mirror/curl
2
0
Fork 0
mirror of https://github.com/curl/curl.git synced 2024-12-27 06:59:43 +08:00
Code Issues Packages Projects Releases Wiki Activity
32,294 Commits 14 Branches 222 Tags 194 MiB
c5e322fb50
Commit Graph

338 Commits

Author SHA1 Message Date
Yang Tse
4594187732 Improve CURL_CHECK_DEF 2008-08-09 17:26:24 +00:00
Yang Tse
3cc40a2584 Add debug tracing for DO_CURL_OFF_T_CHECK 2008-08-08 17:42:58 +00:00
Yang Tse
c11933b3fd Remove some redundancy 2008-08-08 16:53:02 +00:00
Yang Tse
4687699726 Remove potential overquoting 2008-08-08 16:25:07 +00:00
Yang Tse
dabd3cd355 Add debug tracing for CURL_CHECK_DEF 2008-08-08 11:34:08 +00:00
Yang Tse
479466a495 Skip data type check in DO_CURL_OFF_T_CHECK macro when argument is empty. 2008-08-07 19:03:46 +00:00
Yang Tse
ddfa33be79 Fix CURL_CHECK_DEF so that when the expansion of the preprocessor symbol results 
in a set of double-quoted strings, this macro will now return an expansion which
consists of a single double-quoted string result of concatenating all of them.
 2008-08-07 16:07:26 +00:00
Yang Tse
14240e9e10 Initial support of curlbuild.h and curlrules.h which allows 
to have a curl_off_t data type no longer gated to off_t.
 2008-08-07 00:29:08 +00:00
Yang Tse
b4fdccf87a Autoconf 2.62 has changed the behaviour of the AC_AIX macro which we use. 
Prior versions of autoconf defined _ALL_SOURCE if _AIX was defined. But,
autoconf 2.62 version of AC_AIX defines _ALL_SOURCE along with other four
preprocessor symbols no matter if the system is AIX or not. To keep the
traditional behaviour, as well as an uniform one, across autoconf versions
AC_AIX is replaced with our own internal macro.
 2008-08-04 06:48:11 +00:00
Yang Tse
9cef14dfb2 Add a whitespace 2008-08-01 06:21:34 +00:00
Yang Tse
2d15d84a01 Ensure that reserved keyword AC_DEFUN is only used for its purpose. 2008-08-01 06:07:43 +00:00
Yang Tse
1035469662 Removed definition of CURL_CHECK_WORKING_RESOLVER from acinclude.m4 it has 
not been in use since revision 1.81 of configure.in 6 years, 9 months ago.
 2008-08-01 05:24:19 +00:00
Yang Tse
a3498f96ef Reinstate the 'aclocal -I m4' in buildconf and 'ACLOCAL_AMFLAGS = -I m4' way of 
including our local m4/reentrant.m4 file. This even takes care of including the
file in the distribution tarball.
 2008-07-30 03:10:03 +00:00
Yang Tse
52d9a3c34f Tests done using 'aclocal -I m4' in buildconf and 'ACLOCAL_AMFLAGS = -I m4 
in top Makefile.am triggered a problem that prevented aclocal from running
successfully on SunOS 5.10 with GNU m4 1.4.5 and GNU Autoconf 2.61

A tarball which reproduces mentioned problem is the one dated July-28-2008
http://cool.haxx.se/curl-daily/curl-7.19.0-20080728.tar.gz

We actually don't need all the bells and whistles that the above mechanism
provides. We only need to include our m4/reentrant.m4 file in acinclude.m4
so here we go with this simpler mechanism.
 2008-07-29 02:26:21 +00:00
Yang Tse
9b0110f50c Assert that SED and GREP are set 2008-07-27 21:47:22 +00:00
Yang Tse
b377e857bd simplify multi '#' char comment line 2008-07-26 12:11:22 +00:00
Yang Tse
8a323f8354 Another step towards detecting if _REENTRANT is already defined or actually 
needed, and being able to define it if appropriate for further configure tests
as well as for the generated config file.

Introduced reentrant.m4 intended for our reentrant related autotools/m4 macros.
 2008-07-24 15:20:35 +00:00
Yang Tse
e933b28aaf cleanup duplicate line 2008-07-22 19:13:35 +00:00
Yang Tse
1bc490077a Make sure that configure process tests are done with the same _REENTRANT 
setting as the one actually used when finally building the library.
 2008-07-22 19:04:16 +00:00
Yang Tse
7d773abe50 checks for gethostbyaddr_r with 7 and 8 args now also done with -D_REENTRANT 
checks for gethostbyname_r with 5 and 6 args now also done with -D_REENTRANT
 2008-07-22 18:56:46 +00:00
Yang Tse
bffe69a151 when recvfrom prototype uses a void pointer for arguments 2, 5 or 6 this will 
now cause the definition of RECVFROM_TYPE_ARG2_IS_VOID, RECVFROM_TYPE_ARG5_IS_VOID
or RECVFROM_TYPE_ARG6_IS_VOID, as appropriate.
