aboutsummaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2010-01-18Constantine Sapuntzakis enhancements to make memory tracking log file writingYang Tse
of messages atomic, on systems where an fwrite of a memory buffer is atomic.
2010-01-15Store now this file in CVS with unix line endings.Yang Tse
maketgz already converts this file to DOS style with an awk filter.
2010-01-15Added the new protocol source files to the non-autoconf build filesDan Fandrich
2010-01-11ssh_statemach_act() is now modified to loop over the switch() to perform asDaniel Stenberg
much as possible in one go, as long as it doesn't block and hasn't reached the end of the state machine. This avoids spurious -1 returns from curl_multi_fdset() simply because previously it would return from this function without anything in EWOUDLBLOCK and thus basically it wasn't actually waiting for anything!!
2010-01-11After the TCP connect is confirmed in CURLM_STATE_WAITCONNECT and it changesDaniel Stenberg
state, we return CURLM_CALL_MULTI_PERFORM unconditionally then so that we can act faster like in the case the protocol-specific connect doesn't block on anything and we can just persue on the next action immediately. It also then avoids a case where curl_multi_fdset() would return -1.
2010-01-11- Made sure that the progress callback is repeatedly called at a regularDaniel Stenberg
interval even during very slow connects.
2010-01-09struct Curl_sh_entry's 'inuse' member was no longer used and is now removedDaniel Stenberg
2010-01-08- Johan van Selst found and fixed a OpenSSL session ref count leak:Daniel Stenberg
ossl_connect_step3() increments an SSL session handle reference counter on each call. When sessions are re-used this reference counter may be incremented many times, but it will be decremented only once when done (by Curl_ossl_session_free()); and the internal OpenSSL data will not be freed if this reference count remains positive. When a session is re-used the reference counter should be corrected by explicitly calling SSL_SESSION_free() after each consecutive SSL_get1_session() to avoid introducing a memory leak. (http://curl.haxx.se/bug/view.cgi?id=2926284)
2010-01-07removed a parameter from the Curl_http_readwrite_headers() prototype to removeDaniel Stenberg
the need for the struct forward declaration from http.h which caused problems with gcc 2.96 and quite frankly the parameter wasn't necessary anyway
2010-01-07removed Curl_http_header_append() prototype as it isn't used anymore, theDaniel Stenberg
function was moved to http.c and was made static and renamed
2010-01-06- Make sure the progress callback is called repeatedly even during very slowDaniel Stenberg
name resolves when c-ares is used for resolving.
2010-01-06Julien Chaffraix fixed so that the fragment part in an URL is not sent to ↵Claes Jakobsson
the server anymore
2010-01-03- Julien Chaffraix eliminated a duplicated initialization in singlesocket().Kamil Dudka
2010-01-01update copyright year since we are in 2010 nowDaniel Stenberg
2010-01-01- Ingmar Runge enhanced libcurl's FTP engine to support the PRET command. ThisDaniel Stenberg
command is a special "hack" used by the drftpd server, but even though it is a custom extension I've deemed it fine to add to libcurl since this server seems to survive and people keep using it and want libcurl to support it. The new libcurl option is named CURLOPT_FTP_USE_PRET, and it is also usable from the curl tool with --ftp-pret. Using this option on a server that doesn't support this command will make libcurl fail.
2009-12-31turned CURLOPT_MAIL_RCPT into a curl_slist list instead to support multipleDaniel Stenberg
receivers, and made the command line tool thus support the option specified many times
2009-12-30modified to get the EHLO domain from the path part of the URL instead of theDaniel Stenberg
user name
2009-12-30moved the SMTP payload escape function into Curl_smtp_escape_eob and putDaniel Stenberg
it in smtp.c
2009-12-30(SMTP) support DATA better in the server and make sure to "escape" CRLF.CRLFDaniel Stenberg
sequences in uploaded data. The test server doesn't "decode" escaped dot-lines but instead test cases must be written to take them into account. Added test case 803 to verify dot-escaping.
