Daniel Stenberg [Wed, 26 Dec 2007 21:48:52 +0000 (21:48 +0000)]
In an attempt to repeat the problem in bug report #1850730
(http://curl.haxx.se/bug/view.cgi?id=1850730) I wrote up test case 552. The
test is doing a 70K POST with a read callback and an ioctl callback over a
proxy requiring Digest auth. The test case code is more or less identical to
the test recipe code provided by Spacen Jasset (who submitted the bug report).
Daniel Stenberg [Mon, 24 Dec 2007 23:45:48 +0000 (23:45 +0000)]
Gary Maxwell filed bug report #1856628
(http://curl.haxx.se/bug/view.cgi?id=1856628) and provided a fix for the
(small) memory leak in the SSL session ID caching code. It happened when a
previous entry in the cache was re-used.
Yang Tse [Tue, 18 Dec 2007 18:33:24 +0000 (18:33 +0000)]
(http://curl.haxx.se/mail/archive-2007-12/0039.html) reported and fixed
a file truncation problem on Windows build targets triggered when retrying
a download with curl.
Yang Tse [Tue, 18 Dec 2007 18:08:19 +0000 (18:08 +0000)]
MSVC 9.0 (VS2008) does not support Windows build targets prior to WinXP,
and makes wrong asumptions of build target when it isn't specified. So,
if no build target has been defined we will target WinXP when building
with MSVC 9.0 (VS2008).
Daniel Stenberg [Mon, 17 Dec 2007 21:19:42 +0000 (21:19 +0000)]
Mateusz Loskot pointed out that VC++ 9.0 (2008) has the pollfd struct and
defines in the SDK somehow differently so we have to add a define to the
config-win32.h file to make select.h compile nicely.
Daniel Stenberg [Thu, 13 Dec 2007 10:00:06 +0000 (10:00 +0000)]
David Wright filed bug report #1849764
(http://curl.haxx.se/bug/view.cgi?id=1849764) with an included fix. He
identified a problem for re-used connections that previously had sent
Expect: 100-continue and in some situations the subsequent POST (that didn't
use Expect:) still had the internal flag set for its use. David's fix (that
makes the setting of the flag in every single request unconditionally) is
fine and is now used!
Daniel Stenberg [Mon, 10 Dec 2007 22:20:26 +0000 (22:20 +0000)]
grrr, the previous commit was meant to properly make sure that we don't
link any executables when doing debug builds since they kind of assume
symbols provided by libcurl, but it also wrongly included acountry.c
Daniel Stenberg [Sun, 9 Dec 2007 22:31:53 +0000 (22:31 +0000)]
Andrew Moise filed bug report #1847501
(http://curl.haxx.se/bug/view.cgi?id=1847501) and pointed out a memcpy()
that should be memmove() in the convert_lineends() function.
Daniel Stenberg [Sat, 8 Dec 2007 23:01:46 +0000 (23:01 +0000)]
Travelling some 500km by train back and forth on the same day gives you time
to do things you don't otherwise do, but here's the summary of today's work...
Daniel Stenberg [Sat, 8 Dec 2007 22:50:55 +0000 (22:50 +0000)]
All static functions that were previously name Curl_* something no longer
use that prefix as we use that prefix only for library-wide internal global
symbols.
Daniel Stenberg [Thu, 6 Dec 2007 22:36:52 +0000 (22:36 +0000)]
clarify that when curl_multi_timeout() returns -1 it just means that there
is no current timeout. It does not mean wait forever and it does not mean
do not wait at all. It means there is no timeout value known at this point in
time.
Daniel Stenberg [Wed, 5 Dec 2007 21:20:14 +0000 (21:20 +0000)]
Spacen Jasset reported a problem with doing POST (with data read with a
callback) over a proxy when NTLM is used as auth with the proxy. The bug
also concerned Digest and was limited to using callback only. Spacen worked
with us to provide a useful patch. I added the test case 547 and 548 to
verify two variations of POST over proxy with NTLM.
Daniel Stenberg [Wed, 5 Dec 2007 11:08:56 +0000 (11:08 +0000)]
added test548 which uses the lib547 source file, preparing for test547 which
is supposed to repeat the bug report "NTLM proxy authentication with
CURLOPT_READDATA seems broken." posted on the curl-library mailing list on dec
3 2007.
Daniel Stenberg [Mon, 3 Dec 2007 11:48:09 +0000 (11:48 +0000)]
Bug report #1842029 (http://curl.haxx.se/bug/view.cgi?id=1842029) identified
a problem with SSL session caching that prevent it from working, and the
associated fix!
Daniel Stenberg [Mon, 3 Dec 2007 11:39:27 +0000 (11:39 +0000)]
Now libcurl (built with OpenSSL) doesn't return error anymore if the remote
SSL-based server doesn't present a certificate when the request is told to
ignore certificate verification anyway.
Daniel Stenberg [Mon, 3 Dec 2007 09:50:32 +0000 (09:50 +0000)]
removed the ;type= thing for FTP urls through proxy, since that's now only
present when enabled by on option which isn't done by default (and isn't even
available for the curl app atm)
Daniel Stenberg [Sun, 2 Dec 2007 23:38:23 +0000 (23:38 +0000)]
Michal Marek introduced CURLOPT_PROXY_TRANSFER_MODE which is used to control
the appending of the "type=" thing on FTP URLs when they are passed to a
HTTP proxy. Some proxies just don't like that appending (which is done
unconditionally in 7.17.1), and some proxies treat binary/ascii transfers
better with the appending done!
Daniel Stenberg [Thu, 29 Nov 2007 22:14:48 +0000 (22:14 +0000)]
A bug report on the curl-library list showed a HTTP Digest session going on
with a 700+ letter nonce. Previously libcurl only support 127 letter ones
and now I bumped it to 1023.
Daniel Stenberg [Wed, 28 Nov 2007 10:46:40 +0000 (10:46 +0000)]
fix next_lookup() to continue searching even if c-ares failed to load the
/etc/hosts file, pointed out by Erik Kline:
http://daniel.haxx.se/projects/c-ares/mail/c-ares-archive-2007-11/0027.shtml
Yang Tse [Tue, 27 Nov 2007 00:52:30 +0000 (00:52 +0000)]
Explicitly disallow remote hosts to connect to local forwarded ports,
the socks server port in the test suite. This is the default setting
unless a tinkered built ssh is being used.
Yang Tse [Mon, 26 Nov 2007 14:07:09 +0000 (14:07 +0000)]
Providing an explicit bind address besides the port for dynamic application-level
port forwarding, our socks port, prevents ssh from running on some systems.
By default, ssh binds local port forwardings to the loopback address, since this
was the address being given as the explicit bind address, now it isn't given.
Daniel Stenberg [Sat, 24 Nov 2007 23:16:55 +0000 (23:16 +0000)]
struct HandleData is now called struct SingleRequest, and is only for data that
is inited at the start of the DO action. I removed the Curl_transfer_keeper
struct completely, and I had to move out a few struct members (that had to
be set before DO or used after DONE) to the UrlState struct. The SingleRequest
struct is accessed with SessionHandle->req.
One of the biggest reasons for doing this was the bunch of duplicate struct
members in HandleData and Curl_transfer_keeper since it was really messy to
keep track of two variables with the same name and basically the same purpose!