/// <summary>
///
/// </summary>
/// <param name="phrase"></param>
/// <returns></returns>
public static DocumentParser FromPhrase(string phrase)
{
if (phrase.Length <= 0)
{
throw new Exception("The phrase is empty!");
}
// make sure CoreNLP is initialized
if (!VerifyStanfordServiceIsInitialized())
{
throw new Exception("Failed to initialize CoreNLP!");
}
StanfordDocumentFacade doc =
StanfordPipelineService.Instance.Annotate(phrase);
if (doc == null)
{
return(null);
}
DocumentParser docParser = new DocumentParser(doc);
if (!docParser.InitForParsing())
{
throw new Exception("Empty phrase?!");
}
// #fromphrase
docParser._fromFile = false;
return(docParser);
}
//--------------------------------------------------------------------------
/// <summary>
/// This is how you construct a new DocumentParser
/// </summary>
/// <param name="file"></param>
/// <returns></returns>
public static DocumentParser FromFile(FileInfo file)
{
if (!file.Exists)
{
throw new Exception("The document file doesn't exist!");
}
// make sure CoreNLP is initialized
if (!VerifyStanfordServiceIsInitialized())
{
throw new Exception("Failed to initialize CoreNLP!");
}
StanfordDocumentFacade doc =
StanfordPipelineService.Instance.Annotate(file);
if (doc == null)
{
return(null);
}
DocumentParser docParser = new DocumentParser(doc);
if (!docParser.ParseMetaData())
{
throw new Exception("Failed to get document's start mark!");
}
if (!docParser.CheckDocumentValidityAfterMetaData())
{
throw new Exception(
"The document is invalid - either no content or no end mark!");
}
if (!docParser.SetGutenbergId(file))
{
throw new Exception("Failed setting Gutenberg Id!");
}
docParser._fromFile = true;
return(docParser);
}