public EditMeaningViewModel(IKeyedCollection<string, Meaning> meanings, Meaning meaning)
{
_title = "Edit Meaning";
_meanings = meanings;
_meaning = meaning;
_gloss = meaning.Gloss;
_category = meaning.Category;
}
private void AddNewMeaning()
{
var vm = new EditMeaningViewModel(_projectService.Project.Meanings);
if (_dialogService.ShowModalDialog(this, vm) == true)
{
var newMeaning = new Meaning(vm.Gloss, vm.Category);
_projectService.Project.Meanings.Add(newMeaning);
Messenger.Default.Send(new DomainModelChangedMessage(true));
SelectedMeaning = _meanings.Single(s => s.DomainMeaning == newMeaning);
}
}
public void SetUp()
{
_busyService = Substitute.For<IBusyService>();
_analysisService = Substitute.For<IAnalysisService>();
_words = new ObservableList<WordViewModel>
{
new WordViewModel(_busyService, _analysisService, new Word("valid", _meaning)) {IsValid = true},
new WordViewModel(_busyService, _analysisService, new Word("invalid", _meaning)) {IsValid = false}
};
_wordsViewModel = new WordsViewModel(_busyService, new ReadOnlyBindableList<WordViewModel>(_words));
_meaning = new Meaning("gloss", "category");
}
public WordListsVarietyMeaningViewModel(IBusyService busyService, IAnalysisService analysisService, WordViewModel.Factory wordFactory, WordListsVarietyViewModel variety, Meaning meaning)
: base(meaning)
{
_busyService = busyService;
_analysisService = analysisService;
_variety = variety;
_domainWords = new ObservableList<Word>(variety.DomainVariety.Words[meaning]);
_words = new MirroredBindableList<Word, WordViewModel>(_domainWords, word => wordFactory(word), vm => vm.DomainWord);
_domainWords.CollectionChanged += DomainWordsChanged;
_strRep = string.Join(",", _domainWords.Select(word => word.StrRep));
_showInVarietiesCommand = new RelayCommand(ShowInVarieties, () => _domainWords.Count > 0);
}
public static IEnumerable<Cluster<Word>> GenerateCognateSets(this CogProject project, Meaning meaning)
{
var words = new HashSet<Word>();
var noise = new HashSet<Word>();
foreach (VarietyPair vp in project.VarietyPairs)
{
WordPair wp;
if (vp.WordPairs.TryGetValue(meaning, out wp))
{
if (wp.Cognacy)
{
words.Add(wp.Word1);
words.Add(wp.Word2);
noise.Remove(wp.Word1);
noise.Remove(wp.Word2);
}
else
{
if (!words.Contains(wp.Word1))
noise.Add(wp.Word1);
if (!words.Contains(wp.Word2))
noise.Add(wp.Word2);
}
}
}
double min = double.MaxValue;
var distanceMatrix = new Dictionary<UnorderedTuple<Word, Word>, double>();
Word[] wordArray = words.ToArray();
for (int i = 0; i < wordArray.Length; i++)
{
for (int j = i + 1; j < wordArray.Length; j++)
{
Word w1 = wordArray[i];
Word w2 = wordArray[j];
double score = 0;
WordPair wp;
if (w1.Variety != w2.Variety && w1.Variety.VarietyPairs[w2.Variety].WordPairs.TryGetValue(meaning, out wp) && wp.Cognacy
&& wp.GetWord(w1.Variety) == w1 && wp.GetWord(w2.Variety) == w2)
{
score = wp.PredictedCognacyScore;
}
double distance = 1.0 - score;
min = Math.Min(min, distance);
distanceMatrix[UnorderedTuple.Create(w1, w2)] = distance;
}
}
var clusterer = new FlatUpgmaClusterer<Word>((w1, w2) => distanceMatrix[UnorderedTuple.Create(w1, w2)], (1.0 + min) / 2);
return clusterer.GenerateClusters(words).Concat(new Cluster<Word>(noise, true));
}
public bool?GetCognacy(VarietyPair varietyPair, Meaning meaning)
{
Dictionary <Meaning, int> decisions;
if (_lookupDictionary.TryGetValue(UnorderedTuple.Create(varietyPair.Variety1, varietyPair.Variety2),
out decisions))
{
int index;
if (decisions.TryGetValue(meaning, out index))
{
return(Items[index].Cognacy);
}
}
return(null);
}
public bool?GetCognacy(VarietyPair varietyPair, Meaning meaning)
{
Dictionary <Meaning, CognacyDecision> decisions;
if (_lookupDictionary.TryGetValue(UnorderedTuple.Create(varietyPair.Variety1, varietyPair.Variety2),
out decisions))
{
CognacyDecision decision;
if (decisions.TryGetValue(meaning, out decision))
{
return(decision.Cognacy);
}
}
return(null);
}
private CogProject CreateProject()
{
var project = new CogProject(_spanFactory);
var variety1 = new Variety("variety1");
project.Varieties.Add(variety1);
var variety2 = new Variety("variety2");
project.Varieties.Add(variety2);
var variety3 = new Variety("variety3");
project.Varieties.Add(variety3);
var meaning1 = new Meaning("meaning1", null);
project.Meanings.Add(meaning1);
