public void OneSegPerVerse()
{
string pc1 = "Das Buch ist rot. ";
string verse1 = "9";
string pc2 = "Der Herr ist gross.";
string verse2 = "10";
string pc3 = "Ich spreche nicht viel Deutsch.";
ITsStrBldr bldr = m_tsf.MakeString(pc1 + verse1 + pc2 + verse2 + pc3, m_wsVern).GetBldr();
bldr.SetStrPropValue(pc1.Length, pc1.Length + verse1.Length, (int)FwTextPropType.ktptNamedStyle,
ScrStyleNames.VerseNumber);
int ichV2 = pc1.Length + verse1.Length + pc2.Length;
bldr.SetStrPropValue(ichV2, ichV2 + verse2.Length, (int)FwTextPropType.ktptNamedStyle,
ScrStyleNames.VerseNumber);
m_para.Contents = bldr.GetString();
using (ParagraphParser pp = new ParagraphParser(m_para))
{
List<int> eosIndexes;
var segments = pp.CollectSegments(m_para.Contents, out eosIndexes);
Assert.AreEqual(5, segments.Count);
Assert.AreEqual("", ScriptureServices.VerseSegLabel(m_para, 0));
Assert.AreEqual("", ScriptureServices.VerseSegLabel(m_para, 2));
Assert.AreEqual("", ScriptureServices.VerseSegLabel(m_para, 4));
}
}
/// <summary>
/// non-undoable task
/// </summary>
private void DoSetupFixture()
{
var textFactory = Cache.ServiceLocator.GetInstance<ITextFactory>();
var stTextFactory = Cache.ServiceLocator.GetInstance<IStTextFactory>();
m_text = textFactory.Create();
//Cache.LangProject.TextsOC.Add(m_text);
m_stText = stTextFactory.Create();
m_text.ContentsOA = m_stText;
m_para0 = m_stText.AddNewTextPara(null);
m_para0.Contents = TsStringUtils.MakeTss("Xxxhope xxxthis xxxwill xxxdo. xxxI xxxhope.", Cache.DefaultVernWs);
m_para1 = m_stText.AddNewTextPara(null);
m_para1.Contents = TsStringUtils.MakeTss("Xxxcertain xxxto xxxcatch xxxa xxxfrog. xxxCertainly xxxcan. xxxOn xxxLake xxxMonroe.", Cache.DefaultVernWs);
m_para2 = null;
using (ParagraphParser pp = new ParagraphParser(Cache))
foreach (IStTxtPara para in m_stText.ParagraphsOS)
pp.Parse(para);
m_expectedAnOcs = new List<AnalysisOccurrence>();
foreach (IStTxtPara para in m_stText.ParagraphsOS)
foreach (ISegment seg in para.SegmentsOS)
for (int i = 0; i < seg.AnalysesRS.Count; i++)
m_expectedAnOcs.Add(new AnalysisOccurrence(seg, i));
m_expectedAnOcsPara0 = new List<AnalysisOccurrence>();
foreach (ISegment seg in m_para0.SegmentsOS)
for (int i = 0; i < seg.AnalysesRS.Count; i++)
m_expectedAnOcsPara0.Add(new AnalysisOccurrence(seg, i));
}
/// <summary>
/// Dispose ParagraphParser after each test
/// </summary>
public override void TestTearDown()
{
if (m_pp != null)
m_pp.Dispose();
m_pp = null;
base.TestTearDown();
}
private void ParseText()
{
using (var pp = new ParagraphParser(Cache))
{
pp.Parse(m_txtPara);
}
var seg = m_txtPara.SegmentsOS[0];
var wordArray = seg.AnalysesRS.ToArray();
var cwords = wordArray.Length;
m_occurrences = new AnalysisOccurrence[cwords];
for (var i = 0; i < cwords; i++)
m_occurrences[i] = new AnalysisOccurrence(seg, i);
}
/// <summary>
/// Break an occurrence that is a phrase into its constituent wordforms.
/// </summary>
public void BreakPhrase()
{
using (var pp = new ParagraphParser(Paragraph))
{
// This is a new paragraph parser, and we haven't set up any pre-existing analyses, so it doesn't matter
// what we pass for cWfAnalysisPrev.
IList<IAnalysis> wordforms = pp.CollectSegmentForms(GetMyBeginOffsetInPara(), GetMyEndOffsetInPara(), 0, false);
if (wordforms.Count > 1)
{
var oldWordform = Analysis.Wordform;
Segment.AnalysesRS.Replace(Index, 1, wordforms.Cast<ICmObject>());
// Enhance JohnT: for this sort of automatic deletion, I wonder whether we should make
// stronger checks, such as that it has no analysis or glosses?
