8

When I compile the following document:

\documentclass[10pt, oneside, a4paper]{report}
\begin{document}
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
\end{document}

where the number of x's is greater than 32765, Adobe Reader is not able to read the document anymore, and just produces a blank page. Has anybody come across this before?

Improve this question
5
  • 3
    The actual limit seems to be 32767 = 2^{15}-1; other PDF viewers don't choke even with 1073741 characters (Skim on Mac OS X).
    – egreg
    Sep 29, 2011 at 13:07
  • 3
    Have you tried different versions of Adobe Reader/Acrobat?
    – einpoklum
    Sep 29, 2011 at 14:50
  • 16
    How the heck did you come across this behavior?
    – doncherry
    Sep 29, 2011 at 15:04
  • 2
    In the anglophone world, it is standard that the . in 32.765 would be interpreted as a decimal point. That is, 32.765 is less than 33. The typical thousands separator is the comma ,. The numbers here are not so big, so I edited the post to just remove the separator. I believe this is the least ambiguous approach.
    – Seamus
    Sep 29, 2011 at 15:12
  • @doncherry Where I work, we use LaTeX for document generation. One of the unit tests produce a very funky report, including such a string. The reason for the string, is actually to test what happens when generating an excel-sheet, where the maximum number of characters in a cell is larger than 32.767.
    – utdiscant
    Sep 30, 2011 at 17:49

2 Answers 2

Reset to default
12

The implementation limit for the length of strings in content streams for readers on 32-bit architectures is 32767 according to ISO 32000-2008, Table C.1, so 32766 and 32767 should probably also work.

Improve this answer
8
  • 2
    I tried indeed with \romannumeral 32767000 to print a string consisting of 32677 m's and Reader opened it; with 32768 it didn't. Probably Reader is enforcing the limit, while other programs don't (see my comment above).
    – egreg
    Sep 29, 2011 at 17:45
  • 1
    @egreg: It doesn't have to enforce it: the limit is simply the length that any reader has to support - it's free to support longer strings, but producers are warned against creating them.
    – Martin Schröder
    Sep 29, 2011 at 20:02
  • 1
    Said in other words: Adobe people decided to not support longer strings and they have very good reasons.
    – egreg
    Sep 29, 2011 at 20:29
  • @egreg: Well, the implementation limits of 32 bit Adobe software are simply quite old; e.g. floats are limited to singe precision.
    – Martin Schröder
    Sep 29, 2011 at 20:37
  • @egreg: Why would Adobe people enforce such limits? What good reasons do you have in mind?
    – Count Zero
    Sep 30, 2011 at 12:05
1

I tried Adobe Reader X, it is OK. I think you may need to update your software.

Improve this answer
2
  • I used Adobe Reader X myself, and I get the problem.
    – utdiscant
    Sep 30, 2011 at 17:45
  • Seeing the problem with Acrobat X on Win XP
    – einpoklum
    Oct 2, 2011 at 16:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.