public override string Extract(Stream stream, TextExtractorContext context)
{
try
{
// extract text using IFilter
return(SnIFilter.GetText(stream, ".pdf"));
}
catch (OutOfMemoryException ex)
{
SnLog.WriteWarning("Pdf text extract failed with out of memory exception. " + ex,
EventId.Indexing,
properties: new Dictionary <string, object> {
{ "Stream size", stream.Length }
});
return(string.Empty);
}
catch (Exception ex)
{
// log iFilter error only once
if (!_iFilterErrorLogged)
{
SnLog.WriteWarning("Pdf IFilter error: " + ex.Message, EventId.Indexing);
_iFilterErrorLogged = true;
}
}
// fallback to the other mechanism in case the pdf IFilter is missing
var text = new StringBuilder();
try
{
var pdfReader = new PdfReader(stream);
for (var page = 1; page <= pdfReader.NumberOfPages; page++)
{
// extract text using the old version (4.1.6) of iTextSharp
var pageText = ExtractTextFromPdfBytes(pdfReader.GetPageContent(page));
if (string.IsNullOrEmpty(pageText))
{
continue;
}
text.Append(pageText);
}
}
catch (OutOfMemoryException ex)
{
SnLog.WriteWarning("Pdf text extract failed with out of memory exception. " + ex,
EventId.Indexing,
properties: new Dictionary <string, object> {
{ "Stream size", stream.Length }
});
}
return(text.ToString());
}
public override string Extract(Stream stream, TextExtractorContext context)
{
try
{
// IFilter
return(SnIFilter.GetText(stream, ".msg"));
}
catch (Exception ex)
{
SnLog.WriteWarning("Msg IFilter error: " + ex.Message, EventId.Indexing);
}
return(string.Empty);
}