 2008-07-21 00:36:55 +00:00
Yang Tse
facc5f7ec5 Remove showing additional info needed to debug configure 
failure to properly detect recvfrom arg types on Solaris
 2008-07-20 09:51:52 +00:00
Yang Tse
184229b1e5 Added "pointer to void" as another data type to check for the sixth argument of 
function recvfrom as a result of the info additionally logged when running on a
Solaris system.

The compiler error showed that the prototype being used on Solaris was the one
declared in line 427 of "/usr/include/sys/socket.h" as:

function(int,
         pointer to void,
         unsigned int,
         int,
         pointer to struct sockaddr,
         pointer to void) returning int
 2008-07-20 09:46:52 +00:00
Yang Tse
3e488c4f19 Temporarily show additional info needed to debug configure 
failure to properly detect recvfrom arg types on Solaris
 2008-07-19 18:32:27 +00:00
Yang Tse
39f23aec6b RECVFROM_TYPE_ARG2, RECVFROM_TYPE_ARG5 and RECVFROM_TYPE_ARG6 are now defined 
to the data type pointed by its respective argument and not the pointer type.
 2008-07-17 03:07:54 +00:00
Yang Tse
a9dc900515 Configure process now checks availability of recvfrom() socket function and 
finds out its return type and the types of its arguments. Added definitions
for non-configure systems config files, and introduced macro sreadfrom which
will be used on udp sockets as a recvfrom() wrapper.
 2008-07-16 19:16:41 +00:00
Yang Tse
97333deb3f fallback to gettimeofday when monotonic clock is unavailable at run-time 2008-07-02 03:04:56 +00:00
Yang Tse
72b4b3c7f8 make check message wording more precise 2008-06-30 14:10:27 +00:00
Yang Tse
e6c23672b2 fix C style comment 2008-06-29 11:08:52 +00:00
Yang Tse
1058e5fdde John Lightsey filed bug report #1999181: "CLOCK_MONOTONIC always fails on 
some systems" (http://curl.haxx.se/bug/view.cgi?id=1999181). The problem was
that the configure script did not use the _POSIX_MONOTONIC_CLOCK feature test
macro when checking monotonic clock availability. This is now fixed and the
monotonic clock will not be used unless the feature test macro is defined
with a value greater than zero indicating always supported.
 2008-06-29 03:19:20 +00:00
Yang Tse
2c61e9c76a Modified configuration script to actually verify if the compiler is good 
enough at detecting compilation errors or at least it has been properly
configured to do so. Configuration heavily depends on this capability, so
if this compiler sanity check fails the configuration process will now fail.
 2008-06-20 18:09:48 +00:00
Yang Tse
24d41452b0 No longer break out of a shell "for" statement from inside 
AC_FOO_IFELSE macros, otherwise temp files are not removed.

Identation adjustment.
 2008-06-18 04:39:29 +00:00
Dan Fandrich
c1a71ad14f Fixed curl-config --ca which wasn't being exported by configure. 2008-06-13 00:03:12 +00:00
Yang Tse
b39d1e9b9d MSVC does build Windows native targets 2008-06-06 18:44:39 +00:00
Yang Tse
abe2e6ecf7 fix: [action-if-found] part of AC_CHECK_TYPE macro cannot be quoted when empty 2008-05-21 18:24:09 +00:00
Yang Tse
d8efc99217 fix: remove need and definition of HAVE_SOCKLEN_T symbol 2008-05-21 14:04:14 +00:00
Yang Tse
0163b5b8be fix: socklen_t definition comment 2008-05-21 13:57:50 +00:00
Yang Tse
baee3996ab When unable to properly detect gethostbyname_r() usage, configure 
script will simply issue a warning and gethostbyname() will be used.
 2008-05-20 17:30:41 +00:00
Yang Tse
8ad1928d93 update several macros using AC_TRY_LINK with AC_LINK_IFELSE 2008-05-20 15:55:31 +00:00
Yang Tse
862049c490 fix underquoting of AC_LANG_PROGRAM arguments 2008-05-20 04:23:30 +00:00
Yang Tse
d70f33748c update several macros using AC_TRY_LINK with AC_LINK_IFELSE 2008-05-20 01:03:45 +00:00
Yang Tse
9346e55d5a better select() function detection that works 
even when cross compiling a Windows target.
 2008-05-19 16:57:56 +00:00
Yang Tse
8fc2f8ef62 fix socklen_t equivalent detection when cross compiling Windows target 2008-05-19 12:31:11 +00:00
Yang Tse
7358db5c27 minor change for wince-cegcc and wince-mingw32ce support 2008-05-17 01:20:56 +00:00
Yang Tse
2748c64d60 Improve toolchain detection for WinCE cross compilation: 
When cross compiling WinCE with the arm-wince-cegcc-gcc C compiler
symbol __CEGCC__ is defined and the unix-like compatibility layer
is used. For our purposes this is not a native Windows build.

