platform/upstream/curl.git
15 years agoimprove presentation of accepted/rejected debug/optimizer options
Yang Tse [Wed, 8 Oct 2008 19:38:01 +0000 (19:38 +0000)]
improve presentation of accepted/rejected debug/optimizer options

15 years agofix missing null-pointer check
Yang Tse [Wed, 8 Oct 2008 18:32:06 +0000 (18:32 +0000)]
fix missing null-pointer check

15 years agorefactoring of COMPILER_BASIC_OPTS
Yang Tse [Wed, 8 Oct 2008 16:12:10 +0000 (16:12 +0000)]
refactoring of COMPILER_BASIC_OPTS

15 years ago179 - "[PATCH] Colon in username"
Daniel Stenberg [Wed, 8 Oct 2008 10:40:51 +0000 (10:40 +0000)]
179 - "[PATCH] Colon in username"
was committed just now!

15 years ago- Igor Novoseltsev brought a patch that introduced two new options to
Daniel Stenberg [Wed, 8 Oct 2008 10:39:43 +0000 (10:39 +0000)]
- Igor Novoseltsev brought a patch that introduced two new options to
  curl_easy_setopt: CURLOPT_USERNAME and CURLOPT_PASSWORD that sort of
  deprecates the good old CURLOPT_USERPWD since they allow applications to set
  the user name and password independently and perhaps more importantly allow
  both to contain colon(s) which CURLOPT_USERPWD doesn't fully support.

15 years agoInitial attempt to detect SUN C compiler
Yang Tse [Wed, 8 Oct 2008 03:50:45 +0000 (03:50 +0000)]
Initial attempt to detect SUN C compiler

15 years agoCreated test cases 1080 and 1081 to reproduce a problem of
Dan Fandrich [Wed, 8 Oct 2008 03:32:10 +0000 (03:32 +0000)]
Created test cases 1080 and 1081 to reproduce a problem of
CURLINFO_REDIRECT_URL leaking memory and returning incorrect results when
two URLs are requested. Reported by vmpdemo in bug #2152270

15 years agoInitial attempt to detect HP C compiler
Yang Tse [Wed, 8 Oct 2008 02:29:13 +0000 (02:29 +0000)]
Initial attempt to detect HP C compiler

15 years agoAdded const to some pointer variables
Dan Fandrich [Wed, 8 Oct 2008 01:17:51 +0000 (01:17 +0000)]
Added const to some pointer variables

15 years agoSplit off Curl_isxdigit function
Dan Fandrich [Tue, 7 Oct 2008 23:20:06 +0000 (23:20 +0000)]
Split off Curl_isxdigit function

15 years agoChanged the handling of read/write errors in Curl_perform() to allow a
Dan Fandrich [Tue, 7 Oct 2008 23:15:02 +0000 (23:15 +0000)]
Changed the handling of read/write errors in Curl_perform() to allow a
a fresh connection to be made in such cases and the request retransmitted.
This should fix test case 160.  Added test case 1079 in an attempt to
test a similar connection dropping scenario, but as a race condition, it's
hard to test reliably.

15 years agomention the other sf.net bug which does have a sort of fix mentioned
Daniel Stenberg [Tue, 7 Oct 2008 22:53:33 +0000 (22:53 +0000)]
mention the other sf.net bug which does have a sort of fix mentioned

15 years agosome additional classification of the entries
Daniel Stenberg [Tue, 7 Oct 2008 22:49:11 +0000 (22:49 +0000)]
some additional classification of the entries

15 years agoOne fix for CURLINFO_PRIMARY_IP just went in which addressed at least one of
Daniel Stenberg [Tue, 7 Oct 2008 22:13:34 +0000 (22:13 +0000)]
One fix for CURLINFO_PRIMARY_IP just went in which addressed at least one of
the concerns mentioned for 174 and the other guy hasn't returned so I remove
the entry from here now.

15 years ago- Fixed CURLINFO_PRIMARY_IP: When libcurl created a connection to host A then
Daniel Stenberg [Tue, 7 Oct 2008 21:56:56 +0000 (21:56 +0000)]
- Fixed CURLINFO_PRIMARY_IP: When libcurl created a connection to host A then
  the app re-used the handle to do a connection to host B and then again
  re-used the handle to host A, it would not update the info with host A's IP
  address (due to the connection being re-used) but it would instead report
  the info from host B.

15 years agoDocument how to disable tests when certain --disable-* flags are used.
Dan Fandrich [Tue, 7 Oct 2008 20:28:25 +0000 (20:28 +0000)]
Document how to disable tests when certain --disable-* flags are used.

