From: Jurko GospodnetiÄ (jurko.gospodnetic_at_[hidden])
Date: 2008-04-02 05:19:28
>>>> I'm attaching a patch to fix this so please review and grok it for me
>>>> to apply or apply it directly.
>>> It looks like that change introduces a buffer under-run when one passes
>>> in an empty string.
>> Yup, you're right. Sorry. Here's a corrected patch.
>> I tried playing around with the code to get it more succinct but
>> haven't found a way to avoid that explicitly empty/non-empty if. :-)
> This patch does not appear to be in yet. Should it be? Rene, any opinion?
I've had this patch in my sandbox and actively have been actively
using it on at least two developer machines since I posted it to the
mailing list an have not had a single problem with it since.
Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk