public static DiffResultViews DiffTexts(IEnumerable <string> mainTexts, IEnumerable <string> subTexts, out DiffResultDic resultDicMain, out DiffResultDic resultDicSub, double similarThreshold = SIMILAR_THRESHOLD)
{
resultDicMain = new DiffResultDic();
resultDicSub = new DiffResultDic();
Sentences mainSentences = ConvertSentences(mainTexts);
Sentences subSentences = ConvertSentences(subTexts);
GetDiffResultDic(mainSentences: mainSentences, subSentences: subSentences, similarLimit: similarThreshold, resultDicMain: out resultDicMain, resultDicSub: out resultDicSub);
return(ConvertResultToViews(mains: resultDicMain.GetValues(), subs: resultDicSub.GetValues()));
}
static void GetDiffResultDic(Sentences mainSentences, Sentences subSentences, double similarLimit, out DiffResultDic resultDicMain, out DiffResultDic resultDicSub)
{
resultDicMain = new DiffResultDic();
resultDicSub = new DiffResultDic();
int lastIndex = -1, intersectCount;
Sentence sub = new Sentence();
List <string> sameTexts, modifiedTexts;
bool find;
foreach (Sentence main in mainSentences)
{
find = false;
for (int i = lastIndex + 1; i < subSentences.Count; i++)
{
sub = subSentences[i];
// 일단 동등한지 확인
if (string.Equals(main.Text, sub.Text))
{
resultDicMain.Add(main.Index, new DiffResult(index: main.Index, diffType: DiffType.Same, main: main, sub: sub, sameTexts: main.Texts.ToList(), modifiedTexts: new List <string>()));
resultDicSub.Add(sub.Index, new DiffResult(index: sub.Index, diffType: DiffType.Same, main: sub, sub: main, sameTexts: sub.Texts.ToList(), modifiedTexts: new List <string>()));
find = true;
lastIndex = i;
break;
}
// 동등하지 않은 상태에서 유사도 확인
else
{
// 갯수 확인 할 때는 교집합을 하지만, 실제 동일-수정된 텍스트를 찾을 때는 순서가 중요해서 교집합이나 차집합을 하지 않는다.
intersectCount = main.Texts.Intersect(sub.Texts).Count();
if ((double)(intersectCount * 2) / (double)(main.Texts.Length + sub.Texts.Length) > similarLimit)
{
GetSameAndModifiedTexts(mainTexts: main.Texts, subTexts: sub.Texts, sameTexts: out sameTexts, modifiedTexts: out modifiedTexts);
resultDicMain.Add(main.Index, new DiffResult(index: main.Index, diffType: DiffType.Modified, main: main, sub: sub, sameTexts: sameTexts, modifiedTexts: modifiedTexts));
GetSameAndModifiedTexts(mainTexts: sub.Texts, subTexts: main.Texts, sameTexts: out sameTexts, modifiedTexts: out modifiedTexts);
resultDicSub.Add(sub.Index, new DiffResult(index: sub.Index, diffType: DiffType.Modified, main: sub, sub: main, sameTexts: sameTexts, modifiedTexts: modifiedTexts));
find = true;
lastIndex = i;
break;
}
}
}
if (!find)
{
resultDicMain.Add(main.Index, new DiffResult(index: main.Index, diffType: DiffType.Removed, main: main, sub: new Sentence(), sameTexts: new List <string>(), modifiedTexts: main.Texts.ToList()));
}
}
// 위에서 처리하고 남은 right sentence는 add로 처리한다.
foreach (Sentence sentence in subSentences)
{
if (!resultDicSub.ContainsKey(sentence.Index))
{
resultDicSub.Add(sentence.Index, new DiffResult(index: sentence.Index, diffType: DiffType.Added, main: sentence, sub: new Sentence(), sameTexts: new List <string>(), modifiedTexts: sentence.Texts.ToList()));
}
}
// 순서가 꼬였으므로 순서대로 정렬한다.
resultDicSub = new DiffResultDic(resultDicSub.OrderBy(kvp => kvp.Key));
}