public void ExportGuesses()
{
//NOTE: The new test paragraphs need to have all new words w/o duplicates so we can predict the guesses
//xxxcrayzee xxxyouneek xxxsintents.
// copy a text of first paragraph into a new paragraph to generate guesses.
StTxtPara paraGlossed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
StTxtPara paraGuessed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
paraGlossed.Contents.UnderlyingTsString = StringUtils.MakeTss("xxxcrayzee xxxyouneek xxxsintents.", Cache.DefaultVernWs);
paraGuessed.Contents.UnderlyingTsString = paraGlossed.Contents.UnderlyingTsString;
// collect expected guesses from the glosses in the first paragraph.
ParagraphAnnotator paGlossed = new ParagraphAnnotator(paraGlossed);
List <int> expectedGuesses = paGlossed.SetupDefaultWordGlosses();
// then verify we've created guesses for the new text.
ParagraphAnnotator paGuessed = new ParagraphAnnotator(paraGuessed);
bool fDidParse;
ParagraphParser.ParseText(m_text1.ContentsOA, new NullProgressState(), out fDidParse);
paGuessed.LoadParaDefaultAnalyses();
// export the paragraph and test the Display results
m_choices.Add(InterlinLineChoices.kflidWord);
m_choices.Add(InterlinLineChoices.kflidWordGloss);
XmlDocument exportedDoc = ExportToXml();
ValidateExportedParagraph(exportedDoc, m_choices, paraGuessed);
}
public void ExportVariantTypeInformation_LT9374()
{
CheckDisposed();
m_choices.Add(InterlinLineChoices.kflidWord);
m_choices.Add(InterlinLineChoices.kflidMorphemes);
m_choices.Add(InterlinLineChoices.kflidLexEntries);
m_choices.Add(InterlinLineChoices.kflidLexGloss);
m_choices.Add(InterlinLineChoices.kflidLexPos);
XmlDocument exportedDoc = ExportToXml();
IStTxtPara para1 = m_text1.ContentsOA.ParagraphsOS[1] as IStTxtPara;
ValidateExportedParagraph(exportedDoc, m_choices, para1);
// Set alternate case endings.
ParagraphAnnotator ta = new ParagraphAnnotator(para1);
string formLexEntry = "go";
ITsString tssLexEntryForm = StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs);
int clsidForm;
ILexEntry leGo = LexEntry.CreateEntry(Cache,
MoMorphType.FindMorphType(Cache, new MoMorphTypeCollection(Cache), ref formLexEntry, out clsidForm), tssLexEntryForm,
"go.pst", null);
ta.SetVariantOf(0, 1, leGo, "fr. var.");
exportedDoc = this.ExportToXml();
ValidateExportedParagraph(exportedDoc, m_choices, para1);
}
public void ExportPhraseWordGuids()
{
// create two paragraphs with two identical sentences.
// copy a text of first paragraph into a new paragraph to generate guesses.
StTxtPara paraGlossed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
StTxtPara paraGuessed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
paraGlossed.Contents.UnderlyingTsString = StringUtils.MakeTss(
"xxxwordone xxxwordtwo xxxwordthree. xxxwordone xxxwordtwo xxxwordthree.",
Cache.DefaultVernWs);
paraGuessed.Contents.UnderlyingTsString = paraGlossed.Contents.UnderlyingTsString;
// collect expected guesses from the glosses in the first paragraph.
ParagraphAnnotator paGlossed = new ParagraphAnnotator(paraGlossed);
List <int> expectedGuesses = paGlossed.SetupDefaultWordGlosses();
// then verify we've created guesses for the new text.
ParagraphAnnotator paGuessed = new ParagraphAnnotator(paraGuessed);
bool fDidParse;
ParagraphParser.ParseText(m_text1.ContentsOA, new NullProgressState(), out fDidParse);
paGuessed.LoadParaDefaultAnalyses();
// export the paragraph and test the Display results
m_choices.Add(InterlinLineChoices.kflidWord);
m_choices.Add(InterlinLineChoices.kflidWordGloss);
m_choices.Add(InterlinLineChoices.kflidMorphemes);
m_choices.Add(InterlinLineChoices.kflidLexEntries);
m_choices.Add(InterlinLineChoices.kflidLexGloss);
m_choices.Add(InterlinLineChoices.kflidLexPos);
XmlDocument exportedDoc = ExportToXml("elan");
// validate that we included the expected metadata
string exportName = XmlUtils.GetOptionalAttributeValue(exportedDoc.DocumentElement, "exportTarget");
Assert.AreEqual("elan", exportName);
string version = XmlUtils.GetOptionalAttributeValue(exportedDoc.DocumentElement, "version");
Assert.AreEqual("1", version);
ExportedInterlinearReader exportReader = new ExportedInterlinearReader(exportedDoc, m_choices);
ExportedParagraphValidatorForELAN validator = new ExportedParagraphValidatorForELAN(exportReader, paraGlossed);
validator.ValidateParagraphs(paraGlossed, exportReader.GetParaNode(paraGlossed.IndexInOwner));
validator.ValidateParagraphs(paraGuessed, exportReader.GetParaNode(paraGuessed.IndexInOwner));
// only expecting to collect a total of 2 paragraph guids,
// each paragraph with 2 phrase guids (2*2)
// and each phrase with 3 word guids (2*2*3).
validator.ValidateNonrepeatingGuidCount(2 + 2 * 2 + 2 * 2 * 3);
}
public void AlternateCaseAnalyses_Baseline_LT5385()
{
CheckDisposed();
m_choices.Add(InterlinLineChoices.kflidWord);
XmlDocument exportedDoc = ExportToXml();
IStTxtPara para0 = m_text1.ContentsOA.ParagraphsOS[0] as IStTxtPara;
ValidateExportedParagraph(exportedDoc, m_choices, para0);
// Set alternate case endings.
