moparisthebest/curl
1
0
Fork 0
mirror of https://github.com/moparisthebest/curl synced 2024-08-13 17:03:50 -04:00
Code Issues Releases Wiki Activity
5,735 Commits 10 Branches 144 Tags 113 MiB
48b92c163f
Commit Graph

2278 Commits

Author SHA1 Message Date
Daniel Stenberg
5b55f9ecb3 Günter Knauf's NetWare changes. 2004-03-17 12:46:42 +00:00
Daniel Stenberg
fc67cca882 removed the min() macro define 2004-03-16 09:16:38 +00:00
Daniel Stenberg
256a16a8a3 if the global_init() is called from within curl_easy_init() and returns 
an error code, we now make curl_easy_init fail and return NULL.
 2004-03-15 16:28:36 +00:00
Daniel Stenberg
8f77030473 Initial commit of the first attempt to make three new *strerror() functions. 
No protos in the headers yet and no docs.
 2004-03-15 11:51:32 +00:00
Daniel Stenberg
f788f988ea fix signed and unsigned warnings 2004-03-14 18:15:04 +00:00
Daniel Stenberg
d04ffd258b postsize is off_t now, so we typecase it to int before doing normal printf 
with it (knowing it won't be larger than what fits in an int)
 2004-03-13 17:11:42 +00:00
Daniel Stenberg
24cfa7f1bb the postsize is an off_t so use the proper printf format to output the 
content-length when doing multipart posts
 2004-03-13 17:03:17 +00:00
Daniel Stenberg
94a1d09ac7 more variable type fixing for the huge posts 2004-03-12 14:22:16 +00:00
Daniel Stenberg
4ad68ec305 more variable type fixes for the large POST support 2004-03-12 13:06:01 +00:00
Daniel Stenberg
ef776ab893 Made the 'postsize' variable an off_t type to be able to hold large file 
sizes if desired
 2004-03-12 12:07:01 +00:00
Daniel Stenberg
1ebda8fa0e Added CURLOPT_POSTFIELDSIZE_LARGE to offer a large file version of the 
CURLOPT_POSTFIELDSIZE option to allow really big HTTP POSTs.
 2004-03-12 08:55:47 +00:00
Daniel Stenberg
9d064a3927 dl and ulspeed are now curl_off_t so typecast them to double when we 
return their values
 2004-03-11 21:51:55 +00:00
Daniel Stenberg
1d5a914c1c Made max5data() take a curl_off_t size as argument instead of double. Should 
make the progress meter more accurate for large files. Also made the sprintf
usage in that function avoid floating point.
 2004-03-11 21:48:15 +00:00
Daniel Stenberg
e545e33d5f Gisle Vanem's fixes to use CURL_SOCKET_BAD more instead of -1 for sockets. 2004-03-11 13:13:35 +00:00
Daniel Stenberg
326e8b9fc1 don't let the EINTR stuff build on windows 2004-03-11 12:57:04 +00:00
Daniel Stenberg
0d1fc73f21 Use more curl_off_t variables when doing the progress meter calculations and 
argument passing and try to convert to double only when providing data to the
external world.
 2004-03-10 16:20:33 +00:00
Daniel Stenberg
7225b14002 curl_socket_t mistakes cleanup 2004-03-10 16:01:47 +00:00
Daniel Stenberg
85838a8966 turn niflags into a define named NIFLAGS 2004-03-10 15:24:56 +00:00
Daniel Stenberg
b45d50ccc2 Nah, ignore the OS define in here. This is being included by ares and it 
certainly doesn't need the OS define.
 2004-03-10 10:19:32 +00:00
Daniel Stenberg
c39c55cb2d Minor edit to avoid an unreachable break and to remove the extra {} body 
within the switch.
 2004-03-10 09:50:12 +00:00
Daniel Stenberg
cf004cbc7e keep the number of bytes read in a size_t variable 2004-03-10 09:44:42 +00:00
Daniel Stenberg
9948250723 strequal() returns int so we typecast the return to bool when we store the 
result as bool
 2004-03-10 09:41:37 +00:00
Daniel Stenberg
47afa058a8 store times in time_t 2004-03-10 09:36:43 +00:00
Daniel Stenberg
40e892bb36 Jeff Lawson fixed the SSL connection to deal with received signals during the 
connect.
 2004-03-10 08:43:01 +00:00
Daniel Stenberg
bc6f3beb9c Now requires an OS string defined by the config*.h file 2004-03-10 08:15:50 +00:00
Daniel Stenberg
f9e1f32adb make windows builds use the OS string "i386-pc-win32" instead of just "win32" 
to make it more like other OS strings
 2004-03-10 08:14:22 +00:00
Daniel Stenberg
22f4d81f3e Make sure SIZEOF_CURL_OFF_T is defined before we check for it being > 4. 
It is only undefined when this file is included by others (like ares) and
in those cases it doesn't matter.
 2004-03-10 08:08:02 +00:00
Daniel Stenberg
0d1f80ccb5 removed the #if 0'ed pragmas that disable warnings on msvc 2004-03-10 07:04:31 +00:00
Daniel Stenberg
4e87dd89bf moved the curl_socket_t typedef downwards 2004-03-10 07:03:13 +00:00
Daniel Stenberg
ce5805a955 Use curl_socket_t instead of int for holding sockets. The typedefs and 
defines are in setup.h.
 2004-03-09 22:52:50 +00:00
Daniel Stenberg
52d85341ae only build with the windows threading trace code if 
DEBUG_THREADING_GETHOSTBYNAME is defined
 2004-03-09 21:42:43 +00:00
Daniel Stenberg
7b2869338d explicit typecast to visualize that we really want the result of the 
operation as a size_t
 2004-03-09 21:39:50 +00:00
Daniel Stenberg
ab35b2837f Tor Arntsen's fix to a AIX build problem 2004-03-09 21:25:59 +00:00
Daniel Stenberg
bab9f532f2 added HAVE_IOCTLSOCKET here 2004-03-09 08:35:33 +00:00
Daniel Stenberg
702021e099 don't compare signed/unsigned 2004-03-08 16:20:51 +00:00
Daniel Stenberg
2c756fa1d7 Commented the Curl_read() arguments. 2004-03-08 12:37:46 +00:00
Daniel Stenberg
385086b460 size_t/int fix 2004-03-08 12:37:11 +00:00
Daniel Stenberg
72a4715dff strlen() returns size_t 2004-03-08 12:36:54 +00:00
Daniel Stenberg
576be36249 waitconnect() takes the timeout argument as a long 2004-03-08 11:36:47 +00:00
Daniel Stenberg
c52c592f4c store size as size_t 
use %zd when outputting size_t
 2004-03-08 11:33:49 +00:00
Daniel Stenberg
d9ffd2f544 don't use 'register' 
make strtol() returns get stored in long variables
don't mix size_t with int
 2004-03-08 11:28:14 +00:00
Daniel Stenberg
a684c51c9b size_t/int/long fixes 2004-03-08 08:38:29 +00:00
Daniel Stenberg
3fa4ac080e we must not only support long long for Largefile to work, we must have 
a curl_off_t type that is larger than 4 bytes
 2004-03-08 07:46:26 +00:00
Daniel Stenberg
498666bdc3 issue 12 fix 2004-03-05 12:54:18 +00:00
Daniel Stenberg
3d3612e252 another include to prevent warnings 2004-03-05 11:39:19 +00:00
Daniel Stenberg
6717240650 more fixing to make the progress/getinfo stuff to work properly when doing 
file: transfers too
 2004-03-05 09:40:16 +00:00
Daniel Stenberg
ef0a2c714b clearly tell that these are ftp response timeouts 2004-03-05 09:37:48 +00:00
Daniel Stenberg
465753c2de When following to a new URL, we must make sure to call Curl_done() first, 
since the current connection must be taken care of properly before we move
on. Christopher R. Palmer reported a problem he found due to this mistake.
 2004-03-04 16:13:33 +00:00
Daniel Stenberg
e40392ac03 Andrés García's patch to prevent warnings while compiling with mingw, mainly 
because it is now possible to have both WIN32 and HAVE_CONFIG_H defined.
 2004-03-04 15:32:18 +00:00
Daniel Stenberg
1eb9fd6c4d use size_t for the data, but keep the protos use ssize_t to better fit 
with the existing transfer.c code
 2004-03-04 15:25:06 +00:00
Daniel Stenberg
78e47fbb5c include the proper header file too 2004-03-04 15:23:57 +00:00
Daniel Stenberg
2b59e90c54 fix progress data to be updated properly for file: transfers, as reported 
by Jesse Noller
 2004-03-04 15:12:12 +00:00
Daniel Stenberg
59a30e6d9b prevent harmless compiler warning 2004-03-04 12:57:12 +00:00
Daniel Stenberg
2479e06848 David Byron's work on making libcurl only require winsock 1.1 on Windows 
machines.
 2004-03-03 13:32:56 +00:00
Daniel Stenberg
5e7f1a0eaf more variable type cleanups 2004-03-03 13:24:14 +00:00
Daniel Stenberg
eb9383d49b our timeout values are longs while 'tv_sec' is int 2004-03-03 13:17:44 +00:00
Daniel Stenberg
817d17354e make the backup variable of the same kind as the data it backups! ;-) 2004-03-03 13:12:30 +00:00
Daniel Stenberg
88e226c686 some more size_t usage, and two added typecasts when converting from size_t 
to long (MIPSpro warnings)
 2004-03-03 13:11:28 +00:00
Daniel Stenberg
9116dd41bd tv_sec is an int, so we explicitly typecast the result of long - long to 
an int when we assign it.
 2004-03-03 13:07:32 +00:00
Daniel Stenberg
b2ef40731a Tom Bates' adjustment to build on his nsr-tandem-nsk. 2004-03-03 09:27:18 +00:00
Daniel Stenberg
e2f7030202 rename struct FILE to FILEPROTO, to prevent it from causing trouble with 
the plain old FILE typedef.
 2004-03-03 09:25:59 +00:00
Daniel Stenberg
7c85be9435 corrected the reuse_fresh condition 2004-03-02 14:00:44 +00:00
Daniel Stenberg
353f764119 Yet another curl_off_t printf format attempt, we now exclude the %-letter from 
FORMAT_OFF_T to allow additional options to get specified, like with '"%5"
FORMAT_OFF_T'.
 2004-03-02 09:31:18 +00:00
Daniel Stenberg
7ab3b5b3bb use FORMAT_OFF_T instead of CURL_FORMAT_OFF_T to reduce the complexity of 
having to redef that name
 2004-03-02 07:25:39 +00:00
Daniel Stenberg
00d5f886e3 - we switch to simply use FORMAT_OFF_T internally 
- Also, we must not assume that SIZEOF_CURL_OFF_T is defined, as this file
gets included from the ares dir at times and then it isn't defined.
 2004-03-02 07:25:08 +00:00
Daniel Stenberg
cb72a80fe0 Use CURL_FORMAT_OFF_T for printf()inf curl_off_t variables. 2004-03-01 16:28:00 +00:00
Daniel Stenberg
71d81f1f75 Now uses CURL_FORMAT_OFF_T instead of %Od 
Fixed the check for bad resumes. Made test case 99 work and proved a bug
in test case ... eh, was it 32?
 2004-03-01 16:27:34 +00:00
Daniel Stenberg
7983ec84a7 set CURL_VERSION_LARGEFILE if we support large files 2004-03-01 16:24:04 +00:00
Daniel Stenberg
8cb2a68694 define ENABLE_64BIT if we have enabled 64bit large files 
define our internal CURL_FORMAT_OFF_T define, we don't use the global one!
 2004-03-01 16:23:35 +00:00
Daniel Stenberg
485122035c fixed the test code to work 2004-03-01 16:22:17 +00:00
Daniel Stenberg
4d12be9867 removed an unnecessary shift and splut up som weird two-statements-per-line 
code
 2004-03-01 12:54:59 +00:00
Daniel Stenberg
5eeaff8235 Support 'z' for size_t-sized integer printing, as in %zd or %zx. 2004-03-01 12:44:07 +00:00
Daniel Stenberg
7d2e872828 Only consider the fresh-connection option on the first connection made, not 
on followed redirections etc. This should fix the bug #905365, which caused
NTLM to fail with the option set.
 2004-03-01 09:43:42 +00:00
Daniel Stenberg
4dfc0a78e7 Only attempt to send the FTP QUIT command if we actually have a FTP struct. 2004-03-01 07:59:25 +00:00
Daniel Stenberg
82bd583251 in Curl_disonnect(): call the protocol-specific disconnect function before we 
unlink the "current" connection struct from the connection cache.
 2004-03-01 07:19:26 +00:00
Daniel Stenberg
6c78b4b7c0 fixed some more size_t/int/long warnings and removed a few CMC comments 2004-02-27 13:21:14 +00:00
Daniel Stenberg
77268e0649 Joe Halpin made the FTP code send 'QUIT' on the control connection before 
it disconnects the TCP connection, like a good ftp client should!
 2004-02-27 07:08:37 +00:00
Daniel Stenberg
4515d06465 David Byron's fixes to make the latest curl build fine under MSVC 6. 2004-02-26 16:23:28 +00:00
Daniel Stenberg
07de0ff0ff Gisle Vanem's added support calloc()-debugging and outputting mode for 
fopen() as well.
 2004-02-26 14:52:51 +00:00
Daniel Stenberg
7d8cd5906c use calloc instead of malloc and we won't have to memset() the struct 2004-02-26 14:52:16 +00:00
Daniel Stenberg
d571064b65 Clear up int/long/size_t/ssize_t usage a bit 2004-02-26 13:40:43 +00:00
Daniel Stenberg
d687eed33e use %ld when printf()ing long variables (and removed use of 'register') 2004-02-26 12:32:29 +00:00
Daniel Stenberg
be2cdf14f3 Don't call the lock/unlock functions if they are NULL. They can still be 
NULL without violating protocol.
 2004-02-26 11:39:38 +00:00
Daniel Stenberg
f9b2b7940e disable the use of long double, we don't use it 2004-02-25 15:34:05 +00:00
Daniel Stenberg
df94c7aedc Based on a patch by Greg Hewgill I modified how long long is used, as we 
can use a 64bit type with MSVC that is a long long equivalent.
 2004-02-25 14:15:38 +00:00
Daniel Stenberg
f128fae705 fixed some warnings in the (both new and old) base64 usage 2004-02-23 16:09:39 +00:00
Daniel Stenberg
2ae4420869 make newer MSCV7 compilers use _strtoi64() as a strtoll() replacement 2004-02-23 14:24:49 +00:00
Daniel Stenberg
888eec833c Removed getpass from here. The fact this still was present here indicates 
that this file is out of date!
 2004-02-23 12:01:10 +00:00
Daniel Stenberg
2911e63bc2 curl_off_t is 8 bytes big on windows 2004-02-23 11:59:28 +00:00
Daniel Stenberg
34d247bc4e switch the arguments according to the c source 2004-02-23 11:44:03 +00:00
Daniel Stenberg
99bc7e5fd3 oops, the decode() function got its arguments reversed in my cleanup 
operation!
 2004-02-23 11:39:21 +00:00
Daniel Stenberg
f35c28214e use size_t to keep strlen() results 2004-02-23 08:38:49 +00:00
Daniel Stenberg
35fd1365ae simplied how create_hostcache_id() is used, and also its function somewhat 
cleared up some ssize_t/size_t mixups
 2004-02-23 08:35:57 +00:00
Daniel Stenberg
f493081004 adjusted to the modified base64 protos 2004-02-23 08:24:53 +00:00
Daniel Stenberg
e1b5e15431 More size_t cleanups in the base64 functions. 2004-02-23 08:22:43 +00:00
Daniel Stenberg
1aba4c519b The base64 encode function now takes a size_t for size, not an int as 
previously.
 2004-02-23 08:07:55 +00:00
Daniel Stenberg
0ffec4272a the missing part of Gisle Vanem's connect-timeout fix for win32 2004-02-22 22:31:24 +00:00
Daniel Stenberg
4bde770169 added some extra typecasts to prevent compiler warnings when converting 
int to various types
 2004-02-21 15:05:46 +00:00
Daniel Stenberg
9af7e46f72 inlcude krb4.h to get the proto for Curl_krb_kauth() to satisfy picky 
compilers
 2004-02-21 14:57:29 +00:00
Daniel Stenberg
7d844c430c Gisle Vanem brings name resolving timeout possibilities to windows people. 
This works by magicly starting up a new thread that can be killed when the
timeout is reached. testtesttest!
 2004-02-20 16:41:50 +00:00
Daniel Stenberg
44995d6877 Some compilers warn on completely empty source files, we provide a blank 
one to prevent that.
 2004-02-20 16:22:47 +00:00
Daniel Stenberg
2ebc821558 Convert functions to ANSI-style declaration to prevent compiler warnings 2004-02-20 16:18:26 +00:00
Daniel Stenberg
b2e1bf7e7d No longer support Z as a flag to print size_t, it isn't used by libcurl 
and I doubt anyone else uses it.