var meaning2 = new Meaning("meaning2", null);
project.Meanings.Add(meaning2);
var meaning3 = new Meaning("meaning3", null);
project.Meanings.Add(meaning3);
variety1.Words.Add(new Word("word1", meaning1));
variety1.Words.Add(new Word("word2", meaning2));
variety1.Words.Add(new Word("word3", meaning3));
variety2.Words.Add(new Word("word4", meaning1));
variety2.Words.Add(new Word("word5", meaning2));
variety2.Words.Add(new Word("word6", meaning3));
variety3.Words.Add(new Word("word7", meaning1));
variety3.Words.Add(new Word("word8", meaning2));
variety3.Words.Add(new Word("word9", meaning3));
var vpGenerator = new VarietyPairGenerator();
vpGenerator.Process(project);
double score = 1.0;
foreach (VarietyPair vp in project.VarietyPairs)
{
foreach (Meaning meaning in project.Meanings)
{
Word w1 = vp.Variety1.Words[meaning].First();
Word w2 = vp.Variety2.Words[meaning].First();
WordPair wp = vp.WordPairs.Add(w1, w2);
wp.PredictedCognacyScore = score;
wp.PredictedCognacy = true;
score -= 0.1;
}
}
return project;
}
public void RemoveAll(Meaning meaning)
{
_words.Remove(meaning);
}
public MeaningViewModel(Meaning meaning)
: base(meaning)
{
_meaning = meaning;
}
protected Word ParseWordOnce(string wordText, Meaning meaning, CogProject project)
{
Word word;
// We expect to see a lot of duplicates in our input text; save time by memoizing
if (_parsedWords.TryGetValue(wordText, out word))
return word;
try
{
word = ParseWord(wordText, meaning);
}
catch (FormatException e)
{
Errors.Add(e.Message);
return null;
}
project.Segmenter.Segment(word);
_parsedWords.Add(wordText, word);
return word;
}
protected Word ParseWord(string wordText, Meaning meaning)
{
int stemStartIdx = wordText.IndexOf("|", StringComparison.Ordinal);
int stemEndIdx = wordText.LastIndexOf("|", StringComparison.Ordinal) - 1; // -1 because we're going to remove the leading |
if (stemStartIdx != -1 && stemEndIdx < stemStartIdx)
{
// Only way this can happen is if there was only a single "|" in the word
throw new FormatException(String.Format("Words should have either 0 or 2 pipe characters representing word stems. Offending word: {0}", wordText));
}
var word = (stemStartIdx == -1) ?
new Word(wordText, meaning) :
new Word(wordText.Replace("|", ""), stemStartIdx, stemEndIdx - stemStartIdx, meaning);
return word;
}
// Should this be protected? But the unit test needs to call it.
public void SetUpProject()
{
if (ConfigData != null && ConfigFilename != null)
{
Warnings.Add("WARNING: options --config-data and --config-file were both specified. Ignoring --config-file.");
ConfigFilename = null;
}
if (ConfigData == null && ConfigFilename == null)
{
ConfigFilename = FindConfigFilename();
// If ConfigFilename is STILL null at this point, it's because no config files were found at all, so we'll use the default one from the resource.
}
SpanFactory = new ShapeSpanFactory();
SegmentPool = new SegmentPool();
Variety1 = new Variety("variety1");
Variety2 = new Variety("variety2");
Meaning = new Meaning("gloss1", "cat1");
if (ConfigData == null && ConfigFilename == null)
Project = GetProjectFromResource(SpanFactory, SegmentPool);
else if (ConfigData != null)
Project = GetProjectFromXmlString(SpanFactory, SegmentPool, ConfigData);
else if (ConfigFilename != null)
Project = GetProjectFromFilename(SpanFactory, SegmentPool, ConfigFilename);
else // Should never get here given checks above, but let's be safe and write the check anyway
Project = GetProjectFromResource(SpanFactory, SegmentPool);
Project.Meanings.Add(Meaning);
Project.Varieties.Add(Variety1);
Project.Varieties.Add(Variety2);
Project.VarietyPairs.Add(new VarietyPair(Variety1, Variety2));
}
public Word(string strRep, Meaning meaning)
: this(strRep, 0, strRep.Length, meaning)
{
}
public void Import(object importSettingsViewModel, Stream stream, CogProject project)
{
XDocument doc = XDocument.Load(stream, LoadOptions.SetLineInfo);
XElement root = doc.Element("survey");
if (root == null)
throw new ImportException("No survey element.");
var varietyNames = new HashSet<string>();
var varieties = new Dictionary<string, Tuple<Variety, List<Word>>>();
foreach (XElement wordListElem in root.Elements("word_lists").Elements("word_list"))
{
var id = (string) wordListElem.Attribute("id");
if (id == null)
