platform/upstream/libsoup.git
14 years agoAvoid infinite loop trying to connect to host when the network is down.
Dan Winship [Sat, 22 Aug 2009 13:11:45 +0000 (09:11 -0400)]
Avoid infinite loop trying to connect to host when the network is down.

If a connection attempt to a host fails, fail all of the messages
queued for that host.

Previously we would leave them in the queue if there was already an
existing connection to the host (hoping that we'd be able to reuse
that connection for the other messages later), but this meant that if
there was a single connection to a host open, with a message running
on it, and at least one additional queued message to the same host,
and then the network went down, libsoup would get stuck in a loop
repeatedly trying and failing to open a second connection (with the
first connection never getting a chance to time out).

http://bugzilla.gnome.org/show_bug.cgi?id=592492

14 years agoFix SoupLogger to not crash with streamed requests
Dan Winship [Fri, 21 Aug 2009 23:49:42 +0000 (19:49 -0400)]
Fix SoupLogger to not crash with streamed requests

http://bugzilla.gnome.org/show_bug.cgi?id=591857

14 years agoRe-redo not-yet-used connection handling
Dan Winship [Fri, 21 Aug 2009 23:03:56 +0000 (19:03 -0400)]
Re-redo not-yet-used connection handling

Previously we were taking special care to avoid closing connections
that hadn't been used yet, but this is not actually necessary;
run_queue() doesn't prune idle connections until after running through
the entire queue once, so if the new connection is still idle at that
point, it means we no longer have a message to send on it (probably
because the message already got sent on another connection), and so it
is legitimate to close it if needed.

http://bugzilla.gnome.org/show_bug.cgi?id=592084

14 years agoRemove checks for SOUP_MESSAGE_IO_STATUS_CONNECTING, since it never gets set
Dan Winship [Fri, 21 Aug 2009 22:42:55 +0000 (18:42 -0400)]
Remove checks for SOUP_MESSAGE_IO_STATUS_CONNECTING, since it never gets set

The connection-management rewrite removed the code that would set it.
Since it never worked exactly the way it was supposed to anyway, let's
just get rid of it and replace it with something better.

14 years agoFix the definition of soup_message_is_keepalive() for HTTP/1.0
Dan Winship [Thu, 20 Aug 2009 18:14:52 +0000 (14:14 -0400)]
Fix the definition of soup_message_is_keepalive() for HTTP/1.0

Fixes a problem where some requests would reach 100% but never "finish"

14 years agoAdd warning in a "can't happen" case
Benjamin Otte [Mon, 17 Aug 2009 17:29:17 +0000 (19:29 +0200)]
Add warning in a "can't happen" case

This is mostly to indicate that we did not forget error checking.

14 years agoTweak SoupConnection handling a bit wrt "not yet used" connections.
Dan Winship [Tue, 18 Aug 2009 14:10:45 +0000 (10:10 -0400)]
Tweak SoupConnection handling a bit wrt "not yet used" connections.

Should help to avoid the "undead" connections described in bug 592084.

14 years agoNotice closed connections sooner
Dan Winship [Tue, 18 Aug 2009 02:28:08 +0000 (22:28 -0400)]
Notice closed connections sooner

Only works on UNIX; Windows has to wait for the GSocket port.

http://bugzilla.gnome.org/show_bug.cgi?id=578990

14 years agoFix a crash when processing a request with no Host header
Dan Winship [Thu, 13 Aug 2009 22:21:19 +0000 (18:21 -0400)]
Fix a crash when processing a request with no Host header

Pointed out by Andreas Henriksson on IRC

14 years agoAdd SoupPasswordManager and SoupPasswordManagerGNOME
Dan Winship [Wed, 12 Aug 2009 14:23:39 +0000 (10:23 -0400)]
Add SoupPasswordManager and SoupPasswordManagerGNOME

SoupPasswordManager (and some new SoupAuth APIs) provide an interface
for saving passwords, and SoupPasswordManagerGNOME provides an
implementation of that interface using gnome-keyring.

