summaryrefslogtreecommitdiff
path: root/sapi
Commit message (Collapse)AuthorAgeFilesLines
* Support compiling and linking against Darwin / Mac OS X frameworksMarko Karppinen2002-08-062-2/+2
|
* Fix segfault due to unchecked null pointerSascha Schumann2002-07-261-35/+37
|
* - Fixed bug: #18560, the macro was not defined always.foobar2002-07-251-11/+0
|
* Removed some outdated lines from init_request_info().Holger Zimmermann2002-07-231-3/+0
|
* Made PHP_ADD_LIBPATH use the $ld_runpath_switch and changed the DarwinMarko Karppinen2002-07-212-2/+2
| | | | | | link lines in sapi/cli and sapi/cgi to use $NATIVE_RPATHS. This is a tentative fix for #18450.
* refineSascha Schumann2002-07-161-0/+3
|
* integrate the public keep-alive patchSascha Schumann2002-07-142-45/+638
| | | | | | | | | the patch did not handle pipeling at all, so that some code had to be added from Premium thttpd persistent connections are supported, if a script sets the Content-Length header
* thttpd blocks in rare circumstances where it cannot write out a small HTTPSascha Schumann2002-07-142-70/+107
| | | | | | | | | response (happened at customer sites). The response is now written out using the standard state machine. the buffer which is handed to thttpd by php is now simply dealt with as if it were a thttpd generated response (avoids code duplication).
* make the sapi module hand off a buffer to thttpd for final data delivery,Sascha Schumann2002-07-142-77/+524
| | | | | instead of blocking the whole process
* Needed this to be able to test safe-mode stuff on cmd-linefoobar2002-07-131-0/+1
|
* Break up the INSTALL_IT line into comprehensible piecesSascha Schumann2002-07-111-1/+9
| | | | | and fix a bug related to [..] quotes.
* the variable interpolation was already subject to IFS treatment,Sascha Schumann2002-07-111-1/+2
| | | | | | | which resulted in the execution of the foremost directory. we capture the output before setting IFS now
* Use a simple IFS; set step to interprete the httpd -v outputSascha Schumann2002-07-112-2/+12
|
* - Fix the version check again.foobar2002-07-072-3/+2
|
* Moved the cgi link line to sapi/cgi/config.m4 and stoppedMarko Karppinen2002-07-041-0/+10
| | | | | using libtool on Darwin on it
* Generate content-type header, if the user did not specify one of its own.Sascha Schumann2002-07-041-0/+10
| | | | | Noticed by: Edin Kadribasic <edink@proventum.net>
* Add CRLF after complete HTTP reply headerSascha Schumann2002-07-031-0/+1
|
* Use "Status: %d" instead of "HTTP/1.0 %d X" in the CGI SAPI moduleSascha Schumann2002-07-031-6/+18
| | | | | | | | | | | as per the CGI RFC. The status notation is incompatible with Apache's non-parsed-header mode "nph-*", but that has never worked before anyway. One could make the format string configurable. Noticed by: Sebastian Bergmann
* Add sapi_header_op interface which supersedes the sapi_add_header and _exSascha Schumann2002-07-033-24/+14
| | | | | | | | | | | | | | | | | | | | | calls. Revert the change to the sapi_add_header_ex interface. Fix various bugs: 1. header("HTTP/1.0 306 foo"); header("Location: absolute-uri"); did not work in combination with several SAPI modules, because http_status_line was never properly reset. And thus, all SAPI modules which looked at http_status_line ignored the changed http_response_code. 2. The CGI SAPI did not send out the HTTP status line at all, if http_status_line had not been set explicitly by calling header("HTTP/1.0 200 foo");
* - You don't need libtool to build sapi/cli on Darwin.Marko Karppinen2002-07-011-0/+9
| | | | | | | | | | | - We don't need libtool to build sapi/cli on Darwin. - You want the sapi/cli build line to be in sapi/cli, not Makefile.global. - We want the sapi/cli build line to be in sapi/cli, not Makefile.global. - He can go about his business. - You can go about your business. - Move along. - Move along. Move along.