Better preprocessor magic for the O flag (for curl_off_t printing) to prevent
compiler warnings.
 2004-02-20 15:16:31 +00:00
Daniel Stenberg
a39669198f we call the macro CURLMAX() isntead of MAX(), just because it turned up 
it collides with another MAX define on some platforms (like netbsd 1.6.1)
 2004-02-20 08:47:23 +00:00
Daniel Stenberg
2c0c75e345 oops, missed a define when I changed from Curl_ to curlx_ 2004-02-20 07:22:25 +00:00
Daniel Stenberg
c6e5b67197 Doug Porter's patch that changes the order of preferences on how to find the 
default netrc file. We now read and uses HOME _before_ we use getpwuid() to
better allow users to move around HOME to use different .netrc files without
having to rely on even blacker magic.
 2004-02-19 09:22:00 +00:00
Daniel Stenberg
0bf1bd51c4 Remade to use curlx_-prefix. This means this function can be compiled and 
linked separately by the application. This function is not provided by the
libcurl API. It can only be accessed by apps if they compile and use this
particular source code.
 2004-02-19 08:12:13 +00:00
Daniel Stenberg
9efddfedab removed some "jhrg" from comments 2004-02-18 15:28:56 +00:00
Daniel Stenberg
de681d3b8f Made curl_accept() take a 'void *' instead of 'socklen_t *' in the 3rd 
argument to also not force the casual includer to know about the socklen_t
type.
 2004-02-18 12:22:56 +00:00
Daniel Stenberg
f33be3c313 Modified curl_accept() to take a 'void *' in the 2nd argument instead of 
sockaddr *. This has the added benefit that source files that include
memdebug.h doesn't have to know about "sockaddr".
 2004-02-18 12:18:33 +00:00
Daniel Stenberg
62a12b7faf No longer uses the 'ret' variable in the plain ipv4-version of 
my_getaddrinfo() (caused a warning by the IRIX MIPSPro compiler). Also
clarified the situation for the 3-arg version of gethostbyname_r() with a huge
comment.
 2004-02-18 10:05:17 +00:00
Daniel Stenberg
abd7e32c66 simplified and better commented config.h include logic 2004-02-18 08:35:51 +00:00
Daniel Stenberg
e2b6a45cfa Make sure dns cache timeout -1 really means forever, as it is documented to 
be. Simply skip the pruning.
 2004-02-18 07:56:18 +00:00
Daniel Stenberg
09aa165942 support closesocket() for closing sockets as well, as then we can use this 
code fine on ares!
 2004-02-16 16:24:01 +00:00
Daniel Stenberg
f2fbb5f3d5 Make realloc() support NULL as pointer. Made to allow us to use these routines 
to memdebug the ares stuff as well.
 2004-02-16 16:23:19 +00:00
Daniel Stenberg
ccdaa0b51f Make the 'areschannel' get created in the curl_easy_init() and re-use that 
same channel during the whole curl handle's life until curl_easy_cleanup().
 2004-02-16 15:24:22 +00:00
Daniel Stenberg
39926181ea Fix verbosconnect() when ipv6-enabled to not assume that conn->serv_addr 
is a valid pointer, but instead always depend on the passed-in dns pointer.
This happens to be NULL when the connection is re-used...
 2004-02-16 13:14:55 +00:00
Daniel Stenberg
8415b4a271 removed usage of a silly macro instead of the actual functions memcpy 
and memset
 2004-02-16 09:56:18 +00:00
Daniel Stenberg
e4916145ef Jeff Lawson pointed out that we need to check for a '5' in the version field 
to properly work with SOCKS5 proxies. I also included some ascii art describing
the SOCKS5 response, as RFC1928 describes. Jeff provided details in bug
report #741841 and here: http://curl.haxx.se/mail/lib-2004-02/0181.html
 2004-02-16 07:33:30 +00:00
Daniel Stenberg
a7e04cac78 Mark the dns entry 'inuse' properly even when used from the cache. This 
seems to correct some host cache screw-ups I could reproduce.
 2004-02-15 16:57:53 +00:00
Daniel Stenberg
8ddc18a4f9 another case which should use CURLcode and not int 2004-02-15 13:58:57 +00:00
Daniel Stenberg
0d3aa8b7be Use the was_iface variable when binding a socket locally, even if no 
SO_BINDTODEVICE is present, to prevent compiler warnings about the variable
 2004-02-15 13:55:24 +00:00
Daniel Stenberg
0612c275a7 (void) functions we don't check the return code for 2004-02-15 13:48:50 +00:00
Daniel Stenberg
f3d91528a1 CURLcode/int cleanup to reduce IRIX warnings 
Removed some dates/names in the comments.
 2004-02-15 13:48:28 +00:00
Daniel Stenberg
7abf2f44ec in the socks code, make sure we receive Curl_read results in ints and 
Curl_write in CURLcode, to keep the picky compilers happy
 2004-02-13 12:42:37 +00:00
Daniel Stenberg
4d17e77532 use CURLcode, not int, prevents picky compilers to warn 2004-02-13 12:16:24 +00:00
Daniel Stenberg
6a921197e2 the now and start variables were never really used 2004-02-13 12:13:30 +00:00
Daniel Stenberg
fa57a8a78e Ben Greear's SO_BINDTODEVICE patch that binds to a network interface "even 
more" when the previous approach. Known to work on Linux, possibly on other
platforms as well.
 2004-02-13 09:50:23 +00:00
Daniel Stenberg
e7d9ea8919 Greg Hewgill found out 'contentlength' wasn't big enough to hold a large file! 2004-02-13 07:12:40 +00:00
Daniel Stenberg
115e74a8ad I made the same fix here, that Tor already did in the ftp.c code. To make 
sure this doesn't get weird on 64bit archs.
 2004-02-13 07:03:03 +00:00
Daniel Stenberg
1b26fe39f9 Tor Arntsen's fix for the bad (64bit wise) typecast when using gmtime() 2004-02-13 06:59:49 +00:00
Daniel Stenberg
4674f3f115 Make hostcache_fixoffset() take a long for offset, to fully work with 64bit 
archs, also no longer typecast pointers to ints as that is a nono on 64bit
systems.
 2004-02-12 16:02:55 +00:00
Daniel Stenberg
5d64d657de removed the subchar variable, it was only set and never used 2004-02-12 09:53:03 +00:00
Daniel Stenberg
52caba0417 removed the ldaptext variable, it was only set and never used 2004-02-12 09:51:43 +00:00
Daniel Stenberg
ad1a70205f removed the nth variable, it was only set and never used anyway 2004-02-12 09:50:44 +00:00
Daniel Stenberg
d57eed6f22 No longer receive the return code in ConnectionKillOne() that wasn't dealt 
with anyway and thus caused picky compiler to warn.
 2004-02-12 09:48:27 +00:00
Daniel Stenberg
50efb8f33e #if-check for SIGALRM before assuming it is present 2004-02-11 21:11:08 +00:00
Daniel Stenberg
93e32e6f62 Modified the default HTTP Accept: header to only be Accept: */* 2004-02-09 12:46:41 +00:00
Daniel Stenberg
2d0b2f294f Oops. I broke the flow with the previous commit. 2004-02-09 11:40:00 +00:00
Daniel Stenberg
3bdc883c93 some annoying compilers warn about "(void)foo;" lines so we avoid them 2004-02-09 08:34:19 +00:00
Daniel Stenberg
011929cf4d Tor Arntsen's patch for working around a notorious bug in the AIX5 
getaddrinfo() implementation.
 2004-02-09 07:52:36 +00:00
Daniel Stenberg
f56d753507 A custom Host: header is only considered if the request is not made by 
following a location. After discussions with Tim Baker.
 2004-02-06 08:11:58 +00:00
Daniel Stenberg
e5165af5a6 The MIPSPro compiler complains on constructs such as "(void)foo;" so 
we avoid it where possible.
 2004-02-06 07:59:16 +00:00
Daniel Stenberg
83e878420a Make sure DynaGetFunction() returns a function pointer, not a data pointer. 
The standards don't actually allow typecasts between data and functions so
some picky compilers warn about this.
 2004-02-06 07:28:49 +00:00
Daniel Stenberg
1cb2306ae0 Andrés García's updates 2004-02-05 21:03:53 +00:00
Daniel Stenberg
cc90a235f4 if an empty 'transfer-encoding:' header is provided, we switch off the 
chunky coding of uploads
 2004-02-05 15:50:16 +00:00
Daniel Stenberg
dc753b8315 Gisle Vanem fixed a windows compiler warning 2004-02-05 13:25:14 +00:00
Daniel Stenberg
f71139750d options we get as longs need to be typecasted when assigned to prevent 
picky compiler warnings
 2004-02-05 09:38:56 +00:00
Daniel Stenberg
690b3e58b2 fix return type to silence compiler warnings 2004-02-05 09:38:10 +00:00
Daniel Stenberg
0c7cb0c10f prevent warning from that picky MIPSpro compiler 2004-02-05 09:37:46 +00:00
Daniel Stenberg
4816294f52 compiler warning fix, compare struct pointers of the same type 2004-02-05 09:37:04 +00:00
Daniel Stenberg
6587ff440a use the timeout options when waiting for the server to connect when using 
PORT
Provide better error messages to allow debugging if one if the ipv6-related
name functions fail in the ftp_use_port() function. This might help us
diagnose the problems on AIX. Also make sure getaddrinfo() uses NULL and
not "0" for the service argument.
 2004-02-05 09:26:01 +00:00
Daniel Stenberg
791e49134b hide the pack_hostent proto if ipv6 is enabled, as figured out by Tor Arntsen 2004-02-05 08:34:31 +00:00
Daniel Stenberg
556b68f44e the unused quit-function didn't pass a correct variable type to the 
response reading function
 2004-02-03 09:52:32 +00:00
Daniel Stenberg
95dd343244 added the missing ares numerical version initialiser 2004-02-03 06:39:37 +00:00
Daniel Stenberg
0412a201a3 undef accept before defining it, since AIX 5.2 has it as a define! 2004-02-02 21:34:01 +00:00
Daniel Stenberg
c2967c3c39 now provides c-ares version info in both version calls 2004-02-02 16:24:41 +00:00
Daniel Stenberg
886391d145 adjusted to the modified ares_strerror() function 
NOTE that this breaks ares-compatibility, we have now officially taken the
turn into the c-ares path. We will now officially depend on c-ares for asynch
name resolves.
 2004-02-02 16:00:31 +00:00
Daniel Stenberg
e1d541086c Dirk Manske fixed the ares usage even more. We could get a timeout from ares 
as well, and when failing and not getting a timeout we now include the
error message ares can provide us with.
 2004-02-02 15:53:50 +00:00
Daniel Stenberg
b84eaff1d5 set the 'retry' bit to TRUE when the connection is about to be retried, 
this allows the HTTP code to *not* return a failure just because no data
has been received from the server
 2004-02-02 14:49:54 +00:00
Daniel Stenberg
9eb6fc1fb6 clear the sockerror if no error was returned 2004-02-02 14:49:10 +00:00
Daniel Stenberg
dd1f495c2f we include errno.h to truly know if we have ECONNRESET or not 2004-02-02 11:59:42 +00:00
Daniel Stenberg
12fa877f00 Timeout slow ares name lookups. This is based on the patch brought by 
Dirk Manske, but modified by me.
 2004-02-02 10:15:22 +00:00
Daniel Stenberg
580a2fe72b when including ca-bundle.h, don't look in the current dir first, simply use 
