Daniel Stenberg [Wed, 14 Jan 2009 13:08:50 +0000 (13:08 +0000)]
- ares.h no longer uses the HAVE_STRUCT_IN6_ADDR define check, but instead it
now declares the private struct ares_in6_addr for all systems instead of
relying on one possibly not present in the system.
Daniel Stenberg [Tue, 13 Jan 2009 23:45:24 +0000 (23:45 +0000)]
213 - bug #2501457 "Timeouts not working with curl_multi_socket_action()" -
now put in KNOWN_BUGS as entry #62
Daniel Stenberg [Tue, 13 Jan 2009 23:43:23 +0000 (23:43 +0000)]
62. CURLOPT_TIMEOUT does not work properly with the regular multi and
multi_socket interfaces. The work-around for apps is to simply remove the
easy handle once the time is up. See also:
http://curl.haxx.se/bug/view.cgi?id=2501457
Daniel Stenberg [Tue, 13 Jan 2009 23:29:56 +0000 (23:29 +0000)]
Karl Moerder removed wsock32.lib from the vc9 makefiles
Daniel Stenberg [Tue, 13 Jan 2009 23:24:06 +0000 (23:24 +0000)]
Grant Erickson fixed timeouts for TFTP
Daniel Stenberg [Tue, 13 Jan 2009 22:21:14 +0000 (22:21 +0000)]
- Michael Wallner fixed a NULL pointer deref when calling
curl_easy_setup(curl, CURLOPT_COOKIELIST, "SESS") on a CURL handle with no
cookies data.
Daniel Stenberg [Tue, 13 Jan 2009 22:05:48 +0000 (22:05 +0000)]
Stefan Teleman's curlbuild.h fix for SunPro compilers
Daniel Stenberg [Tue, 13 Jan 2009 22:04:43 +0000 (22:04 +0000)]
oops, the previous wasn't meant to get committed but since I did it I thought
I'd just as well enumerate the remainder and commit a "proper" looking one
Daniel Stenberg [Tue, 13 Jan 2009 22:01:39 +0000 (22:01 +0000)]
- Stefan Teleman brought a patch to fix the default curlbuild.h file for the
SunPro compilers.
Phil Blundell [Tue, 13 Jan 2009 11:51:04 +0000 (11:51 +0000)]
- ares__send_query() now varies the retry timeout pseudo-randomly to avoid
packet storms when several queries were started at the same time.
Daniel Stenberg [Tue, 13 Jan 2009 06:44:03 +0000 (06:44 +0000)]
spell-checked a bunch of comments
Daniel Stenberg [Mon, 12 Jan 2009 21:29:23 +0000 (21:29 +0000)]
make this example not only replace an internal header but also add a totally
new and non-standard one
Daniel Stenberg [Mon, 12 Jan 2009 21:22:51 +0000 (21:22 +0000)]
Mohun Biswas clarified
Daniel Stenberg [Sun, 11 Jan 2009 23:48:20 +0000 (23:48 +0000)]
bump year
Daniel Stenberg [Sun, 11 Jan 2009 23:47:01 +0000 (23:47 +0000)]
changed the wording to possibly make some sense
Daniel Stenberg [Sun, 11 Jan 2009 23:41:50 +0000 (23:41 +0000)]
- Based on bug report #2498665 (curl.haxx.se/bug/view.cgi?id=2498665)
by Daniel Black, I've now added magic to the configure script that makes it
use pkg-config to detect gnutls details as well if the existing method
(using libgnutls-config) fails. While doing this, I cleaned up and unified
the pkg-config usage when detecting openssl and nss as well.
Daniel Stenberg [Sun, 11 Jan 2009 22:05:38 +0000 (22:05 +0000)]
credit Karl Moerder properly
Daniel Stenberg [Sun, 11 Jan 2009 13:49:13 +0000 (13:49 +0000)]
- Phil Blundell added the internal function ares__expand_name_for_response()
that is now used by the ares_parse_*_reply() functions instead of the
ares_expand_name() simply to easier return ARES_EBADRESP for the cases where
the name expansion fails as in responses that really isn't expected.
