aboutsummaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2004-11-04delete trailing whitespaceDaniel Stenberg
2004-11-02use ifdef not ifDaniel Stenberg
2004-11-02Paul Nolan fix to make libcurl build nicely on Windows CEDaniel Stenberg
2004-10-27Dan Fandrich's gzip handling fixDaniel Stenberg
2004-10-26Testing to define _REENTRANT unconditionally in here.Daniel Stenberg
2004-10-25Tomas Pospisek filed bug report #1053287 that proved -C - and --fail on aDaniel Stenberg
file that was already completely downloaded caused an error, while it doesn't if you don't use --fail! I added test case 194 to verify the fix. Grrr. CURLOPT_FAILONERROR is now added to the list stuff to remove in libcurl v8 due to all the kludges needed to support it.
2004-10-24Mohun Biswas found out that formposting a zero-byte file didn't work veryDaniel Stenberg
good. I fixed.
2004-10-21Dan Fandrich's better ifdef for include fixDaniel Stenberg
2004-10-19Alexander Krasnostavsky made it possible to make FTP 3rd party transfers withDaniel Stenberg
both source and destination being the same host. It can be useful if you want to move a file on a server or similar.
2004-10-19CURLINFO_NUM_CONNECTS and moreDaniel Stenberg
2004-10-19djgpp has locale.h and setlocale().Gisle Vanem
2004-10-16Alexander Krasnostavsky made the CURLOPT_FTP_CREATE_MISSING_DIRS option workDaniel Stenberg
fine even for third party transfers.
2004-10-16libcurl leaked memory for cookies with the "max-age" field set.Daniel Stenberg
2004-10-16Added Traian Nicolescu's patches for threaded resolver onGisle Vanem
Windows. Plugged some potential handle and memory leaks. Refs. http://curl.haxx.se/mail/lib-2004-10/0134.html http://curl.haxx.se/mail/lib-2004-10/0157.html
2004-10-14Eric Vergnaud pointed out that libcurl didn't treat ?-letters in the user nameDaniel Stenberg
and password fields properly in URLs, like ftp://us?er:pass?word@site.com/. Added test 191 to verify the fix.
2004-10-13Set errno = ENOMEM on faild countcheck().Gisle Vanem
2004-10-12#include "strerror.h" to get the strerror protoDaniel Stenberg
2004-10-12add proper error message when send() failsDaniel Stenberg
2004-10-11removed trailing whitespaceDaniel Stenberg
2004-10-11SO_NOSIGPIPEDaniel Stenberg
2004-10-10another lame attempt to avoid the "warning: will never be executed" warningDaniel Stenberg
by gcc 3.4
2004-10-10MSVC uses 'i64' suffix for 64-bit sizes.Gisle Vanem
2004-10-10If long is 8 bytes we can use strtol() to get 64 bit numbers and won't needDaniel Stenberg
our strtoll() replacement function.
2004-10-10Prevent a longjmp warning by moving the rc assign within Curl_resolv().Daniel Stenberg
Andy Cedilnik reported. Warning on HP-UX?
2004-10-10Use LL suffix for long long constants if the compiler supports it, to preventDaniel Stenberg
warnings.
2004-10-08killed trailing whitespaceDaniel Stenberg
2004-10-08if basename was found, check for a prototype and if none was found, provideDaniel Stenberg
our own in the formdata.c file to prevent warnings on systems without it
2004-10-08prevent warning with comparison between signed and unsignedDaniel Stenberg
2004-10-07use curl_strnequal(), not strncasecmp()Daniel Stenberg
2004-10-07use tld_strerror() only if previously detected, since otherwise we can'tDaniel Stenberg
work with libidn < 0.5.6
2004-10-06Fixed tld_check_name(). idna_to_unicode_lzlz() should never fail,Gisle Vanem
but return FALSE if 'uc_name == NULL' just in case.
2004-10-06USE_LIBIDN: Added Top-level-domain (TLD) check for host->name.Gisle Vanem
Only print a warning if check fails.
2004-10-06Chih-Chung Chang reported that if you use CURLOPT_RESUME_FROM and enabledDaniel Stenberg
CURLOPT_FOLLOWLOCATION, libcurl reported error if a redirect happened even if the new URL would provide the resumed file. Test case 188 added to verify the fix (together with existing test 99).
2004-10-06avoid warnings on systems with this member set constDaniel Stenberg
2004-10-06removed tabs and trailing whitespace from sourceDaniel Stenberg
2004-10-06untabifyDaniel Stenberg
2004-10-05djgpp has basename()Gisle Vanem
2004-10-05minor edit to re-use a variable and to hopefully avoid a (moot) warningDaniel Stenberg
about code that won't be reached
2004-10-05avoid warning for unused variableDaniel Stenberg
2004-10-05use idna_strerror() if it is available (only in libidn 0.5.6 or later)Daniel Stenberg
2004-10-05Only include libgen.h if we have a basename as well.Daniel Stenberg
Mainly meant to deal with the IRIX case which seems to requrie a "-lgen" lib to find the basename function and thus without the gen lib, it finds the header but not the function and our replacement function has a prototype that doesn't match the IRIX one. A different approach would be to make configure detect and use -lgen for the systems that require it.
2004-10-05let our basename() be staticDaniel Stenberg
2004-10-04Made the dns entry remain locked while a connection to the host remains toDaniel Stenberg
allow verbose output during this period. Bertrand Demiddelaer reported and helped fixing.
2004-10-03set async.done to TRUE last in the addrinfo callback to prevent the risk thatDaniel Stenberg
the multi-threaded resolver does wrong
2004-10-03Replaced the use of isspace() with our own version instead since we have mostDaniel Stenberg
data as 'char *' and that makes us pass in negative values if there is 8bit data in the string. Changing to unsigned causes too much warnings or too many required typecasts to the normal string functions.
2004-10-02Gisle Vanem provided code that displays an error message when the (libidnDaniel Stenberg
based) IDN conversion fails. This is really due to a missing suitable function in the libidn API that I hope we can remove once libidn gets a function like this.
2004-10-02removed weird preprocessor juggling not neededDaniel Stenberg
2004-10-01someone should hit meDaniel Stenberg
2004-10-01Aleksandar Milivojevic reported a problem in the Redhat bugzilla (seeDaniel Stenberg
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=134133) and not to anyone involved in the curl project! This happens when you try to curl a file from a proftpd site using SSL. It seems proftpd sends a somewhat unorthodox PASS response code (232 instead of 230). I relaxed the response code check to deal with this and similar cases.
2004-10-01fixed the basename() replacement, reported by GisleDaniel Stenberg