summaryrefslogtreecommitdiff
path: root/ext/com/COM.c
Commit message (Collapse)AuthorAgeFilesLines
* removed a possible NULL pointer referencing inside an error handlerHarald Radi2002-08-051-2/+11
|
* Unify error messagesDerick Rethans2002-06-261-25/+25
|
* removed CONST_EFREE_PERSISTENT so that andi can commit his patchHarald Radi2002-06-211-3/+2
| | | | | # _now_ it should work
* removed CONST_EFREE_PERSISTENT so that andi can commit his patchHarald Radi2002-06-211-4/+11
| | | | | # second try
* removed CONST_EFREE_PERSISTENT so that andi can commit his patchHarald Radi2002-06-211-5/+1
| | | | | | | # this is one of these nice patches where you only remove something :) # i still don't get what CONST_EFREE_PERSISTENT should have done # there but the cvs log says that i can blame zeev :-P
* #forgot somethingHarald Radi2002-05-301-4/+4
|
* @ Added missing AddRef() calls in the COM extension. This shouldHarald Radi2002-05-301-14/+18
| | | | | | | @ fix weird behaviour (in particular with ADODB). (Harald) # waah, this suxx
* Add documentation comment for properties in com_print_typeinfoWez Furlong2002-05-211-0/+8
|
* Enhance com_print_typeinfo.Wez Furlong2002-05-211-80/+133
| | | | | | | | | | | | The main expected use is like this, for figuring out what methods are allowed for a COM object: $ie = new COM("InternetExplorer.Application"); // Prints class definition for IE object com_print_typeinfo($ie, "InternetExplorer.Application", false); // Prints class definition for default IE event handler com_print_typeinfo($ie, "InternetExplorer.Application", true);
* Correct usage of convert_to_string_ex which is not allowed to zval*Wez Furlong2002-05-211-5/+4
|
* Reformat some comments.Wez Furlong2002-05-211-28/+22
|
* Fix a flag, remove an old comment.Wez Furlong2002-05-211-2/+1
|
* - Make sure that COM and VARIANT resources are returned as resourcesWez Furlong2002-05-211-3/+416
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | rather than longs. - Make the IDispatch implementation a bit more generic (and fix my mess of pointers). - Add new com_message_pump() function that acts like an interruptible usleep() that processes COM calls/events. - Add new com_print_typeinfo() function for "decompiling" the typeinfo for an interface into PHP script. This is useful for generating a skeleton for use as an event sink. - Add new com_event_sink() function for sinking events from COM objects. Usage is like this: <?php class IEEventSinker { var $terminated = false; function ProgressChange($progress, $progressmax) { echo "Download progress: $progress / $progressmax\n"; } function DocumentComplete(&$dom, $url) { echo "Document $url complete\n"; } function OnQuit() { echo "Quit!\n"; $this->terminated = true; } } $ie = new COM("InternetExplorer.Application"); $sink =& new IEEventSinker(); com_event_sink($ie, $sink, "DWebBrowserEvents2"); $ie->Visible = true; $ie->Navigate("http://www.php.net"); while(!$sink->terminated) { com_message_pump(4000); } $ie = null; ?>
* integrating wez's patchHarald Radi2002-05-201-15/+3
|
* Added generic COM wrapper for PHP objects.Wez Furlong2002-05-201-4/+35
|
* remove temporary resources immediatelyHarald Radi2002-05-021-17/+22
| | | | | | | return value fix #thanks to alan for 'remote debugging'
* RETURN_NULL() is defined with braces while RETURN_TRUEHarald Radi2002-04-291-11/+11
| | | | | | and RETURN_FALSE are defined without. seems not very consistent ?
* functions returned FALSE in case of error and the oo api returned NULL.Harald Radi2002-04-291-11/+11
| | | | | make them both behave equal (return NULL, as FALSE can be a valid value).