the search path since we want the build-version rather than the one in the
source dir
 2004-02-02 10:13:54 +00:00
Daniel Stenberg
14cb9d296c make sure the connection is closed when it was detected reset! 2004-01-30 12:41:27 +00:00
Daniel Stenberg
df750c236c Somewhat crude attempt at fixing the test 91 failures. I commit this now 
so that the automatic testing hosts will test these changes over the weekend.
 2004-01-30 12:08:18 +00:00
Daniel Stenberg
6b7f169b10 only do the verifyconnect() clear magic on mpeix 2004-01-30 07:51:51 +00:00
Daniel Stenberg
bfe5568c85 added verifyconnect proto and use it correctly in the waitconnect function 2004-01-29 15:37:21 +00:00
Daniel Stenberg
7e45a1ac53 moved the definitions of IOCTL_3_ARGS to setup.h 2004-01-29 15:35:42 +00:00
Daniel Stenberg
76a51ef4b3 1. changed order of two include files to build fine on MPE/iX 
2. now reads the socket error before check connect status, also to make us
   run fine on MPE/iX
 2004-01-29 15:29:47 +00:00
Daniel Stenberg
4d17d6876e Dan Fandrich's cleanup patch to make pedantic compiler options cause less 
warnings. Minor edits by me.
 2004-01-29 13:56:45 +00:00
Daniel Stenberg
958987c214 Chunked-transfers should have an additional CRLF after the final 0 CRLF 
sequence.
 2004-01-28 17:07:22 +00:00
Daniel Stenberg
dc932f739a CONNECT response headers are now passed back as "regular" headers 2004-01-27 12:54:22 +00:00
Daniel Stenberg
e096072745 very big transfers now get nicer progress displayed after 9999 megabytes have 
been transfered!
 2004-01-27 12:25:37 +00:00
Daniel Stenberg
2465ff991f when saving in a cookie jar fails, include the file name in the error message 
to make it easier to track down
 2004-01-26 16:16:44 +00:00
Daniel Stenberg
334c3df09d Proxy username and password on persistant connections could easily get 
messed up. Vincent Bronner detected this.
 2004-01-23 08:36:03 +00:00
Daniel Stenberg
4120b7b0c0 check the arguments to the socks5 function, as the name and password might 
be NULL pointers, and if non-NULL if now support zero-length names/passwords
 2004-01-23 08:29:56 +00:00
Daniel Stenberg
fac1c13895 fixed the progress meter display for files >32 bit, Gisle Vanem reported 2004-01-23 08:02:12 +00:00
Daniel Stenberg
649caa1953 include curl.h for the typedef 2004-01-23 07:44:02 +00:00
Daniel Stenberg
d8ee927be2 define SIZEOF_CURL_OFF_T if not already defined 2004-01-23 07:41:05 +00:00
Daniel Stenberg
3efb90dd75 re-intended the code curl-style 2004-01-22 14:37:06 +00:00
Daniel Stenberg
ad7e26b140 return curl_off_t instead of long long, to work on more platforms 2004-01-22 14:31:46 +00:00
Daniel Stenberg
87eb26a33d it has strtoll 2004-01-22 14:25:54 +00:00
Daniel Stenberg
9a48ca69fc Gisle Vanem's fix, mingw as strtoll 2004-01-22 14:25:19 +00:00
Daniel Stenberg
a259f66fa4 attempt to fix 64bit seeking for Windows, does it work? 2004-01-22 13:11:35 +00:00
Daniel Stenberg
b791e158f0 use curl_off_t instead of off_t! 2004-01-22 12:45:50 +00:00
Daniel Stenberg
e5cd0cc402 use the proper type for formposts, not the deprecated one 2004-01-22 11:54:00 +00:00
Daniel Stenberg
252423a725 added the missing proto for the still unused quit function 2004-01-22 11:53:43 +00:00
Daniel Stenberg
5e2fcc1f74 use the proper timecond defines, not the obsolete ones I've removed! ;-) 
The initial QUIT-sending code is added, but not yet used due to the issues
previously mentioned on the mailing list.
 2004-01-21 08:50:08 +00:00
Daniel Stenberg
e262a7265a use the proper timecond defines, not the obsolete ones! 2004-01-21 08:47:21 +00:00
Daniel Stenberg
61e3d75def Gisle Vanem's patch for variables that "might be used uninitialized" 2004-01-16 09:17:04 +00:00
Daniel Stenberg
27b7220f1b silly me 2004-01-16 07:15:05 +00:00
Daniel Stenberg
a275365c72 Avoid Curl_if2ip() on Interix as well. Fix by Rodney. 2004-01-16 07:09:28 +00:00
Daniel Stenberg
686c70c5b5 use the %dk display for one extra k of progress 2004-01-15 13:08:12 +00:00
Daniel Stenberg
7052daf475 added missing parenthesis 2004-01-15 09:54:30 +00:00
Daniel Stenberg
c5c89862b2 rearranged the connect() call so that there's no interleaved #ifdef, to make 
it compiler better on amigaos
 2004-01-14 09:11:42 +00:00
Daniel Stenberg
2d567af069 fixed the Curl_resolv()'s return code when the looked up host was already 
in the cache (Vincent Bronner)
 2004-01-14 09:05:54 +00:00
Daniel Stenberg
57f67d0241 added a comment about this not being the original getdate.y version, it has 
been modified
 2004-01-13 08:59:53 +00:00
Daniel Stenberg
36f76396ea Diego Casorran's fixes to allow native AmigaOS builds 2004-01-13 08:35:57 +00:00
Daniel Stenberg
a76235c008 Matt Veenstra updated the Mac OS X framework files 2004-01-13 08:04:05 +00:00
Daniel Stenberg
71b3cc08a6 Brian R Duffy made the makefile work to build SSL-enabled curl with Borland 
C++.
 2004-01-13 08:02:01 +00:00
Daniel Stenberg
3a61c98b65 Peter Sylvester brought code that now allows a callback to modified the URL 
even when the multi interface is used, and then libcurl will simulate a
"follow location" to that new URL. Test 509 was added to test this feature.
 2004-01-12 15:26:32 +00:00
Daniel Stenberg
c8ac7a0d95 Dominick Meglio pointed out FTPS should use default port 990 according to 
IANA.
 2004-01-11 22:56:36 +00:00
Daniel Stenberg
44031f32a1 added one assert and a few comments describing how the auth stuff works 2004-01-09 14:03:06 +00:00
Daniel Stenberg
7beac70644 Dominick Meglio updated the win description since we fixed the gettimeofday 
name issue
 2004-01-09 07:45:47 +00:00
Daniel Stenberg
053f6c85ef updated year in the copyright string 2004-01-07 09:19:33 +00:00
Daniel Stenberg
292ef5fabb Dan Fandrich's zlib fix 2004-01-05 22:54:45 +00:00
Daniel Stenberg
9363d94f9b new files for the large file support number parsing 2004-01-05 22:39:46 +00:00
Daniel Stenberg
b60e0fa97e David J Meyer's large file support. 2004-01-05 22:29:29 +00:00
Daniel Stenberg
a0edfb90c2 make our private version of gettimeofday() static 2004-01-04 12:10:14 +00:00
Daniel Stenberg
93a8572928 white space and copyright year changes 2004-01-04 12:09:52 +00:00
Daniel Stenberg
591fe66f18 Dominick Meglio's description how to build ares for libcurl on win32 2004-01-04 12:00:11 +00:00
Daniel Stenberg
84b3c3b569 Make setopt() support CURLOPT_IPRESOLVE... 2003-12-19 08:10:09 +00:00
Daniel Stenberg
4223f602ed djgpp fixes by Gisle Vanem 2003-12-19 08:03:35 +00:00
Daniel Stenberg
8146d46386 Gisle Vanem's minor fixes 2003-12-19 08:03:15 +00:00
Daniel Stenberg
6c6e5522cb redirect from a bad url such as "www.com?moo=foo" to an absolute path didn't 
work, as reported by John McGowan
 2003-12-18 09:19:10 +00:00
Daniel Stenberg
aff9a817b4 David Byron's fix to make libcurl build fine with both .NET and VC6 version 
of MSVC
 2003-12-18 08:56:20 +00:00
Daniel Stenberg
23a6041698 use the curlassert() instead of custom checks 2003-12-16 14:08:53 +00:00
Daniel Stenberg
3788f25eac added a library-wide interface for doing dns cache pruning, and no longer 
make the pruning at each name lookup, only in Curl_done().
 2003-12-15 17:33:49 +00:00
Daniel Stenberg
cfe53768dd if Curl_hash_add() returns NULL, we shall not free the addrinfo field as that 
is made in the hash function in the case of failure (using the already setup
'dtor' function).
 2003-12-15 15:22:10 +00:00
Daniel Stenberg
947e656367 make sure that hash_add() has no allocated resources left in case it 
returns NULL
 2003-12-15 15:21:13 +00:00
Daniel Stenberg
50e7545118 Giuseppe Attardi fixed a really tricky bug 2003-12-15 14:48:37 +00:00
Daniel Stenberg
a025425d06 use the CORRECT file name! ;-) 2003-12-11 07:27:13 +00:00
Daniel Stenberg
8f0abd31c5 Use Curl_transfer() properly. Fixes the bug Gisle Vanem found! 2003-12-10 15:27:59 +00:00
Daniel Stenberg
a13f85fb33 assert that we get a good index to Curl_transfer() 2003-12-10 15:27:27 +00:00
Daniel Stenberg
3b048880ad we now have a 'curlassert' function to use 2003-12-10 15:27:06 +00:00
Daniel Stenberg
9bc97b2d3c updated with the new configure option usage 2003-12-10 14:16:56 +00:00
Daniel Stenberg
68c508976f added the new amiga files to the dist 2003-12-10 13:56:13 +00:00
Daniel Stenberg
4d8a53da2f Diego Casorran's amiga makefiles 2003-12-10 13:54:57 +00:00
Daniel Stenberg
6c56340e1a native AmigaOS support by Diego Casorran 2003-12-10 13:53:35 +00:00
Daniel Stenberg
ac114ed45b do the ares stuff in the configure file 2003-12-10 13:38:51 +00:00
Daniel Stenberg
ee337309cf better human error message when a FTP response can't be read 2003-12-08 13:06:08 +00:00
Daniel Stenberg
0ab2dff69c based on Gisle's comment, I removed the use of syslog() and fixed the 
netdb.h include, then I re-indented some code to use curl-style
 2003-12-04 15:21:49 +00:00
Daniel Stenberg
8ccc1148d9 my contentlength adjustment broke the ftp download range stuff, this makes 
it work again
 2003-12-03 08:26:31 +00:00
Daniel Stenberg
06c38330ba Steve Green fixed Curl_resolv() 2003-12-03 07:55:51 +00:00
Daniel Stenberg
0f4d042d3e Ignore content-length when chunked transfer-encoding is transfered. 2003-12-03 07:52:00 +00:00
Daniel Stenberg
c79de8d86e less complicated code for the MSG_NOSIGNAL detection 2003-12-02 22:05:23 +00:00
Daniel Stenberg
1b99d33b59 fix compiler warnings 2003-12-02 13:40:12 +00:00
Daniel Stenberg
800052dc50 use the HAVE_KRB4 define instead of just KRB4 2003-12-02 13:27:29 +00:00
Daniel Stenberg
2c22feced2 fixes from Gisle Vanem to try 'AUTH SSL' before 'AUTH TLS', edited by me 2003-12-02 13:21:53 +00:00
Daniel Stenberg
2cf209d3f7 If HAVE_MSG_NOSIGNAL is set, we use MSG_NOSIGNAL when we call send() and 
recv() and we no longer attempt to ignore the SIGPIPE signal.
 2003-12-02 10:12:44 +00:00
Daniel Stenberg
d1d417664e Gisle Vanem fixed the check-order for FTPS and FTP. 2003-12-02 06:25:41 +00:00
Daniel Stenberg
cd653f1be9 FTPS now works with active ftp and uploads too. 2003-12-01 10:14:39 +00:00
Daniel Stenberg
012c4ee1fa James Clancy updated the Borland makefiles 2003-11-27 12:24:22 +00:00
Daniel Stenberg
471c30372b Markus Moeller's change to check for HAVE_SPNEGO instead of the previous 2003-11-27 09:52:44 +00:00
Daniel Stenberg
1d977b7848 When basic is the only auth wanted (which it is by default), the auth phase 
is always considered done immediately as Basic needs to extra passes.