When cross compiling WinCE with the arm-wince-mingw32ce-gcc C compiler
symbol __MINGW32CE__ is defined and the unix-like compatibility layer
is not used. For our purposes this _is_ a native Windows build.
 2008-05-14 19:42:17 +00:00
Yang Tse
5e06ec8409 skip checks for Windows specific header files 
when build target is not a native Windows one
 2008-05-14 16:14:35 +00:00
Yang Tse
04d0a84ae5 WinCE cross compilation adjustments: 
HAVE_WINSOCK2_H shall not be defined.
HAVE_WS2TCPIP_H shall not be defined.
 2008-05-14 13:54:36 +00:00
Yang Tse
ed80eb5b0f configure script will now define HAVE_CLOCK_GETTIME_MONOTONIC symbol only 
when function clock_gettime() is available and the monotonic timer is
also available. Otherwise, in some cases, librt or libposix4 could be used
for linking even when finally not using the clock_gettime() function due
to lack of the monotonic clock.
 2008-05-12 02:04:21 +00:00
Yang Tse
b380dd030f Add library checking for clock_gettime() support 2008-05-10 15:46:08 +00:00
Michal Marek
d13be06aaa Removed the generated ca-bundle.h file. The verbatim value of $ca and 
$capath is known to configure, so it can be defined in config.h instead.
 2008-03-31 12:09:43 +00:00
Michal Marek
86cbb23282 - Added --with-ca-path=DIRECTORY configure option to use an openSSL CApath by 
default instead of a ca bundle. The configure script will also look for a
  ca path if no ca bundle is found and no option given.

- Fixed detection of previously installed curl-ca-bundle.crt
 2008-03-20 08:09:23 +00:00
Daniel Stenberg
fb23b85770 - We're no longer providing a very old ca-bundle in the curl tarball. You can 
get a fresh one downloaded and created with 'make ca-bundle' or you can get
  one from here => http://curl.haxx.se/docs/caextract.html if you want a fresh
  new one extracted from Mozilla's recent list of ca certs.

  The configure option --with-ca-bundle now lets you specify what file to use
  as default ca bundle for your build. If not specified, the configure script
  will check a few known standard places for a global ca cert to use.
 2008-02-18 11:35:12 +00:00
Yang Tse
edef367e9c Missing double quotes 2007-10-24 16:40:59 +00:00
Yang Tse
e8d3710aff Add custom check for WINLDAP libraries. 
In CURL_CHECK_LIBS_WINLDAP and CURL_CHECK_LIBS_LDAP, check first with no
additional library even when the optional list of libraries has been given.
 2007-10-19 16:15:42 +00:00
Yang Tse
420ea83ef3 Fix CURL_CHECK_LIBS_LDAP failure when no parameter is given 2007-10-17 18:47:01 +00:00
Yang Tse
5360f88393 Default check for more libraries in CURL_CHECK_LIBS_LDAP, 
and allow parameter specification of libraries to check.
 2007-10-17 13:08:10 +00:00
Yang Tse
fbb5518ab6 Avoid depending on a header file for the definition of NULL 2007-10-15 23:58:11 +00:00
Dan Fandrich
001a2d9b67 Fix LDAP compile error when LDAP is not available. 
Fixed a typo in the LDAP configure code and made sure NULL is defined
in a test programs that need it.
 2007-10-15 16:24:46 +00:00
Yang Tse
95446f694b Fix custom check for LDAP libraries 2007-10-14 23:47:15 +00:00
Yang Tse
4db954f802 Add custom check for LDAP libraries 2007-10-14 21:25:10 +00:00
Yang Tse
a171f60bf7 Add custom checks for lber, ldap, ldapssl and ldap_ssl header files 2007-10-14 02:37:04 +00:00
Yang Tse
07625fe243 Add check for winldap and winber header files 2007-10-13 14:23:15 +00:00
Dan Fandrich
16b95fc773 Enabled a few more gcc warnings with --enable-debug. Renamed a few 
variables to avoid shadowing global declarations.
 2007-09-27 01:45:22 +00:00
Daniel Stenberg
d0edb47896 Patrick Monnerat modified the LDAP code and approach in curl. Starting now, 
the configure script checks for openldap and friends and we link with those
libs just like we link all other third party libraries, and we no longer
dlopen() those libraries. Our private header file lib/ldap.h was renamed to
lib/curl_ldap.h due to this. I set a tag in CVS (curl-7_17_0-preldapfix)
just before this commit, just in case.