14 years agopost-release version bump
Dan Winship [Wed, 12 Aug 2009 14:03:12 +0000 (10:03 -0400)]
post-release version bump

14 years ago2.27.90 LIBSOUP_2_27_90
Dan Winship [Mon, 10 Aug 2009 21:15:15 +0000 (17:15 -0400)]
2.27.90

14 years agoFix up docs
Dan Winship [Mon, 10 Aug 2009 21:15:01 +0000 (17:15 -0400)]
Fix up docs

14 years agoPort SoupAddress to use GResolver, remove soup-dns
Dan Winship [Sat, 27 Sep 2008 22:55:35 +0000 (18:55 -0400)]
Port SoupAddress to use GResolver, remove soup-dns

14 years agoFix error code when trying to https with --disable-ssl
Dan Winship [Sun, 9 Aug 2009 15:23:55 +0000 (11:23 -0400)]
Fix error code when trying to https with --disable-ssl

http://bugzilla.gnome.org/show_bug.cgi?id=590464

14 years agoAdd soup_session_get_feature_for_message, remove _get_proxy_resolver
Dan Winship [Sun, 9 Aug 2009 14:58:54 +0000 (10:58 -0400)]
Add soup_session_get_feature_for_message, remove _get_proxy_resolver

Add a cache to soup_session_get_feature(), and remove the separate
auth_manager and proxy_resolver private vars, and just use
soup_session_get_feature() as needed instead. (Partly inspired by the
fact that the cache branch adds yet another specially-tracked
feature.)

Add soup_session_get_feature_for_message() to get a feature and check
soup_message_disables_feature() on it, and use that to replace
soup_session_get_proxy_resolver().

14 years agoAdd soup_message_disable_feature()
Dan Winship [Sun, 2 Aug 2009 15:35:01 +0000 (11:35 -0400)]
Add soup_message_disable_feature()

This allows you to disable specific SoupSessionFeatures for particular
messages.

Also update tests/sniffing-test to test it out.

http://bugzilla.gnome.org/show_bug.cgi?id=574773

14 years agoSupport "OPTIONS *" in SoupServer
Dan Winship [Tue, 4 Aug 2009 23:34:30 +0000 (19:34 -0400)]
Support "OPTIONS *" in SoupServer

Update soup-message-server-io.c to allow "*" in a Request-Line, and
update SoupServer and SoupAuthDomain methods and docs to deal with it.
For backward-compatibility, requests to "*" are NOT passed to the
default handler; you must explicitly register a handler for "*".

http://bugzilla.gnome.org/show_bug.cgi?id=590751

14 years agoanother out-of-tree build fix
Dan Winship [Sun, 9 Aug 2009 14:19:39 +0000 (10:19 -0400)]
another out-of-tree build fix

14 years agoBuild fixes
Andreas Rottmann [Sun, 9 Aug 2009 13:46:36 +0000 (09:46 -0400)]
Build fixes

- Add -I$(top_builddir) to documentation scanner CFLAGS (fixes "make
  distcheck" run from an out-of-tree build)

14 years agoUse an m4/ directory for aclocal files
Dan Winship [Sun, 9 Aug 2009 13:42:14 +0000 (09:42 -0400)]
Use an m4/ directory for aclocal files

14 years agoAdd automake 1.11 silent-rules support
Dan Winship [Sun, 9 Aug 2009 13:24:19 +0000 (09:24 -0400)]
Add automake 1.11 silent-rules support

14 years agoUse guchar instead of char where the distinction is important.
Dan Winship [Sun, 9 Aug 2009 13:16:48 +0000 (09:16 -0400)]
Use guchar instead of char where the distinction is important.

Noticed by -Wextra

14 years agoRemove a redundant check
Dan Winship [Sun, 9 Aug 2009 13:16:31 +0000 (09:16 -0400)]
Remove a redundant check

14 years agoClarify that an ignored return value is intentional, to appease fortify
Dan Winship [Sun, 9 Aug 2009 13:15:38 +0000 (09:15 -0400)]
Clarify that an ignored return value is intentional, to appease fortify

http://bugzilla.gnome.org/show_bug.cgi?id=591226

14 years agoAdd some g_return_if_fail()s to soup-headers.c
Dan Winship [Sun, 9 Aug 2009 12:53:50 +0000 (08:53 -0400)]
Add some g_return_if_fail()s to soup-headers.c

(partly from "cache" branch)

14 years agosoup-session: fix typo in documentation
Xan Lopez [Wed, 5 Aug 2009 07:50:19 +0000 (10:50 +0300)]
soup-session: fix typo in documentation

14 years agoAdd missing soup-proxy-uri-resolver.h to soup.h
Dan Winship [Sun, 2 Aug 2009 15:34:08 +0000 (11:34 -0400)]
Add missing soup-proxy-uri-resolver.h to soup.h

14 years agoMirror Mozilla behavior for when to follow Content-Length
Benjamin Otte [Tue, 4 Aug 2009 10:11:03 +0000 (12:11 +0200)]
Mirror Mozilla behavior for when to follow Content-Length

Mozilla only honors content length on messages that use keep-alive to
work around servers that send broken Content-Length headers. This patch
mirrors that behavior.