* Better fix it here too..foobar2002-06-291-1/+1
|
* Oops, extra pipe snuck inRasmus Lerdorf2002-06-291-1/+1
|
* Fix for bug #18055Rasmus Lerdorf2002-06-281-1/+2
|
* Mention Apache 2.0.40.Sebastian Bergmann2002-06-281-2/+1
|
* Require 2.0.40foobar2002-06-281-2/+2
|
* - YAAC: I hope this is the last one!Derick Rethans2002-06-281-2/+2
|
* Fixed bug: #17491 (honor INSTALL_ROOT)foobar2002-06-281-3/+5
|
* Fix code which makes wrong assumptions about the return value of snprintf.Sascha Schumann2002-06-261-2/+2
| | | | | The AOLserver module did not use the return value, so simply drop it.
* @ Add runtime Apache2 thread check to ensure we don't run a non-threadedRasmus Lerdorf2002-06-241-0/+10
| | | | | @ PHP inside a threaded Apache2 MPM. (Rasmus)
* @ Turn off ZTS if Apache2 is using the prefork MPM. (Rasmus)Rasmus Lerdorf2002-06-242-2/+8
|
* Made php_admin_value work outside <location> and <directory> blocks.Sander Roobol2002-06-221-2/+2
|
* - Revert WS changesAndi Gutmans2002-06-221-2/+0
|
* - This shouldn't have snuck in.Andi Gutmans2002-06-221-1/+0
|
* - Stop using persist_alloc().Andi Gutmans2002-06-221-6/+20
|
* Fixing the same possible memory leak.Stefan Esser2002-06-213-1/+6
|
* - Added a new parameter to the header() function which overrides the HTTPDerick Rethans2002-06-212-2/+2
| | | | | | | response code. @- Added a new parameter to the header() function which overrides the HTTP @ response code. (Derick)
* @- Fixed bug with Apache which let PHP_AUTH_* variables to be set whenfoobar2002-06-131-1/+1
| | | | | | | @ external basic auth mechanism was used. (Jani) Fixes bugs: #16653, #14534, #14370
* Fix exit code (bug #17733).Edin Kadribasic2002-06-131-1/+1
|
* Disable installing pear when --disable-cli is used since pear installerEdin Kadribasic2002-06-111-1/+2
| | | | | needs cli to function.
* too many flushes are badSascha Schumann2002-06-081-1/+3
|
* Fix buildSascha Schumann2002-06-011-1/+1
|
* - Reapply netware patchAndi Gutmans2002-06-011-1/+1
|
* Put Netware header and comment back in. My CVS revision id got messedRasmus Lerdorf2002-06-011-0/+8
| | | | | up somehow.
* OopsRasmus Lerdorf2002-05-311-1/+1
|
* @ Renamed getallheaders() to apache_request_headers() and keptRasmus Lerdorf2002-05-311-14/+33
| | | | | | | | | @ getallheaders() as an alias to it. Also added apache_response_headers() @ which returns the current response headers from Apache. Renamed getallheaders() to apache_request_headers() and kept getallheaders() as an alias to it. Also added apache_response_headers() which returns the current response headers from Apache.
* Removed ugly code done as part of NetWare change, upon Andi's advice. Now, ↵Venkat Raghavan S2002-05-311-5/+1
| | | | the typecasting is done for all platforms.
* NetWare related changesVenkat Raghavan S2002-05-313-1/+22
|
* That macro is and always was hopelessly broken, which is why it'sCliff Woolley2002-05-301-1/+1
| | | | | now deprecated.
* Merge from cli.Edin Kadribasic2002-05-261-5/+2
|
* Due to the way Zend handles exit() we cannot rely on the return codeEdin Kadribasic2002-05-261-5/+2
| | | | | of php_execute_script.
View Lorry Controller Status