On 6/11/13 11:40 PM, Martin Mikkelsen wrote:
On Thu, May 16, 2013 at 10:06:00AM +0200, Daniel-Constantin Mierla wrote:
I pushed a patch for fixing this issue on master branch -- can you
test and report if works fine for you now?
(Sorry for the late reply)
I'm still on 3.4, but i tested the fe
On Thu, May 16, 2013 at 10:06:00AM +0200, Daniel-Constantin Mierla wrote:
> I pushed a patch for fixing this issue on master branch -- can you
> test and report if works fine for you now?
(Sorry for the late reply)
I'm still on 3.4, but i tested the fe7e4a5152674aa9c81c09dd2fc9938d9e9e762e
commit
Hello,
I pushed a patch for fixing this issue on master branch -- can you test
and report if works fine for you now?
Cheers,
Daniel
On 5/3/13 1:07 AM, Daniel-Constantin Mierla wrote:
Hello,
this patch is fixing only for $var(...), but there are other variables
that can be in the same situa
Hello,
this patch is fixing only for $var(...), but there are other variables
that can be in the same situation. All PV-set functions have to be
updated as well as could be the case for config interpreter and some
core functions.
Anyhow, going for an possible unoptimized operation all the ti
set_var_value() may do memory corruption when a variable is set to the
substring of the same variable, this is atleast the case with the
s.substr, s.select, s.strip and s.striptail transformation functions
which just updates the pointer to its input buffer.
As an example, this would trigger a warn
