aboutsummaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2002-12-16EAGAIN on older (correct) glibc versions indicate a problem and not the needDaniel Stenberg
for a bigger buffer and this is indeed badness for us. Making this work on both old and new glibc versions require an ugly loop that in its worse form cause 45 bad loops when using the correct glibc and a non-resolving host name... :-/ We want a better fix. Badly.
2002-12-13conn->bits.tcpconnect now keeps track of if this connection is connectedDaniel Stenberg
or not
2002-12-13Evan Jordan's fix for a memory leak. Bug report 650989.Daniel Stenberg
2002-12-13make a little work-around for file:// in _is_connected() and voila, now theDaniel Stenberg
multi interface works with file:// URLs fine (previously it crashed). This won't make it work on Windows though...
2002-12-10The initial HTTP request can now be sent in multiple parts, as part of theDaniel Stenberg
regular transfer process. This required some new tweaks, like for example we need to be able to tell the tranfer loop to not chunky-encode uploads while we're transferring the rest of the request...
2002-12-10send_buffer is no more hereDaniel Stenberg
2002-12-09A normal POST now provides data to the main transfer loop via the usualDaniel Stenberg
read callback, and thus won't put a lot of stress on the request sending code (which currently does an ugly loop).
2002-12-09The fread() callback pointer and associated pointer is now stored in theDaniel Stenberg
connectdata struct instead, and is no longer modified within the 'set' struct as previously (which was a really BAAAD thing).
2002-12-05read and write as much as possible until end of data or EWOULDBLOCK beforeDaniel Stenberg
returning back to the select() loop. Consider this a test so far.
2002-12-05changed proto for Curl_krb_kauth()Daniel Stenberg
2002-12-04make WIN32 defined for Borland properly, as told by Alexander J. OssDaniel Stenberg
2002-12-04The waiting for the 226 or 250 line expected to come after a transfer isDaniel Stenberg
complete is now only made for 60 seconds and if no data was received during those 60 seconds, we store a special error message (preparing to make this a special error code) as this most likely means that the control connection has died while we were transferring data.
2002-12-03Curl_GetFTPResponse() takes a different set of parameters and now return aDaniel Stenberg
proper CURLcode. The default timeout for reading one response is now also possible to change while running.
2002-12-03updated to realityDaniel Stenberg
2002-12-02if the PWD reply parser failed, we leaked memoryDaniel Stenberg
2002-12-01more to ignoreDaniel Stenberg
2002-12-01only use Content-Length: header if not transfering data chunkedDaniel Stenberg
2002-11-29let the Curl_FormReader() return 0 when it reaches end of data to that theDaniel Stenberg
chunked transfer work
2002-11-28fix the hash init to call the correct dns cleanup functionDaniel Stenberg
2002-11-28added compareheader protoDaniel Stenberg
2002-11-28compareheader() was moved over to http.c and got a Curl_ prefixDaniel Stenberg
The chunked transfer upload never stopped due to a silly add before we checked for >0!
2002-11-28Moved the compareheader function into this file and added Curl_ prefixDaniel Stenberg
We now check if the chunked transfer-encoding header has been added "by force" and if so, we enabled the chunky upload!
2002-11-26Dan Becker fixed a minor memory leak on persistent connnections usingDaniel Stenberg
FOLLOWLOCATION and CURLOPT_USERPWD.
2002-11-26fixed Curl_freeaddrinfo() to only free addrinfo, and added Curl_freednsinfo()Daniel Stenberg
for freeing single dns cache entries
2002-11-26Removed MFC dependency in Release Build when using VC++ IDEsm
2002-11-24Nedelcho Stanev's work-around for SFU 3.0Daniel Stenberg
2002-11-22bug fix for the problem Juan Ignacio Hervás discovered todayDaniel Stenberg
2002-11-20Kjetil Jacobsen's patch that introduces CURLOPT_PRIVATE and CURLINFO_PRIVATEDaniel Stenberg
for storage and retrieval of private data in the curl handle.
2002-11-15no more dllinit.o usageDaniel Stenberg
2002-11-13dllinit.c is removedDaniel Stenberg
2002-11-13not used and we don't have permission to distribute this!Daniel Stenberg
2002-11-12removed dllinit.c as MSVC doesn't need itDaniel Stenberg
2002-11-11moved the bools in the connectdata struct into the substruct namedDaniel Stenberg
ConnectBits where the other bools already are
2002-11-11The test for DNS cache entries left locked is now only built ifDaniel Stenberg
AGGRESIVE_TEST is also defined, as an addition to MALLOCDEBUG. It doesn't work for multi interface usage and should only be used with careful consideration.
2002-11-11unlock dns cache entries with a function call instead of a variable fiddleDaniel Stenberg
2002-11-11Dave Halbakken added curl_version_infoDaniel Stenberg
2002-11-11Now supports "Transfer-Encoding: chunked" for HTTP PUT operations where theDaniel Stenberg
size of the uploaded file is unknown.
2002-11-07Bug report #634625 identified how curl returned timeout immediately whenDaniel Stenberg
CURLOPT_CONNECTTIMEOUT was used and provided a fix.
2002-11-05Lehel Bernadt's fix to prevent debug message to get sent on errors whenDaniel Stenberg
debug wasn't enabled
2002-11-05ipv4-fixes for the new Curl_dns_entry struct and Curl_resolv() protoDaniel Stenberg
2002-11-05Curl_resolv() now returns a different struct, and it contains a referenceDaniel Stenberg
counter so that the caller needs to decrease that counter when done with the returned data. If compiled with MALLOCDEBUG I've added some extra checking that the counter is decreased before a handle is closed etc.
2002-10-31Disable the DNS cache (by setting the timeout to 0) made libcurl leakDaniel Stenberg
memory. Avery Fay brought the example code that proved this.
2002-10-28when using checkprefix(), the first argument must be the prefix!Daniel Stenberg
2002-10-28Transfer-Encoding: needs 17 bytes passed, not 18Daniel Stenberg
2002-10-28kromJx@crosswinds.net's fix that now uses checkprefix() instead ofDaniel Stenberg
strnequal() when the third argument was strlen(first argument) anyway. This makes it less prone to errors. (Slightly edited by me)
2002-10-28the malloc debug system assumes single threadDaniel Stenberg
2002-10-28Kevin Roth's fixes that use $(RM) instead of @erase and modified SSL versionDaniel Stenberg
2002-10-28prevent compiler warningDaniel Stenberg
2002-10-28prevent compiler warningsDaniel Stenberg
2002-10-23bad headers can come in two kinds, we either treat everything as one bigDaniel Stenberg
badly assumed header, or we think that parts of the buffer is a bad header and the rest is treated as a normal body part