15 years agofix compiler warning: 'dot_4' may be used uninitialized in this function
Yang Tse [Tue, 7 Oct 2008 18:44:32 +0000 (18:44 +0000)]
fix compiler warning: 'dot_4' may be used uninitialized in this function

15 years agofix compiler warning: dereferencing type-punned pointer will break strict-aliasing...
Yang Tse [Tue, 7 Oct 2008 18:28:24 +0000 (18:28 +0000)]
fix compiler warning: dereferencing type-punned pointer will break strict-aliasing rules

15 years agooops, counted one of them twice!
Daniel Stenberg [Tue, 7 Oct 2008 18:14:11 +0000 (18:14 +0000)]
oops, counted one of them twice!

15 years agotwo more bindings
Daniel Stenberg [Tue, 7 Oct 2008 18:10:16 +0000 (18:10 +0000)]
two more bindings

15 years agoadjust ICC_windows settings
Yang Tse [Tue, 7 Oct 2008 17:37:25 +0000 (17:37 +0000)]
adjust ICC_windows settings

15 years agofix VAR_STRIP
Yang Tse [Tue, 7 Oct 2008 16:53:46 +0000 (16:53 +0000)]
fix VAR_STRIP

15 years agoSync up with reality
Yang Tse [Tue, 7 Oct 2008 13:34:59 +0000 (13:34 +0000)]
Sync up with reality

15 years agoInitial attempt to support configure's --(dis|en)able-optimize
Yang Tse [Tue, 7 Oct 2008 13:13:34 +0000 (13:13 +0000)]
Initial attempt to support configure's --(dis|en)able-optimize
option to specify dis(activation) of compiler optimizations.

If option is specified, it will be honored independant of the
--(dis|en)able-debug option.

15 years agorevert change introduced in tftpd.c revision 1.44
Yang Tse [Fri, 3 Oct 2008 03:42:44 +0000 (03:42 +0000)]
revert change introduced in tftpd.c revision 1.44

15 years agofix another error introduced in previously mentioned commit
Yang Tse [Fri, 3 Oct 2008 02:25:52 +0000 (02:25 +0000)]
fix another error introduced in previously mentioned commit

15 years agofix error introduced in previous commit
Yang Tse [Fri, 3 Oct 2008 00:35:22 +0000 (00:35 +0000)]
fix error introduced in previous commit

15 years agotake 2 at fixing compiler warning: argument might be clobbered by `longjmp' or `vfork'
Yang Tse [Thu, 2 Oct 2008 22:58:09 +0000 (22:58 +0000)]
take 2 at fixing compiler warning: argument might be clobbered by `longjmp' or `vfork'

15 years agofix compiler warning: dereferencing type-punned pointer will break strict-aliasing...
Yang Tse [Thu, 2 Oct 2008 14:44:18 +0000 (14:44 +0000)]
fix compiler warning: dereferencing type-punned pointer will break strict-aliasing rules

15 years agofix compiler warning: `variable' might be clobbered by `longjmp' or `vfork'
Yang Tse [Thu, 2 Oct 2008 14:42:57 +0000 (14:42 +0000)]
fix compiler warning: `variable' might be clobbered by `longjmp' or `vfork'

15 years agofix compiler warning: `variable' might be clobbered by `longjmp' or `vfork'
Yang Tse [Thu, 2 Oct 2008 13:28:10 +0000 (13:28 +0000)]
fix compiler warning: `variable' might be clobbered by `longjmp' or `vfork'

15 years agofix comment
Yang Tse [Thu, 2 Oct 2008 04:27:42 +0000 (04:27 +0000)]
fix comment

15 years agodebug option check must be before warnings option check
Yang Tse [Thu, 2 Oct 2008 04:17:36 +0000 (04:17 +0000)]
debug option check must be before warnings option check

15 years agoruntests.pl enables picky compiler warnings unless explicitly disabled
Yang Tse [Thu, 2 Oct 2008 03:59:25 +0000 (03:59 +0000)]
runtests.pl enables picky compiler warnings unless explicitly disabled

15 years agoInitial attempt to support configure's --(dis|en)able-warnings
Yang Tse [Thu, 2 Oct 2008 03:56:55 +0000 (03:56 +0000)]
Initial attempt to support configure's --(dis|en)able-warnings
option to specify dis(activation) of picky compiler warnings.

If option is specified, it will be honored independant of the
--(dis|en)able-debug option.