ParagraphAnnotator ta = new ParagraphAnnotator(para0);
string altCaseForm;
ta.SetAlternateCase(0, 0, StringCaseStatus.allLower, out altCaseForm);
Assert.AreEqual("xxxpus", altCaseForm);
exportedDoc = this.ExportToXml();
ValidateExportedParagraph(exportedDoc, m_choices, para0);
}
/// <exception cref="System.Exception"/>
public static void Main(string[] args)
{
string home = "/home/mjfang/action_grammars/";
// make the first argument one for a base directory
string specificFile = "1PPDevUncollapsed.props";
if (args.Length >= 1)
{
home = args[0];
}
if (args.Length >= 2)
{
specificFile = args[1];
}
System.Console.Out.WriteLine("Base directory: " + home);
Properties props = StringUtils.PropFileToProperties(home + "ExtractQuotesXMLScripts/" + specificFile);
XMLToAnnotation.Data data = XMLToAnnotation.ReadXMLFormat(props.GetProperty("file"));
Properties propsPara = new Properties();
propsPara.SetProperty("paragraphBreak", "one");
ParagraphAnnotator pa = new ParagraphAnnotator(propsPara, false);
pa.Annotate(data.doc);
Properties annotatorProps = new Properties();
annotatorProps.SetProperty("charactersPath", props.GetProperty("charactersPath"));
//"characterList.txt"
annotatorProps.SetProperty("booknlpCoref", props.GetProperty("booknlpCoref"));
annotatorProps.SetProperty("modelPath", props.GetProperty("modelPath"));
//"model.ser");
QuoteAttributionAnnotator qaa = new QuoteAttributionAnnotator(annotatorProps);
qaa.Annotate(data.doc);
ChapterAnnotator ca = new ChapterAnnotator();
ca.Annotate(data.doc);
Train(data, annotatorProps);
}
public void ExportPhraseWordGuids()
{
// create two paragraphs with two identical sentences.
// copy a text of first paragraph into a new paragraph to generate guesses.
StTxtPara paraGlossed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
StTxtPara paraGuessed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
paraGlossed.Contents.UnderlyingTsString = StringUtils.MakeTss(
"xxxwordone xxxwordtwo xxxwordthree. xxxwordone xxxwordtwo xxxwordthree.",
Cache.DefaultVernWs);
paraGuessed.Contents.UnderlyingTsString = paraGlossed.Contents.UnderlyingTsString;
// collect expected guesses from the glosses in the first paragraph.
ParagraphAnnotator paGlossed = new ParagraphAnnotator(paraGlossed);
List<int> expectedGuesses = paGlossed.SetupDefaultWordGlosses();
// then verify we've created guesses for the new text.
ParagraphAnnotator paGuessed = new ParagraphAnnotator(paraGuessed);
bool fDidParse;
ParagraphParser.ParseText(m_text1.ContentsOA, new NullProgressState(), out fDidParse);
paGuessed.LoadParaDefaultAnalyses();
// export the paragraph and test the Display results
m_choices.Add(InterlinLineChoices.kflidWord);
m_choices.Add(InterlinLineChoices.kflidWordGloss);
m_choices.Add(InterlinLineChoices.kflidMorphemes);
m_choices.Add(InterlinLineChoices.kflidLexEntries);
m_choices.Add(InterlinLineChoices.kflidLexGloss);
m_choices.Add(InterlinLineChoices.kflidLexPos);
XmlDocument exportedDoc = ExportToXml("elan");
// validate that we included the expected metadata
string exportName = XmlUtils.GetOptionalAttributeValue(exportedDoc.DocumentElement, "exportTarget");
Assert.AreEqual("elan", exportName);
string version = XmlUtils.GetOptionalAttributeValue(exportedDoc.DocumentElement, "version");
Assert.AreEqual("1", version);
ExportedInterlinearReader exportReader = new ExportedInterlinearReader(exportedDoc, m_choices);
ExportedParagraphValidatorForELAN validator = new ExportedParagraphValidatorForELAN(exportReader, paraGlossed);
validator.ValidateParagraphs(paraGlossed, exportReader.GetParaNode(paraGlossed.IndexInOwner));
validator.ValidateParagraphs(paraGuessed, exportReader.GetParaNode(paraGuessed.IndexInOwner));
// only expecting to collect a total of 2 paragraph guids,
// each paragraph with 2 phrase guids (2*2)
// and each phrase with 3 word guids (2*2*3).
validator.ValidateNonrepeatingGuidCount(2 + 2*2 + 2*2*3);
}
public void ExportGuesses()
{
//NOTE: The new test paragraphs need to have all new words w/o duplicates so we can predict the guesses
//xxxcrayzee xxxyouneek xxxsintents.
// copy a text of first paragraph into a new paragraph to generate guesses.
StTxtPara paraGlossed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
StTxtPara paraGuessed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
paraGlossed.Contents.UnderlyingTsString = StringUtils.MakeTss("xxxcrayzee xxxyouneek xxxsintents.", Cache.DefaultVernWs);
paraGuessed.Contents.UnderlyingTsString = paraGlossed.Contents.UnderlyingTsString;
// collect expected guesses from the glosses in the first paragraph.
ParagraphAnnotator paGlossed = new ParagraphAnnotator(paraGlossed);
List<int> expectedGuesses = paGlossed.SetupDefaultWordGlosses();
// then verify we've created guesses for the new text.