 2007-08-11 20:57:54 +00:00
Daniel Stenberg
5b1bbffdff Peter O'Gorman pointed out (and fixed) that the non-blocking check in 
configure made libcurl use blocking sockets on AIX 4 and 5, while that
wasn't the intention.
 2007-07-30 22:53:18 +00:00
Yang Tse
d8d3dc9302 --enable-sspi only supported on Windows native builds 2007-04-10 18:53:21 +00:00
Yang Tse
305e03905f Cleanup. Warnings related with FD_SET, FD_ISSET, and FD_ZERO macros 
are not icc 9.0 specific.
 2007-04-03 02:45:04 +00:00
Dan Fandrich
59c620bfa5 Daniel Johnson's fix for shared object extension detection on Mac OS X. 2007-03-27 05:10:20 +00:00
Yang Tse
2288094b26 fix wrong macro name introduced in las commit 2007-03-22 18:59:14 +00:00
Yang Tse
273b2b230b Add check for compiler variadic macro support in configuration script 2007-03-22 18:25:38 +00:00
Yang Tse
8920606b8b attempt to keep message length below 80 chars 2007-03-22 14:41:10 +00:00
Yang Tse
d381dd68cf icc 9.0 when compiling its generated code for its own FD_SET, 
FD_ISSET, and FD_ZERO macros emits warnings #1469 and #593.
So for icc 9.0 we also ignore warnings #1469 and #593.
* 593 warns on "variable __d0 was set but never used"
* 1469 warns on "cc clobber ignored"
 2007-02-15 14:02:32 +00:00
Dan Fandrich
23d1041bd8 Added last-resort dynamic library names. 2007-02-13 19:59:58 +00:00
Dan Fandrich
5565f45f5e Properly use libtool macros to fix OpenLDAP library name detection on Darwin. 2007-01-31 23:15:47 +00:00
Yang Tse
73226415fc Added a check in configure that verifies if <signal.h> is available, 
defining HAVE_SIGNAL_H if the header is available.

Added a check in configure that tests if the sig_atomic_t type is
available, defining HAVE_SIG_ATOMIC_T if it is available. Providing
a suitable default in setup_once.h if not available.

Added a check in configure that tests if the sig_atomic_t type is
already defined as volatile, defining HAVE_SIG_ATOMIC_T_VOLATILE
if it is available and already defined as volatile.
 2006-11-22 18:41:34 +00:00
Daniel Stenberg
67e8d22958 Added a check in configure that simply tries to run a program (not when 
cross-compiling) in order to detect problems with run-time libraries that
otherwise would occur when the sizeof tests for curl_off_t would run and
thus be much more confusing to users. The check of course should run after
all lib-checks are done and before any other test is used that would run an
executable built for testing-purposes.
 2006-10-16 08:30:54 +00:00
Yang Tse
32ac4edeed Check for struct timeval at configuration time 2006-10-13 01:35:14 +00:00
Yang Tse
7240acdebc Being unable to link or find out recv() or send() args types is a fatal error. 2006-08-04 00:39:34 +00:00
Yang Tse
ae8a01ead6 Avoid the risk of a false positive detection of MSG_NOSIGNAL when cross compiling a Windows target. 2006-07-31 18:41:29 +00:00
Yang Tse
77b3bc239d First step trying to avoid the multiple header inclusion and recursion nightmare. 
Reintroduce checking for HAVE_MSG_NOSIGNAL in configure script, so that we don't depend on header inclusion order for a valid check.
 2006-07-28 14:19:02 +00:00
Yang Tse
13616f8f96 Simplify check for NEED_MALLOC_H, and make more explicit that NEED_MALLOC_H shall be defined if <malloc.h> header file must be included even when including <stdlib.h>. 2006-07-25 13:49:49 +00:00
Yang Tse
4d95d23d99 Define NEED_MALLOC_H if including <stdlib.h> is not enough for proper compilation and <malloc.h> must also be included. 2006-07-11 20:40:38 +00:00
Yang Tse
8ef454dcbe Finally get rid of CURL_CHECK_HEADERS_ONCE since it adds very little value and has portability issues. 
Change some shell if...then...fi tests into case...esac tests which demand less resources.
 2006-07-07 17:34:40 +00:00
Yang Tse
551a041283 Substitution of the literal '-' is only done if it's the first or last character. 2006-07-07 14:03:17 +00:00
Yang Tse
03288943af Using backslashes and slashes in the strings of the sed 'y' command shall be avoided since its interpretation is not the same across platforms. 
Now we use the sed 's' command with a bracket expression.