If option is not specified, it will follow --(dis|en)able-debug
setting, whose default is disabled if not specified.

15 years ago"make clean" now cleans out the docs and tests directories, too.
Dan Fandrich [Wed, 1 Oct 2008 18:29:13 +0000 (18:29 +0000)]
"make clean" now cleans out the docs and tests directories, too.

15 years agoFixed some compiler warnings with gcc
Dan Fandrich [Wed, 1 Oct 2008 17:34:24 +0000 (17:34 +0000)]
Fixed some compiler warnings with gcc

15 years agoremove unnecessary typecast, otherwise triggering compiler warning:
Yang Tse [Tue, 30 Sep 2008 18:59:02 +0000 (18:59 +0000)]
remove unnecessary typecast, otherwise triggering compiler warning:
dereferencing type-punned pointer will break strict-aliasing rules

15 years agoAdded "HTTP proxy" keyword
Dan Fandrich [Tue, 30 Sep 2008 17:18:27 +0000 (17:18 +0000)]
Added "HTTP proxy" keyword

15 years agofix compiler warning: dereferencing type-punned pointer will break strict-aliasing...
Yang Tse [Tue, 30 Sep 2008 17:04:05 +0000 (17:04 +0000)]
fix compiler warning: dereferencing type-punned pointer will break strict-aliasing rules

15 years agofix compiler warning: declaration of 'err' shadows a previous local
Yang Tse [Tue, 30 Sep 2008 15:10:27 +0000 (15:10 +0000)]
fix compiler warning: declaration of 'err' shadows a previous local

15 years agofix compiler warning: function declaration isn't a prototype
Yang Tse [Tue, 30 Sep 2008 15:06:03 +0000 (15:06 +0000)]
fix compiler warning: function declaration isn't a prototype

15 years agoInclude latest parameter definitions in ILE/RPG bindings
Patrick Monnerat [Tue, 30 Sep 2008 14:51:28 +0000 (14:51 +0000)]
Include latest parameter definitions in ILE/RPG bindings

15 years agofix compiler warning: enumerated type mixed with another type
Yang Tse [Tue, 30 Sep 2008 12:50:52 +0000 (12:50 +0000)]
fix compiler warning: enumerated type mixed with another type

15 years agonow compiler warnings are activated for all gcc builds, not only debug ones.
Yang Tse [Tue, 30 Sep 2008 12:39:54 +0000 (12:39 +0000)]
now compiler warnings are activated for all gcc builds, not only debug ones.

15 years ago- The libcurl FTP code now returns CURLE_REMOTE_FILE_NOT_FOUND error when SIZE
Daniel Stenberg [Tue, 30 Sep 2008 09:51:58 +0000 (09:51 +0000)]
- The libcurl FTP code now returns CURLE_REMOTE_FILE_NOT_FOUND error when SIZE
  gets a 550 response back for the cases where a download (or NOBODY) is
  wanted. It still allows a 550 as response if the SIZE is used as part of an
  upload process (like if resuming an upload is requested and the file isn't
  there before the upload). I also modified the FTP test server and a few test
  cases accordingly to match this modified behavior.

15 years agoUse CFLAGS for icc linker options instead of LDFLAGS,
Yang Tse [Tue, 30 Sep 2008 02:59:35 +0000 (02:59 +0000)]
Use CFLAGS for icc linker options instead of LDFLAGS,
otherwise gethostbyname() is not detected.

15 years agoFixed a couple of compile warnings.
Dan Fandrich [Mon, 29 Sep 2008 23:45:43 +0000 (23:45 +0000)]
Fixed a couple of compile warnings.

15 years agoMade the month days table static const
Dan Fandrich [Mon, 29 Sep 2008 23:22:41 +0000 (23:22 +0000)]
Made the month days table static const

15 years agoFixed a compile problem with --disable-proxy
Dan Fandrich [Mon, 29 Sep 2008 22:45:25 +0000 (22:45 +0000)]
Fixed a compile problem with --disable-proxy

15 years agoAdded "HTTP proxy" and "proxytunnel" keywords where applicable
Dan Fandrich [Mon, 29 Sep 2008 22:44:04 +0000 (22:44 +0000)]
Added "HTTP proxy" and "proxytunnel" keywords where applicable

15 years ago177 - "[Patch] Disable proxy support" was applied
Daniel Stenberg [Mon, 29 Sep 2008 21:49:31 +0000 (21:49 +0000)]
177 - "[Patch] Disable proxy support" was applied

15 years agospell!
Daniel Stenberg [Mon, 29 Sep 2008 21:49:10 +0000 (21:49 +0000)]
spell!

