aboutsummaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2008-03-06Regression fix:Yang Tse
select/poll calls will only be retried upon EINTR failures as it previously was in lib/select.c revision 1.29 In this way Curl_socket_ready() and Curl_poll() will again fail on any select/poll errors different than EINTR.
2008-03-04Mike Protts brought a patch that makes resumed transfers work with SFTP.Daniel Stenberg
2008-03-01- Anatoli Tubman found and fixed a crash with Negotiate authentication used onDaniel Stenberg
a re-used connection where both requests used Negotiate.
2008-02-28fixed commented define for SSPI.Gunter Knauf
2008-02-27another small change to the makefiles to simplify rules.Gunter Knauf
2008-02-26some more minor makefile changes; removed useless dist target.Gunter Knauf
2008-02-26fixed install target to create a ca-bundle.crt since we have no longer one ↵Gunter Knauf
in the project.
2008-02-26Added support for server name indication (RFC 4366).Gunter Knauf
Patch submitted by Kaspar Brand.
2008-02-25- Kaspar Brand made GnuTLS-built libcurl properly acknowledge the option thatDaniel Stenberg
forces it to prefer SSLv3.
2008-02-23- Sam Listopad provided a patch in feature-request #1900014Daniel Stenberg
http://curl.haxx.se/bug/feature.cgi?id=1900014 that makes libcurl (built to use OpenSSL) support a full chain of certificates in a given PKCS12 certificate.
2008-02-21- Zmey Petroff found a crash when libcurl accessed a NULL pointer, whichDaniel Stenberg
happened if you set the connection cache size to 1 and for example failed to login to an FTP site. Bug report #1896698 (http://curl.haxx.se/bug/view.cgi?id=1896698)
2008-02-21assert that the *connp is a non-NULL pointer when Curl_done() is calledDaniel Stenberg
2008-02-20oops, fixed to buildDaniel Stenberg
2008-02-20- Based on initial work done by Gautam Kachroo to address a bug, we now keepDaniel Stenberg
better control at the exact state of the connection's SSL status so that we know exactly when it has completed the SSL negotiation or not so that there won't be accidental re-uses of connections that are wrongly believed to be in SSL-completed-negotiate state.
2008-02-20- We no longer support setting the CURLOPT_URL option from inside a callbackDaniel Stenberg
such as the CURLOPT_SSL_CTX_FUNCTION one treat that as if it was a Location: following. The patch that introduced this feature was done for 7.11.0, but this code and functionality has been broken since about 7.15.4 (March 2006) with the introduction of non-blocking OpenSSL "connects". It was a hack to begin with and since it doesn't work and hasn't worked correctly for a long time and nobody has even noticed, I consider it a very suitable subject for plain removal. And so it was done.
2008-02-19applied patch to disable SSLv2 by default; discussion:Gunter Knauf
http://sourceforge.net/tracker/index.php?func=detail&aid=1767276&group_id=976&atid=350976 Submitted by Kaspar Brand.
2008-02-19fix compiler warnings: 'statement is unreachable'Yang Tse
2008-02-19fix compiler warnings: 'enumerated type mixed with another type'Yang Tse
2008-02-18https_getsock() should be static all over (and did some fixed indenting)Daniel Stenberg
2008-02-18added check symbol for linking with POSIX prelude.Gunter Knauf
2008-02-18fix compiler warnings:Yang Tse
'enumerated type mixed with another type' and 'variable was set but never used'
2008-02-18- We're no longer providing a very old ca-bundle in the curl tarball. You canDaniel Stenberg
get a fresh one downloaded and created with 'make ca-bundle' or you can get one from here => http://curl.haxx.se/docs/caextract.html if you want a fresh new one extracted from Mozilla's recent list of ca certs. The configure option --with-ca-bundle now lets you specify what file to use as default ca bundle for your build. If not specified, the configure script will check a few known standard places for a global ca cert to use.