An example for such a page is
http://sourceforge.net/apps/wordpress/sourceforge

15 years agoRe-fix _SOUP_ATOMIC_INTERN_STRING to not cause strict-aliasing warnings
Dan Winship [Fri, 31 Jul 2009 20:39:40 +0000 (16:39 -0400)]
Re-fix _SOUP_ATOMIC_INTERN_STRING to not cause strict-aliasing warnings

http://bugzilla.gnome.org/show_bug.cgi?id=588771

15 years ago2.27.5 LIBSOUP_2_27_5
Dan Winship [Tue, 28 Jul 2009 02:39:04 +0000 (22:39 -0400)]
2.27.5

15 years agoDon't crash when a request is redirected to an invalid/non-http URI
Dan Winship [Tue, 28 Jul 2009 01:27:27 +0000 (21:27 -0400)]
Don't crash when a request is redirected to an invalid/non-http URI

Make redirect_handler() reject such responses, and also make soup-uri
g_return_if_fail rather than crashing. Add a regression test.

http://bugzilla.gnome.org/show_bug.cgi?id=528882

15 years agoDo not discard the request body after sending, even if !accumulate
Dan Winship [Mon, 27 Jul 2009 17:06:19 +0000 (13:06 -0400)]
Do not discard the request body after sending, even if !accumulate

We can't discard the request body until the message is completely
finished, because it may be necessary to send it a second time due to
redirect, authentication, or premature connection close.

In particular, if the request body did get discarded, then when
soup-message-io tried to send it the second time, it would see a non-0
content-length but no body chunks, and so would assume that the
application was intending to stream the body, and so would pause it
and wait for the first chunk. Since the application was not actually
intending to stream the request, this would result in the message
getting stuck forever.

http://bugzilla.gnome.org/show_bug.cgi?id=584645

15 years agoBuild fixes
Andreas Rottmann [Sun, 21 Jun 2009 01:21:49 +0000 (21:21 -0400)]
Build fixes

- Remove acinclude.m4, which just hosted a stale copy of
  GTK_DOC_CHECK, which in turn broke "make distcheck" when used in
  combination with the gtk-doc.make installed by newer gtkdocize.

- Fix #include in soup-path-map.h.

15 years agoAllow version 1 CA certificates in clients
Patrick Ohly [Thu, 23 Jul 2009 14:03:34 +0000 (16:03 +0200)]
Allow version 1 CA certificates in clients

This is done by setting the GNUTLS_VERIFY_ALLOW_X509_V1_CA_CRT
flag while setting up the credentials and parsing the CA certificate
file. This is necessary because without it, some releases of gnutls
reject certificates that are currently in use by major sites, including
Google.

Setting the flag is safe in libsoup because the file is meant to
contain only CA certificates. The main weakness around version 1
certs is that personal certificates could be mistaken for CA certs,
which shouldn't happen if libsoup is used as intended.

Servers don't have the possibility to verify their clients, so this
problem doesn't apply to them.

http://bugzilla.gnome.org/show_bug.cgi?id=589323

15 years agoFix cut+pasto in win32 gnutls code
Dan Winship [Mon, 13 Jul 2009 15:31:49 +0000 (11:31 -0400)]
Fix cut+pasto in win32 gnutls code

15 years ago2.27.4 LIBSOUP_2_27_4
Dan Winship [Mon, 13 Jul 2009 14:21:38 +0000 (10:21 -0400)]
2.27.4

15 years agodistcheck fixes for sniffing-test
Dan Winship [Mon, 13 Jul 2009 15:04:34 +0000 (11:04 -0400)]
distcheck fixes for sniffing-test

15 years agoMake gcc happy with body handling in sniffing-test
Gustavo Noronha Silva [Mon, 13 Jul 2009 14:40:21 +0000 (15:40 +0100)]
Make gcc happy with body handling in sniffing-test

15 years agoFix HTML or feed algorithm to not go past resource_length
Gustavo Noronha Silva [Mon, 13 Jul 2009 14:36:10 +0000 (15:36 +0100)]
Fix HTML or feed algorithm to not go past resource_length

15 years agoFix SSL on Windows
Dan Winship [Mon, 13 Jul 2009 00:26:43 +0000 (20:26 -0400)]
Fix SSL on Windows

The connection timeout changes had broken SSL on Windows because they
used read/write rather than recv/send.