Daniel Stenberg [Sun, 11 Jan 2009 13:34:05 +0000 (13:34 +0000)]
- Karl M brought the patch that creates vc9 Makefiles, and I made 'maketgz'
now use the actual makefile targets to do the VC8 and VC9 makefiles.
Daniel Stenberg [Sat, 10 Jan 2009 22:10:57 +0000 (22:10 +0000)]
- Emil Romanus fixed:
When using the multi interface over HTTP and the server returns a Location
header, the running easy handle will get stuck in the CURLM_STATE_PERFORM
state, leaving the external event loop stuck waiting for data from the
ingoing socket (when using the curl_multi_socket_action stuff). While this
bug was pretty hard to find, it seems to require only a one-line fix. The
break statement on line 1374 in multi.c caused the function to skip the call
to multistate().
How to reproduce this bug? Well, that's another question. evhiperfifo.c in
the examples directory chokes on this bug only _sometimes_, probably
depending on how fast the URLs are added. One way of testing the bug out is
writing to hiper.fifo from more than one source at the same time.
Daniel Stenberg [Thu, 8 Jan 2009 22:53:37 +0000 (22:53 +0000)]
Only care about the first line of libtool --version as otherwise this script
get fooled by libtool 2, and at least libtool 1.5.2x do output the interesting
version string in the first line.
Daniel Stenberg [Thu, 8 Jan 2009 10:32:54 +0000 (10:32 +0000)]
"204 - Rob C's NSS fix" is done and since we enter feature freeze now I moved
the remaining three issues to next release since they are about new things and
not just fixing bugs
Dan Fandrich [Thu, 8 Jan 2009 01:09:29 +0000 (01:09 +0000)]
Added a few more issues
Dan Fandrich [Thu, 8 Jan 2009 00:31:49 +0000 (00:31 +0000)]
Unified much of the SessionHandle initialization done in Curl_open() and
curl_easy_reset() by creating Curl_init_userdefined(). This had the side effect
of fixing curl_easy_reset() so it now also resets CURLOPT_FTP_FILEMETHOD and
CURLOPT_SSL_SESSIONID_CACHE
Dan Fandrich [Wed, 7 Jan 2009 21:57:26 +0000 (21:57 +0000)]
Always use nocheck="yes" for consistency
Dan Fandrich [Wed, 7 Jan 2009 19:39:35 +0000 (19:39 +0000)]
Created a CURLMIN macro to match CURLMAX
Daniel Stenberg [Wed, 7 Jan 2009 14:12:01 +0000 (14:12 +0000)]
fix compiler warnings
Daniel Stenberg [Wed, 7 Jan 2009 14:10:35 +0000 (14:10 +0000)]
- Rob Crittenden did once again provide an NSS update:
I have to jump through a few hoops now with the NSS library initialization
since another part of an application may have already initialized NSS by the
time Curl gets invoked. This patch is more careful to only shutdown the NSS
library if Curl did the initialization.
It also adds in a bit of code to set the default ciphers if the app that
call NSS_Init* did not call NSS_SetDomesticPolicy() or set specific
ciphers. One might argue that this lets other application developers get
lazy and/or they aren't using the NSS API correctly, and you'd be right.
But still, this will avoid terribly difficult-to-trace crashes and is
generally helpful.
Daniel Stenberg [Tue, 6 Jan 2009 22:30:40 +0000 (22:30 +0000)]
four fresh items to deal with!
Daniel Stenberg [Thu, 1 Jan 2009 20:50:59 +0000 (20:50 +0000)]
- 'reconf' is removed since we rather have users use 'buildconf'
Daniel Stenberg [Wed, 31 Dec 2008 20:36:08 +0000 (20:36 +0000)]
credit Bas
Daniel Stenberg [Wed, 31 Dec 2008 20:35:46 +0000 (20:35 +0000)]
- Bas Mevissen reported curl.haxx.se/bug/view.cgi?id=2479030 pointing
out that 'reconf' didn't properly point out the m4 subdirectory when running
aclocal.
Gunter Knauf [Tue, 30 Dec 2008 15:33:09 +0000 (15:33 +0000)]
sync'd address with other tests.