DeleteWordformIfPossible(oldWordform);
}
}
}
/// <summary>
/// Parse all the paragraphs in the text.
/// </summary>
public static void ParseText(IStText sttext)
{
using (var parser = new ParagraphParser(sttext.Cache))
{
foreach (IStTxtPara para in sttext.ParagraphsOS)
parser.Parse(para);
}
}
/// <summary>
/// Parse a single paragraph with the specified options.
/// </summary>
public static void ParseParagraph(IStTxtPara para, ParagraphParserOptions options)
{
if (para.ParseIsCurrent)
return;
using (var pp = new ParagraphParser(para.Cache))
{
pp.ParseWithOptions(para, options);
}
}
internal SegmentMaker(ITsString text, ILgWritingSystemFactory wsf, ParagraphParser pp)
: base(text, wsf)
{
m_paraParser = pp;
}
/// <summary>
/// Retrieve the wordforms collected during the last parsing session.
/// </summary>
/// <param name="cache"></param>
/// <returns></returns>
public static Set<int> WordformsFromLastParseSession(FdoCache cache)
{
Set<int> parsedWordforms = null;
using (ParagraphParser pp = new ParagraphParser(cache))
{
parsedWordforms = new Set<int>(pp.WordformIdOccurrencesTable);
}
return parsedWordforms;
}
/// <summary>
/// Create a new ParagraphParser for each test
/// </summary>
public override void TestSetup()
{
base.TestSetup();
m_pp = new ParagraphParser(m_para);
}
public void TwoSegsPerVerse()
{
string pc1 = "Das Buch ist rot. ";
string pc2 = "Das Maedchen ist schoen.";
string verse1 = "9";
string pc3 = "Der Herr ist gross.";
string pc4 = "Ich spreche nicht viel Deutsch.";
string verse2 = "10";
string pc5 = "Was ist das?";
string pc6 = "Wie gehts?";
ITsStrBldr bldr = m_tsf.MakeString(pc1 + pc2 + verse1 + pc3 + pc4 + verse2 + pc5 + pc6, m_wsVern).GetBldr();
bldr.SetStrPropValue(pc1.Length + pc2.Length, pc1.Length + pc2.Length + verse1.Length, (int)FwTextPropType.ktptNamedStyle,
ScrStyleNames.VerseNumber);
int ichEndV1 = pc1.Length + pc2.Length + verse1.Length + pc3.Length + pc4.Length;
bldr.SetStrPropValue(ichEndV1, ichEndV1 + verse2.Length, (int)FwTextPropType.ktptNamedStyle,
ScrStyleNames.VerseNumber);
m_para.Contents = bldr.GetString();
using (ParagraphParser pp = new ParagraphParser(m_para))
{
List<int> eosIndexes;
var segments = pp.CollectSegments(m_para.Contents, out eosIndexes);
Assert.AreEqual(8, segments.Count);
Assert.AreEqual("a", ScriptureServices.VerseSegLabel(m_para, 0));
Assert.AreEqual("b", ScriptureServices.VerseSegLabel(m_para, 1));
Assert.AreEqual("a", ScriptureServices.VerseSegLabel(m_para, 3));
Assert.AreEqual("b", ScriptureServices.VerseSegLabel(m_para, 4));
Assert.AreEqual("a", ScriptureServices.VerseSegLabel(m_para, 6));
Assert.AreEqual("b", ScriptureServices.VerseSegLabel(m_para, 7));
}
}
private IList<ISegment> GetSegments(ITsStrBldr bldr, IScrTxtPara para)
{
para.Contents = bldr.GetString();
using (ParagraphParser pp = new ParagraphParser(para))
{
List<int> eosIndexes;
var segments = pp.CollectSegments(para.Contents, out eosIndexes);
return segments;
}
}
/// ------------------------------------------------------------------------------------
/// <summary>
/// Fixes the paragraph's analysis.