Based on a patch from Fridrich Strba.
http://bugzilla.gnome.org/show_bug.cgi?id=587910

15 years agoMake SOUP_SESSION_TIMEOUT also affect async connection
Mark Nauwelaerts [Mon, 13 Jul 2009 00:19:16 +0000 (20:19 -0400)]
Make SOUP_SESSION_TIMEOUT also affect async connection

http://bugzilla.gnome.org/show_bug.cgi?id=588177

15 years agoFix a content-sniffer crash with redirected URLs
Dan Winship [Sun, 12 Jul 2009 16:22:45 +0000 (12:22 -0400)]
Fix a content-sniffer crash with redirected URLs

SoupMessageHeaders was not properly cleaning up the cached
Content-Type when a request was redirected, which meant that if the
redirected-to URL had no Content-Type header, then
soup_message_headers_get_content_type() would return the Content-Type
of the original request, but
soup_message_headers_get_one("Content-Type") would return NULL,
causing the sniffer to crash.

http://bugzilla.gnome.org/show_bug.cgi?id=588335

15 years agoMiscellaneous content-sniffing fixes
Dan Winship [Fri, 10 Jul 2009 17:11:07 +0000 (13:11 -0400)]
Miscellaneous content-sniffing fixes

Remove the should_sniff_content flag; if there is a sniffer, we always
sniff.

Clean up soup-message-io a bit to require fewer special cases, and fix
a few potential leaks/crashes where it wasn't dealing with
IF_CANCELLED_OR_PAUSED correctly. (Although this looks like a large
patch, it actually results in a much smaller diff relative to 2.26.

Fix tests/sniffing-test to do chunked encoding correctly, make sure to
return data in chunks that are smaller than the sniffing buffer size,
and make resources/mbox larger than the sniffing buffer size so it
takes multiple reads to get enough data (when using chunked encoding).
Also add tests for empty response bodies (from Gustavo).

http://bugzilla.gnome.org/show_bug.cgi?id=587907

15 years agoActually test chunked encoding, and fix some hangs
Gustavo Noronha Silva [Fri, 10 Jul 2009 13:29:28 +0000 (14:29 +0100)]
Actually test chunked encoding, and fix some hangs

The test was not really testing chunked encoding, because of incorrect
handling of the query string in the SoupURI, now it does. This has
exposed a couple hangs.

This change also avoids hanging on chunked encoding messages on the
sniffing test by calling soup_message_body_complete(), and by
sanitizing IO read state and read_length inconditionally when we reach
got_body, during message IO.

15 years agoFix the SOUP_METHOD_* and SOUP_URI_SCHEME_* defines to be threadsafe
Dan Winship [Fri, 3 Jul 2009 01:28:53 +0000 (21:28 -0400)]
Fix the SOUP_METHOD_* and SOUP_URI_SCHEME_* defines to be threadsafe

Also fix/optimize a few other things in URI scheme canonicalization

15 years agoImplement content sniffing
Gustavo Noronha Silva [Wed, 17 Jun 2009 23:53:17 +0000 (20:53 -0300)]
Implement content sniffing

The implementation is based on the draft spec on Content-Type
Processing Model (draft-abarth-mime-sniff-01). It is a spinoff from
the HTML5 spec.

Soup now provides a SoupContentSniffer session feature, which hooks
into the message I/O, and delays emissions of the got-chunk signal to
be able to figure out the Content-Type of messages from the actual
content received, in some cases.

GIO is also used to sniff content, whenever the spec allows further
sniffing.

http://bugzilla.gnome.org/show_bug.cgi?id=572589

15 years agoRequire explicit --disable-ssl to build without GNUTLS support
Dan Winship [Sun, 28 Jun 2009 14:00:00 +0000 (10:00 -0400)]
Require explicit --disable-ssl to build without GNUTLS support

In general, if someone is trying to build without GNUTLS installed
it's an accident, so bail out with an error.

Inspired by http://bugzilla.gnome.org/show_bug.cgi?id=584955

15 years agoFix POST 301 redirect behavior
Dan Winship [Mon, 22 Jun 2009 21:52:39 +0000 (17:52 -0400)]
Fix POST 301 redirect behavior

We were doing POST->GET translation on 302 and 303, but we need to do
it on 301 as well to be compatible.

http://bugzilla.gnome.org/show_bug.cgi?id=586692

15 years ago2.27.2 LIBSOUP_2_27_2
Dan Winship [Tue, 16 Jun 2009 15:39:22 +0000 (11:39 -0400)]
2.27.2

15 years agoRemove -r flag from tests/get
Dan Winship [Tue, 9 Jun 2009 17:36:12 +0000 (13:36 -0400)]
Remove -r flag from tests/get

The requirement that it imposed--that we "guess" the name of the index
file when fetching a URL that ended with "/"--is more annoying than
the usefulness of being able to test multiple downloads at once.

