I've been playing with the source from bbrun 1.1 to make it stop segfaulting, and i think i've got it fixed.
Get the patch at: http://furt.com/code/misc/bbrun_patch.tgz I tracked down the segfault bug to readHistory and writeHistory() (thanks to the mailing list emails that mentioned deleting .bbrun_history to get it to stop segfaulting, which made me focus my attention first on anything related to history). i found out that there was no ptr==NULL check before a strcmp(historyItem, ptr) in readHistory, and a similar bug in writeHistory which, i assume, was causing it to segfault for all of us. I also reworked the history file re-writing and handling of duplicates in a way that seems to help ward off corruption of .bbrun_history a bit better. Whenever the .bbrun_history file would get corrupted by a blank line or a gibberish entry (or even too many lines), it would segfault. I am not fluent in C/C++, so please be gentle if my code isn't as good as it could be... Please let me know what you all think. There's a README file in the tgz (along with source code comments) that should fill you all in on what i did, why i did it, and why it should help stop the segfaults. Have a happy christmas all :) -- [scott] :: "ein kalter Tod für den sprecher von Lügen"
bbrun_patch.tgz
Description: GNU Zip compressed data