Gunter Knauf [Tue, 30 Dec 2008 09:16:38 +0000 (09:16 +0000)]
trial with an address which has a valid domain part but invalid host to avoid dns redirections.
Gunter Knauf [Tue, 30 Dec 2008 08:16:24 +0000 (08:16 +0000)]
added HAVE_SOCKADDR_IN6_SIN6_SCOPE_ID to ares Makefile.netware and sync'd with other Makefile.netware.
Gunter Knauf [Tue, 30 Dec 2008 08:05:38 +0000 (08:05 +0000)]
changed HAVE_SIN6_SCOPE_ID define to HAVE_SOCKADDR_IN6_SIN6_SCOPE_ID since just found that ares already uses this define.
Gunter Knauf [Tue, 30 Dec 2008 07:31:04 +0000 (07:31 +0000)]
added HAVE_SIN6_SCOPE_ID define to all non-configure platforms which seem to be IPv6-aware.
Daniel Stenberg [Mon, 29 Dec 2008 22:25:50 +0000 (22:25 +0000)]
- Phil Lisiecki filed bug report #2413067
(http://curl.haxx.se/bug/view.cgi?id=2413067) that identified a problem that
would cause libcurl to mark a DNS cache entry "in use" eternally if the
subsequence TCP connect failed. It would thus never get pruned and refreshed
as it should've been.
Daniel Stenberg [Mon, 29 Dec 2008 21:26:11 +0000 (21:26 +0000)]
further clarifcation based on input from Anthony Bryan
Daniel Stenberg [Sun, 28 Dec 2008 22:00:36 +0000 (22:00 +0000)]
- Peter Korsgaard fixed building libcurl with "configure --with-ssl
--disable-verbose".
Daniel Stenberg [Sun, 28 Dec 2008 21:56:56 +0000 (21:56 +0000)]
Anthony Bryan's man page cleanup in language and spelling
Daniel Stenberg [Sun, 28 Dec 2008 21:43:34 +0000 (21:43 +0000)]
Anthony Bryan reported quirks, I updated
Daniel Stenberg [Sun, 28 Dec 2008 21:35:53 +0000 (21:35 +0000)]
Anthony Bryan reported and I corrected two typos.
Daniel Stenberg [Sun, 28 Dec 2008 21:31:55 +0000 (21:31 +0000)]
The CURLOPT_KRBLEVEL description wasn't properly formatted and thus was
corrupted in ouputs. Another report from Anthony Bryan.
Daniel Stenberg [Sun, 28 Dec 2008 21:29:13 +0000 (21:29 +0000)]
Anthony Bryan reported this outputs wrong in the PDF and I've now tried to
escape these letters what I think is the correct way.
Gunter Knauf [Sun, 28 Dec 2008 05:49:39 +0000 (05:49 +0000)]
silent 'unused' warnings.
Daniel Stenberg [Sat, 27 Dec 2008 23:25:31 +0000 (23:25 +0000)]
libssh2 0.19 became 1.0 instead
Gunter Knauf [Sat, 27 Dec 2008 04:30:36 +0000 (04:30 +0000)]
added check for sin6_scope_id member in struct sockaddr_in6.
Daniel Stenberg [Thu, 25 Dec 2008 15:09:18 +0000 (15:09 +0000)]
193 removed, no work has been started on this and I personally don't care that
much about it
Gisle Vanem [Mon, 22 Dec 2008 18:46:12 +0000 (18:46 +0000)]
Remove trailing #undef value. Typo?
Daniel Stenberg [Mon, 22 Dec 2008 13:21:23 +0000 (13:21 +0000)]
libssh2_sftp_seek2 was just renamed to libssh2_sftp_seek64 ...
Daniel Stenberg [Mon, 22 Dec 2008 13:12:36 +0000 (13:12 +0000)]
- Given a recent enough libssh2, libcurl can now seek/resume with SFTP even
on file indexes beyond 2 or 4GB.
Daniel Stenberg [Mon, 22 Dec 2008 13:07:13 +0000 (13:07 +0000)]
- Anthony Bryan provided a set of patches that cleaned up manual language,
corrected spellings and more.