15 years agoadd a HACKING file with some initial notes on code style and correctness
Dan Winship [Mon, 8 Jun 2009 13:55:15 +0000 (09:55 -0400)]
add a HACKING file with some initial notes on code style and correctness

15 years agoAdd "Content-Length: 0" to requests with a Content-Type but no body
Dan Winship [Sun, 7 Jun 2009 03:14:05 +0000 (23:14 -0400)]
Add "Content-Length: 0" to requests with a Content-Type but no body

Some servers get annoyed if a POST/PUT declares a Content-Type but
contains no body, so make sure to send a 0-length body (which is
different) instead.

15 years agoImplement SoupProxyURIResolver, to replace SoupProxyResolver
Dan Winship [Sat, 6 Jun 2009 22:57:51 +0000 (18:57 -0400)]
Implement SoupProxyURIResolver, to replace SoupProxyResolver

Simplifies implementations, allows for non-http proxy resolution, and
allows authentication information to be passed.

http://bugzilla.gnome.org/show_bug.cgi?id=580051

15 years agoSoupConnection/SoupSession reorg and cleanup
Dan Winship [Sat, 6 Jun 2009 22:55:48 +0000 (18:55 -0400)]
SoupConnection/SoupSession reorg and cleanup

Move CONNECT handling from SoupConnection to SoupSession so it works
more like other requests. Generally clean up and simplify
SoupConnection, and simplify the coordination between SoupSession and
its subclasses.

15 years agoFix hostname resolution behavior
Dan Winship [Sat, 6 Jun 2009 22:52:30 +0000 (18:52 -0400)]
Fix hostname resolution behavior

Previously we went to some effort to resolve the message URI hostname
to an IP address before figuring out proxies/connections, but this
turns out to be wrong for multiple reasons:

  1. Some hosts that send all requests via proxy don't even have a
     working DNS config.
     (http://bugzilla.gnome.org/show_bug.cgi?id=577532)

  2. Apparently no one expects hostnames in requests to be matched
     against IP addresses in proxy ignore lists anyway.

  3. The big web browsers all implement connection limits on a
     per-hostname basis, not a per-IP basis, and some web pages take
     advantage of this by using multiple aliases for the same host
     to get around the connection limit.

Also update tests/proxy-test to verify that the hostname is not
resolved when passing a request to a proxy.

15 years agoMore temporary SSL tweaking
Dan Winship [Thu, 4 Jun 2009 00:13:38 +0000 (20:13 -0400)]
More temporary SSL tweaking

Disable TLS 1.0 too, since some sites don't even support that (I'm
looking at you, PayPal). Eventually we'll want to make this be more
clever, and have it first try TLS 1.1 with extensions, and then fall
back to SSL 3.0 if the server gets confused by the new stuff. But that
will take some rewriting.

15 years agoInclude the (optional) algorithm param in Digest auth responses
Dan Winship [Tue, 2 Jun 2009 01:37:55 +0000 (21:37 -0400)]
Include the (optional) algorithm param in Digest auth responses

Apple's CalDAV server apparently considers it mandatory.
http://bugzilla.gnome.org/show_bug.cgi?id=583091

15 years agoUse POSIXly-correct "test a = b" rather than bash "test a == b"
Dan Winship [Tue, 2 Jun 2009 01:26:37 +0000 (21:26 -0400)]
Use POSIXly-correct "test a = b" rather than bash "test a == b"

Stupid bash. http://bugzilla.gnome.org/show_bug.cgi?id=583911

15 years agoUpdate REQUIRED_AUTOMAKE_VERSION to match reality
Dan Winship [Tue, 2 Jun 2009 01:21:18 +0000 (21:21 -0400)]
Update REQUIRED_AUTOMAKE_VERSION to match reality

http://bugzilla.gnome.org/show_bug.cgi?id=583942

15 years agoDon't emit "authenticate" until a message is actually sent and fails.
Dan Winship [Thu, 21 May 2009 21:36:17 +0000 (18:36 -0300)]
Don't emit "authenticate" until a message is actually sent and fails.

In some cases, we're reasonably sure it's going to fail if we don't
emit authenticate, but forcing callers to deal with both the
pre-sending and post-sending cases in their authenticate handlers
makes writing them tricky. (In particular, calling
soup_message_io_pause() in the pre-sending case won't work, since the
message hasn't actually started yet.) This also solves the problem
that we were previously emitting "authenticate" both before and after
the message was sent in some cases.