/// </summary>
/// <param name="para">The paragraph.</param>
/// ------------------------------------------------------------------------------------
private static void FixParaAnalysis(IScrTxtPara para)
{
// If it has any word-level analysis, we need to reparse the whole text.
if ((from segment in para.SegmentsOS where segment.AnalysesRS.Count > 0 select segment).FirstOrDefault() == null)
{
// No analyses; just resegment it.
using (ParagraphParser parser = new ParagraphParser(para))
{
parser.CollectPreExistingParaAnnotations();
SegmentMaker segmentMaker = new SegmentMaker(para.Contents, para.Cache.WritingSystemFactory, parser);
segmentMaker.Run();
if (segmentMaker.Segments.Count < para.SegmentsOS.Count)
{
// The paragraph has more segments than it should have, so remove any
// extras that are floating around.
for (int i = para.SegmentsOS.Count - 1; i >= segmentMaker.Segments.Count; i--)
para.SegmentsOS.RemoveAt(i);
}
}
}
else
{
// Reparse the whole thing.
ParagraphParser.ParseParagraph(para);
}
}
internal void DoDataSetup()
{
var textFactory = Cache.ServiceLocator.GetInstance<ITextFactory>();
var stTextFactory = Cache.ServiceLocator.GetInstance<IStTextFactory>();
Text = textFactory.Create();
//Cache.LangProject.TextsOC.Add(Text);
StText = stTextFactory.Create();
Text.ContentsOA = StText;
Para0 = (StTxtPara)StText.AddNewTextPara(null);
var wfFactory = Cache.ServiceLocator.GetInstance<IWfiWordformFactory>();
var wsVern = Cache.DefaultVernWs;
/* A a a a. */
IWfiWordform A = wfFactory.Create(TsStringUtils.MakeTss("A", wsVern));
IWfiWordform a = wfFactory.Create(TsStringUtils.MakeTss("a", wsVern));
Words_para0.Add(A);
Words_para0.Add(a);
Words_para0.Add(a);
Words_para0.Add(a);
Para0.Contents = TsStringUtils.MakeTss(
Words_para0[0].Form.BestVernacularAlternative.Text + " " +
Words_para0[1].Form.BestVernacularAlternative.Text + " " +
Words_para0[2].Form.BestVernacularAlternative.Text + " " +
Words_para0[3].Form.BestVernacularAlternative.Text + ".", wsVern);
/* b B. */
IWfiWordform b = wfFactory.Create(TsStringUtils.MakeTss("b", wsVern));
IWfiWordform B = wfFactory.Create(TsStringUtils.MakeTss("B", wsVern));
Words_para0.Add(b);
Words_para0.Add(B);
var bldr = Para0.Contents.GetIncBldr();
bldr.AppendTsString(TsStringUtils.MakeTss(
" " + Words_para0[4].Form.BestVernacularAlternative.Text + " " +
Words_para0[5].Form.BestVernacularAlternative.Text + ".", wsVern));
Para0.Contents = bldr.GetString();
using (ParagraphParser pp = new ParagraphParser(Cache))
{
foreach (IStTxtPara para in StText.ParagraphsOS)
pp.Parse(para);
}
}
private void Setup2ndText()
{
var textFactory = Cache.ServiceLocator.GetInstance<ITextFactory>();
var stTextFactory = Cache.ServiceLocator.GetInstance<IStTextFactory>();
var text2 = textFactory.Create();
//Cache.LangProject.TextsOC.Add(text2);
m_stText2 = stTextFactory.Create();
text2.ContentsOA = m_stText2;
m_para2 = m_stText2.AddNewTextPara(null);
m_para2.Contents = TsStringUtils.MakeTss("Small one segment paragraph.", Cache.DefaultVernWs);
using (ParagraphParser pp = new ParagraphParser(Cache))
foreach (IStTxtPara para in m_stText2.ParagraphsOS)
pp.Parse(para);
m_expectedAnOcsPara2 = new List<AnalysisOccurrence>();
foreach (IStTxtPara para in m_stText2.ParagraphsOS)
foreach (ISegment seg in para.SegmentsOS)
for (int i = 0; i < seg.AnalysesRS.Count; i++)
m_expectedAnOcsPara2.Add(new AnalysisOccurrence(seg, i));
}
private IStTxtPara MakeSimpleParsedText()
{
var textFactory = Cache.ServiceLocator.GetInstance<ITextFactory>();
var stTextFactory = Cache.ServiceLocator.GetInstance<IStTextFactory>();
var text = textFactory.Create();
//Cache.LangProject.TextsOC.Add(text);
var stText = stTextFactory.Create();
text.ContentsOA = stText;
var para0 = stText.AddNewTextPara(null);
para0.Contents =
TsStringUtils.MakeTss("the book is red. the pages in the book are the color of the paper.",
Cache.DefaultVernWs);
using (ParagraphParser pp = new ParagraphParser(Cache))
foreach (IStTxtPara para in stText.ParagraphsOS)
pp.Parse(para);
return para0;
}