ParagraphAnnotator paGuessed = new ParagraphAnnotator(paraGuessed);
bool fDidParse;
ParagraphParser.ParseText(m_text1.ContentsOA, new NullProgressState(), out fDidParse);
paGuessed.LoadParaDefaultAnalyses();
// export the paragraph and test the Display results
m_choices.Add(InterlinLineChoices.kflidWord);
m_choices.Add(InterlinLineChoices.kflidWordGloss);
XmlDocument exportedDoc = ExportToXml();
ValidateExportedParagraph(exportedDoc, m_choices, paraGuessed);
}
public void ExportVariantTypeInformation_LT9374()
{
CheckDisposed();
m_choices.Add(InterlinLineChoices.kflidWord);
m_choices.Add(InterlinLineChoices.kflidMorphemes);
m_choices.Add(InterlinLineChoices.kflidLexEntries);
m_choices.Add(InterlinLineChoices.kflidLexGloss);
m_choices.Add(InterlinLineChoices.kflidLexPos);
XmlDocument exportedDoc = ExportToXml();
IStTxtPara para1 = m_text1.ContentsOA.ParagraphsOS[1] as IStTxtPara;
ValidateExportedParagraph(exportedDoc, m_choices, para1);
// Set alternate case endings.
ParagraphAnnotator ta = new ParagraphAnnotator(para1);
string formLexEntry = "go";
ITsString tssLexEntryForm = StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs);
int clsidForm;
ILexEntry leGo = LexEntry.CreateEntry(Cache,
MoMorphType.FindMorphType(Cache, new MoMorphTypeCollection(Cache), ref formLexEntry, out clsidForm), tssLexEntryForm,
"go.pst", null);
ta.SetVariantOf(0, 1, leGo, "fr. var.");
exportedDoc = this.ExportToXml();
ValidateExportedParagraph(exportedDoc, m_choices, para1);
}
public void AlternateCaseAnalyses_Baseline_LT5385()
{
CheckDisposed();
m_choices.Add(InterlinLineChoices.kflidWord);
XmlDocument exportedDoc = ExportToXml();
IStTxtPara para0 = m_text1.ContentsOA.ParagraphsOS[0] as IStTxtPara;
ValidateExportedParagraph(exportedDoc, m_choices, para0);
// Set alternate case endings.
ParagraphAnnotator ta = new ParagraphAnnotator(para0);
string altCaseForm;
ta.SetAlternateCase(0, 0, StringCaseStatus.allLower, out altCaseForm);
Assert.AreEqual("xxxpus", altCaseForm);
exportedDoc = this.ExportToXml();
ValidateExportedParagraph(exportedDoc, m_choices, para0);
}
public virtual void Annotate(Annotation annotation)
{
bool perDocumentCharacterMap = false;
if (buildCharacterMapPerAnnotation)
{
if (annotation.ContainsKey(typeof(CoreAnnotations.MentionsAnnotation)))
{
EntityMentionsToCharacterMap(annotation);
}
}
// 0. pre-preprocess the text with paragraph annotations
// TODO: maybe move this out, definitely make it so that you can set paragraph breaks
Properties propsPara = new Properties();
propsPara.SetProperty("paragraphBreak", "one");
ParagraphAnnotator pa = new ParagraphAnnotator(propsPara, false);
pa.Annotate(annotation);
// 1. preprocess the text
// a) setup coref
IDictionary <int, string> pronounCorefMap = QuoteAttributionUtils.SetupCoref(CorefPath, characterMap, annotation);
//annotate chapter numbers in sentences. Useful for denoting chapter boundaries
new ChapterAnnotator().Annotate(annotation);
// to incorporate sentences across paragraphs
QuoteAttributionUtils.AddEnhancedSentences(annotation);
//annotate depparse of quote-removed sentences
QuoteAttributionUtils.AnnotateForDependencyParse(annotation);
Annotation preprocessed = annotation;
// 2. Quote->Mention annotation
IDictionary <string, QMSieve> qmSieves = GetQMMapping(preprocessed, pronounCorefMap);
foreach (string sieveName in qmSieveList.Split(","))
{
qmSieves[sieveName].DoQuoteToMention(preprocessed);
}
// 3. Mention->Speaker annotation
IDictionary <string, MSSieve> msSieves = GetMSMapping(preprocessed, pronounCorefMap);
foreach (string sieveName_1 in msSieveList.Split(","))
{
msSieves[sieveName_1].DoMentionToSpeaker(preprocessed);
}
// see if any speaker's could be matched to a canonical entity mention
foreach (ICoreMap quote in QuoteAnnotator.GatherQuotes(annotation))
{
int firstSpeakerTokenIndex = quote.Get(typeof(QuoteAttributionAnnotator.MentionBeginAnnotation));
if (firstSpeakerTokenIndex != null)
{
CoreLabel firstSpeakerToken = annotation.Get(typeof(CoreAnnotations.TokensAnnotation))[firstSpeakerTokenIndex];
int entityMentionIndex = firstSpeakerToken.Get(typeof(CoreAnnotations.EntityMentionIndexAnnotation));
if (entityMentionIndex != null)
{
// set speaker string
ICoreMap entityMention = annotation.Get(typeof(CoreAnnotations.MentionsAnnotation))[entityMentionIndex];
int canonicalEntityMentionIndex = entityMention.Get(typeof(CoreAnnotations.CanonicalEntityMentionIndexAnnotation));
if (canonicalEntityMentionIndex != null)
{
ICoreMap canonicalEntityMention = annotation.Get(typeof(CoreAnnotations.MentionsAnnotation))[canonicalEntityMentionIndex];
// add canonical entity mention info to quote
quote.Set(typeof(QuoteAttributionAnnotator.CanonicalMentionAnnotation), canonicalEntityMention.Get(typeof(CoreAnnotations.TextAnnotation)));
// set first and last tokens of canonical entity mention
IList <CoreLabel> canonicalEntityMentionTokens = canonicalEntityMention.Get(typeof(CoreAnnotations.TokensAnnotation));
CoreLabel canonicalEntityMentionFirstToken = canonicalEntityMentionTokens[0];
CoreLabel canonicalEntityMentionLastToken = canonicalEntityMentionTokens[canonicalEntityMentionTokens.Count - 1];
quote.Set(typeof(QuoteAttributionAnnotator.CanonicalMentionBeginAnnotation), canonicalEntityMentionFirstToken.Get(typeof(CoreAnnotations.TokenBeginAnnotation)));
quote.Set(typeof(QuoteAttributionAnnotator.CanonicalMentionEndAnnotation), canonicalEntityMentionLastToken.Get(typeof(CoreAnnotations.TokenBeginAnnotation)));
}
}
}
}
}
public void CheckValidGuessesAfterInsertNewWord_LT8467()
{
//NOTE: The new test paragraphs need to have all new words w/o duplicates so we can predict the guesses
//xxxcrayzee xxxyouneek xxxsintents.
