aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2007-10-04I enabled test 1009 and made the --local-port use a wide range to reduce theDaniel Stenberg
risk of failures.
2007-10-04Kim Rinnewitz reported that --local-port didn't work with TFTP transfers.Daniel Stenberg
This happened because the tftp code always uncondionally did a bind() without caring if one already had been done and then it failed. I wrote a test case (1009) to verify this, but it is a bit error-prone since it will have to pick a fixed local port number and since the tests are run on so many different hosts in different situations I add it in disabled state.
2007-10-04Removed a piece of redundant code (process_answer already takes care of it).Steinar H. Gunderson
2007-10-04Another timeout fix in ares_getnameinfo().Steinar H. Gunderson
2007-10-04Send the timeout count in ares_getnameinfo().Steinar H. Gunderson
2007-10-04Moved the NULL check for channel upwards in ares_destroy().Steinar H. Gunderson
2007-10-04Clarified the comment over ares_cancel.Steinar H. Gunderson
2007-10-04On error, close "log/server.response"Yang Tse
2007-10-03If TCP_NODELAY is not defined we can't disable the Nagle algorithmYang Tse
2007-10-03Cleanup no longer used macrosYang Tse
2007-10-03Fix compiler warning: local variable may be used without having been initializedYang Tse
2007-10-03Upgrade OS400 wrappers and RPG copy file according to latest code updatesPatrick Monnerat
2007-10-03Fix issue related with the use of ares_timeout() result.Yang Tse
2007-10-03exported symbols must use lowercase "curl_", and I also fixed two compilerDaniel Stenberg
warnings, one C99 thing and the bad pointer sent to the callback
2007-10-03recount contributors after the 7.17.0 releaseDaniel Stenberg
2007-10-03people from the 7.17.0 announcementDaniel Stenberg
2007-10-03"97 - check ip callback", checkDaniel Stenberg
2007-10-03Alexey Pesternikov introduced CURLOPT_OPENSOCKETFUNCTION andDaniel Stenberg
CURLOPT_OPENSOCKETDATA to set a callback that allows an application to replace the socket() call used by libcurl. It basically allows the app to change address, protocol or whatever of the socket. (I also did some whitespace indent/cleanups in lib/url.c which kind of hides some of these changes, sorry for mixing those in.)
2007-10-03I renamed the CURLE_SSL_PEER_CERTIFICATE error code toDaniel Stenberg
CURLE_PEER_FAILED_VERIFICATION (standard CURL_NO_OLDIES style), and made this return code get used by the previous SSH MD5 fingerprint check in case it fails.
2007-10-03Based on a patch brought by Johnny Luong, libcurl now offersDaniel Stenberg
CURLOPT_SSH_HOST_PUBLIC_KEY_MD5 and the curl tool --hostpubmd5. They both make the SCP or SFTP connection verify the remote host's md5 checksum of the public key before doing a connect, to reduce the risk of a man-in-the-middle attack.
2007-10-02"99 - curl_easy_close()" seems to have gone uninterestingDaniel Stenberg
2007-10-02102, checkDaniel Stenberg
2007-10-02Fix memory leak under low memory conditions.Yang Tse
2007-10-02Avoid a segfault when generating a DNS "Transaction ID" in internalYang Tse
function init_id_key() under low memory conditions.
2007-10-02Renamed a couple of global variables to avoid shadowing warningsYang Tse
2007-10-02Fix compiler warningYang Tse
2007-10-02<winber.h> needed for Windows LDAP client 32 API supportYang Tse
2007-10-02Linking with wldap32.lib needed for Windows LDAP client 32 API supportYang Tse
2007-10-02Add ares_llist.c and ares_llist.h to MSCV project file.Yang Tse
2007-10-02known bug #46: chunked-encoded CONNECT responses from a http proxy now works.Daniel Stenberg
Added test case 1008 to verify. Note that #47 is still there.
2007-10-02Disable the Nagle algorithm and send back responses in small chunks in anDaniel Stenberg
attempt to force smaller bits to get read by clients.
2007-10-02document --post301, based on the phrasing in curl_easy_setopt.3 forDaniel Stenberg
CURLOPT_POST301 written by Philip Langdale
2007-10-02CURLOPT_POST301 section, added by Philip LangdaleDaniel Stenberg
2007-10-02Fixed the problem where next_lookup would use 'status' uninitialized. NowDaniel Stenberg
it gets passed the initial value as an argument.
2007-10-02Avoid inline C99ism, and move c-ares routines for managing doubly-linked lists.Yang Tse
2007-10-01ares_strerror() segfaulted if the input error number was out of the currentlyDaniel Stenberg
supported range.
2007-10-01Prevent ares_strerror() from segfaulting if an invalid error code is passedDaniel Stenberg
in as argument!
2007-09-30Added test536 that was accidentally missing. I also wrote up a new makefileDaniel Stenberg
target called 'filecheck' so that if you run 'make filecheck' in this directory it'll check if the local files are also mentioned in the Makefile.am so that they are properly included in release archives!
2007-09-30Alex Fishman reported a curl_easy_escape() problem that was made theDaniel Stenberg
function do wrong on all input bytes that are >= 0x80 (decimal 128) due to a signed / unsigned mistake in the code. I fixed it and added test case 543 to verify.
2007-09-30Fix compiler warningYang Tse
2007-09-30check availability of <netinet/tcp.h>Yang Tse
2007-09-30Fix missing right parenthesisYang Tse
2007-09-30Fix comparison between signed and unsignedYang Tse
2007-09-30improve portability, defining MAXDNAME and MAXCDNAMEYang Tse
2007-09-30Fix a memory leak that I recently inadvertedly introduced.Steinar H. Gunderson
2007-09-29Use ISDIGIT instead of isdigit; fixes a gcc warning.Steinar H. Gunderson
2007-09-29Immanuel Gregoire fixed a problem with persistent transfers over SFTP - theDaniel Stenberg
previous proto struct was kept.
2007-09-29Port the TCP socket fix made in ares_fds() to ares_getsock() as well.Steinar H. Gunderson
2007-09-29Previously, processing a large batch of timeouts was O(n^2) in the number ofSteinar H. Gunderson
outstanding queries, and processing a DNS response packet was O(n) in the number of outstanding queries. To speed things up in Google, we added a few circular, doubly-linked lists of queries that are hash-bucketed based on the attributes we care about, so most important operations are now O(1). It might be that the number of buckets are higher than most people would need, but on a quick calculation it should only be 100kB or so even on a 64-bit system, so I've let it stay as-is.
2007-09-29We should standarise on C comments.Gisle Vanem