platform/upstream/curl.git
20 years agonew mirror and one that changed domain
Daniel Stenberg [Tue, 20 Apr 2004 07:52:16 +0000 (07:52 +0000)]
new mirror and one that changed domain

20 years agomake the first response get a "connection: close" header as that is how
Daniel Stenberg [Tue, 20 Apr 2004 07:36:09 +0000 (07:36 +0000)]
make the first response get a "connection: close" header as that is how
most 1.1 connections will be closed. The case where it gets closed anyway
is subject for a new separate test.

20 years agochanged the logging to work when the logfile is removed during testing
Daniel Stenberg [Mon, 19 Apr 2004 15:20:30 +0000 (15:20 +0000)]
changed the logging to work when the logfile is removed during testing

20 years agoMake the server include "Connection: close" in the headers of the replies
Daniel Stenberg [Mon, 19 Apr 2004 08:41:36 +0000 (08:41 +0000)]
Make the server include "Connection: close" in the headers of the replies
it actually will close. This is after all what HTTP 1.1 says a server should
do.

20 years agoGisle Vanem corrected a mistake in a recent progress fix
Daniel Stenberg [Mon, 19 Apr 2004 07:18:26 +0000 (07:18 +0000)]
Gisle Vanem corrected a mistake in a recent progress fix

20 years agoGisle Vanem: patches to make sws.c compile under MingW/MSVC is
Daniel Stenberg [Sat, 17 Apr 2004 11:38:41 +0000 (11:38 +0000)]
Gisle Vanem: patches to make sws.c compile under MingW/MSVC is
attached. And some cosmetic fixes.

20 years agoGisle made a failed connect output the reason for it
Daniel Stenberg [Sat, 17 Apr 2004 11:33:47 +0000 (11:33 +0000)]
Gisle made a failed connect output the reason for it

20 years agoNDEBUG should not be defined when CURLDEBUG is
Daniel Stenberg [Sat, 17 Apr 2004 11:33:14 +0000 (11:33 +0000)]
NDEBUG should not be defined when CURLDEBUG is

20 years agoadding issue 36 to be fixed before release
Daniel Stenberg [Fri, 16 Apr 2004 07:33:53 +0000 (07:33 +0000)]
adding issue 36 to be fixed before release

20 years agodon't display the . and .. files when dumping the log/ contents
Daniel Stenberg [Fri, 16 Apr 2004 07:02:17 +0000 (07:02 +0000)]
don't display the . and .. files when dumping the log/ contents

20 years agoremade the logging function to better deal with removed logfiles during
Daniel Stenberg [Fri, 16 Apr 2004 07:01:47 +0000 (07:01 +0000)]
remade the logging function to better deal with removed logfiles during
the execution of the tests

20 years agoissue 31 - fix windows multi interface is now corrected
Daniel Stenberg [Thu, 15 Apr 2004 15:05:48 +0000 (15:05 +0000)]
issue 31 - fix windows multi interface is now corrected

20 years agoclarified that select() timeouts should always remain rather short
Daniel Stenberg [Thu, 15 Apr 2004 15:05:12 +0000 (15:05 +0000)]
clarified that select() timeouts should always remain rather short

20 years agomy test suite edits
Daniel Stenberg [Thu, 15 Apr 2004 14:53:47 +0000 (14:53 +0000)]
my test suite edits

20 years agochange the log format to look similar to sws
Daniel Stenberg [Thu, 15 Apr 2004 13:55:37 +0000 (13:55 +0000)]
change the log format to look similar to sws
added various logging info

20 years agoclean the log directory between each single test, so that we can better
Daniel Stenberg [Thu, 15 Apr 2004 13:37:19 +0000 (13:37 +0000)]
clean the log directory between each single test, so that we can better
display all logs and only logs with relevant data when a test fails and -p
is used.

