iTextSharp exception: PDF header signature not found

Question

I'm using iTextSharp to read the contents of PDF documents:

PdfReader reader = new PdfReader(pdfPath);
using (StringWriter output = new StringWriter())
{
    for (int i = 1; i <= reader.NumberOfPages; i++)
        output.WriteLine(PdfTextExtractor.GetTextFromPage(reader, i, new SimpleTextExtractionStrategy()));

    reader.Close();
    pdfText = output.ToString();
}

99% of the time it works just fine. However, there is this one PDF file that will sometimes throw this exception:

PDF header signature not found. StackTrace: at
iTextSharp.text.pdf.PRTokeniser.CheckPdfHeader() at
iTextSharp.text.pdf.PdfReader.ReadPdf() at
iTextSharp.text.pdf.PdfReader..ctor(String filename, Byte[]> ownerPassword) at
Reader.PDF.DownloadPdf(String url) in

What's annoying is that I can't always reproduce the error. Sometimes it works, sometimes it doesn't. Has anyone encountered this problem?

Solution 1

After some research, I've found that this problem relates to either a file being corrupted during PDF generation, or an error related to an object in the document that doesn't conform to the PDF standard as implemented in iTextSharp. It also seems to happen only when you read from a PDF file from disk.

I have not found a complete solution to the problem, but only a workaround. What I've done is read the PDF document using the PdfReader itextsharp object and see if an error or exception happens before reading the file in a normal operation.

So running something similar to this:

private bool IsValidPdf(string filepath)
{
    bool Ret = true;

    PdfReader reader = null;

    try
    {
        reader = new PdfReader(filepath);
    }
    catch
    {
        Ret = false;
    }

    return Ret;
}

Solution 2

I found it was because I was calling new PdfReader(pdf) with the PDF stream position at the end of the file. By setting the position to zero it resolved the issue.

Before:

// Throws: InvalidPdfException: PDF header signature not found.
var pdfReader = new PdfReader(pdf);

After:

// Works correctly.
pdf.Position = 0;
var pdfReader = new PdfReader(pdf);

Comments

[nishanth]

It took me a while, but I finally figured out that the file was indeed corrupted. The blame goes to the website that created the PDF, and not a bug with iTextSharp. Thanks for taking the time to answer my question.

[JoshBerke]

I can confirm this can occur if you load the reader from a steam as well as from disk:)

[Anthony Horne]

Turns out I was looking at the wrong file in my case. The filename was referring to one of the assets (images) I was previously using, i.e. a jpg is not a pdf, doh :) , so it was in fact a corrupted PDF (or not one at all). Thanks - got me on the right path.

[Guilherme Batista]

This answer helped me a lot today. I recommend to put a finally block with reader.Close() to prevent files beeing locked by the process

[KMR]

This does not provide an answer to the question. Once you have sufficient reputation you will be able to comment on any post; instead, provide answers that don't require clarification from the asker. - From Review