// copy a text of first paragraph into a new paragraph to generate guesses.
IStTxtPara paraGlossed = m_text1.ContentsOA.AddNewTextPara("Normal");
IStTxtPara paraGuessed = m_text1.ContentsOA.AddNewTextPara("Normal");
paraGlossed.Contents = TsStringUtils.MakeTss("xxxcrayzee xxxyouneek xxxsintents.", Cache.DefaultVernWs);
paraGuessed.Contents = paraGlossed.Contents;
// collect expected guesses from the glosses in the first paragraph.
ParagraphAnnotator paGlossed = new ParagraphAnnotator(paraGlossed);
ParagraphParser.ParseText(m_text1.ContentsOA);
IList<IWfiGloss> expectedGuesses = paGlossed.SetupDefaultWordGlosses();
// then verify we've created guesses for the new text.
ParagraphAnnotator paGuessed = new ParagraphAnnotator(paraGuessed);
IList<IWfiGloss> expectedGuessesBeforeEdit = expectedGuesses;
ValidateGuesses(expectedGuessesBeforeEdit, paraGuessed);
// now edit the paraGuessed and expected Guesses.
paraGuessed.Contents = TsStringUtils.MakeTss("xxxcrayzee xxxguessless xxxyouneek xxxsintents.", Cache.DefaultVernWs);
IList<IWfiGloss> expectedGuessesAfterEdit = new List<IWfiGloss>(expectedGuesses);
// we don't expect a guess for the inserted word, so insert 0 after first twfic.
expectedGuessesAfterEdit.Insert(1, null);
// Note: we need to use ParseText rather than ReparseParagraph, because it uses
// code to Reuse dummy annotations.
ParagraphParser.ParseText(m_text1.ContentsOA);
ValidateGuesses(expectedGuessesAfterEdit, paraGuessed);
}
public void CheckValidGuessesAfterInsertNewWord_LT8467()
{
//NOTE: The new test paragraphs need to have all new words w/o duplicates so we can predict the guesses
//xxxcrayzee xxxyouneek xxxsintents.
// copy a text of first paragraph into a new paragraph to generate guesses.
StTxtPara paraGlossed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
StTxtPara paraGuessed = m_text1.ContentsOA.ParagraphsOS.Append(new StTxtPara()) as StTxtPara;
paraGlossed.Contents.UnderlyingTsString = StringUtils.MakeTss("xxxcrayzee xxxyouneek xxxsintents.", Cache.DefaultVernWs);
paraGuessed.Contents.UnderlyingTsString = paraGlossed.Contents.UnderlyingTsString;
// collect expected guesses from the glosses in the first paragraph.
ParagraphAnnotator paGlossed = new ParagraphAnnotator(paraGlossed);
List<int> expectedGuesses = paGlossed.SetupDefaultWordGlosses();
// then verify we've created guesses for the new text.
ParagraphAnnotator paGuessed = new ParagraphAnnotator(paraGuessed);
bool fDidParse;
ParagraphParser.ParseText(m_text1.ContentsOA, new NullProgressState(), out fDidParse);
paGuessed.LoadParaDefaultAnalyses();
List<int> expectedGuessesBeforeEdit = new List<int>(expectedGuesses);
ValidateGuesses(expectedGuessesBeforeEdit, paraGuessed);
// now edit the paraGuessed and expected Guesses.
paraGuessed.Contents.UnderlyingTsString = StringUtils.MakeTss("xxxcrayzee xxxguessless xxxyouneek xxxsintents.", Cache.DefaultVernWs);
List<int> expectedGuessesAfterEdit = new List<int>(expectedGuesses);
// we don't expect a guess for the inserted word, so insert 0 after first twfic.
expectedGuessesAfterEdit.Insert(1, 0);
// Note: we need to use ParseText rather than ReparseParagraph, because it uses
// code to Reuse dummy annotations.
ParagraphParser.ParseText(m_text1.ContentsOA, new NullProgressState(), out fDidParse);
paGuessed.LoadParaDefaultAnalyses();
ValidateGuesses(expectedGuessesAfterEdit, paraGuessed);
}
public void ConvertDummyToReal()
{
CheckDisposed();
(m_fdoCache.LangProject.WordformInventoryOA as WordformInventory).SuspendUpdatingConcordanceWordforms = true;
//xxxpus xxxyalola xxxnihimbilira. xxxnihimbilira xxxpus xxxyalola. xxxhesyla xxxnihimbilira.
