Commit graph

565 commits

Author SHA1 Message Date
Rasmus Lerdorf
76a2d2538c Heads up! I have moved the headers_only and response_code checks out of
SAPI and down into the individual SAPI modules.  I have made the
appropriate changes in all the SAPI modules, but please verify these.
The reason for this change is that Apache sometimes will feed PHP
a request_method of GET but have r->header_only set to true.  This happens
in an ErrorDocument redirect.  In this same scenario we want to preserve
the status code as well instead of just overwriting it with a 200 and
losing this information.  For now the other sapi modules act exactly as
before since they probably do not make this distinction, and they may
not even have a valid response code this early in the request.
@ Fix HEAD request bug on an Apache ErrorDocument redirect and preserve
@ the status code across the redirect as well.  (Rasmus)
2000-08-02 22:48:45 +00:00
Zeev Suraski
52ff887db5 Made ob_start() and friends reentrant. It's now possible to implement this
long-requested functionality, now that output buffering is re-entrant:

function eval_ret($code)
{
	ob_start();
	eval($code);
	$retval = ob_get_contents();
	ob_end_clean();
	return $retval;
}
2000-07-29 14:46:09 +00:00
Hartmut Holzgraefe
1a7f851985 added '-l' option to command line / CGI version for 'syntax check only'
plus a little cleanup and rearranging in command line option parsing

@ CGI aka. command line version has now an option '-l' for syntax check
@   without execution (Hartmut)
2000-07-28 15:21:20 +00:00
Stanislav Malyshev
4c6ffdb1c2 make -c in stand-alone mode work again.
Setting ini_path after php_module_startup doesn't do much good -
since php_module_startup reads .ini.
# This fix is very ugly. Everyone is welcome to make better fix
# that won't report errors twice and won't scan argument 3 times
2000-07-13 09:17:39 +00:00
Andi Gutmans
85422c3c60 - Extra newline crept in somehow 2000-06-29 13:34:58 +00:00
Andi Gutmans
1f828a6e88 - Fix problem with DICARD_PATH being defined in windows 2000-06-29 13:31:52 +00:00
Andrei Zmievski
4010b422dd Separate plain name returned by php_sapi_module() and pretty name
used for output.
2000-06-26 18:05:55 +00:00
Andi Gutmans
fd11a7e13e - Pray this is an OK fix for the DISCARD_PATH junk 2000-06-26 17:57:03 +00:00
Stanislav Malyshev
378440f7cf Right name for the function persist_alloc 2000-06-26 16:26:20 +00:00
Andi Gutmans
91ed009ae2 - Try two in fixing the problem with memory leaks reporting path_translated 2000-06-26 16:17:58 +00:00
Zeev Suraski
effa42696e - Fix problem with filename being freed before possible memory leaks are
- reported by the memory manager
2000-06-26 15:55:59 +00:00
Stanislav Malyshev
5be54abf24 If we freed it, we might not touch it
# BTW, this is not a real fix - we still might use it
# when reporting memory leaks, but at least we won't reference
# unitialized memory
2000-06-26 12:26:40 +00:00
Andi Gutmans
17ce56825d - Beautify a bit 2000-06-15 22:46:17 +00:00
Andi Gutmans
fcefeee914 - Try and fix crash with CGI.
- Please check out the comment I wrote in init_request_info(). There was a
- lot of code there which was running for no apparent reason. Should it be
- removed or should it actually run?
2000-06-15 22:40:52 +00:00
Andi Gutmans
4c99114a25 - Unncessary SLS_FETCH(); 2000-06-15 21:34:00 +00:00
Sascha Schumann
6caf633cf7 Proper casts 2000-06-13 09:17:06 +00:00
Andi Gutmans
77f4e1c8c7 - This seems to be a bug. Anyone know who wrote this code? Should it stay
- in?
2000-06-10 09:05:22 +00:00
Sascha Schumann
102c6ce4b8 Move main.h to php_main.h. 2000-06-05 23:24:42 +00:00
Andi Gutmans
8c82566b48 - Fix problem in sessions module under Win32. Need to use ; as an options
- delimiter in save_path instead of :.
2000-05-27 16:38:49 +00:00
Zeev Suraski
1895a09749 Only fetch globals if it's necessary (which it usually won't be) 2000-05-25 14:43:30 +00:00
Thies C. Arntzen
e319b5f6a5 @- ignore_user_abort() & friends should now work in CGI mode as well.
@  (Patch by daniel.braun@ercom.fr)
2000-05-25 07:01:34 +00:00
Hartmut Holzgraefe
22bfb799a7 -i, -v, -h and -? execution pathes didn't flush the output buffer
on exit, resultiong in no output at all when "output-buffering=On"
( BUG id#3420 )
2000-05-22 22:49:18 +00:00
Zeev Suraski
e043439ff6 Update the license with the new clause 6 2000-05-18 15:34:45 +00:00
Andi Gutmans
0d194ec80b - Fix problem with sapi_globals being used before it is initialized. 2000-05-05 07:21:48 +00:00
Andi Gutmans
198537b70f - Make Windows compile again. 2000-05-05 07:18:50 +00:00
Zeev Suraski
0b33ac7627 - Add -z and -g switches to the CGI 2000-05-04 13:42:21 +00:00
Andi Gutmans
9aba6a94d3 - Another PHP_WIN32 patch 2000-04-20 13:43:48 +00:00
Andi Gutmans
f9e9cab7ac - Change to PHP_WIN32 2000-04-20 13:23:19 +00:00
Thies C. Arntzen
0540bbe5af improve fix 2000-04-20 07:35:22 +00:00
Frank M. Kromann
429cf808e6 SIGPIPE is unknown on WIN32 platforms 2000-04-19 17:28:23 +00:00
Thies C. Arntzen
19a9577901 ignore SIGPIPE in standalone-mode so that one can actually use fsockopen() *and* have the remote side close the socket whithout PHP dying! 2000-04-19 15:37:43 +00:00
Sascha Schumann
bb09ee9505 Use sprintf everywhere and define sprintf to php_sprintf, if sprintf is broken. 2000-03-12 17:18:21 +00:00
Sascha Schumann
caeaaf4d98 Fix build on systems which lack snprintf or vsnprintf.
PR: #3786
2000-03-12 15:47:34 +00:00
Zeev Suraski
1261271839 Allow the INI callbacks to know at what stage PHP is 2000-02-26 15:36:23 +00:00
Thies C. Arntzen
b7e53b12f6 compile-fix 2000-02-21 14:01:19 +00:00
Zeev Suraski
86fa2aade2 the pipe is breaking all the time 2000-02-19 23:41:32 +00:00
Zeev Suraski
80bdd19e03 Fine tune Andi's patch 2000-02-11 15:59:30 +00:00
Andi Gutmans
f9e0d94a65 - Baby steps with PHP_WIN32 :) 2000-02-11 12:56:54 +00:00
Zeev Suraski
6723bdd920 request_info.c is giving up 2000-02-10 20:03:17 +00:00
Zeev Suraski
6d0ab6166c SAPIfication, Episode VI: Return of the SAPI
Remove mostly all references to APACHE and CGI_BINARY from the code.