Add a regression test for this case (written by Gustavo Noronha, who
figured out the circumstances where the bug was happening), and tweak
the existing regression tests to match the new rules.

http://bugzilla.gnome.org/show_bug.cgi?id=583462

15 years agoTemporarily redo the TLS compatibility patch
Dan Winship [Mon, 18 May 2009 14:04:39 +0000 (10:04 -0400)]
Temporarily redo the TLS compatibility patch

It broke tests/ssl-test on machines that support %SSL3_RECORD_VERSION.
For now, we will just not use that flag, and instead always disable
TLS 1.1. The gnutls problem has been reported upstream.

15 years agoDon't set AI_CANONNAME when calling getaddrinfo()
Dan Winship [Mon, 18 May 2009 13:46:30 +0000 (09:46 -0400)]
Don't set AI_CANONNAME when calling getaddrinfo()

We weren't using the information it generated, so it had no effect
besides slowing things down with additional DNS traffic. Pointed out
by Christophe Gillette on libsoup-list.

15 years agoFix a warning introduced in fix for bug 582002
Dan Winship [Fri, 15 May 2009 14:26:34 +0000 (10:26 -0400)]
Fix a warning introduced in fix for bug 582002

15 years agoFix cflags to work with older versions of gcc
Dan Winship [Sun, 19 Apr 2009 14:46:40 +0000 (10:46 -0400)]
Fix cflags to work with older versions of gcc

Check that gcc supports -Wmissing-include-dirs and -Wundef rather than
using them unconditionally.

http://bugzilla.gnome.org/show_bug.cgi?id=578851

15 years agoFix the header/body boundary detection logic
Dan Winship [Sat, 9 May 2009 18:34:31 +0000 (14:34 -0400)]
Fix the header/body boundary detection logic

The change to support LF LF in addition to CRLF CRLF broke this subtly
so that if the first network read returned a packet that ended at a
CR, then the following LF would be mistakenly interpreted as ending
the headers.

http://bugzilla.gnome.org/show_bug.cgi?id=582002

15 years agoAttempt better compatibility with non-TLS1.1 https servers.
Dan Winship [Fri, 8 May 2009 22:11:20 +0000 (18:11 -0400)]
Attempt better compatibility with non-TLS1.1 https servers.

Try to use the (not-yet-widely-deployed) SSL3_RECORD_VERSION option in
gnutls to make it handshake according to the backward-compatibility
recommendations in the TLS spec. If that fails, fall back to just
disabling TLS1.1. Should hopefully help us to work better with servers
that freak out if you even suggest using TLS1.1.

Fix based on a Pidgin patch from Martin von Gagern.

http://bugzilla.gnome.org/show_bug.cgi?id=581342

15 years agoAdd another missing G_GNUC_NULL_TERMINATED and a G_GNUC_PRINTF
Dan Winship [Fri, 8 May 2009 20:59:36 +0000 (16:59 -0400)]
Add another missing G_GNUC_NULL_TERMINATED and a G_GNUC_PRINTF

15 years agoMark soup_uri_set_query_from_fields with G_GNUC_NULL_TERMINATED
Ross Burton [Thu, 7 May 2009 15:11:39 +0000 (16:11 +0100)]
Mark soup_uri_set_query_from_fields with G_GNUC_NULL_TERMINATED

http://bugzilla.gnome.org/show_bug.cgi?id=581754

15 years agoHandle PUT methods in soup_form_request_for_data
Ross Burton [Fri, 8 May 2009 13:04:19 +0000 (14:04 +0100)]
Handle PUT methods in soup_form_request_for_data

http://bugzilla.gnome.org/show_bug.cgi?id=581860

15 years agoNEWS: fix a bug number that was typoed
Dan Winship [Tue, 5 May 2009 14:06:16 +0000 (10:06 -0400)]
NEWS: fix a bug number that was typoed

15 years ago2.27.1 LIBSOUP_2_27_1
Dan Winship [Tue, 5 May 2009 01:27:15 +0000 (21:27 -0400)]
2.27.1

15 years agoFix typo in documentation: TIME should be TYPE.
Pierre-Luc Beaudoin [Fri, 24 Apr 2009 14:51:16 +0000 (10:51 -0400)]
Fix typo in documentation: TIME should be TYPE.

15 years agoFix a crash when cancelling a message from a "restarted" handler
Dan Winship [Tue, 28 Apr 2009 16:23:03 +0000 (12:23 -0400)]
Fix a crash when cancelling a message from a "restarted" handler

SoupSessionAsync was mistakenly leaving its own "restarted" handler connected
even after "finished" was emitted. misc-test was supposed to catch that, but
it was using SoupSessionSync rather than Async. Changed it to use Async since
Async adds some signal handlers of its own, while Sync does not.