* don't set CLSCTX_REMOTE_SERVER if NULL is passed as servernameHarald Radi2002-04-271-9/+4
|
* Missing break; causes fallthrough which actually causes heap corruption in ↵Alan Brown2002-04-181-4/+8
| | | | the debugging version despite being just plain wrong. Also placed a default "Unavailable" message when the object does not populate the EXCEPINFO structure. Also removed a minor memory leak.
* @ fixed a bug that caused php to crash in php_COM_get_ids_of_names() ↵Harald Radi2002-03-151-0/+2
| | | | (Harald, Paul)
* Update headers.Sebastian Bergmann2001-12-111-2/+2
|
* proto fixesHartmut Holzgraefe2001-12-051-1/+1
|
* Fixed two protos.Egon Schmid2001-12-021-3/+3
|
* Not all components populate every field in the ExceptInfo structure. Thus we ↵Alan Brown2001-11-091-9/+21
| | | | sometimes would try to convert NULL strings and see php_OLECHAR_to_char errors while displaying Exception information. This version is a little smarter about the member derefencing and the resulting error string.
* * zend_module_entry change: apino, debug and zts are moved first,Stig Bakken2001-10-111-1/+18
| | | | | | see README.EXTENSIONS file for upgrade help. @Introduced extension version numbers (Stig)
* fixed Z_* conversion errorsHarald Radi2001-09-261-2/+2
|
* Fixing Win32 build...Frank M. Kromann2001-09-261-2/+2
|
* no messageHarald Radi2001-09-251-32/+139
|
* Back-substitute for Z_* macro's. If it breaks some extension (the script ↵Jeroen van Wolffelaar2001-09-251-6/+6
| | | | isn't optimal, it parses for example var->zval.value incorrect) please let me know.
* removed VARIANT module and put the VARIANT class intoHarald Radi2001-09-241-7/+28
| | | | | | the COM module. also fixed a few bugs.
* IDispatchs got released if they were passed to another componentHarald Radi2001-09-101-1/+11
|
* com_*() functions returned an oo-resource instead of an resource id,Harald Radi2001-09-081-7/+3
| | | | | thus subsequent com_*() calls to com-returnvalues failed.
* Fixed some protos.Egon Schmid2001-09-041-2/+3
|
* hum, wasn't $foo->bar 's property type OE_IS_METHOD ?Harald Radi2001-08-201-0/+6
|
* removed CoInitialize() and CoUninitialize()Harald Radi2001-08-151-7/+7
|
* fixed com_invoke retvalHarald Radi2001-08-141-2/+2
|
* - Avoid using malloc()Zeev Suraski2001-08-141-18/+24
| | | | | - Improve and fix leaks in the typelib constants registration mechanism
* fixed bug in com_load_typelibHarald Radi2001-08-141-3/+5
|
* Whitespace and API updates (please keep code in the repository in K&R styleZeev Suraski2001-08-141-432/+244
| | | | | like the CODING_STANDARDS ask, guys...)
* Add missing breakZeev Suraski2001-08-141-0/+1
|
* Whenever typelibrary constants were loaded, there was a GP Fault at process ↵Alan Brown2001-08-141-10/+9
| | | | exit as shutdown_memory_manager seemed to delete constants that were supposed to have been deleted earlier. It seems as though CONST_PERSISTENT is a bad thing to use. Resetting that bit on the constant creation still seems to work and the GP Fault at exit is gone.
* hresult is returned if no return type is specifiedHarald Radi2001-08-141-0/+7
|
* make next() return a single value instead of an array with one elementHarald Radi2001-08-141-0/+9
|
* merged from EXPERIMENTALHarald Radi2001-08-131-193/+311
| | | | | lots of cleanup work
* WhitespaceZeev Suraski2001-08-111-22/+22
|
* Some more TSRMLS_FETCH workZeev Suraski2001-07-311-11/+8
|
* More TSRMLS_FETCH annihilationZeev Suraski2001-07-311-1/+2
|
* Zend compatibility patchZeev Suraski2001-07-301-1/+1
|
* More TSRMLS_FETCH work, and a bit of cleanupZeev Suraski2001-07-301-1/+0
|
View Lorry Controller Status