Daniel Stenberg [Sat, 20 Dec 2008 22:51:57 +0000 (22:51 +0000)]
malloc+memset => calloc
Daniel Stenberg [Sat, 20 Dec 2008 22:47:49 +0000 (22:47 +0000)]
same procedure, simpler code
Daniel Stenberg [Sat, 20 Dec 2008 22:10:42 +0000 (22:10 +0000)]
197 - IIS-bug in Digest
The curl tool parts are postponed to a later time
201 - "bug: header data output to the body callback function after set header"
Was probably not a bug, I asked about it but I didn't get any response.
202 - "hangs up of application above libcurl" - problems with the multi_socket
Fixes from Igor have been committed and there's currently no pending ones.
Daniel Stenberg [Sat, 20 Dec 2008 22:03:22 +0000 (22:03 +0000)]
- Igor Novoseltsev fixed a bad situation for the multi_socket() API when doing
pipelining, as libcurl could then easily get confused and A) work on the
handle that was not "first in queue" on a pipeline, or even B) tell the app
to REMOVE a socket while it was in use by a second handle in a pipeline. Both
errors caused hanging or stalling applications.
Daniel Stenberg [Sat, 20 Dec 2008 21:48:34 +0000 (21:48 +0000)]
make the debug/helper function output to stderr as that makes it play nicer
in combination with infof() calls
Daniel Stenberg [Sat, 20 Dec 2008 17:16:45 +0000 (17:16 +0000)]
remove outdated stuff
Daniel Stenberg [Fri, 19 Dec 2008 22:58:22 +0000 (22:58 +0000)]
- curl_multi_timeout() could return a timeout value of 0 even though nothing
was actually ready to get done, as the internal time resolution is higher
than the returned millisecond timer. Therefore it could cause applications
running on fast processors to do short bursts of busy-loops.
curl_multi_timeout() will now only return 0 if the timeout is actually
alreay triggered.
Daniel Stenberg [Fri, 19 Dec 2008 22:27:11 +0000 (22:27 +0000)]
oops, removing debug output that wasn't supposed to be there
Daniel Stenberg [Fri, 19 Dec 2008 21:14:52 +0000 (21:14 +0000)]
- Using the libssh2 0.19 function libssh2_session_block_directions(), libcurl
now has an improved ability to do right when the multi interface (both
"regular" and multi_socket) is used for SCP and SFTP transfers. This should
result in (much) less busy-loop situations and thus less CPU usage with no
speed loss.
Daniel Stenberg [Fri, 19 Dec 2008 19:03:55 +0000 (19:03 +0000)]
expand the CURLOPT_POSTREDIR explanation
Daniel Stenberg [Wed, 17 Dec 2008 12:32:41 +0000 (12:32 +0000)]
- SCP and SFTP with the multi interface had the same flaw: the 'DONE'
operation didn't complete properly if the EAGAIN equivalent was returned but
libcurl would simply continue with a half-completed close operation
performed. This ruined persistent connection re-use and cause some
SSH-protocol errors in general. The correction is unfortunately adding a
blocking function - doing it entirely non-blocking should be considered for
a better fix.
Gisle Vanem [Tue, 16 Dec 2008 09:12:46 +0000 (09:12 +0000)]
Updated dependencies based on "gcc -MM".
Gisle Vanem [Tue, 16 Dec 2008 09:03:58 +0000 (09:03 +0000)]
Added the use of Watt-32 tcp/ip stack for Win32 target.
If USE_WATT32=1 one needs to use stack-based calls (-3s).
So to keep the makefile nice and clean, specify -3s for
Winsock target too (there's hardly any speed-gain using -3r).
Gisle Vanem [Tue, 16 Dec 2008 08:59:45 +0000 (08:59 +0000)]
Added the use of Watt-32 tcp/ip stack for Win32 targets.
Gisle Vanem [Tue, 16 Dec 2008 08:48:44 +0000 (08:48 +0000)]
Added the use of Watt-32 tcp/ip stack for Win32 targets.
Added USE_WATT32 unconditionally for MSDOS targets since
it's the only option. Adjusted the text for '--wdebug'.