 2006-07-07 12:59:45 +00:00
Yang Tse
05edd48ad0 Fix excessive escaping. 2006-07-07 05:39:13 +00:00
Yang Tse
266ab95557 Fix CURL_CHECK_HEADERS_ONCE 2006-07-07 04:42:11 +00:00
Yang Tse
0a4bba565c Fix MinGW/MSYS support in CURL_CHECK_FUNC_RECV and CURL_CHECK_FUNC_SEND. 2006-07-06 15:51:03 +00:00
Yang Tse
bc2f0c7dcb Prevent definition of HAVE_WINxxx_H symbols and avoid inclusion of Windows headers when compiled with Cygwin in POSIX emulation mode. 2006-07-05 23:10:37 +00:00
Yang Tse
55329b56cb Use a more descriptive var name. 2006-07-04 17:19:15 +00:00
Yang Tse
7e43d06b60 Get qualifier of arg 2 for send() apart into SEND_QUAL_ARG2. 2006-07-04 16:54:10 +00:00
Yang Tse
01fa02d0b5 Find out return types and argument types for functions recv() and send() at configuration stage. 2006-07-04 13:03:32 +00:00
Yang Tse
bec1977137 Use CURL_CHECK_FUNC_GETNAMEINFO results in CURL_CHECK_NI_WITHSCOPEID 2006-07-03 15:32:12 +00:00
Yang Tse
4c08eb4b11 Make CURL_CHECK_NI_WITHSCOPEID actually try to compile NI_WITHSCOPEID when cross-compiling. 2006-07-02 23:09:46 +00:00
Yang Tse
0163730437 Fix shell globbing in CURL_CHECK_FUNC_GETNAMEINFO 2006-07-02 01:21:54 +00:00
Yang Tse
39745ac38e Fix shell globbing in CURL_CHECK_FUNC_GETNAMEINFO 2006-07-02 01:17:37 +00:00
Yang Tse
a1c6d5861a Get some debug info 2006-07-01 17:07:12 +00:00
Yang Tse
589c4596d8 Get some debug info 2006-07-01 15:01:48 +00:00
Yang Tse
7a6d7fca42 Avoid shell globbing 2006-07-01 12:53:24 +00:00
Yang Tse
5305c9f1e8 Get qualifier of arg 1 for getnameinfo apart. Take 3. 2006-07-01 11:21:38 +00:00
Yang Tse
b1022ea4c1 Get qualifier of arg 1 for getnameinfo apart. 2006-07-01 03:07:07 +00:00
Yang Tse
2e0ad842d0 Remove experimental notice from CURL_CHECK_FUNC_GETNAMEINFO 2006-06-30 00:22:38 +00:00
Daniel Stenberg
2df622fd14 detect ICC and pass on "-we 147" so that the configure checks for function 
arguments work properly - and the option is not harmful for the rest of the
curl build either!
 2006-04-19 11:11:10 +00:00
Daniel Stenberg
65afc576ea configure no longer warns on "missing" if the current path contains a space 2006-01-18 10:00:36 +00:00
Yang Tse
e4388643f1 Checking for function getnameinfo and its arguments is finally 
done in one single function CURL_CHECK_FUNC_GETNAMEINFO which
will only define HAVE_GETNAMEINFO if the function has been found
AND the type of its arguments has been properly been detected
 2005-12-21 20:44:54 +00:00
Yang Tse
bc4208201c Undefine HAVE_GETNAMEINFO if unable to find 
proper types to use for getnameinfo args
 2005-12-21 17:51:08 +00:00
Yang Tse
a0d69d52a1 Undefine HAVE_GETNAMEINFO if unable to find 
proper types to use for getnameinfo args
 2005-12-21 17:20:22 +00:00
Yang Tse
e3657644d6 In CURL_FUNC_GETNAMEINFO_ARGTYPES, when cross-compiling a windows 
target use calling convention WSAAPI for getnameinfo() prototype.
Checking type DWORD as argument 4 and 6 of getnameinfo not needed.
 2005-12-20 23:49:14 +00:00
Yang Tse
df2b1251a0 fix ioctlsocket detection 2005-12-20 20:29:53 +00:00
Daniel Stenberg
d551412a32 added our standard source header 2005-12-20 09:02:36 +00:00
Daniel Stenberg
6de67a134e fix closing parentheses 2005-12-20 08:51:48 +00:00
Yang Tse
fd0d560b47 Fix quoting 2005-12-20 03:23:49 +00:00
Yang Tse
7b488a3bef Use native type SOCKET instead of int when testing functionality of ioctlsocket on Windows 2005-12-20 00:27:59 +00:00
Yang Tse
f448168501 Ooops 2005-12-19 23:32:52 +00:00
Yang Tse
898bb397b1 Add checking for type DWORD as argument 4 and 6 of getnameinfo 2005-12-19 21:45:43 +00:00
Yang Tse
74ed5b5ebd Adjust more windows header includes 2005-12-19 21:38:10 +00:00
Yang Tse
d6eb1a7b98 Fix guard detection of _WIN32_WINNT for MingW in CURL_FUNC_GETNAMEINFO_ARGTYPES 2005-12-19 05:57:17 +00:00
Yang Tse
4ff56b15e9 Add check for 'unsigned int' as type of arguments 4 and 6 of getnameinfo 2005-12-19 05:32:33 +00:00
Yang Tse
5fd8dd2dce When checking the type of the first argument of getnameinfo do it in the 
following order: 'struct sockaddr *' 'const struct sockaddr *' 'void *'.