IStTxtPara para = (IStTxtPara)m_text1.ContentsOA.ParagraphsOS[(int)Text1ParaIndex.SimpleSegmentPara];
ParagraphAnnotator ta = new ParagraphAnnotator(para);
// 1. convert a twfic dummy wordform to real one.
int hvoCba0_0 = ta.GetSegmentForm(0,0) ; // xxxpus
ICmBaseAnnotation dummyCba0_0 = CmBaseAnnotation.CreateFromDBObject(Cache, hvoCba0_0);
Assert.IsTrue(dummyCba0_0.IsDummyObject);
int hvoDummyWordform0_0 = dummyCba0_0.InstanceOfRAHvo;
IWfiWordform dummyWordform0_0 = WfiWordform.CreateFromDBObject(Cache, hvoDummyWordform0_0);
Assert.IsTrue(dummyWordform0_0.IsDummyObject);
// Lookup the form and see if we can find it.
hvoDummyWordform0_0 = Cache.LangProject.WordformInventoryOA.GetWordformId(dummyWordform0_0.Form.GetAlternativeTss(Cache.DefaultVernWs));
Assert.AreEqual(dummyWordform0_0.Hvo, hvoDummyWordform0_0);
IWfiWordform realWordform0_0 = CmObject.ConvertDummyToReal(Cache, dummyWordform0_0.Hvo) as IWfiWordform;
Assert.IsTrue(realWordform0_0.IsRealObject);
Assert.AreEqual(dummyWordform0_0.Form.VernacularDefaultWritingSystem,
realWordform0_0.Form.VernacularDefaultWritingSystem);
// Make sure we can still lookup the form.
int hvoRealWordform0_0 = Cache.LangProject.WordformInventoryOA.GetWordformId(realWordform0_0.Form.GetAlternativeTss(Cache.DefaultVernWs));
Assert.AreEqual(realWordform0_0.Hvo, hvoRealWordform0_0);
// Enhance: We could check virtual properties for wordform, e.g. wordform Occurrences.
// 2. convert its twfic annotation to a real one.
ICmBaseAnnotation realCba0_0 = CmObject.ConvertDummyToReal(Cache, hvoCba0_0) as ICmBaseAnnotation;
FdoValidator validator = new FdoValidator(para);
validator.CompareCbas(dummyCba0_0.Hvo, realCba0_0.Hvo,
String.Format("Para{0}/Segment{1}/SegForm{2}", para.Hvo, 0, 0));
// 3. convert a twfic annotation with a dummy wordform instanceOf to a real one,
int hvoCba0_1 = ta.GetSegmentForm(0, 1); // xxxyalola
ICmBaseAnnotation dummyCba0_1 = CmBaseAnnotation.CreateFromDBObject(Cache, hvoCba0_1);
Assert.IsTrue(dummyCba0_1.IsDummyObject);
Assert.IsTrue(Cache.IsDummyObject(dummyCba0_1.InstanceOfRAHvo));
IWfiWordform dummyWordform0_1 = WfiWordform.CreateFromDBObject(Cache, dummyCba0_1.InstanceOfRAHvo) as IWfiWordform;
ICmBaseAnnotation realCba0_1 = CmObject.ConvertDummyToReal(Cache, hvoCba0_1) as ICmBaseAnnotation;
Assert.IsTrue(realCba0_1.IsRealObject);
validator.CompareCbas(dummyCba0_1.Hvo, realCba0_1.Hvo,
String.Format("Para{0}/Segment{1}/SegForm{2}", para.Hvo, 0, 1));
// and confirm the wordform is now real as well.
IWfiWordform realWordform0_1 = realCba0_1.InstanceOfRA as IWfiWordform;
Assert.IsNotNull(realWordform0_1);
Assert.IsTrue(realWordform0_1.IsRealObject);
Assert.AreEqual(dummyWordform0_1.Form.VernacularDefaultWritingSystem,
realWordform0_1.Form.VernacularDefaultWritingSystem);
// 3. Try converting a dummy cba with an analysis to a real one,
// and check to make sure we haven't altered the basic analysis
// of another occurrence of the same wordform.
int hvoCba1_0 = ta.GetSegmentForm(1, 0); // xxxnihimbilira
ICmBaseAnnotation dummyCba1_0 = CmBaseAnnotation.CreateFromDBObject(Cache, hvoCba1_0);
Assert.IsTrue(dummyCba1_0.IsDummyObject);
int hvoDummyWordform1_0 = dummyCba1_0.InstanceOfRAHvo;
IWfiWordform dummyWordform1_0 = WfiWordform.CreateFromDBObject(Cache, hvoDummyWordform1_0);
Assert.IsTrue(dummyWordform1_0.IsDummyObject);
// this will establish a real wfiGloss analysis and convert cba to real one.
string wordgloss;
int hvoWordGloss = ta.SetDefaultWordGloss(1, 0, out wordgloss);
hvoCba1_0 = ta.GetSegmentForm(1, 0); // xxxnihimbilira
Assert.IsFalse(Cache.IsDummyObject(hvoCba1_0));
Assert.IsTrue(Cache.IsValidObject(hvoCba1_0));
int hvoCba2_1 = ta.GetSegmentForm(2, 1); // xxxnihimbilira (third instance)
ICmBaseAnnotation dummyCba2_1 = CmBaseAnnotation.CreateFromDBObject(Cache, hvoCba2_1);
Assert.IsTrue(dummyCba2_1.IsDummyObject);
int hvoDummyWordform2_1 = dummyCba2_1.InstanceOfRAHvo;
// the basic class of the InstanceOf should still be the same.
Assert.AreEqual(WfiWordform.kClassId, Cache.GetClassOfObject(hvoDummyWordform2_1),
"Cba2_1.InstanceOf should be WfiWordform.");
// Instance of, should be real, however.
Assert.IsFalse(Cache.IsDummyObject(hvoDummyWordform2_1), "Cba2_1.InstanceOf should not be dummy object.");
// See if we can get the guess for this twfic.
ta.LoadParaDefaultAnalyses();
int tagTwficDefault = StTxtPara.TwficDefaultFlid(Cache);
int hvoGuess2_1 = Cache.GetObjProperty(hvoCba2_1, tagTwficDefault);
Assert.IsTrue(Cache.IsValidObject(hvoGuess2_1), "Cba2_1 should have a default analysis (guess).");
// Convert it to a real, and then see if we still have the guess.