Fixes #380193, reported by Gustavo Noronha

15 years agoFix mbox to be a mailto
Dan Winship [Tue, 28 Apr 2009 14:37:51 +0000 (10:37 -0400)]
Fix mbox to be a mailto

15 years agoVerify that soup_auth_is_for_proxy() is returning the right thing
Dan Winship [Tue, 28 Apr 2009 14:37:01 +0000 (10:37 -0400)]
Verify that soup_auth_is_for_proxy() is returning the right thing

15 years agoBelatedly move Changelog to Changelog.pre-git, update for that
Dan Winship [Fri, 24 Apr 2009 15:01:13 +0000 (11:01 -0400)]
Belatedly move Changelog to Changelog.pre-git, update for that

15 years agoAdd a missing file to .gitignore
Dan Winship [Fri, 24 Apr 2009 15:00:47 +0000 (11:00 -0400)]
Add a missing file to .gitignore

15 years agoheader-parsing.c: Add tests for new bad-header-parsing rules
Dan Winship [Sun, 19 Apr 2009 15:25:14 +0000 (11:25 -0400)]
header-parsing.c: Add tests for new bad-header-parsing rules

15 years agoAdd an "expect_warning" variable to the test framework
Dan Winship [Sun, 19 Apr 2009 15:24:43 +0000 (11:24 -0400)]
Add an "expect_warning" variable to the test framework

This lets a test tell the log handler that it is expecting to cause a
warning, so it shouldn't be treated as an error

15 years agosoup_message_headers_append: require @name to be non-empty
Dan Winship [Sun, 19 Apr 2009 15:22:11 +0000 (11:22 -0400)]
soup_message_headers_append: require @name to be non-empty

15 years agoBug 579318 – stupid servers that return garbled headers
Dan Winship [Sun, 19 Apr 2009 15:22:49 +0000 (11:22 -0400)]
Bug 579318 – stupid servers that return garbled headers

soup_headers_parse: don't pass bad header lines to
soup_message_headers_append, since it will emit warnings in that case
now. Also, don't bail out because one line of the headers is bad.
Instead, just skip the bad line and continue with the next good line.

15 years agoMark soup_message_headers_get() deprecated
Dan Winship [Sun, 19 Apr 2009 14:40:41 +0000 (10:40 -0400)]
Mark soup_message_headers_get() deprecated

(in favor of soup_message_headers_get_one() and _get_list())

15 years agoAdd more tests to tests/date.c
Dan Winship [Sat, 18 Apr 2009 15:40:10 +0000 (11:40 -0400)]
Add more tests to tests/date.c

Conversion between formats
non-UTC timestamps
floating timestamps

15 years agoFix soup_date_to_string() to deal with non-UTC times correctly
Dan Winship [Sat, 18 Apr 2009 15:37:26 +0000 (11:37 -0400)]
Fix soup_date_to_string() to deal with non-UTC times correctly

soup_date_new_from_string: fix setting of UTC vs floating for ISO8601
and RFC2822 timestamps. Allow a time of "24:00:00" (per ISO).

soup_date_to_string: coerce @date to UTC for HTTP and cookie dates,
output correct offset for RFC2822, ISO compact, and ISO full.

15 years agoBug 579055 – SOUP_DATE_RFC2822 not handled in soup_date_to_string()
Enrico Tröger [Sat, 18 Apr 2009 13:33:39 +0000 (09:33 -0400)]
Bug 579055 – SOUP_DATE_RFC2822 not handled in soup_date_to_string()

15 years agoAdd a regression test for SOUP_SESSION_TIMEOUT
Dan Winship [Sat, 18 Apr 2009 12:34:22 +0000 (08:34 -0400)]
Add a regression test for SOUP_SESSION_TIMEOUT

15 years agoBug 578928 – SOUP_SESSION_TIMEOUT does not work with https
Dan Winship [Sat, 18 Apr 2009 12:31:48 +0000 (08:31 -0400)]
Bug 578928 – SOUP_SESSION_TIMEOUT does not work with https

soup-gnutls.c: Fix error-code logic to return G_IO_STATUS_AGAIN even
with synchronous connections, if the underlying socket operation
returns EAGAIN.

