aboutsummaryrefslogtreecommitdiff
path: root/lib/url.c
AgeCommit message (Collapse)Author
2002-01-28Steve Marx helped us realize that we shouldn't treat customrequest as aDaniel Stenberg
request of its own, it just changes the keyword of a request.
2002-01-08Add support for DNS cache timeouts via the CURLOPT_DNS_CACHE_TIMEOUT option.Sterling Hughes
The default cache timeout for this is 60 seconds, which is arbitrary and completely subject to change :)
2002-01-07Make cach'ing work with threads now, there are now three cases:Sterling Hughes
- Use a global dns cache (via setting the tentatively named, CURLOPT_DNS_USE_GLOBAL_CACHE option to true) - Use a per-handle dns cache, by default - Use a pooled dns cache when in the "multi" interface
2002-01-04an unconditional occurance of inet_ntoa() now uses inet_ntoa_r() on allDaniel Stenberg
platforms that have such a function. This affects multi-thread running libcurls on IPv4 systems that have VERBOSE switched on. The previous version was risking that another thread overwrote the data before it was read out in this thread. There could possibly also be a slight risk that the data isn't zero terminated for a short while and thus could cause the thread to crash...
2002-01-03merged the multi-dev branch back into MAIN againDaniel Stenberg
2002-01-03Sterling Hughes' provided initial DNS cache source code.Daniel Stenberg
2001-12-20*cool* fix by Björn Stenberg, makes proxy transfers work better...! :-)Daniel Stenberg
2001-12-17Götz Babin-Ebell's OpenSSL ENGINE patchDaniel Stenberg
2001-12-05Jon Travis suggested fix. when CURLOPT_HTTPGET is used we must assignDaniel Stenberg
set.upload to FALSE or else we might still get an upload if the previous operation was an upload!
2001-12-02more careful re-use of connections when SSL is used over proxiesDaniel Stenberg
2001-11-29disabling EPSV is now possibleDaniel Stenberg
2001-11-28CURLOPT_FTP_USE_EPSV can now be set to FALSE to prevent libcurl fromDaniel Stenberg
attempting to use EPSV before the standard PASV.
2001-11-12made CURLOPT_HTTPPROXYTUNNEL work for plain HTTP as wellDaniel Stenberg
2001-11-07we use signal() to ignore signals only as long as we have to, and we nowDaniel Stenberg
restore the previous (if any) signal handler properly on return.
2001-11-07get the previous struct keep_sigactDaniel Stenberg
2001-11-06myalarm() is history, we now use HAVE_ALARM and we now do our very best toDaniel Stenberg
1 - restore the previous sigaction struct as soon as we are about to shut off our timeout 2 - restore the previous alarm() timeout, in case an application or similar had it running before we "borrowed" it for a while. No, this does not fix the multi-thread problem you get with alarm(). This patch should correct bug report #478780: //sourceforge.net/tracker/?func=detail&atid=100976&aid=478780&group_id=976 If not, please post details!
2001-11-01ConnectionExists() now returns FALSE immediately if it finds a connectionDaniel Stenberg
that is dead, because it can only find one entry anyway and if that is dead there won't be any other entry that matches
2001-10-31Removed the SocketIsDead() stuff for SSL again as it doesn't work. We mustDaniel Stenberg
rely on the new go-ahead-and-try mechanism that I just added to Transfer()
2001-10-31nonblock => Curl_nonblock, remade the check for a live SSL connection (again)Daniel Stenberg
2001-10-30Added an additional SSL check for a dead socket before we re-use an SSLDaniel Stenberg
connection. The simple socket-check is not enough in these cases.
2001-10-29fixed conn->name error on connection re-use and enlarged the 'gname' arrayDaniel Stenberg
to hold 512 bytes (for user+password+hostname)
2001-10-19now counts header size return from server and if nothing is returned from aDaniel Stenberg
HTTP server we return error
2001-10-12Curl_tvdiff() now returns a millisecond diff, no double like beforeDaniel Stenberg
2001-10-11looks nicer and is better compatible with older vim versionsSterling Hughes
2001-10-10cookiejar now enables the cookie engineDaniel Stenberg
2001-10-09added the option CURLOPT_HTTP_VERSION that can specify which HTTP versionDaniel Stenberg
libcurl should use in its request
2001-10-09ignore SIGPIPE, as that can be actually get sent when we write to a socketDaniel Stenberg
2001-10-04getaddrinfo() cleanupsDaniel Stenberg
2001-10-02added port number in informational connect messageDaniel Stenberg
2001-10-02IPv6 adjustments, connect()ing to bad ports still don't work properly forDaniel Stenberg
IPv6
2001-10-02major connecting updatesDaniel Stenberg
2001-10-01pick the correct timeout before the connecthost callDaniel Stenberg
2001-10-01conn->hp is now conn->hostaddrDaniel Stenberg
changed the Curl_connethost() proto again
2001-10-01moved the myalarm() usage, and now makes sure to switch it off after theDaniel Stenberg
name resolving, as that should be the *ONLY* section in libcurl that may take a while in a synchronous call.
2001-10-01introduced non-blocking connectsDaniel Stenberg
2001-09-28more transparant support for IPv6 name resolvingDaniel Stenberg
2001-09-18fixed bug report #462600, following a Location: when the initial URL didn'tDaniel Stenberg
have a protocol:// part did wrong
2001-09-18the stuff formerly done in Curl_http_close is now done in Curl_closeDaniel Stenberg
2001-09-12ConnectionKillOne() _can_ return -1 as an indication of errorDaniel Stenberg
This is T. Bharath's fix
2001-09-11CURLOPT_SSL_CIPHER_LIST supportDaniel Stenberg
2001-09-07Added formatting sections for emacs and vimSterling Hughes
2001-09-03use the LIBCURL_NAME instead of the "hardcoded" stringDaniel Stenberg
2001-08-30Major rename and redesign of the internal "backbone" structs. Details willDaniel Stenberg
be posted in a minute to the libcurl list.
2001-08-29cookie jar adjustmentsDaniel Stenberg
2001-08-28Added SSL session ID caching, moved some SSL code from url.c to ssluse.cDaniel Stenberg
2001-08-23When setting *_URL or *_PROXY in *_setopt(), it is important that we checkDaniel Stenberg
and possibly free the existing pointer first, and then clear the "allocated" bit. We previously mistakenly could free the new pointer passed to us by the friendly user...!
2001-08-22CURLOPT_FTPASCII is the old name, CURLOPT_TRANSFERTEXT is the newDaniel Stenberg
2001-08-15IPv6-addresses can have dots too!Daniel Stenberg
2001-08-15extract IPv6-style specified IP-addresses properlyDaniel Stenberg
2001-08-15non-public functions should not use CURL * as arguments, so I changed themDaniel Stenberg
to use 'struct UrlData *' instead