ICmBaseAnnotation realCba2_1 = CmObject.ConvertDummyToReal(Cache, hvoCba2_1) as ICmBaseAnnotation;
int hvoGuess2_1_real = Cache.GetObjProperty(realCba2_1.Hvo, tagTwficDefault);
Assert.AreEqual(hvoGuess2_1, hvoGuess2_1_real, "Cba2_1 has unexpected twfic default after conversion.");
(m_fdoCache.LangProject.WordformInventoryOA as WordformInventory).SuspendUpdatingConcordanceWordforms = false;
}
//[Ignore("FWC-16 - this test causes NUnit to hang. Need to investigate further.")]
public void LT6967_LimitingAutoSearch()
{
// first initialization should not load or clear matches.
Assert.AreEqual(0, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
m_concordanceControl.SelectLineOption(ConcordanceControl.ConcordanceLines.kBaseline);
m_concordanceControl.MatchCase = true;
m_concordanceControl.SearchOption = MockConcordanceControl.ConcordanceSearchOption.Anywhere;
m_concordanceControl.SearchText = "XXX"; // should get all our twfics.
List<int> results = m_concordanceControl.Search();
m_window.ProcessPendingItems();
Assert.AreEqual(6, results.Count);
Assert.AreEqual(1, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
m_concordanceControl.ResetCounters();
// Do a refresh
// the list should reload with the same results.
this.MasterRefresh();
Assert.AreEqual(1, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(6, m_concordanceControl.Results().Count);
m_concordanceControl.ResetCounters();
// switch tools and come back, without modifying anything.
// The record list should reload, but not the virtual property.
m_window.ActivateTool("wordListConcordance");
m_window.ProcessPendingItems();
SwitchToConcordanceTool();
Assert.AreEqual(0, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(6, m_concordanceControl.Results().Count);
m_concordanceControl.ResetCounters();
// switch tools, and modify the text.
// the record list should clear, expecting the user to click 'Search' to do the search again.
/// interlinearEdit
m_window.ActivateTool("interlinearEdit");
m_window.ProcessPendingItems();
IStTxtPara para1 = m_text1.ContentsOA.ParagraphsOS[1] as IStTxtPara;
using (new UndoRedoTaskHelper(Cache, "ConcordanceControlTests - LT6967_LimitingAutoSearch.ModifyParagraph",
"ConcordanceControlTests - LT6967_LimitingAutoSearch.ModifyParagraph"))
{
// does this issue a prop change?
ITsStrFactory tsf = TsStrFactoryClass.Create();
para1.Contents.UnderlyingTsString = tsf.MakeString(para1.Contents.Text + " XXXinsertedTextZZZ", Cache.DefaultVernWs);
}
SwitchToConcordanceTool();
Assert.AreEqual(0, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(1, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(0, m_concordanceControl.Results().Count);
m_concordanceControl.ResetCounters();
// Do a refresh
// we should still have no results, even though we have text in the search line.
this.MasterRefresh();
Assert.AreEqual("XXX", m_concordanceControl.SearchText);
Assert.AreEqual(0, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(0, m_concordanceControl.Results().Count);
m_concordanceControl.ResetCounters();
// Redo our search.
m_concordanceControl.SelectLineOption(ConcordanceControl.ConcordanceLines.kWord);
results = m_concordanceControl.Search();
m_window.ProcessPendingItems();
Assert.AreEqual(7, results.Count);
Assert.AreEqual(1, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
m_concordanceControl.ResetCounters();
using (new UndoRedoTaskHelper(Cache, "ConcordanceControlTests - LT6967_LimitingAutoSearch.MakeAndBreakPhrases",
"ConcordanceControlTests - LT6967_LimitingAutoSearch.MakeAndBreakPhrases"))
{
ParagraphAnnotator ta = new ParagraphAnnotator(m_text1.ContentsOA.ParagraphsOS[1] as IStTxtPara);
// 2. Make a basic phrase
// XXXlocoZZZ, XXXsegmentZZZ?? [ZZZamazingXXX wonderfulXXXzzzcounselor]
ta.MergeAdjacentAnnotations(1, 0);
results = m_concordanceControl.Results(); // the merge should delete an annotation.
Assert.AreEqual(6, results.Count);
Assert.AreEqual(0, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(7, m_concordanceControl.Clerk.ListSize); // we haven't refreshed our browse view yet.
m_concordanceControl.ResetCounters();
results = m_concordanceControl.Search(); // the merge should result in new search results.
m_window.ProcessPendingItems();
Assert.AreEqual(6, results.Count);
Assert.AreEqual(1, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(6, m_concordanceControl.Clerk.ListSize); // we haven't refreshed our browse view yet.
m_concordanceControl.ResetCounters();
}
Cache.Undo(); // Undo MakePhrase
// Test that the Undo resulted in the old results set.
m_window.ProcessPendingItems();
results = m_concordanceControl.Results(); // the undo should delete our phrase annotation.
// results count will be zero if Undo cleared our virtual property
//Assert.AreEqual(6, results.Count);
Assert.AreEqual(0, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(6, m_concordanceControl.Clerk.ListSize); // we haven't refreshed our browse view yet.
m_concordanceControl.ResetCounters();
results = m_concordanceControl.Search();
Assert.AreEqual(1, m_concordanceControl.SearchForMatchesCount);
Assert.AreEqual(0, m_concordanceControl.ClearMatchesCount);
Assert.AreEqual(7, results.Count);
Assert.AreEqual(7, m_concordanceControl.Clerk.ListSize); // we haven't refreshed our browse view yet.
m_concordanceControl.ResetCounters();
}
void SetupTexts()
{
// First make a regular text.
using (new UndoRedoTaskHelper(Cache, "ConcordanceControlTests - SetupTexts()", "ConcordanceControlTests - SetupTexts()"))
{
m_text1 = Cache.LangProject.TextsOC.Add(new Text());
m_text1.ContentsOA = new StText();
StTxtPara para0 = new StTxtPara();
StTxtPara para1 = new StTxtPara();
m_text1.ContentsOA.ParagraphsOS.Append(para0);
m_text1.ContentsOA.ParagraphsOS.Append(para1);
ITsStrFactory tsf = TsStrFactoryClass.Create();
// 1 2 3 4 5 6
// 0123456789012345678901234567890123456789012345678901234567890123456789
// XXXXsecrecyZZZ; XXXsentenceZZZ!!