This fix corrects bug report #848371.
 2003-11-25 09:05:15 +00:00
Daniel Stenberg
45614d54c6 allow the end-of-headers from a proxy response to CONNECT end with a CRCR 
as well as a CRLF
 2003-11-24 16:17:55 +00:00
Daniel Stenberg
d8cf2d42c0 prevent warning for non-SSL builds 2003-11-24 11:44:04 +00:00
Daniel Stenberg
96d7a95ce4 windows build fix based on a patch by Gisle Vanem 2003-11-24 11:41:26 +00:00
Daniel Stenberg
d5d55a856e adjusted code to the new socket fields in the sessionhandle struct 2003-11-24 09:04:09 +00:00
Daniel Stenberg
1e98727c55 FTPS support added as RFC2228 and the murray-ftp-auth-ssl draft describe it 2003-11-24 07:15:37 +00:00
Daniel Stenberg
af9dc9f128 Gaz Iqbal fixed a range string memory leak! 2003-11-24 07:10:02 +00:00
Daniel Stenberg
4874aa6986 fixed wrong comment 2003-11-24 07:08:07 +00:00
Daniel Stenberg
3fed12a203 combine the two identical cases 2003-11-20 14:16:18 +00:00
Daniel Stenberg
f481b3e1dc bail out properly on a 407 when we can't authenticate basic, bug report 
#845247
 2003-11-20 09:53:42 +00:00
Daniel Stenberg
edce444551 If a partial file has been read or written, force a closure the existing 
connection to avoid re-use of it, since we cannot know in what state it is.
 2003-11-19 14:36:42 +00:00
Daniel Stenberg
7bee4140e8 respect bits.close even if an error already is set, but make sure that the 
existing error is the one returned back
 2003-11-19 14:35:40 +00:00
Daniel Stenberg
a0a5668dbd do ares_destroy() on the ares handle when we're done in Curl_is_resolved() 2003-11-15 10:29:43 +00:00
Daniel Stenberg
dfe0118033 Mathias Axelsson found a case where we free()d the server certificate twice! 2003-11-15 10:00:20 +00:00
Daniel Stenberg
a701f3dddf Siddhartha Prakash Jain found a case with a bad resolve that we didn't 
properly bail out from, when using ares.
 2003-11-14 11:56:15 +00:00
Daniel Stenberg
43bb20461f Peter Sylvester found a flaw in the connect code for ipv6-enabled hosts. 
I guess it seldomly happens on linux and that's why it wasn't found before.
He used Solaris to notice it.