 2005-12-18 20:24:35 +00:00
Yang Tse
8a3280a2de MingW guards getnameinfo, getaddrinfo and freeaddrinfo with _WIN32_WINNT >= 0x0501 2005-12-18 06:07:10 +00:00
Yang Tse
939d368d5f Check first arg of getnameinfo with and without const qualifier. 2005-12-17 17:33:14 +00:00
Yang Tse
47c06fa308 Avoid breaking configure due to CURL_FUNC_GETNAMEINFO_ARGTYPES failure, since at this point nothing depends on it. 2005-12-17 02:41:33 +00:00
Yang Tse
b7f447f8d8 const qualifier in getnameinfo check 2005-12-17 02:32:55 +00:00
Yang Tse
03af76b631 Since there is no proof of the existence of a platform which would 
justify checking for socklen_t in more than one function, the code
used to find a valid socklen_t replacement is simplified back. The
only function that will be used to find a socklen_t replacement is
getpeername, as it has been since revision 1.4 of curl/acinclude.m4
 2005-12-16 23:15:33 +00:00
Yang Tse
d6ffb4c177 TYPE_SOCKLEN_T completely replaced by CURL_CHECK_TYPE_SOCKLEN_T. 
CURL_FUNC_GETNAMEINFO_ARGTYPES now also checks first argument.
All related changes taken to cares configuration scripts.
 2005-12-16 18:18:00 +00:00
Yang Tse
fea5ddf585 More quotes 2005-12-16 07:28:56 +00:00
Yang Tse
b9f39c2711 Fix copy paste bug 2005-12-16 06:50:03 +00:00
Yang Tse
5acac0309a Test CURL_CHECK_TYPE_SOCKLEN_T 2005-12-16 05:05:55 +00:00
Yang Tse
78febad718 Oops. Wrong double quotes 2005-12-16 04:54:56 +00:00
Yang Tse
a6a5bba0a9 Experimental check for socklen_t CURL_CHECK_TYPE_SOCKLEN_T 2005-12-16 04:18:44 +00:00
Yang Tse
51581c034d Tests to check the availability of compilable and 
valid windows.h winsock.h winsock2.h and ws2tcpip.h
header files:
CURL_CHECK_HEADER_WINDOWS
CURL_CHECK_HEADER_WINSOCK
CURL_CHECK_HEADER_WINSOCK2
CURL_CHECK_HEADER_WS2TCPIP
 2005-12-15 19:39:44 +00:00
Yang Tse
7f5d092223 Some preprocessors have problems if the # character isn't at position 1. 2005-12-14 21:09:30 +00:00
Yang Tse
2645782f89 Determine the correct type to be passed to four of the getnameinfo' function's arguments, and define those types in GETNAMEINFO_TYPE_ARG2', GETNAMEINFO_TYPE_ARG46', and GETNAMEINFO_TYPE_ARG7'. 2005-12-14 19:00:44 +00:00
Daniel Stenberg
be524fed38 on windows (mingw32) the sockaddr_storage struct is in winsock2.h 2005-11-12 19:11:02 +00:00
Daniel Stenberg
c7b7bf32f8 removed the unreachable code warning from gcc debug builds, even the most 
recent gcc versions give far too many false positives for this to be valuable
 2005-08-19 15:07:50 +00:00
Daniel Stenberg
ef35519696 removed some inaccurate comments about the TYPE_IN_ADDR_T check 2005-08-19 07:02:24 +00:00
Dan Fandrich
e81d4ac8c8 Add -Wdeclaration-after-statement to gcc to detect accidental C99-style 
variable declarations.
 2005-07-20 21:58:23 +00:00
Dan Fandrich
b37a0b0512 Detect (or at least infer) glibc-style strerror_r even when cross-compiling. 2005-06-30 04:53:05 +00:00
Daniel Stenberg
b9e5302f27 adjusted the strerror_r test more, use _REENTRANT instead of _THREAD_SAFE 
when looking for the prototype
 2005-05-18 15:15:00 +00:00
Daniel Stenberg
dd7b6aef39 The configure check for strerror_r() failed to detect the proper API at 
times, like on my HP-UX 10.20 tests. And then lib/strerror.c badly assumed
the glibc version if the posix define wasn't set (since it _had_ found a
strerror_r).
 2005-05-18 13:24:23 +00:00
Daniel Stenberg
07e58aaa79 Sort of "fixed" KNOWN_BUGS #4: curl now builds IPv6 enabled on AIX 4.3. At 
least it should no longer cause a compiler error. However, it does not have
AI_NUMERICHOST so we cannot getaddrinfo() any numerical addresses with it (we
use that for FTP PORT/EPRT)! So, I modified the configure check that checks if
the getaddrinfo() is working, to use AI_NUMERICHOST since then it'll fail on
AIX 4.3 and it will automatically build with IPv6 support disabled.