15 years ago- Daniel Egger provided a patch that allows you to disable proxy support in
Daniel Stenberg [Mon, 29 Sep 2008 21:46:04 +0000 (21:46 +0000)]
- Daniel Egger provided a patch that allows you to disable proxy support in
  libcurl to somewhat reduce the size of the binary. Run configure
  --disable-proxy.

15 years agoRemoved unneeded includes of signal.h and setjmp.h
Dan Fandrich [Mon, 29 Sep 2008 21:44:50 +0000 (21:44 +0000)]
Removed unneeded includes of signal.h and setjmp.h

15 years agoMoved all signal-based name resolution timeout handling into a single new
Dan Fandrich [Mon, 29 Sep 2008 21:02:22 +0000 (21:02 +0000)]
Moved all signal-based name resolution timeout handling into a single new
Curl_resolv_timeout function to reduce coupling.

15 years agouse ac_cv_compiler and ac_cv_compiler_num to keep compiler ID and version number
Yang Tse [Mon, 29 Sep 2008 19:14:52 +0000 (19:14 +0000)]
use ac_cv_compiler and ac_cv_compiler_num to keep compiler ID and version number

15 years agoTemporary icc adjustment:
Yang Tse [Mon, 29 Sep 2008 16:09:31 +0000 (16:09 +0000)]
Temporary icc adjustment:

  Disable floating point optimizations

15 years ago- Ian Lynagh provided a patch that now makes CURLOPT_RANGE work fine for SFTP
Daniel Stenberg [Mon, 29 Sep 2008 13:21:21 +0000 (13:21 +0000)]
- Ian Lynagh provided a patch that now makes CURLOPT_RANGE work fine for SFTP
  downloads!

15 years ago175 - "No info on REDIRECT_URL in multi-mode" by Max Ivanov is fixed!
Daniel Stenberg [Mon, 29 Sep 2008 13:16:16 +0000 (13:16 +0000)]
175 - "No info on REDIRECT_URL in multi-mode" by Max Ivanov is fixed!

15 years agocredit Maxim Ivanov
Daniel Stenberg [Mon, 29 Sep 2008 12:36:47 +0000 (12:36 +0000)]
credit Maxim Ivanov

15 years ago- Bug #2107803 (http://curl.haxx.se/bug/view.cgi?id=2107803) "no
Daniel Stenberg [Mon, 29 Sep 2008 12:22:10 +0000 (12:22 +0000)]
- Bug #2107803 (curl.haxx.se/bug/view.cgi?id=2107803) "no
  CURLINFO_REDIRECT_URL in multi mode" also contained a patch that fixed the
  problem.

15 years agomade Curl_read_plain() return an 'int' instead of CURLcode since it actually
Daniel Stenberg [Mon, 29 Sep 2008 11:13:37 +0000 (11:13 +0000)]
made Curl_read_plain() return an 'int' instead of CURLcode since it actually
returns -1 in EAGAIN cases and that's not valid CURLcode

15 years agothree more patches in the pipe
Daniel Stenberg [Mon, 29 Sep 2008 06:07:20 +0000 (06:07 +0000)]
three more patches in the pipe

15 years agoFixed a few comment typos
Dan Fandrich [Fri, 26 Sep 2008 19:55:12 +0000 (19:55 +0000)]
Fixed a few comment typos

15 years agoif a test execution is aborted due to a signal, and the process
Yang Tse [Fri, 26 Sep 2008 18:56:56 +0000 (18:56 +0000)]
if a test execution is aborted due to a signal, and the process
returns no other exitcode, use 2000 + signal number as return code

15 years agosiginterrupt, signal and sigsetjmp checks
Yang Tse [Fri, 26 Sep 2008 18:28:32 +0000 (18:28 +0000)]
siginterrupt, signal and sigsetjmp checks

15 years agomore
Daniel Stenberg [Fri, 26 Sep 2008 18:17:01 +0000 (18:17 +0000)]
more

15 years agouse shift right. Division result is not integral.
Yang Tse [Fri, 26 Sep 2008 14:08:33 +0000 (14:08 +0000)]
use shift right. Division result is not integral.

15 years agoAvoid the use of the '? :' operator inside the call to our
Yang Tse [Fri, 26 Sep 2008 11:21:22 +0000 (11:21 +0000)]
Avoid the use of the '? :' operator inside the call to our
test-server logging function. It doesn't work on some systems.

