/// <summary>This will do the escaping on an input file.</summary>
/// <remarks>
/// This will do the escaping on an input file. Input file should already be tokenized,
/// with tokens separated by whitespace. <br />
/// Usage: java edu.stanford.nlp.process.PTBEscapingProcessor fileOrUrl
/// </remarks>
/// <param name="args">Command line argument: a file or URL</param>
public static void Main(string[] args)
{
if (args.Length != 1)
{
System.Console.Out.WriteLine("usage: java edu.stanford.nlp.process.PTBEscapingProcessor fileOrUrl");
return;
}
string filename = args[0];
try
{
IDocument <string, Word, Word> d;
// initialized below
if (filename.StartsWith("http://"))
{
IDocument <string, Word, Word> dpre = new BasicDocument <string>(WhitespaceTokenizer.Factory()).Init(new URL(filename));
IDocumentProcessor <Word, Word, string, Word> notags = new StripTagsProcessor <string, Word>();
d = notags.ProcessDocument(dpre);
}
else
{
d = new BasicDocument <string>(WhitespaceTokenizer.Factory()).Init(new File(filename));
}
IDocumentProcessor <Word, IHasWord, string, Word> proc = new Edu.Stanford.Nlp.Process.PTBEscapingProcessor <Word, string, Word>();
IDocument <string, Word, IHasWord> newD = proc.ProcessDocument(d);
foreach (IHasWord word in newD)
{
System.Console.Out.WriteLine(word);
}
}
catch (Exception e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
/// <summary>For internal debugging purposes only.</summary>
public static void Main(string[] args)
{
new BasicDocument <string>();
IDocument <string, Word, Word> htmlDoc = BasicDocument.Init("top text <h1>HEADING text</h1> this is <p>new paragraph<br>next line<br/>xhtml break etc.");
System.Console.Out.WriteLine("Before:");
System.Console.Out.WriteLine(htmlDoc);
IDocument <string, Word, Word> txtDoc = new Edu.Stanford.Nlp.Process.StripTagsProcessor <string, Word>(true).ProcessDocument(htmlDoc);
System.Console.Out.WriteLine("After:");
System.Console.Out.WriteLine(txtDoc);
IDocument <string, Word, IList <Word> > sentences = new WordToSentenceProcessor <Word>().ProcessDocument(txtDoc);
System.Console.Out.WriteLine("Sentences:");
System.Console.Out.WriteLine(sentences);
}
/// <summary>This will print out some text, recognizing tags.</summary>
/// <remarks>
/// This will print out some text, recognizing tags. It can be used to
/// test tag breaking. <br /> Usage: <code>
/// java edu.stanford.nlp.process.WordToTaggedWordProcessor fileOrUrl
/// </code>
/// </remarks>
/// <param name="args">Command line argument: a file or URL</param>
public static void Main(string[] args)
{
if (args.Length != 1)
{
System.Console.Out.WriteLine("usage: java edu.stanford.nlp.process.WordToTaggedWordProcessor fileOrUrl");
System.Environment.Exit(0);
}
string filename = args[0];
try
{
IDocument <IHasWord, Word, Word> d;
if (filename.StartsWith("http://"))
{
IDocument <IHasWord, Word, Word> dpre = new BasicDocument <IHasWord>().Init(new URL(filename));
IDocumentProcessor <Word, Word, IHasWord, Word> notags = new StripTagsProcessor <IHasWord, Word>();
d = notags.ProcessDocument(dpre);
}
else
{
d = new BasicDocument <IHasWord>().Init(new File(filename));
}
IDocumentProcessor <Word, IHasWord, IHasWord, Word> proc = new Edu.Stanford.Nlp.Process.WordToTaggedWordProcessor <Word, IHasWord, Word>();
IDocument <IHasWord, Word, IHasWord> sentd = proc.ProcessDocument(d);
// System.out.println(sentd);
int i = 0;
foreach (IHasWord w in sentd)
{
System.Console.Out.WriteLine(i + ": " + w);
i++;
}
}
catch (Exception e)
{
Sharpen.Runtime.PrintStackTrace(e);
}
}
/// <summary>Loads a text or html file from a file path or URL.</summary>
/// <remarks>
/// Loads a text or html file from a file path or URL. Treats anything
/// beginning with <tt>http:\\</tt>,<tt>.htm</tt>, or <tt>.html</tt> as an
/// html file, and strips all tags from the document
/// </remarks>
public virtual void LoadFile(string filename)
{
if (filename == null)
{
return;
}
File file = new File(filename);
string urlOrFile = filename;
// if file can't be found locally, try prepending http:// and looking on web
if (!file.Exists() && filename.IndexOf("://") == -1)
{
urlOrFile = "http://" + filename;
}
else
{
// else prepend file:// to handle local html file urls
if (filename.IndexOf("://") == -1)
{
urlOrFile = "file://" + filename;
}
}
// TODO: why do any of this instead of just reading the file? THIS SHOULD BE UPDATED FOR 2017!
// Also, is this working correctly still?
// load the document
IDocument <object, Word, Word> doc;
try
{
if (urlOrFile.StartsWith("http://") || urlOrFile.EndsWith(".htm") || urlOrFile.EndsWith(".html"))
{
// strip tags from html documents
IDocument <object, Word, Word> docPre = new BasicDocument <object>().Init(new URL(urlOrFile));
IDocumentProcessor <Word, Word, object, Word> noTags = new StripTagsProcessor <object, Word>();
doc = noTags.ProcessDocument(docPre);
}
else
{
doc = new BasicDocument <object>(Edu.Stanford.Nlp.Parser.UI.ParserPanel.GetTokenizerFactory()).Init(new InputStreamReader(new FileInputStream(filename), encoding));
}
}
catch (Exception e)
{
JOptionPane.ShowMessageDialog(this, "Could not load file " + filename + "\n" + e, null, JOptionPane.ErrorMessage);
Sharpen.Runtime.PrintStackTrace(e);
SetStatus("Error loading document");
return;
}
// load the document into the text pane
StringBuilder docStr = new StringBuilder();
foreach (Word aDoc in doc)
{
if (docStr.Length > 0)
{
docStr.Append(' ');
}
docStr.Append(aDoc.ToString());
}
textPane.SetText(docStr.ToString());
dataFileLabel.SetText(urlOrFile);
HighlightSentence(0);
forwardButton.SetEnabled(endIndex != textPane.GetText().Length - 1);
// scroll to top of document
textPane.SetCaretPosition(0);
SetStatus("Done");
}
public static Document FromBasic(BasicDocument basicDocument)
{
return(new Document(basicDocument));
}
private Document(BasicDocument document) : base(document)
{
}
public static void ShouldHaveOneResultMatching(this List <BasicDocument> result, BasicDocument expected)
{
result.Should().HaveCount(1);
result.First().Id.Should().Be(expected.Id);
}