I took the opportunity to rewrite the Curl_connecthost() slightly to feature
less duplicate code in the two different versions (ipv4/ipv6).
 2003-11-13 13:28:40 +00:00
Daniel Stenberg
1e251a64f3 Default Content-Type for formparts has changed to "application/octet-stream". 
This seems more appropriate, and I believe mozilla and the likes do this.

.html files now get text/html as Content-Type.

Pointed out in bug report #839806.
 2003-11-13 07:43:18 +00:00
Daniel Stenberg
776f0bd95e don't log failed socket() calls 2003-11-13 07:33:51 +00:00
Daniel Stenberg
ad77f760cf Added CURLOPT_NETRC_FILE. 2003-11-11 14:30:43 +00:00
Daniel Stenberg
a50d2a45d7 Generate the ca-bundle.h file in the build dir and not in the source dir, 
which works nicely since the build dir is before the source dir in the
include path.
 2003-11-10 09:28:22 +00:00
Daniel Stenberg
17a834bc63 After Sébastien Willemijns' bug report, we now check the separators properly 
in the 229-reply servers respond on a EPSV command and bail out better if
the reply string is not valid RFC2428-compliant.
 2003-11-10 08:12:53 +00:00
Daniel Stenberg
f1445ea228 remove the struct fields for the deprecated passwd prompt callback 2003-11-07 13:26:00 +00:00
Daniel Stenberg
b1aa8eb7ef the resource file itself 2003-11-07 07:20:15 +00:00
Daniel Stenberg
967cb4d75e removed the resource lines I shouldn't have added in the firt place 2003-11-07 07:13:56 +00:00
Daniel Stenberg
41be41358e CURLDEBUG, no underscore 2003-11-06 16:04:19 +00:00
Daniel Stenberg
ecb818805b Gisle Vanem's libcurl.dll version info patch applied 2003-11-06 15:09:30 +00:00
Daniel Stenberg
bd4c081157 unitialized variable fix, reported by both Marty Kuhrt and benjamin gerard 2003-11-06 07:55:45 +00:00
Daniel Stenberg
927e392b31 Mac OS X framework build files, contributed by Matt Veenstra 2003-11-06 07:42:02 +00:00
Daniel Stenberg
aeb5edbf07 Moved the TRUE and FALSE #defines to lib/setup.h instead, as they don't 
belong in the public header file. They are not in our name space so we
should not set them globally.
 2003-11-06 07:21:32 +00:00
Daniel Stenberg
c45f43ce32 Troels Walsted Hansen fixed the MSVC makefiles to build fine on Windows. 2003-11-04 06:59:25 +00:00
Daniel Stenberg
0eb037a849 spellfix in the generated comment by Kevin Roth 2003-11-04 06:52:26 +00:00
Daniel Stenberg
a717afc3a7 gmtime() really can't even return NULL, can it? 2003-10-31 21:43:22 +00:00
Daniel Stenberg
a7aacac5e8 assume that MDTM returns the timestamp as UTC 2003-10-31 21:36:43 +00:00
Daniel Stenberg
693df0fa34 silly me, I was meaning to do this change already as discussed on the libcurl 
list, we get the time in GMT and not localtime
 2003-10-30 09:13:04 +00:00
Daniel Stenberg
fb26b2bd98 curl --head now reports info "headers" on file:// URLs as well 2003-10-30 09:08:16 +00:00
Daniel Stenberg
46accd62d1 when getting headers only, stop pretending the reported time is GMT, as it 
isn't. In fact, it might be, but we have no idea.
 2003-10-30 09:06:24 +00:00
Daniel Stenberg
c964aedf3b Only compile the ipv6-section for ipv6-enabled libcurls. Should save 
us some trouble.
 2003-10-30 07:13:13 +00:00
Daniel Stenberg
e9c835ad06 David Hull made the file: URL parser also accept the somewhat sloppy file 
syntax: file:/path. I added test case 203 to verify this.
 2003-10-29 09:53:21 +00:00
Daniel Stenberg
25613503cb Another glibc resolve name fix 2003-10-28 13:06:15 +00:00
Daniel Stenberg
1cdc66d927 overlapping memory chunks with strcpy(), detected by the friendly valgrind 2003-10-28 09:17:15 +00:00
Daniel Stenberg
203f66d0cb James Bursa's fix to prevent failf() to write outside its buffer boundary 2003-10-27 06:35:47 +00:00
Daniel Stenberg
4cccceb034 snprintf() made a single-byte buffer overflow, as it could write a zero 
outside its given buffer. Discovered and reported by James Bursa.
 2003-10-26 15:37:45 +00:00
Daniel Stenberg
a5e22867c7 Resuming a download of an already downloaded document, that is trying to get 
a range of a document beyond its size, caused libcurl to "hang" until the
server closed the connection and then it returned error 18.