20 years agoftp->dirs[] is no longer terminated with a zero entry but ftp->dirdepth
Daniel Stenberg [Thu, 15 Apr 2004 10:43:40 +0000 (10:43 +0000)]
ftp->dirs[] is no longer terminated with a zero entry but ftp->dirdepth
should be used

20 years agoremoved the fixed dir depth limit in the FTP code
Daniel Stenberg [Thu, 15 Apr 2004 07:52:39 +0000 (07:52 +0000)]
removed the fixed dir depth limit in the FTP code

20 years agotwo ipresolve fixes
Daniel Stenberg [Wed, 14 Apr 2004 12:13:32 +0000 (12:13 +0000)]
two ipresolve fixes

20 years agoformat fix
Daniel Stenberg [Wed, 14 Apr 2004 12:13:21 +0000 (12:13 +0000)]
format fix

20 years agoasking for CURL_IPRESOLVE_V6 when ipv6 addresses can't be resolved will
Daniel Stenberg [Wed, 14 Apr 2004 12:10:44 +0000 (12:10 +0000)]
asking for CURL_IPRESOLVE_V6 when ipv6 addresses can't be resolved will
now cause the resolve function to return NULL immediately

20 years agoGisle Vanem made the -4/-6 actually get set too
Daniel Stenberg [Wed, 14 Apr 2004 12:00:53 +0000 (12:00 +0000)]
Gisle Vanem made the -4/-6 actually get set too

20 years agoCurl_wait_for_resolv() could hang due to the bad timeout timer resolution and
Daniel Stenberg [Wed, 14 Apr 2004 11:43:26 +0000 (11:43 +0000)]
Curl_wait_for_resolv() could hang due to the bad timeout timer resolution and
some bad thinking on my part.

20 years agoseveral changes
Daniel Stenberg [Wed, 14 Apr 2004 07:07:30 +0000 (07:07 +0000)]
several changes

20 years agodisplay interesting log files on failure, if -p is used
Daniel Stenberg [Wed, 14 Apr 2004 07:04:45 +0000 (07:04 +0000)]
display interesting log files on failure, if -p is used

20 years agoenable verbose as well
Daniel Stenberg [Wed, 14 Apr 2004 06:53:34 +0000 (06:53 +0000)]
enable verbose as well

20 years agoAdded test case 511 in an attempt to repeat bug report #934666 "storage leak
Daniel Stenberg [Wed, 14 Apr 2004 06:30:37 +0000 (06:30 +0000)]
Added test case 511 in an attempt to repeat bug report #934666 "storage leak
in ftp.c", but it shows no leaking.

20 years agominor format fix
Daniel Stenberg [Tue, 13 Apr 2004 14:34:12 +0000 (14:34 +0000)]
minor format fix

20 years agoeh, these can't be used for poll()! ;-)
Daniel Stenberg [Tue, 13 Apr 2004 14:31:49 +0000 (14:31 +0000)]
eh, these can't be used for poll()! ;-)

20 years agoadditional info
Daniel Stenberg [Tue, 13 Apr 2004 14:27:47 +0000 (14:27 +0000)]
additional info

20 years agoGisle Vanem's fix that makes the multi interface work on Windows again even
Daniel Stenberg [Tue, 13 Apr 2004 13:59:12 +0000 (13:59 +0000)]
Gisle Vanem's fix that makes the multi interface work on Windows again even
when not using ares.

20 years agospell fixes
Daniel Stenberg [Tue, 13 Apr 2004 11:03:02 +0000 (11:03 +0000)]
spell fixes

20 years agomore news
Daniel Stenberg [Tue, 13 Apr 2004 10:58:44 +0000 (10:58 +0000)]
more news

20 years agoproper typecast to prevent compiler warning
Daniel Stenberg [Tue, 13 Apr 2004 10:42:32 +0000 (10:42 +0000)]
proper typecast to prevent compiler warning

20 years agoremoved the BUGS section since it offers nothing good
Daniel Stenberg [Tue, 13 Apr 2004 09:08:52 +0000 (09:08 +0000)]
removed the BUGS section since it offers nothing good

20 years ago1.2.0
Daniel Stenberg [Tue, 13 Apr 2004 07:44:26 +0000 (07:44 +0000)]
1.2.0

