PDFsharp & MigraDoc Foundation https://forum.pdfsharp.net/ |
|
Bug + patch: Missing parenthesis causes infinite loop https://forum.pdfsharp.net/viewtopic.php?f=3&t=3411 |
Page 1 of 1 |
Author: | Gerben Vos [ Thu Aug 04, 2016 2:49 pm ] | ||
Post subject: | Bug + patch: Missing parenthesis causes infinite loop | ||
What happens: http://www.stillhq.com/pdfdb/000590/data.pdf causes pdfsharp to loop infinitely until it runs out of memory. Cause: Because of http://forum.pdfsharp.net/viewtopic.php?f=3&t=3410, pdfsharp fails to emit a closing parenthesis in a content stream. Then, the scanning for the end of the string results in an infinite loop. Regardless of the previous bug, a missing end-of-string parenthesis should never result in an infinite loop. Patch: See attachment. Patch is against 1.50-beta3b. Note: I constructed the diff partially by hand, because CLexer.cs was inadvertently stored as UTF-16 in our local repository. I hope I made no mistakes. The patch also fixes an error message.
|
Author: | Gerben Vos [ Tue Aug 09, 2016 1:33 pm ] | ||
Post subject: | Re: Bug + patch: Missing parenthesis causes infinite loop | ||
This program, based on the issue submission template, is enough to reproduce the problem, also for the other issues I posted. PDF files are not included; they are linked in the issues themselves.
|
Author: | Thomas Hoevel [ Mon Aug 15, 2016 2:08 pm ] |
Post subject: | Re: Bug + patch: Missing parenthesis causes infinite loop |
Thanks for the feedback. I think we already fixed that endless loop issue. Using the latest internal build, I cannot replicate the problem. With respect to "cast" vs. "casted", I didn't change the comments as I consider "casted" correct in the context: http://english.stackexchange.com/a/94570 |
Author: | Gerben Vos [ Mon Aug 15, 2016 2:37 pm ] |
Post subject: | Re: Bug + patch: Missing parenthesis causes infinite loop |
Okay, in that case I'm looking forward to a new public build, as I'm sure are many others... I did a quick scan of all while loops in the source, and found about 10 that looked somewhat suspicious, but in that case I guess I'll hold off with further tests in that area unless we hit on a problematic PDF again. Please note that, apart from that unimportant change in two comments, I also fixed the format string of an error message. |
Page 1 of 1 | All times are UTC |
Powered by phpBB® Forum Software © phpBB Group https://www.phpbb.com/ |