2008-02-17rephrased commentDaniel Stenberg
2008-02-17In Curl_done() if premature is TRUE, it means this connection was said to beDaniel Stenberg
DONE before the entire request operation is complete and thus we can't know in what state it is for re-using, so we're forced to close it. In a perfect world we can add code that keep track of if we really must close it here or not, but currently we have no such detail knowledge. Jerome Muffat-Meridol helped us work this out.
2008-02-17don't do the GOT_NOTHING error check if the DONE function was called withDaniel Stenberg
premature set TRUE, which means it was done before the request comleted. It could then very well not have received any data.
2008-02-17added a comment about the ignoring of the Curl_done() return codeDaniel Stenberg
2008-02-16oops, that was debug code not meant to be committed like this...Daniel Stenberg
2008-02-16fix warnings about shadowingDaniel Stenberg
2008-02-16seems that curently we dont need the imports from (l)ldapx.imp.Gunter Knauf
2008-02-16re-ordered the module dependency list;Gunter Knauf
removed unsused ldap module dependency since the module didnt autounload from protected address space.
2008-02-15- Made the gnutls code path not even try to get the server cert if no peerDaniel Stenberg
verification is requested. Previously it would even return failure if gnutls failed to get the server cert even though no verification was asked for. - Fix my Curl_timeleft() leftover mistake in the gnutls code
2008-02-15mention that we explicitly ignore the return codeDaniel Stenberg
2008-02-15log SSH public key authentication failure and reasonYang Tse
2008-02-15- Pooyan McSporran found and fixed a flaw where you first would do a normalDaniel Stenberg
http request and then you'd reuse the handle and replace the Accept: header, as then libcurl would send two Accept: headers!
2008-02-15fixed version var.Gunter Knauf
2008-02-15moved info block up before help block so that it can also be displayed ↵Gunter Knauf
before help option; trial to add a version number.
2008-02-11Yang Tse pointed out a few remaining quirks from my timeout refactoring fromDaniel Stenberg
Feb 7 that didn't abort properly on timeouts. These are actually old problems but now they should be fixed.
2008-02-11open pipe to openssl commandline instead of writing into temp file.Gunter Knauf
2008-02-11Fixed unused variable warning.Dan Fandrich
2008-02-11added strict to make sure all vars are properly defined;Gunter Knauf
added -t switch to make text info of CAs optional; added -q switch to be really quiet.
2008-02-10Bug report #1888932 (http://curl.haxx.se/bug/view.cgi?id=1888932) pointsYang Tse
out and provides test program that demonstrates that libcurl might not set error description message for error CURLE_COULDNT_RESOLVE_HOST for Windows threaded name resolver builds. Fixed now.
2008-02-10added -b switch to provide a backup functionality for existing ca-bundle.crt ↵Gunter Knauf
file.
2008-02-09fixed another wrong var in error message.Gunter Knauf
2008-02-09fixed wrong var in error message.Gunter Knauf
2008-02-08- Mike Hommey filed and fixed bug report #1889856Daniel Stenberg
(http://curl.haxx.se/bug/view.cgi?id=1889856): When using the gnutls ssl layer, cleaning-up and reinitializing curl ends up with https requests failing with "ASN1 parser: Element was not found" errors. Obviously a regression added in 7.16.3.
2008-02-08include mk-ca-bundle.pl in the tarballsDaniel Stenberg
2008-02-08use argument to specify output filename if present.Gunter Knauf
2008-02-08fixed regex to fetch certdata.txt version since it was replaced by CVS (argh!)Gunter Knauf
added a switch to display certdata.txt version header.
2008-02-08added Perl script to create a fresh ca-bundle.crt.Gunter Knauf
2008-02-07- Refactored a lot of timeout code into a few functions in an attempt to makeDaniel Stenberg
them all use the same (hopefully correct) logic to make it less error-prone and easier to introduce library-wide where it should be used.