[poppler] poppler really slow when reading some documents

[Leonard Rosenthol]

At 10:43 PM 1/4/2006, Jonathan Blandford wrote:
>The algorithm you described will catch this particular instance, but
>might not catch other such errors (such as a missing ')'.)

         True.

         The best thing is actually to have your object parser be 
aware of the xref.  If you go past the end of the object (as defined 
by the values in the xref), then you know you went to far.  So you 
can then back up and try again (via a variety of different approaches).


>I'm wondering if it also makes sense to put a cap on string sizes for
>certain fields.

         You could do that also, but be aware that the max string 
size for PDF 1.5 and 1.6 is now HUGE...


Leonard

---------------------------------------------------------------------------
Leonard Rosenthol                            <mailto:leonardr at pdfsages.com>
Chief Technical Officer                      <http://www.pdfsages.com>
PDF Sages, Inc.                              215-938-7080 (voice)
                                              215-938-0880 (fax)