Hi,

I'm wondering if anyone has come across this
problem. I use PDFlib-4.0.1 with PHP-4.0.6
on Linux to create piecharts, barcharts,
trend charts, etc. I can open these PDFs just
fine from the command line using xpdf,
acroread, gs, gimp, etc. I can convert the
PDFs to Postscript just fine with pdftops,
again, when executed from the command line.
However, if I try to use exec() or system()
from within a PHP script to do these exact
same things, I get the error message:

Error (0): PDF file is damaged - attempting to reconstruct xref table...
Error: Couldn't find trailer dictionary
Error: Couldn't read xref table

I used a number of different techniques
(including file(<pdf filename>) and
readfile(<pdf filename>)) to try to figure
out what is going on. It appears that any
attempt to access the file from within a
PHP script in this fashion causes the file
to be truncated before the xref table is
reached, hence the error. The xref table is
fine in the original file. I don't see any
reason for this to be happening, though. Is
it possible that the makers of PDF lib have
done something to the file to keep a user
from performing such operations from within
a PHP script?

Any help anyone can provide would be MUCH
appreciated. Thanks.

Michael Mann

    I should add that PDF files created with FPDF
    or from some other source do not behave this
    way, and I've made sure that the PDF files
    were properly closed before trying to access
    them in any way.

      Write a Reply...