2009/11/25, Larry Lansing wrote:
Attached is a patch to make ares_parse_srv_reply successfully parse an
SRV reply containing non-SRV answers, such as CNAMEs. This code makes
no attempt to follow or validate the CNAME chain, it just skips over
non-SRV answers. This is probably enough for
2009/11/26, Yang Tse wrote:
Fixed in CVS now, as well as ares_parse_txt_reply.
Daily snapshots dated 25 November 2009 or later will contain this fix.
No, no. Wrong date. :-(
Daily snapshots that will have the fix are dated 26 November 2009 or later.
--
-=[Yang]=-
Thank you for the fast response!
On Wed, Nov 25, 2009 at 6:47 PM, Yang Tse yangs...@gmail.com wrote:
2009/11/26, Yang Tse wrote:
Fixed in CVS now, as well as ares_parse_txt_reply.
Daily snapshots dated 25 November 2009 or later will contain this fix.
No, no. Wrong date. :-(
Daily
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 10/27/2009 07:47 PM, Yang Tse wrote:
Hi Jakub,
It seems we still lack in CVS an ares_parse_srv_reply man page for the
provided function.
Cheers,
Yang,
I am confused now. Both my CVS tree and the CVS web UI[1] show that
there is a man
2009/10/28, Jakub Hrozek wrote:
Both my CVS tree and the CVS web UI[1] show that
there is a man page. Did I miss something?
Aha! I did, not you.
Some missing references in Makefile.inc were preventing it from being
included in the distribution tarball, as well as corresponding .pdf
format man
Hi Jakub,
It seems we still lack in CVS an ares_parse_srv_reply man page for the
provided function.
Cheers,
--
-=[Yang]=-
Hi Daniel,
Daniel Stenberg schrieb:
On Fri, 4 Sep 2009, Larry Lansing wrote:
I have test code that proves that it successfully parses replies with
single and multiple SRV response records. (Tested it against a
capture from wireshark.)
Thanks. And thanks Jakub Hrozek too of course.
On Sat, 5 Sep 2009, Guenter wrote:
I've changed u_int16_t to unsigned short since we even use unsigned short
for ports, and it would be the first place within ares and curl where we
rely on such type; if we want a type here we should probably consider to
define our own ares_uint16_t or
Greetings.
I have an immediate use for this patch. Is there anything in particular
holding it up? Anything I can do to help?
Jakub Hrozek wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 08/04/2009 03:48 PM, Jakub Hrozek wrote:
On 08/03/2009 01:57 PM, Daniel Stenberg wrote:
Sorry
On Fri, 4 Sep 2009, Larry Lansing wrote:
I have an immediate use for this patch. Is there anything in particular
holding it up? Anything I can do to help?
You can tell us what you think about the patch and if you have tried it how it
works etc.
--
/ daniel.haxx.se
On Fri, 4 Sep 2009, Larry Lansing wrote:
I have test code that proves that it successfully parses replies with single
and multiple SRV response records. (Tested it against a capture from
wireshark.)
Thanks. And thanks Jakub Hrozek too of course.
Applied and committed just now.
--
/
On 09/04/2009 11:56 PM, Daniel Stenberg wrote:
On Fri, 4 Sep 2009, Larry Lansing wrote:
I have test code that proves that it successfully parses replies with
single and multiple SRV response records. (Tested it against a
capture from wireshark.)
Thanks. And thanks Jakub Hrozek too of
On Wed, 15 Jul 2009, Jakub Hrozek wrote:
OK, I'm attaching a new version of the patch which does not use memcpy at
all. Hope the code is clearer now.
Daniel, can you see any other issues with the patch? Is c-ares upstream
interested in merging this functionality?
Sorry for the delay. I am
13 matches
Mail list logo