20 years agoremove an long time #defined struct member and use the actual "real" name
Daniel Stenberg [Tue, 13 Apr 2004 07:37:28 +0000 (07:37 +0000)]
remove an long time #defined struct member and use the actual "real" name
instead to make it easier to find/read

20 years agoMoved the 'tcp_nodelay' member to the proper 'UserDefined' struct within the
Daniel Stenberg [Tue, 13 Apr 2004 07:16:26 +0000 (07:16 +0000)]
Moved the 'tcp_nodelay' member to the proper 'UserDefined' struct within the
sessionhandle to make the duphandle() function work as supposed. Also tried
to start document functions the doxygen way (in the headers of the functions).
Can't make it work though...

20 years agomove issue 35 (hostip.c cleanup) forward, I don't feel like doing that now
Daniel Stenberg [Tue, 13 Apr 2004 06:13:41 +0000 (06:13 +0000)]
move issue 35 (hostip.c cleanup) forward, I don't feel like doing that now

20 years agofull-test passes -p to runtests as well to get more details in case of failure
Daniel Stenberg [Tue, 13 Apr 2004 05:58:19 +0000 (05:58 +0000)]
full-test passes -p to runtests as well to get more details in case of failure

20 years agoInitial support for dumping the contents of the files in log/ when failing
Daniel Stenberg [Tue, 13 Apr 2004 05:57:50 +0000 (05:57 +0000)]
Initial support for dumping the contents of the files in log/ when failing
when -p is used. For easier bug-hunting of autobuild failures. This still
only shows what files that are present in log/, as I believe we need to
filter which files we show on a failure.

20 years agosomewhat safer typecasting in case sizeof(long) != sizeof(void *) (is there
Daniel Stenberg [Mon, 12 Apr 2004 06:55:25 +0000 (06:55 +0000)]
somewhat safer typecasting in case sizeof(long) != sizeof(void *) (is there
even such platforms?)

20 years agoDavid Byron's patch for MSVC builds with zlib
Daniel Stenberg [Sun, 11 Apr 2004 20:25:10 +0000 (20:25 +0000)]
David Byron's patch for MSVC builds with zlib

20 years agoupdated to not include the msvc-generated files
Daniel Stenberg [Sun, 11 Apr 2004 06:33:02 +0000 (06:33 +0000)]
updated to not include the msvc-generated files

20 years agowhen checked out from CVS, run this to generate the proper scripts
Daniel Stenberg [Sun, 11 Apr 2004 06:32:41 +0000 (06:32 +0000)]
when checked out from CVS, run this to generate the proper scripts

20 years agoDirk Manske increased the resolution for what the CURLINFO_*_TIME return.
Daniel Stenberg [Fri, 9 Apr 2004 09:36:31 +0000 (09:36 +0000)]
Dirk Manske increased the resolution for what the CURLINFO_*_TIME return.

20 years agowe're working on 1.2.0 now
Daniel Stenberg [Thu, 8 Apr 2004 18:10:03 +0000 (18:10 +0000)]
we're working on 1.2.0 now

20 years agoadded test 159, use --ntlm together with -0
Daniel Stenberg [Wed, 7 Apr 2004 15:01:11 +0000 (15:01 +0000)]
added test 159, use --ntlm together with -0

20 years agogetting only a 100 Continue response and nothing else, when talking HTTP,
Daniel Stenberg [Wed, 7 Apr 2004 14:27:54 +0000 (14:27 +0000)]
getting only a 100 Continue response and nothing else, when talking HTTP,
is now treated as an error by libcurl

20 years ago--limit-rate using -d or -F does not work
Daniel Stenberg [Wed, 7 Apr 2004 14:03:13 +0000 (14:03 +0000)]
--limit-rate using -d or -F does not work

20 years agorenamed the strtoofft() macro to curlx_strtoofft() to adjust to the curlx_*
Daniel Stenberg [Wed, 7 Apr 2004 07:30:40 +0000 (07:30 +0000)]
renamed the strtoofft() macro to curlx_strtoofft() to adjust to the curlx_*
concept, and added lib/README.curlx to explain details about it

20 years agoUse curl_off_t for the limit rate values to support REALLY huge values on
Daniel Stenberg [Wed, 7 Apr 2004 07:23:52 +0000 (07:23 +0000)]
Use curl_off_t for the limit rate values to support REALLY huge values on
such platforms that support large files.