// XXXlocoZZZ, XXXsegmentZZZ?? ZZZamazingXXX wonderfulXXXzzzcounselor!!
para0.Contents.UnderlyingTsString = tsf.MakeString("XXXXsecrecyZZZ; XXXsentenceZZZ!!", Cache.DefaultVernWs);
para1.Contents.UnderlyingTsString = tsf.MakeString("XXXlocoZZZ, XXXsegmentZZZ?? ZZZamazingXXX wonderfulXXXzzzcounselor!!", Cache.DefaultVernWs);
// add scripture
m_newBook1 = new ScrBook();
Cache.LangProject.TranslatedScriptureOA.ScriptureBooksOS.Append(m_newBook1);
m_newBook1.TitleOA = new StText();
m_newBook1.TitleOA.ParagraphsOS.Append(new StTxtPara());
(m_newBook1.TitleOA.ParagraphsOS[0] as StTxtPara).Contents.UnderlyingTsString = tsf.MakeString("XXXnewBook1zzz.Title", Cache.DefaultVernWs);
IScrSection newSection1_0 = m_newBook1.SectionsOS.Append(new ScrSection());
newSection1_0.ContentOA = new StText();
StTxtPara paraSection1_0 = new StTxtPara();
(newSection1_0.ContentOA as StText).ParagraphsOS.Append(paraSection1_0);
paraSection1_0.Contents.UnderlyingTsString = tsf.MakeString("ZZZnewBook1.Section0.Introduction1XXX", Cache.DefaultVernWs);
IScrSection newSection1_1 = m_newBook1.SectionsOS.Append(new ScrSection());
newSection1_1.ContentOA = new StText();
StTxtPara paraSection1_1 = new StTxtPara();
(newSection1_1.ContentOA as StText).ParagraphsOS.Append(paraSection1_1);
paraSection1_1.Contents.UnderlyingTsString = tsf.MakeString("XXXnewBook1.Section1.1:1-1:20ZZZ", Cache.DefaultVernWs);
// section.VerseRefEnd = book.CanonicalNum * 1000000 + 1 * 1000 + (introSection ? 0 : 1);
newSection1_1.VerseRefEnd = 70 * 1000000 + 1 * 1000 + 1;
// setup some basic analyses for the texts.
string formLexEntry = "XXXlexEntry1";
ITsString tssLexEntryForm = StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs);
int clsidForm;
StringUtils.ReassignTss(ref tssLexEntryForm, StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs));
ILexEntry lexEntry1_Entry = LexEntry.CreateEntry(Cache,
MoMorphType.FindMorphType(Cache, new MoMorphTypeCollection(Cache), ref formLexEntry, out clsidForm), tssLexEntryForm,
"XXXlexEntry1.sense1", null);
ILexSense lexEntry1_Sense1 = lexEntry1_Entry.SensesOS[0];
ILexSense lexEntry1_Sense2 = LexSense.CreateSense(lexEntry1_Entry, null, "XXXlexEntry1.sense2");
ParagraphAnnotator tapara0 = new ParagraphAnnotator(para0);
ParagraphAnnotator tapara1 = new ParagraphAnnotator(para1);
ParagraphAnnotator taSection1_0 = new ParagraphAnnotator(paraSection1_0);
ParagraphAnnotator taSection1_1 = new ParagraphAnnotator(paraSection1_1);
// currently setup mono-morphemic search
ArrayList morphForms = new ArrayList();
formLexEntry = "XXXXsecrecyZZZ";
StringUtils.ReassignTss(ref tssLexEntryForm, StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs));
ILexEntry lexEntry2_Entry = LexEntry.CreateEntry(Cache,
MoMorphType.FindMorphType(Cache, new MoMorphTypeCollection(Cache), ref formLexEntry, out clsidForm), tssLexEntryForm,
"XXXlexEntry2.sense1", null);
morphForms.Add(lexEntry2_Entry.LexemeFormOA);
IWfiAnalysis wfiAnalysis = tapara0.BreakIntoMorphs(0, 0, morphForms);
m_wfXXXXsecrecyZZZ = (wfiAnalysis as WfiAnalysis).Owner as IWfiWordform;
ILexSense lexEntry2_Sense1 = lexEntry2_Entry.SensesOS[0];
tapara0.SetMorphSense(0, 0, 0, lexEntry2_Sense1);
formLexEntry = "XXXsegmentZZZ";
StringUtils.ReassignTss(ref tssLexEntryForm, StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs));
ILexEntry lexEntry3_Entry = LexEntry.CreateEntry(Cache,
MoMorphType.FindMorphType(Cache, new MoMorphTypeCollection(Cache), ref formLexEntry, out clsidForm), tssLexEntryForm,
"XXXlexEntry3.sense1", null);
morphForms[0] = lexEntry3_Entry.LexemeFormOA;
tapara1.BreakIntoMorphs(0, 2, morphForms);
ILexSense lexEntry3_Sense1 = lexEntry3_Entry.SensesOS[0];
tapara1.SetMorphSense(0, 2, 0, lexEntry3_Sense1);
formLexEntry = "ZZZamazingXXX";
StringUtils.ReassignTss(ref tssLexEntryForm, StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs));
ILexEntry lexEntry4_Entry = LexEntry.CreateEntry(Cache,