Gisle Vanem [Tue, 16 Dec 2008 08:25:55 +0000 (08:25 +0000)]
Added the use of Watt-32 tcp/ip stack for Win32 targets.
Daniel Stenberg [Mon, 15 Dec 2008 23:04:51 +0000 (23:04 +0000)]
- libssh2_sftp_last_error() was wrongly used at some places in libcurl which
made libcurl sometimes not properly abort problematic SFTP transfers.
Daniel Stenberg [Fri, 12 Dec 2008 12:38:47 +0000 (12:38 +0000)]
196 is gone
Daniel Stenberg [Fri, 12 Dec 2008 12:21:11 +0000 (12:21 +0000)]
- More work with Igor Novoseltsev to first fix the remaining stuff for
removing easy handles from multi handles when the easy handle is/was within
a HTTP pipeline. His bug report #2351653
(http://curl.haxx.se/bug/view.cgi?id=2351653) was also related and was
eventually fixed by a patch by Igor himself.
Daniel Stenberg [Fri, 12 Dec 2008 12:04:10 +0000 (12:04 +0000)]
Make the getoff all pipelines function only set _inuse to FALSE if the
specified data pointer was head.
Daniel Stenberg [Fri, 12 Dec 2008 08:36:56 +0000 (08:36 +0000)]
credit Mark Karpeles for his report and work
Yang Tse [Fri, 12 Dec 2008 03:24:59 +0000 (03:24 +0000)]
Patrick Monnerat fixed a build regression, introduced in 7.19.2, affecting
OS/400 compilations with IPv6 enabled.
Daniel Stenberg [Thu, 11 Dec 2008 23:55:20 +0000 (23:55 +0000)]
Removed 200 as it wasn't a bug after all
Added 203 "dns cache memory leak and TTL failure after failed conn"
Daniel Stenberg [Thu, 11 Dec 2008 23:52:56 +0000 (23:52 +0000)]
- Bug report #2416182 titled "crash in ConnectionExists when using
duphandle+curl_mutli" (http://curl.haxx.se/bug/view.cgi?id=2416182) showed
that curl_easy_duphandle() wrongly also copied the pointer to the connection
cache, which was plain wrong and caused a segfault if the handle would be
used in a different multi handle than the handle it was duplicated from.
Daniel Stenberg [Thu, 11 Dec 2008 23:27:22 +0000 (23:27 +0000)]
Dan F pointed out that we don't need to scan IPv6 addresses for '%'-letters
in the parse_remote_port() function as the scope id has already been stripped
from the string.
Daniel Stenberg [Thu, 11 Dec 2008 22:22:46 +0000 (22:22 +0000)]
- Keshav Krity found out that libcurl failed to deal with dotted IPv6
addresses if they were very long (>39 letters) due to a too strict address
validity parser. It now accepts addresses up to 45 bytes long.
Patrick Monnerat [Thu, 11 Dec 2008 19:20:03 +0000 (19:20 +0000)]
_ Use getaddrinfo/getnameinfo ascii wrappers on OS400.
_ Adjust OS400 make script for non-CVS distributions.
_ Upgrade ILE/RPG binding.
_ Define CURL_HIDDEN_SYMBOLS on OS400, since only CURL_EXTERN-marked symbols are exported.
Dan Fandrich [Thu, 11 Dec 2008 03:01:38 +0000 (03:01 +0000)]
Mention what makes this test special
Daniel Stenberg [Wed, 10 Dec 2008 23:13:31 +0000 (23:13 +0000)]
- Internet Explorer had a broken HTTP digest authentication before v7 and
there are servers "out there" that relies on the client doing this broken
Digest authentication. Apache even comes with an option to work with such
broken clients.
The difference is only for URLs that contain a query-part (a '?'-letter and
text to the right of it).
libcurl now supports this quirk, and you enable it by setting the
CURLAUTH_DIGEST_IE bit in the bitmask you pass to the CURLOPT_HTTPAUTH or
CURLOPT_PROXYAUTH options. They are thus individually controlled to server
and proxy.