2009-12-30Oops, should have removed 'not' in previous commit.Yang Tse
2009-12-30VMS specific preprocessor symbol checking adjustmentsYang Tse
2009-12-30Replaced stricmp() usage with our portable strequal()Yang Tse
2009-12-29move HTTP-specific functions to http.c where they belongDaniel Stenberg
2009-12-25first shot at actually doing the SMTP upload as well, not doing the properDaniel Stenberg
end-of-body treatment
2009-12-20remove some unused codeDaniel Stenberg
2009-12-20free the allocated mailbox name at disconnectDaniel Stenberg
2009-12-19fixed a precaution check in the cookie code, pointed out by Julien ChaffraixDaniel Stenberg
2009-12-17Fix compilation failureYang Tse
2009-12-17uh, assign the bool it points to properlyDaniel Stenberg
2009-12-17Stop overloading the conn->protocol field with the PROT_MISSING bit. ItDaniel Stenberg
really didn't belong there and had no real point.
2009-12-17Remove pointless storing of the protocol as a string within the connectdataDaniel Stenberg
struct, and instead use the already stored string in the handler struct.
2009-12-17- David Byron fixed Curl_ossl_cleanup to actually call ENGINE_cleanup whenDaniel Stenberg
available.
2009-12-16Follow-up fix for the proxy fix I did for Jon Nelson's bug. It turned out IDaniel Stenberg
was a bit too quick and broke test case 1101 with that change. The order of some of the setups is sensitive. I now changed it slightly again.
2009-12-14- Jon Nelson found a regression that turned out to be a flaw in how libcurlDaniel Stenberg
detects and uses proxies based on the environment variables. If the proxy was given as an explicit option it worked, but due to the setup order mistake proxies would not be used fine for a few protocols when picked up from '[protocol]_proxy'. Obviously this broke after 7.19.4. I now also added test case 1106 that verifies this functionality. (http://curl.haxx.se/bug/view.cgi?id=2913886)
2009-12-14Fix compiler warningYang Tse
2009-12-14Fix compiler warningsYang Tse
2009-12-12split out more protocol-specific structs from urldata.h into their ownDaniel Stenberg
protocol-specific header files
2009-12-12introducing IMAP, POP3 and SMTP support (still lots of polish left to do)Daniel Stenberg
2009-12-11Prevent rewinding unless pipelining.Yang Tse
See http://curl.haxx.se/mail/lib-2009-12/0107.html
2009-12-11Removed function prototype without implementationYang Tse
2009-12-10- Constantine Sapuntzakis figured out a case which would lead to libcurlDaniel Stenberg
accessing alredy freed memory and thus crash when using HTTPS (with OpenSSL), multi interface and the CURLOPT_DEBUGFUNCTION and a certain order of cleaning things up. I fixed it. (http://curl.haxx.se/bug/view.cgi?id=2891591)
2009-12-10minor indent changeDaniel Stenberg
2009-12-07- Martin Storsjo made libcurl use the Expect: 100-continue header for postsDaniel Stenberg
with unknown size. Previously it was only used for posts with a known size larger than 1024 bytes.
2009-12-02lib/nss.c: avoid use of uninitialized valueKamil Dudka
2009-12-01- If the Expect: 100-continue header has been set by the application throughDaniel Stenberg
curl_easy_setopt with CURLOPT_HTTPHEADER, the library should set data->state.expect100header accordingly - the current code (in 7.19.7 at least) doesn't handle this properly. Martin Storsjo provided the fix!
2009-11-28s/socklen_t/curl_socklen_t/gYang Tse
2009-11-28Fix compiler warning: unused variable `data'Yang Tse
2009-11-27- Markus Koetter provided a polished and updated version of Chad Monroe's TFTPDaniel Stenberg
rework patch that now integrates TFTP properly into libcurl so that it can be used non-blocking with the multi interface and more. BLKSIZE also works. The --tftp-blksize option was added to allow setting the TFTP BLKSIZE from the command line.
2009-11-26 - Extended and fixed the change I did on Dec 11 for the the progressDaniel Stenberg
meter/callback during FTP command/response sequences. It turned out it was really lame before and now the progress meter SHOULD get called at least once per second.
2009-11-24refreshedDaniel Stenberg