15 years agoBug 574414 – Make SOUP_SESSION_TIMEOUT work with SoupSessionAsync
Dan Winship [Sat, 18 Apr 2009 12:29:28 +0000 (08:29 -0400)]
Bug 574414 – Make SOUP_SESSION_TIMEOUT work with SoupSessionAsync

soup-socket.c: when adding an IO watch to wait for an async socket to
become readable/writable, add a timeout as well, and fail if the
timeout times out.

soup-connection.c: Pass the TIMEOUT flag to async sockets too

15 years agoAdd .gitignore
Dan Winship [Fri, 17 Apr 2009 14:43:34 +0000 (10:43 -0400)]
Add .gitignore

15 years agoAdd libsoup.doap
Dan Winship [Fri, 17 Apr 2009 14:13:51 +0000 (10:13 -0400)]
Add libsoup.doap

15 years ago2.26.1 LIBSOUP_2_26_1
Dan Winship [Mon, 13 Apr 2009 22:15:27 +0000 (22:15 +0000)]
2.26.1

* configure.in: 2.26.1

* NEWS: update

svn path=/trunk/; revision=1275

15 years ago2.26.1
Dan Winship [Mon, 13 Apr 2009 22:15:08 +0000 (22:15 +0000)]
2.26.1

* configure.in: 2.26.1

* NEWS: update

svn path=/trunk/; revision=1274

15 years agofix use of async context here
Dan Winship [Mon, 13 Apr 2009 21:50:14 +0000 (21:50 +0000)]
fix use of async context here

* libsoup/soup-proxy-resolver-gnome.c (get_proxy_async): fix
use of async context here

svn path=/trunk/; revision=1273

15 years ago Bug 578809 – warnings in soup_address_equal_by_ip
Dan Winship [Mon, 13 Apr 2009 20:52:07 +0000 (20:52 +0000)]
Bug 578809 – warnings in soup_address_equal_by_ip

* libsoup/soup-session.c (get_host_for_message): don't try to look
up unresolved addresses in the hosts hash; just return NULL. The
"cancel other messages for the bad host" code in connect_result()
will loop over the whole queue, including messages with unresolved
addresses.

svn path=/trunk/; revision=1272

15 years ago Bug 578746 – http_proxy env var set to "http://:80" in session
Dan Winship [Mon, 13 Apr 2009 14:04:54 +0000 (14:04 +0000)]
Bug 578746 – http_proxy env var set to ":80" in session
despite proxy mode "none"

* libsoup/soup-proxy-resolver-gnome.c (update_proxy_settings):
Don't set anything if proxy_mode is NONE, and properly ignore the
proxy host if it's empty.

svn path=/trunk/; revision=1271

15 years ago Bug 578645 - crash in Rhythmbox Music Player
Dan Winship [Sat, 11 Apr 2009 12:45:57 +0000 (12:45 +0000)]
Bug 578645 - crash in Rhythmbox Music Player

* libsoup/soup-proxy-resolver-gnome.c
(soup_proxy_resolver_gnome_init): unlock id.lock before freeing
it. (Fixes the case where SoupProxyResolverGNOME is first
initialized from a thread other than the main thread, while the
default main loop is running.)

svn path=/trunk/; revision=1270

15 years ago2.26.0.9, aka "I can't believe it's not 2.26.1!", a pre-release to give LIBSOUP_2_26_0_9
Dan Winship [Wed, 8 Apr 2009 18:07:48 +0000 (18:07 +0000)]
2.26.0.9, aka "I can't believe it's not 2.26.1!", a pre-release to give

* configure.in: 2.26.0.9, aka "I can't believe it's not 2.26.1!",
a pre-release to give the proxy resolver fix a bit of extra
testing before next week.

* NEWS: Update

svn path=/trunk/; revision=1268

15 years ago Re-fix GNOME proxy resolution (qv bug 571527)
Dan Winship [Wed, 8 Apr 2009 17:16:52 +0000 (17:16 +0000)]
Re-fix GNOME proxy resolution (qv bug 571527)

* libsoup/soup-proxy-resolver-gnome.c: New and improved GNOME
proxy resolver; gets information out of GConf in a thread-safe
manner, and then passes it on to libproxy via environment
variables, so that libproxy won't try to access GConf itself, but
we still can use it for PAC, WPAD, and ignore_hosts.

* libsoup/soup-proxy-resolver-gconf.c:
* libsoup/soup-proxy-resolver-libproxy.c: gone now

* libsoup/soup-gnome-features.c: update for the fact that
SoupProxyResolverGNOME is actually a real type now, not
compile-time-defined alias

svn path=/trunk/; revision=1267

15 years ago Fix ISO 8601 parsing to accept either "." or ","
Dan Winship [Wed, 8 Apr 2009 14:23:15 +0000 (14:23 +0000)]
Fix ISO 8601 parsing to accept either "." or ","

* libsoup/soup-date.c (parse_iso8601_date): accept either "." or
"," as decimal separator, per the spec. qv glib bug 578369.

svn path=/trunk/; revision=1265