Daniel Stenberg [Tue, 9 Dec 2008 22:49:51 +0000 (22:49 +0000)]
use the new URL
Daniel Stenberg [Tue, 9 Dec 2008 22:41:36 +0000 (22:41 +0000)]
start over on the 1.6.1 release...
Daniel Stenberg [Tue, 9 Dec 2008 22:30:59 +0000 (22:30 +0000)]
add space
Dan Fandrich [Tue, 9 Dec 2008 22:00:18 +0000 (22:00 +0000)]
Documented CURLOPT_CONNECT_ONLY as being useful only on HTTP URLs.
Dan Fandrich [Tue, 9 Dec 2008 21:59:57 +0000 (21:59 +0000)]
Added test cases 1089 and 1090 to test --write-out after a redirect to
test a report that the size didn't work, but these test cases pass.
Daniel Stenberg [Tue, 9 Dec 2008 15:02:37 +0000 (15:02 +0000)]
- Ken Hirsch simplified how libcurl does FTPS: now it doesn't assume any
particular state for the control connection like it did before for implicit
FTPS (libcurl assumed such control connections to be encrypted while some
FTPS servers such as FileZilla assumes such connections to be clear
mode). Use the CURLOPT_USE_SSL option to set your desired level.
Gisle Vanem [Tue, 9 Dec 2008 14:39:08 +0000 (14:39 +0000)]
Fix for Win32 targets using Watt-32.
Dan Fandrich [Tue, 9 Dec 2008 01:02:28 +0000 (01:02 +0000)]
C89 compilers (like Minix' ACK) only need to handle 31 functions arguments
so split a long sprintf into two calls to get below that number.
Daniel Stenberg [Mon, 8 Dec 2008 20:21:36 +0000 (20:21 +0000)]
195 - SIZE returning 550 must not abort the transfer
Fixed!
Daniel Stenberg [Mon, 8 Dec 2008 20:20:51 +0000 (20:20 +0000)]
- Fred Machado posted about a weird FTP problem on the curl-users list and when
researching it, it turned out he got a 550 response back from a SIZE command
and then I fell over the text in RFC3659 that says:
The presence of the 550 error response to a SIZE command MUST NOT be taken
by the client as an indication that the file cannot be transferred in the
current MODE and TYPE.
In other words: the change I did on September 30th 2008 and that has been
included in the last two releases were a regression and a bad idea. We MUST
NOT take a 550 response from SIZE as a hint that the file doesn't exist.
Gisle Vanem [Mon, 8 Dec 2008 16:12:11 +0000 (16:12 +0000)]
Added needed defines for Watt-32 on Windows.
Gisle Vanem [Mon, 8 Dec 2008 16:11:16 +0000 (16:11 +0000)]
Undefine 'optarg', 'optind' and 'opterr' when using Watt-32
(to get correct linkage on Windows).
Gisle Vanem [Mon, 8 Dec 2008 16:09:21 +0000 (16:09 +0000)]
ares_writev() shall not be exported when using Watt-32 (has writev).
Added _USE_32BIT_TIME_T to avoid runtime warning. Applies to
VC-2008+ only.
Gisle Vanem [Mon, 8 Dec 2008 14:58:56 +0000 (14:58 +0000)]
Removed unneeded defines HAVE_SIGNAL_H, HAVE_SIG_ATOMIC_T,
RETSIGTYPE and HAVE_PROCESS_H.
Daniel Stenberg [Mon, 8 Dec 2008 14:26:28 +0000 (14:26 +0000)]
one more
Daniel Stenberg [Mon, 8 Dec 2008 14:24:12 +0000 (14:24 +0000)]
192 - "infinite loop during GSS authentication" bug #2221237
Fixed!
Daniel Stenberg [Mon, 8 Dec 2008 13:52:20 +0000 (13:52 +0000)]
- Christian Krause filed bug #2221237
(http://curl.haxx.se/bug/view.cgi?id=2221237) that identified an infinite
loop during GSS authentication given some specific conditions. With his
patience and great feedback I managed to narrow down the problem and
eventually fix it although I can't test any of this myself!
Daniel Stenberg [Thu, 4 Dec 2008 12:54:43 +0000 (12:54 +0000)]
the initial version of the ares_set_socket_callback man page