- Apache include files are no longer included by any PHP code, except for the Apache SAPI module.
- No server specific code is in any of the base PHP code.

Still left to be done:
- Eliminate any references to APACHE from the few remaining modules.
- Move request_info.c's logic to SAPI
- Modify the regex function names, and globals, so that we can always
  include them, without having to fear any interference with Apache;
  Always use the bundled regex library
2000-02-10 19:41:21 +00:00
Zeev Suraski
5bdd18caf1 Yes, you guessed it... More cleanup 2000-02-10 18:44:33 +00:00
Zeev Suraski
0fd89aa20d More cleanup - move getenv() to SAPI 2000-02-10 17:55:01 +00:00
Zeev Suraski
7d926a0e0c More cleanup... 2000-02-10 17:26:57 +00:00
Andrei Zmievski
681f352a9d Should be 'deactivate' rather than the opposite. 2000-02-10 16:48:22 +00:00
Zeev Suraski
073b148167 More abstraction 2000-02-10 16:44:59 +00:00
Zeev Suraski
59b53ea2c8 Always use getopt with CGI, never use it for anything else 2000-02-10 16:27:28 +00:00
Zeev Suraski
404bbe1f26 Move the logging mechanism to SAPI 2000-02-10 15:55:10 +00:00
Zeev Suraski
b2449f929c - Change the argument order of php_register_variable() to something more
intuitive.
- Make the authentication variables be a part of the HTTP_SERVER_VARS[] array
2000-01-29 11:55:44 +00:00
Zeev Suraski
99f079a349 - A few fixes
- Added register_argv_argc directive to allow disabling of argv/argc
2000-01-28 18:29:37 +00:00
Zeev Suraski
9ab35ae393 Tried to centralize global variable registration as much as possible:
- Added $HTTP_ENV_VARS[] and $HTTP_SERVER_VARS[] support, which similarly
  to $HTTP_GET_VARS[], contain environment and server variables.  Setting
  register_globals to Off will now also prevent registration of the
  environment and server variables into the global scope (Zeev)
- Renamed gpc_globals to register_globals (Zeev)
- Introduced variables_order that deprecates gpc_order, and allows control
  over the server and environment variables, in addition to GET/POST/Cookies
  (Zeev)
2000-01-28 17:24:53 +00:00