This is bad. This way, we don't return any error at all, which isn't nice
either, as we need to alert the app somehow that the request range was out
of size.
 2003-10-24 21:54:34 +00:00
Daniel Stenberg
fcfc6de918 Based on David Hull's fix in bug report 804599, we now check for solaris 
and gcc, to set the -mimpure-text link flag for linking the lib better.
 2003-10-24 21:26:04 +00:00
Daniel Stenberg
656d4b4c20 if malloc fails, we must bail out nicely 2003-10-24 12:58:59 +00:00
Daniel Stenberg
7247041e21 Improved checks for bad memory situations and proper behaviour when we 
run out of memory.
 2003-10-24 12:58:29 +00:00
Daniel Stenberg
017a78d0c5 better bailing-out cleanup if a malloc fails in the DNS cache 2003-10-24 12:57:23 +00:00
Daniel Stenberg
14f795816d Georg Horn's fixes to do different CA cert verifications. They can now be 
done even if the result is ignored, as some sites seem to require that.
 2003-10-23 07:44:55 +00:00
Daniel Stenberg
f232baf4d2 Dylan Ellicott added the release-libcurl-ssl-dll target 2003-10-22 11:26:41 +00:00
Daniel Stenberg
b439e8ffb7 Do the auth stuff at the end-of-headers and not at the start-of-body, as 
we might not get a body when we get a 401 with a set of WWW-Authenticate:
headers. This fixes the problem Kevin Roth detected in 7.10.8-pre4 and pre5.
Verified by test case 91.
 2003-10-22 11:15:48 +00:00
Daniel Stenberg
640d6a5564 updated for mingw32 2003-10-21 14:14:25 +00:00
Daniel Stenberg
d1c08e1d85 conn->user and conn->passwd will now always be set to point to something. 
If not user or password is set, they will point to a "" string.
 2003-10-21 06:34:15 +00:00
Daniel Stenberg
b83c7ade10 allow no user and password for proxy too, Basic 2003-10-20 20:32:08 +00:00
Daniel Stenberg
7d6584e140 allow no user and no passwd when doing basic auth 2003-10-20 20:30:40 +00:00
Daniel Stenberg
31b6bb6a41 lock the DNS cache properly before adding an entry when using asynch DNS 2003-10-20 08:45:33 +00:00
Daniel Stenberg
d9f961ed2d better error output on SSL errors when receiving data - Georg Horn patch 2003-10-19 17:38:51 +00:00
Daniel Stenberg
c7f5476416 fixed the bug my previous change introduced 2003-10-19 05:42:49 +00:00
Daniel Stenberg
381c6c5d52 minor fix to not shadow a variable 2003-10-18 20:38:18 +00:00
Daniel Stenberg
9df8f9a63d builds warning-free with -Wshadow now 2003-10-18 20:35:03 +00:00
Daniel Stenberg
5eee801d06 don't shadow 'socket' 2003-10-18 20:34:33 +00:00
Daniel Stenberg
80a06403e4 fixed gcc -Wshadow warnings 2003-10-18 20:28:53 +00:00
Daniel Stenberg
7ad69f59c7 removed gcc -Wshadow warning 2003-10-18 20:24:54 +00:00
Daniel Stenberg
83b82e4472 memory leak 2003-10-18 20:24:20 +00:00
Daniel Stenberg
65dab79ca5 gcc -Wshadow complaints fixed 2003-10-18 20:14:33 +00:00
Daniel Stenberg
f0089b6291 compiler warning fix 2003-10-18 20:14:21 +00:00
Daniel Stenberg
d10508e08c prevent compiler warnings with relly picky compiler options 2003-10-18 20:13:54 +00:00
Daniel Stenberg
630995f40a Dominick Meglio pointed out these files should be removed from here 2003-10-18 19:40:58 +00:00
Daniel Stenberg
ce5db9a86e Dominick Meglio implemented CURLOPT_MAXFILESIZE and --max-filesize. 2003-10-17 13:11:00 +00:00
Daniel Stenberg
94568f884d typecasts to prevent warnings 2003-10-17 12:44:54 +00:00
Daniel Stenberg
9935d33b8e make empty names look fine in verbose output 2003-10-17 09:29:30 +00:00
Daniel Stenberg
d42c6b7e78 make no user or no password just mean blank fields, not aborted operation 2003-10-17 09:29:06 +00:00
Daniel Stenberg
679654bd47 o the name and password arrays are 256 bytes, so let's accept that lengthy 
input
o have ->passwd and ->name be NULL if no name/passwd was given
o only set default user+password for FTP if no userpwd was given
 2003-10-17 09:28:00 +00:00
Daniel Stenberg
9e3f54431d made the code deal with empty name and password 2003-10-17 09:26:28 +00:00
Daniel Stenberg
260c353577 show info text (verbose) about auth type and user name in use 2003-10-17 07:05:26 +00:00
Daniel Stenberg
22adcb9cd1 password promting support removed from libcurl 2003-10-16 14:08:59 +00:00
Daniel Stenberg
597c1fe6bc rewritten alternative name check 2003-10-16 13:44:34 +00:00
Daniel Stenberg
7201a5a290 the previous fix left this compiler error 2003-10-15 20:37:00 +00:00
Daniel Stenberg
c6a0bb99af bad license situation for the altname patch 2003-10-15 14:42:11 +00:00
Daniel Stenberg
a9af971c59 Avoid doing getsockopt() on Windows to verify connects. It seems that this 
hogs Windows machines when libcurl is being used multi-threaded (with > ~50
threads). Andrew Fuller helped us verify and test this.

According to a MSDN web page on connect(), it returns 0 when the connect
is done and thus we don't need the getsockopt() call anyway on Windows.
 2003-10-15 11:42:57 +00:00
Daniel Stenberg
b71ca15598 Kimmo Kinnunen fixed a crash with duphandle() when CURLDEBUG was set 2003-10-14 13:10:05 +00:00
Daniel Stenberg
749f5387c1 Gisle Vanem's IPv6-on-Windows patch applied! 2003-10-14 12:00:45 +00:00
Daniel Stenberg
7d6e2a3adb added more mirrored versions of the ares 1.1.1 package 2003-10-13 13:10:05 +00:00
Daniel Stenberg
43cc844d3f Updated build instruction since I now offer a complete ares-package with all 
the patches already applied. No need to apply any patches manually anymore.

ares 1.1.1 is still compatible.
 2003-10-13 13:07:50 +00:00
Daniel Stenberg
8341e8e502 Clear the connection pointer after the async resolve has failed. 
This cures the problem reported by Giuseppe Attardi on October 12, 2003.
 2003-10-13 12:21:56 +00:00
Daniel Stenberg
c9d154ca7f Removed extra LDFLAGS assignment. Pointed out in debian bug report #212086 
(http://bugs.debian.org/212086). Patch by Domenico Andreoli.
 2003-10-13 00:12:06 +00:00
Daniel Stenberg
022996e5c7 Dirk Manske made the share-locking around DNS lookups a bit "looser" so that 
multiple DNS lookups can run simultaneously faster. The downside is that
resolving the same host name now can be made at once from multiple threads,
but the upside is that threads now don't alwys have to wait for the others'
resolves. Test case 506 updated accordingly.
 2003-10-12 13:57:41 +00:00
Daniel Stenberg
d1a3c9944e Lachlan O'Dea fixed a resume problem: "If I set CURLOPT_RESUME_FROM, perform 
an HTTP download, then reset CURLOPT_RESUME_FROM to 0, the next download still
has a Range header with a garbage value." bug report #820502
 2003-10-09 20:04:47 +00:00
Daniel Stenberg
e4adbf4017 Dominick Meglio fixed this to build fine on MSVC 2003-10-09 08:19:01 +00:00
Daniel Stenberg
867c189ef7 bail out hard if ipv6 and ares are both enabled at the same time 2003-10-08 19:56:44 +00:00
Daniel Stenberg
5c52cacc98 - Frank Ticheler provided a patch that fixes how libcurl connects to multiple 
addresses, if one of them fails (ipv4-code).
 2003-10-08 13:32:43 +00:00
Daniel Stenberg
e612f73374 include file fixes 2003-10-08 13:07:08 +00:00
Daniel Stenberg
ff5b6ff528 fixed to build fine without ssl 2003-10-08 13:06:50 +00:00
Daniel Stenberg
6494889e3b Neil Dunbar provided a patch that now makes libcurl check SSL 
subjectAltNames when matching certs. This is apparently detailed in RFC2818
  as the right thing to do. I had to add configure checks for inet_pton() and
  our own (strictly speaking, code from BIND written by Paul Vixie) provided
  code for the function for platforms that miss it.
 2003-10-07 21:46:47 +00:00
Daniel Stenberg
b61c25af45 make sure that a regular POST using the read callback works 2003-10-07 14:39:32 +00:00
Daniel Stenberg
17e13acd10 BeOS-adjustments, as provided by Shard 2003-10-05 15:04:09 +00:00
Daniel Stenberg
5564e9a321 weird typo removed 2003-10-05 15:03:37 +00:00
Daniel Stenberg
a0118fb390 just re-indented some code 2003-10-05 15:03:21 +00:00
Daniel Stenberg
57b6202eaa New code for BeOS-style non-blocking sockets, provided by Shard and 
Jeremy Friesner.
 2003-10-05 15:02:58 +00:00
Daniel Stenberg
780652e5b6 Vincent Bronner pointed out that if you set CURLOPT_COOKIE to NULL, it still 
sends off the previously set cookie. This also concerns CURLOPT_REFERER.
This fix should stop this.
 2003-10-04 15:48:15 +00:00
Daniel Stenberg
45ef092cc2 Jon Turner mentioned this bug fix to correct how libcurl deals with paths 
after a failed transfer.
 2003-10-04 15:25:02 +00:00
Daniel Stenberg
343291ce37 Based on a patch provided by Siddhartha Prakash Jain. In Curl_resolv() when 
my_getaddrinfo() has been called (and wait has been set to TRUE), we check
if the name already is resolved and if so don't return wait status to the
parent. This can happen with IP-only names.
 2003-10-04 14:50:45 +00:00
Daniel Stenberg
8aa2894bfb failing to resolve a name using ares must Curl_disconnect() the handle 
properly or risk getting into trouble!
 2003-09-21 23:10:47 +00:00
Daniel Stenberg
2297eca103 Added CURLOPT_IPRESOLVE support 2003-09-19 15:16:47 +00:00
Daniel Stenberg
09ccfcdcd4 Markus Moeller's SPNEGO patch applied, with my edits, additions and minor 
cleanups.
 2003-09-19 12:56:22 +00:00
Daniel Stenberg
951e2a6545 If a connection is closed down and it had some kind of NTLM involved, we 
reset the authentication state to make it restart on the next connection.