throw new ImportException(string.Format("A \"word_list\" element is missing an \"id\" attribute. Line: {0}", ((IXmlLineInfo) wordListElem).LineNumber));
if (varieties.ContainsKey(id))
throw new ImportException(string.Format("The ID of a \"word_list\" element is not unique. Line: {0}", ((IXmlLineInfo) wordListElem).LineNumber));
XElement nameElem = wordListElem.Element("name");
if (nameElem == null)
throw new ImportException(string.Format("A \"word_list\" element is missing a \"name\" element. Line: {0}", ((IXmlLineInfo) wordListElem).LineNumber));
var name = ((string) nameElem).Trim();
if (string.IsNullOrEmpty(name))
throw new ImportException(string.Format("A blank variety name is not allowed. Line: {0}", ((IXmlLineInfo) nameElem).LineNumber));
if (varietyNames.Contains(name))
throw new ImportException(string.Format("The variety name, \"{0}\", is not unique. Line: {1}", name, ((IXmlLineInfo) wordListElem).LineNumber));
varietyNames.Add(name);
var variety = new Variety(name);
varieties[id] = Tuple.Create(variety, new List<Word>());
}
var meanings = new Dictionary<string, Meaning>();
foreach (XElement glossElem in root.Elements("glosses").Elements("gloss"))
{
XElement nameElem = glossElem.Element("name");
if (nameElem == null)
throw new ImportException(string.Format("A \"gloss\" element is missing a \"name\" element. Line: {0}", ((IXmlLineInfo) glossElem).LineNumber));
var gloss = ((string) nameElem).Trim();
if (string.IsNullOrEmpty(gloss))
throw new ImportException(string.Format("A blank gloss is not allowed. Line: {0}", ((IXmlLineInfo) nameElem).LineNumber));
var pos = (string) glossElem.Element("part_of_speech");
if (meanings.ContainsKey(gloss))
throw new ImportException(string.Format("The gloss, \"{0}\", is not unique. Line: {1}", gloss, ((IXmlLineInfo) nameElem).LineNumber));
var meaning = new Meaning(gloss, pos);
meanings[gloss] = meaning;
foreach (XElement transElem in glossElem.Elements("transcriptions").Elements("transcription"))
{
XElement wordListIdElem = transElem.Element("word_list_id");
if (wordListIdElem == null)
throw new ImportException(string.Format("A \"transcription\" element is missing a \"word_list_id\" element. Line: {0}", ((IXmlLineInfo) transElem).LineNumber));
var varietyID = (string) wordListIdElem;
var wordform = (string) transElem.Element("name");
if (wordform != null)
{
wordform = wordform.Trim();
if (!string.IsNullOrEmpty(wordform))
{
Tuple<Variety, List<Word>> variety;
if (varieties.TryGetValue(varietyID, out variety))
{
foreach (string w in wordform.Split(','))
{
string str = w.Trim();
variety.Item2.Add(new Word(str, meaning));
}
}
}
}
}
}
project.Meanings.ReplaceAll(meanings.Values);
using (project.Varieties.BulkUpdate())
{
project.Varieties.Clear();
foreach (Tuple<Variety, List<Word>> variety in varieties.Values)
{
variety.Item1.Words.AddRange(variety.Item2);
project.Varieties.Add(variety.Item1);
}
}
}
public static IEnumerable <Cluster <Word> > GenerateCognateSets(this CogProject project, Meaning meaning)
{
var words = new HashSet <Word>();
var noise = new HashSet <Word>();
foreach (VarietyPair vp in project.VarietyPairs)
{
WordPair wp;
if (vp.WordPairs.TryGet(meaning, out wp))
{
if (wp.Cognacy)
{
words.Add(wp.Word1);
words.Add(wp.Word2);
noise.Remove(wp.Word1);
noise.Remove(wp.Word2);
}
else
{
if (!words.Contains(wp.Word1))
{
noise.Add(wp.Word1);
}
if (!words.Contains(wp.Word2))
{
noise.Add(wp.Word2);
}
}
}
}
double min = double.MaxValue;
var distanceMatrix = new Dictionary <UnorderedTuple <Word, Word>, double>();
Word[] wordArray = words.ToArray();
for (int i = 0; i < wordArray.Length; i++)
{
for (int j = i + 1; j < wordArray.Length; j++)
{
Word w1 = wordArray[i];
Word w2 = wordArray[j];
double score = 0;
WordPair wp;
if (w1.Variety != w2.Variety && w1.Variety.VarietyPairs[w2.Variety].WordPairs.TryGet(meaning, out wp) && wp.Cognacy &&
wp.GetWord(w1.Variety) == w1 && wp.GetWord(w2.Variety) == w2)
{
score = wp.PredictedCognacyScore;
}
double distance = 1.0 - score;
min = Math.Min(min, distance);
distanceMatrix[UnorderedTuple.Create(w1, w2)] = distance;
}
}
var clusterer = new FlatUpgmaClusterer <Word>((w1, w2) => distanceMatrix[UnorderedTuple.Create(w1, w2)], (1.0 + min) / 2);
return(clusterer.GenerateClusters(words).Concat(new Cluster <Word>(noise, true)));
}
