public QuoteAttributionAnnotator(Properties props)
{
// settings
// these paths go in the props file
// fields
Verbose = PropertiesUtils.GetBool(props, "verbose", false);
Timing timer = null;
CorefPath = props.GetProperty("booknlpCoref", null);
if (CorefPath == null && Verbose)
{
log.Err("Warning: no coreference map!");
}
ModelPath = props.GetProperty("modelPath", DefaultModelPath);
CharactersFile = props.GetProperty("charactersPath", null);
if (CharactersFile == null && Verbose)
{
log.Err("Warning: no characters file!");
}
qmSieveList = props.GetProperty("QMSieves", DefaultQmsieves);
msSieveList = props.GetProperty("MSSieves", DefaultMssieves);
if (Verbose)
{
timer = new Timing();
log.Info("Loading QuoteAttribution coref [" + CorefPath + "]...");
log.Info("Loading QuoteAttribution characters [" + CharactersFile + "]...");
}
// loading all our word lists
FamilyWordList = props.GetProperty("familyWordsFile", FamilyWordList);
AnimacyWordList = props.GetProperty("animacyWordsFile", AnimacyWordList);
GenderWordList = props.GetProperty("genderNamesFile", GenderWordList);
familyRelations = QuoteAttributionUtils.ReadFamilyRelations(FamilyWordList);
genderMap = QuoteAttributionUtils.ReadGenderedNounList(GenderWordList);
animacyList = QuoteAttributionUtils.ReadAnimacyList(AnimacyWordList);
if (characterMap != null)
{
characterMap = QuoteAttributionUtils.ReadPersonMap(CharactersFile);
}
else
{
buildCharacterMapPerAnnotation = true;
}
// use Stanford CoreNLP coref to map mentions to canonical mentions
useCoref = PropertiesUtils.GetBool(props, "useCoref", useCoref);
if (Verbose)
{
timer.Stop("done.");
}
}
private void Extract()
{
log.Info("content type: " + editorPane.GetContentType());
if (!editorPane.GetContentType().Equals("text/html"))
{
DefaultStyledDocument doc = (DefaultStyledDocument)editorPane.GetDocument();
string text = null;
try
{
text = doc.GetText(0, doc.GetLength());
}
catch (Exception e)
{
log.Err(e);
}
string labeledText = classifier.ClassifyWithInlineXML(text);
taggedContents = labeledText;
ICollection <string> tags = classifier.Labels();
string background = classifier.BackgroundSymbol();
StringBuilder tagPattern = new StringBuilder();
foreach (string tag in tags)
{
if (background.Equals(tag))
{
continue;
}
if (tagPattern.Length > 0)
{
tagPattern.Append('|');
}
tagPattern.Append(tag);
}
Pattern startPattern = Pattern.Compile("<(" + tagPattern + ")>");
Pattern endPattern = Pattern.Compile("</(" + tagPattern + ")>");
string finalText = labeledText;
Matcher m = startPattern.Matcher(finalText);
while (m.Find())
{
int start = m.Start();
finalText = m.ReplaceFirst(string.Empty);
m = endPattern.Matcher(finalText);
if (m.Find())
{
int end = m.Start();
string tag_1 = m.Group(1);
finalText = m.ReplaceFirst(string.Empty);
IAttributeSet attSet = GetAttributeSet(tag_1);
try
{
string entity = Sharpen.Runtime.Substring(finalText, start, end);
doc.SetCharacterAttributes(start, entity.Length, attSet, false);
}
catch (Exception ex)
{
log.Err(ex);
System.Environment.Exit(-1);
}
log.Info(tag_1 + ": " + Sharpen.Runtime.Substring(finalText, start, end));
}
else
{
log.Info("Couldn't find end pattern!");
}
m = startPattern.Matcher(finalText);
}
editorPane.Revalidate();
editorPane.Repaint();
}
else
{
string untaggedContents = editorPane.GetText();
if (untaggedContents == null)
{
untaggedContents = string.Empty;
}
taggedContents = classifier.ClassifyWithInlineXML(untaggedContents);
ICollection <string> tags = classifier.Labels();
string background = classifier.BackgroundSymbol();
StringBuilder tagPattern = new StringBuilder();
foreach (string tag in tags)
{
if (background.Equals(tag))
{
continue;
}
if (tagPattern.Length > 0)
{
tagPattern.Append('|');
}
tagPattern.Append(tag);
}
Pattern startPattern = Pattern.Compile("<(" + tagPattern + ")>");
Pattern endPattern = Pattern.Compile("</(" + tagPattern + ")>");
string finalText = taggedContents;
Matcher m = startPattern.Matcher(finalText);
while (m.Find())
{
string tag_1 = m.Group(1);
Color col = tagToColorMap[tag_1];
if (col != null)
{
string color = ColorToHTML(col);
string newTag = "<span style=\"background-color: " + color + "; color: white\">";
finalText = m.ReplaceFirst(newTag);
int start = m.Start() + newTag.Length;
Matcher m1 = endPattern.Matcher(finalText);
if (m1.Find(m.End()))
{
string entity = Sharpen.Runtime.Substring(finalText, start, m1.Start());
log.Info(tag_1 + ": " + entity);
}
else
{
log.Warn("Failed to find end for " + tag_1);
}
finalText = m1.ReplaceFirst("</span>");
m = startPattern.Matcher(finalText);
}
}
// System.out.println(finalText);
editorPane.SetText(finalText);
editorPane.Revalidate();
editorPane.Repaint();
}
// log.info(finalText);
saveTaggedAs.SetEnabled(true);
}