//Parse document(option=['document'->.docx, .pdf] or ['txt'->.txt])
//Removes punctuation and returns the words of the document in an array of strings
public string[] GetText(string path, string option)
{
string text = null;
try
{
ParserContext context = new ParserContext(path);
if (option.Equals("txt"))
{
ITextParser parser = ParserFactory.CreateText(context);
text = parser.Parse().ToString().ToLower().Replace('\n', ' ').Replace('\r', ' ')
.Replace('\t', ' ');
}
else if (option.Equals("document"))
{
IDocumentParser parser = ParserFactory.CreateDocument(context);
text = parser.Parse().ToString().ToLower().Replace('\n', ' ').Replace('\r', ' ')
.Replace('\t', ' ');
}
}
catch (Exception e)
{
Console.WriteLine("Exception found");
Console.WriteLine(e.Message);
}
text = RemovePunctuation(text);
string[] words = text.Split(default(Char[]), StringSplitOptions.RemoveEmptyEntries);
return(words);
}
public void TestParseSimpleDocumentFromWord()
{
ParserContext context = new ParserContext(TestDataSample.GetWordPath("SampleDoc.docx"));
IDocumentParser parser = ParserFactory.CreateDocument(context);
ToxyDocument doc = parser.Parse();
Assert.AreEqual(7, doc.Paragraphs.Count);
Assert.AreEqual("I am a test document", doc.Paragraphs[0].Text);
Assert.AreEqual("This is page 1", doc.Paragraphs[1].Text);
Assert.AreEqual("I am Calibri (Body) in font size 11", doc.Paragraphs[2].Text);
Assert.AreEqual("\n", doc.Paragraphs[3].Text);
Assert.AreEqual("This is page two", doc.Paragraphs[4].Text);
Assert.AreEqual("It’s Arial Black in 16 point", doc.Paragraphs[5].Text);
Assert.AreEqual("It’s also in blue", doc.Paragraphs[6].Text);
}
public void TestParseDocumentWithTable()
{
ParserContext context = new ParserContext(TestDataSample.GetWordPath("simple-table.docx"));
IDocumentParser parser = ParserFactory.CreateDocument(context);
ToxyDocument doc = parser.Parse();
Assert.AreEqual(8, doc.Paragraphs.Count);
Assert.AreEqual("This is a Word document that was created using Word 97 – SR2. It contains a paragraph, a table consisting of 2 rows and 3 columns and a final paragraph.",
doc.Paragraphs[0].Text);
Assert.AreEqual("This text is below the table.", doc.Paragraphs[1].Text);
Assert.AreEqual("Cell 1,1", doc.Paragraphs[2].Text);
Assert.AreEqual("Cell 1,2", doc.Paragraphs[3].Text);
Assert.AreEqual("Cell 1,3", doc.Paragraphs[4].Text);
Assert.AreEqual("Cell 2,1", doc.Paragraphs[5].Text);
Assert.AreEqual("Cell 2,2", doc.Paragraphs[6].Text);
Assert.AreEqual("Cell 2,3", doc.Paragraphs[7].Text);
}
public string ExtractText(string filePath, string extension)
{
ParserContext c = new ParserContext(filePath);
try
{
IDocumentParser parser = ParserFactory.CreateDocument(c);
ToxyDocument result = parser.Parse();
return(result.ToString());
}
catch (InvalidDataException)
{
Console.Error.WriteLine($"'{filePath}' is supported but don't have the required extension.");
var newFilePath = $"{filePath}.{extension}";
Console.Error.WriteLine($"Creating a copy in '{newFilePath}' and using that to read.");
File.Copy(filePath, newFilePath);
return(ExtractText(newFilePath, extension));
}
catch (Exception e)
{
Console.Error.WriteLine("{0} Exception caught error with {1}.", e, filePath);
return(null);
}
}