This of course because NTLM is connection-oriented, whilst all other
authentication schemes are not.
 2003-09-15 22:33:18 +00:00
Daniel Stenberg
58b1437cae When we issue a HTTP request, first make sure if the authentication phase 
is over or not, as if it isn't we shall not begin any PUT or POST operation.

This cures bug report #805853, and test case 88 verifies it!
 2003-09-15 21:11:22 +00:00
Daniel Stenberg
c611ac958e new ares patch uploaded 2003-09-14 22:42:08 +00:00
Daniel Stenberg
071fc74c0e reverted the change that blanks LIBS 2003-09-14 21:24:20 +00:00
Daniel Stenberg
14597475b1 Jeff Pohlmeyer did some marvelous debugging to track this one down. We MUST 
NOT free the existing hash entry when we try to add a new one that matches
an existing entry. We now instead free the new one, and make the parent
function use the old entry's struct instead.
 2003-09-14 21:17:54 +00:00
Daniel Stenberg
de3281a3a8 blank the LIBS variable, as we don't need any libs when we link the lib. 2003-09-12 20:45:39 +00:00
Daniel Stenberg
fd42247cac Tim Bartley's patch that makes the GSSNEGOTIATE option work for Microsoft's 
"Negotiate" authentication as well.
 2003-09-11 22:21:11 +00:00
Daniel Stenberg
0efcb57623 For easy handles within multi handles, we share the DNS cache always. 2003-09-11 22:14:29 +00:00
Daniel Stenberg
6fde147273 #ifdef 0'ed Curl_llist_insert_prev and Curl_llist_remove_next, as they are 
not used by any code in libcurl!
 2003-09-11 22:13:37 +00:00
Daniel Stenberg
cc497fd153 don't use a blank proxy 2003-09-11 22:05:23 +00:00
Daniel Stenberg
c1f31180a2 Curl_is_resolved(): FD_ZERO the file descriptors before we call ares_fds(). 
Problem tracked down by Bjorn Reese.
 2003-09-11 21:27:19 +00:00
Daniel Stenberg
e8c762981a do not use 'long' to store 4 bytes, as 64bit architectures have 64bit longs. 2003-09-07 15:00:10 +00:00
Daniel Stenberg
adcbe03aeb Uses less macros. #ifdef'ed out unused functions. Edited slightly to be 
more in the same style as other curl source code. The only actual code change
is an added check after a malloc() call.
 2003-09-05 12:44:35 +00:00
Daniel Stenberg
df760a0a4b no user or password set, bail out 2003-09-04 13:31:49 +00:00
Daniel Stenberg
f2057eb744 ah, this change broke multiple test cases 2003-09-04 13:07:09 +00:00
Daniel Stenberg
2956fe97a0 Jörg pointed out that I did this fix wrongly, now authwant is cleared 
properly if no auth is available
 2003-09-04 11:34:09 +00:00
Daniel Stenberg
1f9b0e70ab Based on Joerg Mueller-Tolk's patch, this introduces support for 
CURLINFO_HTTPAUTH_AVAIL and CURLINFO_PROXYAUTH_AVAIL
 2003-09-04 10:55:20 +00:00
Daniel Stenberg
8fae12b2f1 1. store the httpproxycode in the loop after it is parsed 
2. made Curl_http_auth_act() reset 'authavail' unconditionally
 2003-09-04 10:18:12 +00:00
Daniel Stenberg
7410210f19 avoid proxy (and other) auth if no user+password is given 2003-09-04 10:08:53 +00:00
Daniel Stenberg
40e6b12cca added the missing share-functions, pointed out by Edward Chan 2003-09-04 10:04:15 +00:00
Daniel Stenberg
f2d422235b Early Ehlinger's CURLOPT_FTP_RESPONSE_TIMEOUT patch applied. 2003-09-03 22:02:40 +00:00
Daniel Stenberg
52ceab5e41 Re-arranged code to make the proxy-CONNECT loop able to do some of the 
authentication negotiations needed for NTLM, Digest etc.
 2003-09-03 21:51:28 +00:00
Daniel Stenberg
cafcc242e6 Joerg Mueller-Tolk's fix to better deal with user+passwords for proxies 2003-09-03 21:47:41 +00:00
Daniel Stenberg
2ad0a7d20f minor fix to make better bail-out check 2003-09-03 21:44:59 +00:00
Daniel Stenberg
481094db90 warn if no CN is available if verify is only set to 1 2003-09-03 20:47:17 +00:00
Daniel Stenberg
f7db3023a8 wrap at 80 cols 2003-09-02 07:48:58 +00:00
Daniel Stenberg
070cd2b52f Henrik Storner's update to make libcurl work with OpenLDAP 2.1.22 (current). 
Also reported to work with OpenLDAP 2.0.26.
 2003-09-01 08:23:31 +00:00
Daniel Stenberg
5629eeaec2 The error buffer was not getting filled when Curl_wait_for_resolv() fails. 
Jeff Pohlmeyer fixed.
 2003-09-01 08:21:08 +00:00
Daniel Stenberg
46c2bc4bb6 typecase to please the compiler gods 2003-08-29 08:43:21 +00:00
Daniel Stenberg
3938e0f58b typecast to prevent compiler warning 2003-08-29 08:43:06 +00:00
Daniel Stenberg
fa162e8feb CURLOPT_BUFFERSIZE must not be smaller than 0 (zero) as that is impossible 
for us to deal with
 2003-08-28 11:28:55 +00:00
Daniel Stenberg
8feb5110b5 only do chmod if there's a file present 2003-08-28 10:39:36 +00:00
Daniel Stenberg
32899f8309 added the strcasecmp() proto here (moved from setup.h), as this is the 
only file within libcurl to use that function
 2003-08-24 14:29:06 +00:00
Daniel Stenberg
dc7c6d1030 removed the strcasecmp() proto from here, and moved it to the strequal.c 
file instead
 2003-08-24 14:28:15 +00:00
Daniel Stenberg
4459d3b434 have cwd_and_mkd prefixed with ftp_ to make it appear as a ftp-only 
function
 2003-08-22 12:35:18 +00:00
Daniel Stenberg
7e7d89471a use proper type to prevent compiler warning 2003-08-22 12:25:47 +00:00
Daniel Stenberg
e67d15b50e Vincent Sanders provided a fix for name resolving when linked with uClibc. 2003-08-21 12:07:47 +00:00
Daniel Stenberg
5974309d10 CURLINFO_RESPONSE_CODE replaces CURLINFO_HTTP_CODE 2003-08-20 15:42:24 +00:00
Daniel Stenberg
d00e4a377b store the FTP response code in the httpcode variable 2003-08-20 15:41:45 +00:00
Daniel Stenberg
afc6ade2b9 removed the goto and re-indented slightly 2003-08-20 15:40:21 +00:00
Daniel Stenberg
734e8d6229 transfer fix for multi interface 2003-08-20 15:39:49 +00:00
Daniel Stenberg
61629d2c86 made curl_multi_info_read() set 'msgs_in_queue' to 0 even when it returns 
NULL!
 2003-08-20 13:49:46 +00:00
Daniel Stenberg
b1da98170f make sure the 'done' variable is always set to something in the 
Curl_is_resolved() function
 2003-08-19 23:23:45 +00:00
Daniel Stenberg
fb3eee5f0b Respect HAVE_LONGLONG to support 'long long' 2003-08-19 15:37:07 +00:00
Daniel Stenberg
a8c78cbbb0 CRYPTO_cleanup_all_ex_data() is not present in all OpenSSL versions so 
we need to check for its presence in the configure script
 2003-08-19 09:56:16 +00:00
Daniel Stenberg
33cf204e9a don't set done==TRUE if the host name doesn't resolve 2003-08-19 09:29:59 +00:00
Daniel Stenberg
dafc652f63 Loren Kirkby pointed out that we need to call CRYPTO_cleanup_all_ex_data() 
when we cleanup the SSL stuff to not leak any memory.