 2005-05-02 14:06:27 +00:00
Daniel Stenberg
a31ddd363b Now configure checks for struct sockaddr_storage and the ftp code tries 
to survive without it if not found. AIX 4.3 targetted adjustment.
 2005-05-02 11:56:15 +00:00
Daniel Stenberg
4e555c7b32 When cross-compiling, we do some better checking for the NI_WITHSCOPEID option instead of just assuming it is present. 2005-03-22 18:02:06 +00:00
Dan Fandrich
1f68fa19c7 Finally fixed the LDAP library searching bug on libtool ver. 1.5 2005-03-15 04:47:07 +00:00
Dan Fandrich
228e627475 Use the libtool variables better to make LDAP library search work on 
more platforms.
 2005-03-14 19:37:07 +00:00
Dan Fandrich
b5112096d7 Work around a bug in libtool ver. 1.5 during LDAP library detection. 2005-03-11 23:07:40 +00:00
Dan Fandrich
205f8b266c Fixed LDAP library file name bug (KNOWN_BUGS #1). configure now auto-detects 
the correct dynamic library names by default, and provides override switches
--with-ldap-lib, --with-lber-lib and --without-lber-lib.  Added
CURL_DISABLE_LDAP to platform-specific config files to disable LDAP
support on those platforms that probably don't have dynamic OpenLDAP
libraries available to avoid compile errors.
 2005-03-11 05:28:07 +00:00
Daniel Stenberg
58f4af7973 Marcin Konicki provided two configure fixes and a source fix to make curl 
build out-of-the-box on BeOS.
 2004-12-22 22:28:10 +00:00
Dan Fandrich
0e59182945 gcc 2.7 can't handle a few warning options that gcc 2.95 can. 2004-12-20 18:20:17 +00:00
Dan Fandrich
38181fbc74 Fix the --enable-debug compiler warning options for older versions of gcc. 2004-12-15 21:09:53 +00:00
Daniel Stenberg
18dd8154e2 Reverted the previous change and redid it differently as it seemed to not 
work. This is supposed to detect cross-compiling and alert the user, and not
do the POSIX-check for strerror_r() if it already detected a glibc-compatible
strerror_r().
 2004-06-07 08:30:36 +00:00
Daniel Stenberg
e0b436f8e1 1 - do better when cross-compiling when checking for strerror_r() - alert 
the user.
2 - don't check for POSIX-style if glibc-style is found first
 2004-06-04 19:01:58 +00:00
Daniel Stenberg
55c015c136 typo AGAIN 2004-05-06 11:02:50 +00:00
Daniel Stenberg
bde6c6685e When using the icc compiler, we also ignore remark #1418 "external definition 
with no prior declaration" since this is a habit we have in the code.
 2004-05-06 10:58:57 +00:00
Daniel Stenberg
6f74820cfc if no strerror_r prototype is found, we provide our own to prevent picky 
compilers to warn
 2004-05-06 10:49:40 +00:00
Daniel Stenberg
aae521d086 gcc 3.4 now uses the -Wunreachable-code option, I believe we can make older 
ones use this too...
 2004-05-05 06:59:27 +00:00
Daniel Stenberg
1fc7ff878e Applied Joe Halpin's bugfixes to the NI_WITHSCOPEID test program. 2004-04-01 06:53:11 +00:00
Daniel Stenberg
5d27f50f2f HAVE_NI_WITHSCOPEID spelled right! 2004-03-31 20:13:53 +00:00
Daniel Stenberg
40d9855df2 Andrs Garca fixed a warning in the ioctlsocket() usage. 2004-03-31 10:34:53 +00:00
Daniel Stenberg
ecf7adba15 modified the NI_WITHSCOPEID to use an AF_INET6 socket immediately and 
added some more debug output to make it easier to detect failure reasons
in the autobuild logs
 2004-03-31 10:31:08 +00:00
Daniel Stenberg
a2ea0abf7f Added CURL_CHECK_NI_WITHSCOPEID that checks if NI_WITHSCOPEID exists and 
works. No code actually uses the HAVE_NI_WITHSCOPEID (that a positive test
results in), but this is still only for testing purposes.
 2004-03-30 10:35:54 +00:00
Daniel Stenberg
7461592a16 strerror_r() detection changes: 
1. Try with _THREAD_SAFE instead of _REENTRANT, as AIX seems to require it
   and if _REENTRANT is required we should already have it set since one of
   the previous tests.
2. Added API-detection for what kind of strerror_r() that is provided. The
   POSIX style or the glibc style.

Tor Arntsen provided the necessary feedback these changes are based upon.