20 years agomention getting windows builds to work after rearrangements
Daniel Stenberg [Tue, 6 Apr 2004 15:30:57 +0000 (15:30 +0000)]
mention getting windows builds to work after rearrangements

20 years agoMoved long-standing issues over from TODO-RELEASE to the more long-term TODO
Daniel Stenberg [Tue, 6 Apr 2004 15:29:01 +0000 (15:29 +0000)]
Moved long-standing issues over from TODO-RELEASE to the more long-term TODO
file.

20 years agothe memory leak on windows have been addressed
Daniel Stenberg [Tue, 6 Apr 2004 15:22:21 +0000 (15:22 +0000)]
the memory leak on windows have been addressed

20 years agorequire ssl since ntlm needs it
Daniel Stenberg [Tue, 6 Apr 2004 15:16:07 +0000 (15:16 +0000)]
require ssl since ntlm needs it

20 years agoNew authentication code added, particularly noticable when doing POST or PUT
Daniel Stenberg [Tue, 6 Apr 2004 15:14:10 +0000 (15:14 +0000)]
New authentication code added, particularly noticable when doing POST or PUT
with Digest or NTLM. libcurl will now use HEAD to negotiate the authentication
and when done perform the requested POST.

20 years agomention the man page updates from the other day
Daniel Stenberg [Tue, 6 Apr 2004 15:12:50 +0000 (15:12 +0000)]
mention the man page updates from the other day

20 years agothe pack_hostent() proto isn't used/needed with ipv6 is enabled.
Daniel Stenberg [Tue, 6 Apr 2004 15:09:43 +0000 (15:09 +0000)]
the pack_hostent() proto isn't used/needed with ipv6 is enabled.

time to restructure this source file!

20 years agoGisle Vanem's fix for bug item #927979 reported by Nathan O'Sullivan.
Daniel Stenberg [Tue, 6 Apr 2004 14:51:14 +0000 (14:51 +0000)]
Gisle Vanem's fix for bug item #927979 reported by Nathan O'Sullivan.

Good enough?

20 years agoAdded the curl source header and changed some comments
Daniel Stenberg [Tue, 6 Apr 2004 14:07:04 +0000 (14:07 +0000)]
Added the curl source header and changed some comments

20 years agoimproved --limit-rate functionality, partly by the new use of curlx_tvnow()
Daniel Stenberg [Tue, 6 Apr 2004 12:06:05 +0000 (12:06 +0000)]
improved --limit-rate functionality, partly by the new use of curlx_tvnow()

20 years agoGisle Vanem caught me breaking the windows version of Curl_strerror()
Daniel Stenberg [Tue, 6 Apr 2004 12:02:36 +0000 (12:02 +0000)]
Gisle Vanem caught me breaking the windows version of Curl_strerror()

20 years agoprovide these functions as curlx_* ones as this enables the curl app to
Daniel Stenberg [Tue, 6 Apr 2004 10:15:10 +0000 (10:15 +0000)]
provide these functions as curlx_* ones as this enables the curl app to
re-use these sources and functions for subsecond resolution timing

20 years agoup-to-date with reality
Daniel Stenberg [Tue, 6 Apr 2004 07:59:11 +0000 (07:59 +0000)]
up-to-date with reality

20 years agoadded HAVE_GETTIMEOFDAY, we need it for better time resolution
Daniel Stenberg [Tue, 6 Apr 2004 07:49:57 +0000 (07:49 +0000)]
added HAVE_GETTIMEOFDAY, we need it for better time resolution

20 years agotypecasts to please picky compilers checking the printf() format string
Daniel Stenberg [Tue, 6 Apr 2004 07:48:29 +0000 (07:48 +0000)]
typecasts to please picky compilers checking the printf() format string