15 years agoattempt to avoid HTTP server startup misdetection on some systems
Yang Tse [Fri, 26 Sep 2008 01:08:24 +0000 (01:08 +0000)]
attempt to avoid HTTP server startup misdetection on some systems

15 years agofix potential buffer overflow in test-server logging function
Yang Tse [Fri, 26 Sep 2008 00:17:01 +0000 (00:17 +0000)]
fix potential buffer overflow in test-server logging function

15 years agogive proper credit!
Daniel Stenberg [Thu, 25 Sep 2008 22:35:57 +0000 (22:35 +0000)]
give proper credit!

15 years ago- Fixed the HTTP Digest auth code to not behave badly when getting a blank realm
Daniel Stenberg [Thu, 25 Sep 2008 14:09:22 +0000 (14:09 +0000)]
- Fixed the HTTP Digest auth code to not behave badly when getting a blank realm
  with realm="". http://curl.haxx.se/bug/view.cgi?id=2126435

15 years agoHAVE_INET_PTON will only be defined when an IPv6 capable working
Yang Tse [Wed, 24 Sep 2008 19:13:01 +0000 (19:13 +0000)]
HAVE_INET_PTON will only be defined when an IPv6 capable working
inet_pton function is available.

15 years agoconfigure script now checks availability of the inet_ntop() function,
Yang Tse [Wed, 24 Sep 2008 16:43:40 +0000 (16:43 +0000)]
configure script now checks availability of the inet_ntop() function,
and when not crosscompiling verifies if it is IPv6 capable.

HAVE_INET_NTOP will only be defined when an IPv6 capable working
inet_ntop function is available.

15 years agoHAVE_INET_NTOP will only be defined when an IPv6 capable working
Yang Tse [Wed, 24 Sep 2008 16:43:12 +0000 (16:43 +0000)]
HAVE_INET_NTOP will only be defined when an IPv6 capable working
inet_ntop function is available.

15 years agoOfficial OpenSSL released versions up to and including 0.9.8i as well as
Yang Tse [Wed, 24 Sep 2008 13:55:23 +0000 (13:55 +0000)]
Official OpenSSL released versions up to and including 0.9.8i as well as
2008-09-24 stable snapshot have a buf_mem_st.length structure member with
'int' data type.

OpenSSL un-released 0.9.9 CVS version has a buf_mem_st.length structure member
with 'size_t' data type since 2007-Oct-09.

These 4 typecasts should silence compiler warnings in all cases.

15 years agotemporarily enable CURL_CHECK_FUNC_INET_NTOA_R
Yang Tse [Wed, 24 Sep 2008 12:38:11 +0000 (12:38 +0000)]
temporarily enable CURL_CHECK_FUNC_INET_NTOA_R

15 years agontoa() and inet_ntoa_r() no longer used
Yang Tse [Wed, 24 Sep 2008 12:22:16 +0000 (12:22 +0000)]
ntoa() and inet_ntoa_r() no longer used

15 years ago4.16 My HTTP POST or PUT requests are slow!
Daniel Stenberg [Wed, 24 Sep 2008 07:50:46 +0000 (07:50 +0000)]
4.16 My HTTP POST or PUT requests are slow!

15 years agoAdded "5.14 Using C++ non-static functions for callbacks?"
Daniel Stenberg [Wed, 24 Sep 2008 07:39:42 +0000 (07:39 +0000)]
Added "5.14 Using C++ non-static functions for callbacks?"

15 years agoRe-enable test #100 on i686 icc autobuilds
Yang Tse [Wed, 24 Sep 2008 01:29:23 +0000 (01:29 +0000)]
Re-enable test #100 on i686 icc autobuilds

15 years agoMake sure not to dereference the wrong UrlState proto union member when
Dan Fandrich [Wed, 24 Sep 2008 01:08:01 +0000 (01:08 +0000)]
Make sure not to dereference the wrong UrlState proto union member when
switching from one protocol to another in a single request (e.g.
redirecting from HTTP to FTP as in test 1055) by resetting
state.expect100header before every request.