I wish this was documented anywhere.
 2003-08-19 07:51:09 +00:00
Daniel Stenberg
394faaeb97 setting WRITEFUNCTION or READFUNCTION to NULL will now reset the callback 
pointers to the internal default functions
 2003-08-17 13:32:37 +00:00
Daniel Stenberg
a10581d459 Possible code for large file support, added within #if 0 so far. 2003-08-14 22:42:18 +00:00
Daniel Stenberg
acbcd68d89 Curl_SSL_InitSessions can return error, so check the return code and bail 
out if necessary
 2003-08-14 15:06:36 +00:00
Daniel Stenberg
4281470fca Curl_llist_destroy() checks the input for non-NULL 2003-08-14 15:06:08 +00:00
Daniel Stenberg
68a4aa6773 new proto for Curl_hash_init 2003-08-14 15:05:25 +00:00
Daniel Stenberg
905b160097 1. check allocs 
2. don't leave allocated memory behind when returning error
 2003-08-14 15:05:13 +00:00
Daniel Stenberg
52596c339b return failure when the host cache creation fails 2003-08-14 15:02:25 +00:00
Daniel Stenberg
3454319c17 prevent memory leak when going out of memory 2003-08-14 14:20:03 +00:00
Daniel Stenberg
02c78ecf81 allow out-of-memory testing by setting a limit. That number of memory 
allocation calls will succeed, the following will return NULL!
 2003-08-14 14:19:36 +00:00
Daniel Stenberg
a3c1248214 Bugfix from Serge Semashko that fixes a bug introduced when we applied his 
NTLM patch. Test case 84 and 85 verify this.
 2003-08-12 08:20:16 +00:00
Daniel Stenberg
27619fc450 Added support for CURLINFO_HTTP_CONNECTCODE 2003-08-11 23:15:41 +00:00
Daniel Stenberg
96fecba190 bindlocal works for Windows! 2003-08-11 23:15:13 +00:00
Daniel Stenberg
3eb4ae031c Set the CURL_VERSION_ASYNCHDNS bit if USE_ARES is defined. 2003-08-11 23:13:09 +00:00
Daniel Stenberg
c7be232fee added include "http.h" to prevent a warning 2003-08-11 14:55:30 +00:00
Daniel Stenberg
84ed5e755a use safefree instead 2003-08-11 12:30:21 +00:00
Daniel Stenberg
acfa131c8c memory leak fixed when re-using connections with proxy user+passwd 2003-08-11 12:25:30 +00:00
Daniel Stenberg
7a19923afa Serge Semashko added CURLOPT_PROXYAUTH support, and now NTLM for proxies 
work.
 2003-08-11 11:47:45 +00:00
Daniel Stenberg
96e217b496 the new cookie functions that require 'data' passed in 2003-08-11 09:56:06 +00:00
Daniel Stenberg
2dd1518d63 support sending off cookies without contents 2003-08-11 09:55:48 +00:00
Daniel Stenberg
168703b7bf Added some infof() calls, that require the data pointer so now several 
cookie functions need that.

I also fixed the cookie loader to properly load and deal with cookies without
contents (or rather with a blank content).
 2003-08-11 09:55:11 +00:00
Daniel Stenberg
8dd069604c Dirk Manske's bugfix for the share stuff 2003-08-11 06:30:02 +00:00
Daniel Stenberg
5dadbd094e don't claim the PASV connect is connected unless it *really* is! 2003-08-10 17:11:41 +00:00
Daniel Stenberg
514a8739b6 make sure the string is long enough 2003-08-08 17:56:47 +00:00
Daniel Stenberg
9273096a8a David Byron's fix for file:// URLs with drive letters included. 2003-08-08 17:12:04 +00:00
Daniel Stenberg
686c6133f8 chmod the cabundle file before we attempt to write to it, to make 
make distcheck run fine
 2003-08-08 11:13:18 +00:00
Daniel Stenberg
7954eee639 re-arranged the cwd/mkd stuff a bit 2003-08-08 09:55:16 +00:00
Daniel Stenberg
f9f1f0e316 Early Ehlinger's CURLOPT_FTP_CREATE_MISSING_DIRS patch was applied 2003-08-08 09:13:19 +00:00
Daniel Stenberg
6af73f417a use 644 for the chmod 2003-08-07 06:43:11 +00:00
Daniel Stenberg
32468a0072 argh, it wasn't *that* easy to generate the ca-bundle header in the build 
dir instead of the source dir, reverting that change
 2003-08-06 23:59:15 +00:00
Daniel Stenberg
bbe23945e4 fix the treatment of the variable width specifier '*', which caused a bug 
in the urlglobbing just now, fixed in the debian bug tracker as Bug#203827
 2003-08-06 22:32:47 +00:00
Daniel Stenberg
77ba0d3686 generate the ca-bundle.h in the build dir, and also make sure to chmod 
the file in the dist-hook to make distcheck run fine
 2003-08-06 22:14:16 +00:00
Daniel Stenberg
c704d1545c include "share.h" for the cookie sharing 2003-08-06 15:26:24 +00:00
Daniel Stenberg
62b65a5f20 make it build without ares support 
make sure it set async false even when using ipv6 (made test case 20 fail
before)
 2003-08-06 15:26:02 +00:00
Daniel Stenberg
256b9f31e1 more fix 2003-08-06 13:22:29 +00:00
Daniel Stenberg
a3037e1173 updated the ares instruction 2003-08-06 13:21:19 +00:00
Daniel Stenberg
f3e7a5d755 LDFLAGS fix to make the GSSAPI build again 2003-08-05 15:22:15 +00:00
Daniel Stenberg
5f0cba7775 added README.ares 2003-08-05 14:54:15 +00:00
Daniel Stenberg
673759fe7e how to build with ares 2003-08-05 14:52:31 +00:00
Daniel Stenberg
b73612392d ares awareness/usage/support added. If configure --enable-ares is used, we 
build libcurl to use ares for asynch name resolves.
 2003-08-05 14:40:59 +00:00
Daniel Stenberg
1e7e53c87e clean up the dir tree hierarchy in *_done() to make persistant connection 
FTP use the correct directories!

Reported in bug report #783116
 2003-08-05 13:04:10 +00:00
Daniel Stenberg
98ee12bc35 Jan Sundin reported a case where curl ignored a cookie that browsers don't, 
which turned up to be due to the number of dots in the 'domain'. I've now
  made curl follow the the original netscape cookie spec less strict on that
  part.
 2003-08-04 23:05:57 +00:00
Daniel Stenberg
41ae97e710 Dirk Manske's patch that introduces cookie support to the share interface. 2003-08-04 15:02:42 +00:00
Daniel Stenberg
f72ba7f79d Mark Fletcher provided an excellent bug report that identified a problem 
with FOLLOWLOCATION and chunked transfer-encoding, as libcurl would not
 properly ignore the body contents of 3XX response that included the
 Location: header.
 2003-08-03 22:18:14 +00:00
Daniel Stenberg
db9f87f697 When proxy authentication is used in a CONNECT request (as used for all SSL 
connects and otherwise enforced tunnel-thru-proxy requests), the same
authentication header is also wrongly sent to the remote host.

The name and password can then be captured by an evil host and possibly get
used for malicious purposes.
 2003-08-02 23:36:35 +00:00
Daniel Stenberg
d3e512c738 Jörg Müller-Tolk updated this to build fine with 7.10.6 2003-07-30 14:19:44 +00:00
Daniel Stenberg
e3b4dd08ff Daniel Noguerol made the ftp code output "Accept-Ranges: bytes" in similar 
style like other faked HTTP headers when NOBODY and HEADER are used.
 2003-07-30 07:51:33 +00:00
Daniel Stenberg
1c35cbcc07 Reverted the 'filetime' struct field back to a 'long' as time_t is sometimes 
unsigned and we want this to be able to hold -1 for illegal/unset values.
 2003-07-30 07:22:28 +00:00
Daniel Stenberg
48a580e609 clear http->send_buffer when we have freed the memory it pointed to 2003-07-28 10:21:57 +00:00
Daniel Stenberg
d9246ff24d François Pons brought a patch that once again made curl deal with ftp and 
"double slash" as indicating the root directory. In the RFC1738-fix of April
30, that ability was removed (since it is not the "right" way).
 2003-07-28 08:50:02 +00:00
Daniel Stenberg
428f41bd12 having it in CVS causes us problems *grrr* 2003-07-25 09:46:07 +00:00
Daniel Stenberg
99c32e460f Andrés García updated with the added files etc 2003-07-25 08:59:55 +00:00
Daniel Stenberg
83f249cf65 With an unknown CA path, we undef the variable. To build properly without 
SSL/CA.
 2003-07-25 08:47:34 +00:00
Daniel Stenberg
f0278ca114 Removed #include <sys/resource.h>, as pointed out by Henry Bland we don't 
need it.
 2003-07-25 08:30:58 +00:00
Daniel Stenberg
e9f63bf4e8 When we re-use an existing connection we must make sure that we don't 
accidentally re-use the connect_addr field, as that might no longer be
around. Fix verified by Tracy Boehrer who basicly debugged and tracked down
this problem.
 2003-07-23 17:06:21 +00:00