MoMorphType.FindMorphType(Cache, new MoMorphTypeCollection(Cache), ref formLexEntry, out clsidForm), tssLexEntryForm,
"XXXlexEntry4.sense1", null);
morphForms[0] = lexEntry4_Entry.LexemeFormOA;
tapara1.BreakIntoMorphs(1, 0, morphForms);
ILexSense lexEntry4_Sense1 = lexEntry4_Entry.SensesOS[0];
tapara1.SetMorphSense(1, 0, 0, lexEntry4_Sense1);
//XXXlocoZZZ
morphForms[0] = "XXXlocoZZZ";
tapara1.BreakIntoMorphs(0, 0, morphForms);
tapara1.SetMorphSense(0, 0, 0, lexEntry1_Sense2);
formLexEntry = paraSection1_0.Contents.Text;
StringUtils.ReassignTss(ref tssLexEntryForm, StringUtils.MakeTss(formLexEntry, Cache.DefaultVernWs));
ILexEntry lexEntry5_Entry = LexEntry.CreateEntry(Cache,
MoMorphType.FindMorphType(Cache, new MoMorphTypeCollection(Cache), ref formLexEntry, out clsidForm), tssLexEntryForm,
"XXXlexEntry5.sense1", null);
morphForms[0] = lexEntry5_Entry.LexemeFormOA;
taSection1_0.BreakIntoMorphs(0, 0, morphForms);
ILexSense lexEntry5_Sense1 = lexEntry5_Entry.SensesOS[0];
taSection1_0.SetMorphSense(0, 0, 0, lexEntry5_Sense1);
morphForms[0] = paraSection1_1.Contents.Text;
taSection1_1.BreakIntoMorphs(0, 0, morphForms); // won't match on LexEntry
taSection1_1.SetMorphSense(0, 0, 0, lexEntry1_Sense2); // will match on LexGloss
string gloss;
tapara0.SetDefaultWordGloss(0, 0, out gloss);
tapara1.SetDefaultWordGloss(0, 2, out gloss);
tapara1.SetDefaultWordGloss(1, 0, out gloss);
taSection1_1.SetDefaultWordGloss(0, 0, out gloss);
StTxtPara.TwficInfo infoCba0_0_0 = new StTxtPara.TwficInfo(Cache, tapara0.GetSegmentForm(0, 0));
StTxtPara.TwficInfo infoCba1_0_2 = new StTxtPara.TwficInfo(Cache, tapara1.GetSegmentForm(0, 2));
StTxtPara.TwficInfo infoCbaScr_0_0 = new StTxtPara.TwficInfo(Cache, taSection1_1.GetSegmentForm(0, 0));
int segDefn_literalTranslation = Cache.GetIdFromGuid(LangProject.kguidAnnLiteralTranslation);
int segDefn_freeTranslation = Cache.GetIdFromGuid(LangProject.kguidAnnFreeTranslation);
int segDefn_note = Cache.GetIdFromGuid(LangProject.kguidAnnNote);
BaseFreeformAdder ffAdder = new BaseFreeformAdder(Cache);
ICmIndirectAnnotation freeTrans0 = ffAdder.AddFreeformAnnotation(infoCba0_0_0.SegmentHvo, segDefn_freeTranslation);
freeTrans0.Comment.SetAlternative("Para0.Segment0: XXXFreeform translation.", Cache.DefaultAnalWs);
ICmIndirectAnnotation literalTrans0 = ffAdder.AddFreeformAnnotation(infoCba0_0_0.SegmentHvo, segDefn_literalTranslation);
literalTrans0.Comment.SetAlternative("Para0.Segment0: XXXLiteral translation.", Cache.DefaultAnalWs);
ICmIndirectAnnotation note0 = ffAdder.AddFreeformAnnotation(infoCba0_0_0.SegmentHvo, segDefn_note);
note0.Comment.SetAlternative("Para0.Segment0: XXXNote.", Cache.DefaultAnalWs);
ICmIndirectAnnotation freeTrans1 = ffAdder.AddFreeformAnnotation(infoCba1_0_2.SegmentHvo, segDefn_freeTranslation);
freeTrans1.Comment.SetAlternative("Para1.Segment0: XXXFreeform translation.", Cache.DefaultAnalWs);
ICmIndirectAnnotation literalTrans1 = ffAdder.AddFreeformAnnotation(infoCba1_0_2.SegmentHvo, segDefn_literalTranslation);
literalTrans1.Comment.SetAlternative("Para1.Segment0: XXXLiteral translation.", Cache.DefaultAnalWs);
ICmIndirectAnnotation note1 = ffAdder.AddFreeformAnnotation(infoCba1_0_2.SegmentHvo, segDefn_note);
note1.Comment.SetAlternative("Para1.Segment0: XXXNote.", Cache.DefaultAnalWs);
// Scripture
ICmIndirectAnnotation freeTransScr1 = ffAdder.AddFreeformAnnotation(infoCbaScr_0_0.SegmentHvo, segDefn_freeTranslation);
freeTransScr1.Comment.SetAlternative("Scr1.Para0.Segment0: XXXFreeform translation.", Cache.DefaultAnalWs);
ICmIndirectAnnotation literalTransScr1 = ffAdder.AddFreeformAnnotation(infoCbaScr_0_0.SegmentHvo, segDefn_literalTranslation);
literalTransScr1.Comment.SetAlternative("Scr1.Para0.Segment0: XXXLiteral translation.", Cache.DefaultAnalWs);
ICmIndirectAnnotation noteScr1 = ffAdder.AddFreeformAnnotation(infoCbaScr_0_0.SegmentHvo, segDefn_note);
noteScr1.Comment.SetAlternative("Scr1.Para0.Segment0: XXXNote.", Cache.DefaultAnalWs);
}
}