20 years agoone change, three bugs, one credit
Daniel Stenberg [Tue, 6 Apr 2004 06:24:37 +0000 (06:24 +0000)]
one change, three bugs, one credit

20 years agothe last couple of days
Daniel Stenberg [Tue, 6 Apr 2004 06:24:06 +0000 (06:24 +0000)]
the last couple of days

20 years agotwo issues to fix before 7.11.2, one issue to fix befor 7.12.0
Daniel Stenberg [Tue, 6 Apr 2004 06:18:48 +0000 (06:18 +0000)]
two issues to fix before 7.11.2, one issue to fix befor 7.12.0

20 years agoremove the general use of sys_nerr
Daniel Stenberg [Tue, 6 Apr 2004 06:06:10 +0000 (06:06 +0000)]
remove the general use of sys_nerr

20 years agogenerated files, no need to keep in CVS
Daniel Stenberg [Mon, 5 Apr 2004 12:38:54 +0000 (12:38 +0000)]
generated files, no need to keep in CVS

20 years agoprepend the man3 dir to the file name to work better.
Daniel Stenberg [Fri, 2 Apr 2004 11:04:34 +0000 (11:04 +0000)]
prepend the man3 dir to the file name to work better.
Robin Kay pointed this out.

20 years agoedits, mainly to make the generated html output nicer
Daniel Stenberg [Fri, 2 Apr 2004 09:56:01 +0000 (09:56 +0000)]
edits, mainly to make the generated html output nicer

20 years agominor edits
Daniel Stenberg [Fri, 2 Apr 2004 09:50:42 +0000 (09:50 +0000)]
minor edits

20 years agoDirk Manske's feedback:
Daniel Stenberg [Fri, 2 Apr 2004 07:32:03 +0000 (07:32 +0000)]
Dirk Manske's feedback:
* bring back subsecond resolution to CURLINFO_TOTAL_TIME
* Fix the Curl_pgrsDone() so that the final progress update is shown properly

20 years agoAndrés García's updated mingw makefiles
Daniel Stenberg [Fri, 2 Apr 2004 07:18:13 +0000 (07:18 +0000)]
Andrés García's updated mingw makefiles

20 years agoif select returns -1, bail out of the loop
Daniel Stenberg [Fri, 2 Apr 2004 06:40:31 +0000 (06:40 +0000)]
if select returns -1, bail out of the loop

20 years agoOnly check that the c-ares lib is valid if we don't use the "embedded"
Daniel Stenberg [Thu, 1 Apr 2004 10:26:59 +0000 (10:26 +0000)]
Only check that the c-ares lib is valid if we don't use the "embedded"
directory. The provided ares dir is probably up-to-date, but more importantly
it is often not built yet at the time when this configure script runs.

20 years agoWhen ares is enabled, we now check for the ares_cancel function to verify
Daniel Stenberg [Thu, 1 Apr 2004 09:10:33 +0000 (09:10 +0000)]
When ares is enabled, we now check for the ares_cancel function to verify
that we use a library that is recent enough to build with the latest libcurl.

20 years agoDirk Manske's fix that makes sure we cancel the ares resolve when we time out
Daniel Stenberg [Thu, 1 Apr 2004 08:40:36 +0000 (08:40 +0000)]
Dirk Manske's fix that makes sure we cancel the ares resolve when we time out
from a name resolve. Without this, we leak memory!

20 years agoDirk Manske's ares_cancel() function was added.
Daniel Stenberg [Thu, 1 Apr 2004 08:25:58 +0000 (08:25 +0000)]
Dirk Manske's ares_cancel() function was added.

