<Damin> Anyone taken a look at http://bugs.digium.com/view.php?id=6476 yet?
ùíù SignOff __deg__: #asterisk-dev (Read error: 110 (Connection timed out))
<Damin> BOO!
ð jsmith/#asterisk-dev hides behind a desk
<jsmith> Damin: Looks like a valid bug, but I'm not sure what it would take to fix it.
<Damin> jsmith: 2 bytes?
<Damin> - char dest[80] = "";
<Damin> + char dest[256] = "";
<jcollie> would be nice if we could eliminate fixed-size buffers like these
<Damin> jcollie: Yes.. it would. Fixed sized buffers waste memory when you scale them up a lot, and in cases where you are
looking at doing embedded platforms, memory is always at a premium..
<Damin> jcollie: Be nice to actually size the buffer based on the size of the string..
<kpfleming> we are rapidly eliminating fixed-size buffers
<jcollie> plus in extreme cases, they are always a few bytes too small :)
<Damin> jcollie: But that has it's own set of issues..
<kpfleming> see all the new stringfields in chan_sip and ast_channel
<Damin> kpfleming: I see.. looks like a work of art! ;)
ùíù oej [n=oej@apollo.webway.se] has joined #asterisk-dev
ùíù mode/#asterisk-dev [+o oej] by ChanServ
<kpfleming> it's better
ùíù evilbuny [n=evilbunn@cpe-24-95-54-117.columbus.res.rr.com] has joined #asterisk-dev
<kpfleming> if the strings are written over they waste memory, but not as much as fixed-size buffers do
<Damin> Shh.. evilbunny has joined.. stop talking about it now!
<MikeJ[Laptop]> heh
<Damin> kpfleming: Yep.. I guess the issue w/ this bug is that there is already an 80 char fixed length buffer, and a
simple chnage to the size fixed the problem.
<Damin> kpfleming: So for expediency sake, just upping the size of the buffer makes things work..
<evilbuny> kpfleming: the new enum code makes the txtcidname stuff defunct
<Damin> kpfleming: But the longer term fix, as you elaborated, is to use a dynamic buffer size..
<evilbuny> and it uses a number of fixed buffer sizes
<Damin> kpfleming: And evilbunny knows Enum! ;)