Mailing List Archive

Re: strlcat/strlcpy patch
On Tue, 06 Apr 2004 19:17:13 -0600
Phil Oleson <oz@nixil.net> wrote:

> Sir,
> did you revert the patch because of clashing symbols? I'd like
> to
> see that patch
> stay in the tree in some form.. I was thinking that it would likely
> need to be in libclamav/ dir rather than the shared/ directory.

It has been temporarily reverted due to problems in strlcpy logic in a
few places. I will review the patch more carefully and apply it back on
days.

--
oo ..... Tomasz Kojm <tkojm@clamav.net>
(\/)\......... http://www.ClamAV.net/gpg/tkojm.gpg
\..........._ 0DCA5A08407D5288279DB43454822DC8985A444B
//\ /\ Wed Apr 7 03:46:25 CEST 2004
Re: strlcat/strlcpy patch [ In reply to ]
Tomasz Kojm wrote:

>On Tue, 06 Apr 2004 19:17:13 -0600
>Phil Oleson <oz@nixil.net> wrote:
>
>
>
>>Sir,
>> did you revert the patch because of clashing symbols? I'd like
>> to
>>see that patch
>>stay in the tree in some form.. I was thinking that it would likely
>>need to be in libclamav/ dir rather than the shared/ directory.
>>
>>
>
>It has been temporarily reverted due to problems in strlcpy logic in a
>few places. I will review the patch more carefully and apply it back on
>days.
>
>
>
Thanks,

The main reason I was asking was that I was working on a sprintf ->
snprintf
conversion patch today, and the big reversion just surprised me.

oh.. btw, in looking at some of the code in libclamav, I must say that
there is quite
a bit of abuse of const pointers in the code.. should I look into this
further and
make up a 2nd patchset to work on correcting this? I didn't touch it
when I was
doing the conversion, as I didn't want to confuse the patch any... Plus
I wanted to
take a better look to see the bigger picture on which functions really
want protected
strings and those that can have const'ness removed earlier in the stack,
or not.


Phil.

--
It's a frail sad line between optimism and delusion.