20 years agoedited slightly, point out our new mailinglist
Daniel Stenberg [Thu, 1 Apr 2004 08:25:23 +0000 (08:25 +0000)]
edited slightly, point out our new mailinglist

20 years agoremind us about cvs tagging when we've built a release archive
Daniel Stenberg [Thu, 1 Apr 2004 08:23:26 +0000 (08:23 +0000)]
remind us about cvs tagging when we've built a release archive

20 years agoremoved my previously attempted fix for ares timeouts, not needed
Daniel Stenberg [Thu, 1 Apr 2004 07:04:58 +0000 (07:04 +0000)]
removed my previously attempted fix for ares timeouts, not needed

20 years agoApplied Joe Halpin's bugfixes to the NI_WITHSCOPEID test program.
Daniel Stenberg [Thu, 1 Apr 2004 06:53:11 +0000 (06:53 +0000)]
Applied Joe Halpin's bugfixes to the NI_WITHSCOPEID test program.

20 years agoDominick Meglio man page fixes
Daniel Stenberg [Thu, 1 Apr 2004 06:10:56 +0000 (06:10 +0000)]
Dominick Meglio man page fixes

20 years agoUse the new HAVE_NI_WITHSCOPEID define instead of merely checking for the
Daniel Stenberg [Wed, 31 Mar 2004 21:33:52 +0000 (21:33 +0000)]
Use the new HAVE_NI_WITHSCOPEID define instead of merely checking for the
existance of NI_WITHSCOPEID since some platforms have that define but still
can't function with it set.

20 years agoissue 30, digest re-negotiate works now!
Daniel Stenberg [Wed, 31 Mar 2004 21:04:26 +0000 (21:04 +0000)]
issue 30, digest re-negotiate works now!

20 years agorecent changes
Daniel Stenberg [Wed, 31 Mar 2004 21:03:55 +0000 (21:03 +0000)]
recent changes

20 years agoupdates and David Byron's spellfix
Daniel Stenberg [Wed, 31 Mar 2004 21:01:20 +0000 (21:01 +0000)]
updates and David Byron's spellfix

20 years agoRoy Shan fixed a case that prevented ares name resolve timeouts to occur.
Daniel Stenberg [Wed, 31 Mar 2004 20:50:01 +0000 (20:50 +0000)]
Roy Shan fixed a case that prevented ares name resolve timeouts to occur.

20 years agowe're working on 7.11.2-CVS right now
Daniel Stenberg [Wed, 31 Mar 2004 20:22:28 +0000 (20:22 +0000)]
we're working on 7.11.2-CVS right now

20 years agoHAVE_NI_WITHSCOPEID spelled right!
Daniel Stenberg [Wed, 31 Mar 2004 20:13:53 +0000 (20:13 +0000)]
HAVE_NI_WITHSCOPEID spelled right!

20 years agoRemove the elapsed time from the most recent select() only.
Daniel Stenberg [Wed, 31 Mar 2004 13:19:41 +0000 (13:19 +0000)]
Remove the elapsed time from the most recent select() only.

20 years agoThe asynch name resolve methods now all use CURL_TIMEOUT_RESOLVE for
Daniel Stenberg [Wed, 31 Mar 2004 12:55:24 +0000 (12:55 +0000)]
The asynch name resolve methods now all use CURL_TIMEOUT_RESOLVE for
the specific time to wait for a resolve. The definition is at the top of
this source file.

20 years agoDirk Manske found out the Curl_wait_for_resolv() timed out too early.
Daniel Stenberg [Wed, 31 Mar 2004 12:45:26 +0000 (12:45 +0000)]
Dirk Manske found out the Curl_wait_for_resolv() timed out too early.

20 years agoadded swsbounce
Daniel Stenberg [Wed, 31 Mar 2004 12:24:08 +0000 (12:24 +0000)]
added swsbounce

20 years agoadded include to fix warning
Daniel Stenberg [Wed, 31 Mar 2004 11:55:56 +0000 (11:55 +0000)]
added include to fix warning

20 years ago* Fixed a memory leak when doing repeated re-negotiations.
Daniel Stenberg [Wed, 31 Mar 2004 11:55:07 +0000 (11:55 +0000)]
* Fixed a memory leak when doing repeated re-negotiations.
* Made the incoming line parser more forgiving to allow "name=contents" pairs
where the contents isn't within double quotes.
* Made the digest code return CURLDIGEST_BADALGO if a requested algorithm
isn't supported by the code.