 2004-03-25 12:15:00 +00:00
Daniel Stenberg
242be55771 added check for strerror_r() 2004-03-24 08:45:58 +00:00
Daniel Stenberg
78e0ba82b0 ignore the icc warning 981 "operands are evaluated in unspecified order" 2004-03-10 09:52:04 +00:00
Daniel Stenberg
7e6043f04f we ignore the ICC warning 1419 as well 2004-03-08 11:37:12 +00:00
Daniel Stenberg
6789dfc5f5 Perhaps -Wundef is better on gcc versions after 2.95, since the autobuild 
on FreeBSD gives us lots of warnings in system headers and I suspect this
option is what causes them!
 2004-03-01 13:10:35 +00:00
Daniel Stenberg
193a652e6a When setting aggressive pedantic compiler options, display what options 
that were set. For easier debugging/changing of this.
 2004-03-01 13:02:34 +00:00
Daniel Stenberg
85af357d81 Make icc ignore "invalid format string conversion" warnings as well. They 
appear because of our home-grown option '%Od' for the curl_off_t output.
 2004-02-26 12:40:09 +00:00
Daniel Stenberg
0943f33438 Use __INTEL_COMPILER instead of __ICC to the cpp to detect the Intel icc 
compiler!
 2004-02-26 11:37:14 +00:00
Daniel Stenberg
6bd2de0c42 set debug options when using the icc compiler 2004-02-25 14:14:30 +00:00
Daniel Stenberg
4fa58560bf Moved most of the set-debug-options-depending-on-compiler logic to the new 
CURL_CC_DEBUG_OPTS function in acinclude.m4
 2004-02-25 10:19:02 +00:00
Daniel Stenberg
4c70c3e820 Added escaping of the function name in the AC_DEFUN() macros, so that 
automake 1.8+ stops complaining!
 2004-01-15 12:48:13 +00:00
Daniel Stenberg
5f4fb857f0 Added proper #ifdef's for the #include files when checking for in_addr_t, 
which made the test start working nicely on BeOS. Shard verified it.
 2003-10-30 07:08:45 +00:00
Daniel Stenberg
bf20ed5984 O_NONBLOCK does not work on BeOS according to Shard 2003-10-06 22:03:11 +00:00
Daniel Stenberg
0bbe184f1f improved the non-block check a lot: 
* several tests ALWAYS failed due to bad test programs
* the IoctlSocket() test now is linked on test as it could otherwise compile
  fine on many systems that doesn't support it
* added a test for the BeOS way of doing non-blocking sockets (based on a
  patch from Shard)
 2003-10-05 08:35:55 +00:00
Daniel Stenberg
de9b76cef0 change the order of the in_addr_t tests, so that 'unsigned long' is tested 
for first, as it seems to be what many systems use
 2003-05-15 21:13:36 +00:00
Daniel Stenberg
b9c60df04b added description in all AC_DEFINE() calls 2003-01-20 15:16:56 +00:00
Daniel Stenberg
325391aef9 Albert Chin: 
Forgot one case. On HP-UX 11.00, gethostbyname_r() is properly defined
if -D_REENTRANT is used. Without it, the compiler still accepts the
function prototype but gives a warning about hostent_data going out of
scope. This is because struct hostent_data is not declared. So, we
force an error by trying to set a variable to the struct.
 2002-01-31 07:53:20 +00:00
Daniel Stenberg
d220389647 Stoned Elipot's patch for the in_addr_t test 2002-01-14 07:53:09 +00:00
Daniel Stenberg
820de919b6 now sets a type for in_addr_t even if it isn't found in the #include files 
like on my linux box
 2001-11-28 23:14:20 +00:00
Daniel Stenberg
d09b436937 Added an in_addr_t check 2001-11-28 13:16:56 +00:00
Daniel Stenberg
6ebac3dc76 now we make sure that NULL is defined in the gethostbyname_r() compiles 
as it turned out they aren't everywhere, and that causes compiles to fail
and then we don't find the proper function call!
 2001-11-08 14:48:50 +00:00
Daniel Stenberg
c3363f833c Albert Chin's improved gethostbyname_r() tests 2001-10-04 14:05:12 +00:00
Daniel Stenberg
6ca45beaed Added non-blocking sockets test 2001-10-01 08:58:03 +00:00
Daniel Stenberg
a1cec0e49a Albert Chin's neat configure/package fixes 2001-08-20 07:10:28 +00:00
Daniel Stenberg
5c6859e537 not used in autoconf version 2.50 2001-05-29 19:28:50 +00:00
Daniel Stenberg
8755c44d40 Albert Chin's patch posted to the mailing list 19 Apr 2001 2001-04-19 06:01:48 +00:00
Daniel Stenberg
76229d3da1 largefile.m4 was turned into acinclude.m4 2000-08-24 19:16:26 +00:00