15 years agoRemove usage of inet_ntoa and inet_ntoa_r
Yang Tse [Tue, 23 Sep 2008 19:17:19 +0000 (19:17 +0000)]
Remove usage of inet_ntoa and inet_ntoa_r

15 years agoReplace inet_ntoa and inet_ntoa_r with Curl_inet_ntop
Yang Tse [Tue, 23 Sep 2008 19:16:56 +0000 (19:16 +0000)]
Replace inet_ntoa and inet_ntoa_r with Curl_inet_ntop

15 years agoAvoid 'warning: comparison between signed and unsigned'.
Gisle Vanem [Tue, 23 Sep 2008 13:16:36 +0000 (13:16 +0000)]
Avoid 'warning: comparison between signed and unsigned'.

15 years ago172 was applied, 173 is now being discussed
Daniel Stenberg [Tue, 23 Sep 2008 11:31:57 +0000 (11:31 +0000)]
172 was applied, 173 is now being discussed

15 years agoremove the entire comment now when we no longer set TZ
Daniel Stenberg [Tue, 23 Sep 2008 11:02:56 +0000 (11:02 +0000)]
remove the entire comment now when we no longer set TZ

15 years ago- Introducing Jamie Lokier's function for date to epoch conversion used in the
Daniel Stenberg [Tue, 23 Sep 2008 11:00:01 +0000 (11:00 +0000)]
- Introducing Jamie Lokier's function for date to epoch conversion used in the
  date parser function. This makes our function less dependent on system-
  provided functions and instead we do all the magic ourselves. We also no
  longer depend on the TZ environment variable.

15 years agoissue #171 is now in CVS
Daniel Stenberg [Tue, 23 Sep 2008 10:27:41 +0000 (10:27 +0000)]
issue #171 is now in CVS

15 years ago- Rob Crittenden brought a patch to "add some locking for thread-safety to NSS
Daniel Stenberg [Tue, 23 Sep 2008 10:27:04 +0000 (10:27 +0000)]
- Rob Crittenden brought a patch to "add some locking for thread-safety to NSS
  implementation".

15 years ago- Made the SOCKS code use the new Curl_read_plain() function to fix the bug
Daniel Stenberg [Mon, 22 Sep 2008 23:12:00 +0000 (23:12 +0000)]
- Made the SOCKS code use the new Curl_read_plain() function to fix the bug
  Markus Moeller reported: http://curl.haxx.se/mail/archive-2008-09/0016.html

- recv() errors other than those equal to EAGAIN now cause proper
  CURLE_RECV_ERROR to get returned. This made test case 160 fail so I've now
  disabled it until we can figure out another way to exercise that logic.

15 years agoremoved issue #162 as mentioned on the mailing list http://curl.haxx.se/mail/archive...
Daniel Stenberg [Mon, 22 Sep 2008 22:27:43 +0000 (22:27 +0000)]
removed issue #162 as mentioned on the mailing list curl.haxx.se/mail/archive-2008-09/0086.html

15 years agobetter description
Daniel Stenberg [Mon, 22 Sep 2008 21:33:18 +0000 (21:33 +0000)]
better description

15 years ago- Michael Goffioul filed bug report #2107377 "Problem with mutli + GnuTLS +
Daniel Stenberg [Mon, 22 Sep 2008 20:42:13 +0000 (20:42 +0000)]
- Michael Goffioul filed bug report #2107377 "Problem with mutli + GnuTLS +
  proxy" (http://curl.haxx.se/bug/view.cgi?id=2107377) that showed how a multi
  interface using program didn't work when built with GnuTLS and a CONNECT
  request was done over a proxy (basically test 502 over a proxy to a HTTPS
  site). It turned out the ssl connect function would get called twice which
  caused the second call to fail.

15 years agoRe-enable tests #1 on i686 icc autobuilds
Yang Tse [Mon, 22 Sep 2008 18:55:42 +0000 (18:55 +0000)]
Re-enable tests #1 on i686 icc autobuilds

15 years agoicc adjustments for icc 9.0 and prior versions:
Yang Tse [Mon, 22 Sep 2008 18:42:48 +0000 (18:42 +0000)]
icc adjustments for icc 9.0 and prior versions:

  Disable remark #279: controlling expression is constant

      Remark triggered mostly on va_arg() and FD_ZERO() macros.

15 years agoArgument to CURLMOPT_MAXCONNECTS must be a long
Dan Fandrich [Mon, 22 Sep 2008 17:27:24 +0000 (17:27 +0000)]
Argument to CURLMOPT_MAXCONNECTS must be a long

15 years agoFixed test 539 to handle an out of memory condition that shows up now
Dan Fandrich [Mon, 22 Sep 2008 17:20:29 +0000 (17:20 +0000)]
Fixed test 539 to handle an out of memory condition that shows up now
that memdebug.h is included in the test programs.