/// <summary>
/// Parse the document searching for sentences where the entity found.
/// Returns a csv line with the file, the entity the sentence and the sintax analisis of the sentences
/// </summary>
/// <param name="text">Document text</param>
/// <param name="entity">Entity.</param>
/// <param name="origFile">Original file.</param>
public static List <string[]> Parse(string text, string entity, string origFile, string language)
{
var results = new List <string[]>();
//Load spanish models.
var modelsDirectory = StanfordEnv.PARSER_MODELS;
var lexparserDirectory = modelsDirectory + StanfordEnv.GetParserLanguageFiles(language);
var lp = LexicalizedParser.loadModel(lexparserDirectory);
string[] splittedText = SplitText(text);
List <string> entityLines = GetEntitiesLines(splittedText, entity);
foreach (var line in entityLines)
{
//Parser sentence.
var tokenizerFactory = PTBTokenizer.factory(new CoreLabelTokenFactory(), "");
var sent2Reader = new java.io.StringReader(line);
var rawWords2 = tokenizerFactory.getTokenizer(sent2Reader).tokenize();
sent2Reader.close();
var tree2 = lp.apply(rawWords2);
results.Add(new string[] { origFile, entity, line, tree2.ToString() });
}
return(results);
}
public static string MediaWikiToXHTML(string markup)
{
string retVal = null;
using (XhtmlPrinter printer = new XhtmlPrinter())
{
java.io.Reader rdr = new java.io.StringReader(markup);
// org.wikimodel.wem.WikiPrinter wp = new org.wikimodel.wem.WikiPrinter ();
// var listener = new org.wikimodel.wem.xwiki.XWikiSerializer(wp);
// var listener = new org.wikimodel.wem.xwiki.XWikiSerializer(printer);
org.wikimodel.wem.IWemListener listener = new org.wikimodel.wem.xhtml.PrintListener(printer);
org.wikimodel.wem.mediawiki.MediaWikiParser mep =
new org.wikimodel.wem.mediawiki.MediaWikiParser();
mep.parse(rdr, listener);
retVal = printer.Text;
rdr.close();
rdr = null;
listener = null;
mep = null;
} // End Using printer
return(retVal);
}
public bool Run()
{
string code = editor.Document.TextContent;
// check if we already have compiled this code before
int hashCode = code.GetHashCode();
if (database != null && this.hashCode == hashCode)
{
return(false);
}
this.hashCode = hashCode;
java.io.StringReader reader = new java.io.StringReader(code);
try
{
java.io.StringWriter writer = new java.io.StringWriter();
try
{
java.io.PrintWriter log = new java.io.PrintWriter(writer);
try
{
Database db;
db = JPortal.run(name, reader, log);
if (db != null)
{
database = db;
}
}
finally
{
log.flush();
log.close();
}
string result = writer.ToString();
char[] sep = { '\n' };
string[] lines = result.Split(sep);
for (int i = 0; i < lines.Length; i++)
{
string line = lines[i].Trim();
if (line.Length == 0)
{
continue;
}
form.LogInfo = line;
}
}
finally
{
writer.close();
}
}
finally
{
reader.close();
}
return(database != null);
}
public static List <string> ExtractNounsFromSemantics(string sentence)
{
string assemblyPath = Assembly.GetExecutingAssembly().GetName().CodeBase;
string projectPath = Directory.GetParent(new Uri(Path.GetDirectoryName(Path.GetDirectoryName(Path.GetDirectoryName(assemblyPath)))).LocalPath).FullName;
string modelsDirectory = Path.GetFullPath(projectPath + @"\Parser\CoreNLP-3.9.1-Models\edu\stanford\nlp\models");
// Loading english PCFG parser from file
LexicalizedParser lp = LexicalizedParser.loadModel(modelsDirectory + @"\lexparser\englishPCFG.ser.gz");
// This shows loading and using an explicit tokenizer
var tokenizerFactory = PTBTokenizer.factory(new CoreLabelTokenFactory(), "");
var sent2Reader = new java.io.StringReader(sentence);
var rawWords = tokenizerFactory.getTokenizer(sent2Reader).tokenize();
sent2Reader.close();
var tree = lp.apply(rawWords);
return(tree.toArray().Cast <LabeledScoredTreeNode>().Where(n => n.isLeaf() && nounLabels.Contains(n.parent(tree).label().value())).Select(n => n.label().ToString()).ToList());
}