;
}
attachToPrevious = true;
break;
default: // 助数词
alsoAttachToLemma = true;
attachToPrevious = true;
break;
}
break;
case "接続詞的":
partOfSpeech = PartOfSpeech.接続詞;
break;
case "動詞非自立的":
partOfSpeech = PartOfSpeech.動詞;
grammar = Grammar.Nominal; // not using.
break;
}
}
break;
case "接頭詞":
// TODO: "elaborate this when we have the "main part" feature for words?"
partOfSpeech = PartOfSpeech.接頭詞;
break;
case "助動詞":
// Refers to line 290.
partOfSpeech = PartOfSpeech.助詞;
List <string> tokushuList = new() { "特殊・タ", "特殊・ナイ", "特殊・タイ", "特殊・マス", "特殊・ヌ" };
if (previous is null || !previous.GetPartsOfSpeechSection1().Equals("係助詞") &&
tokushuList.Contains(current.GetConjugatedForm()))
{
attachToPrevious = true;
}
