[Dillo-dev]html parser stress testing?
Hello, (i just noticed this has been reported already, but anyway) a bit ago somebody reported a interesting experiment on bugtraq: he wrote a script that generates random malformed html and used it to test the stability (and security, in case bugs could be exploited to execute code!) of various browsers: http://www.securityfocus.com/archive/1/378632/2004-10-15/2004-10-21/0 this would probably also be interesting to test dillo with (running in a debugger) to weed out bugs. and just a bit after i thought that, somebody reported the first: http://home.eggdrop.ch/cgi-bin/remangle.cgi?0x702d9be9 <TABLE><TD COLSPAN=400192481> -> segfault - Thorben
On Wed, Oct 20, Thorben Thuermer wrote:
Hello,
(i just noticed this has been reported already, but anyway) a bit ago somebody reported a interesting experiment on bugtraq: he wrote a script that generates random malformed html and used it to test the stability (and security, in case bugs could be exploited to execute code!) of various browsers: http://www.securityfocus.com/archive/1/378632/2004-10-15/2004-10-21/0
this would probably also be interesting to test dillo with (running in a debugger) to weed out bugs.
Yes, I downloaded the programs and tested it locally, and found dillo sometimes hang. It would be useful to collect such cases. BTW, If you wonder, where the logs to to (stderr in the code), it is /var/log/apache/error.log for Apache.
and just a bit after i thought that, somebody reported the first: http://home.eggdrop.ch/cgi-bin/remangle.cgi?0x702d9be9 <TABLE><TD COLSPAN=400192481> -> segfault
Fixed in CVS. See changes in dw_table.c. Sebastian
participants (2)
-
Sebastian Geerken -
Thorben Thuermer