Commit Graph

233 Commits

Author SHA1 Message Date
Andi Gutmans
5418b59681 - Fix NULL handling in ARRAY opcode and resolve memory leak 2000-09-14 20:30:06 +00:00
Zeev Suraski
5286b3971c Make compile_string() accept a description of the code 2000-09-12 19:47:25 +00:00
Andi Gutmans
27de50012b - Use emalloc() for opened_path now. This was a potential leak before.
- This patch has potential to break stuff but I tested it as much as I
- could. Fixes should be easy.
2000-08-31 22:24:20 +00:00
Andi Gutmans
fbbc249f2b - Revert patch from 9/7/2000 which seems to have broken unset().
- I hope what made me do this patch doesn't appear again.
2000-08-24 19:03:35 +00:00
Andi Gutmans
c698e14635 - Beautify 2000-08-19 15:16:53 +00:00
Zeev Suraski
7795aca4ca Fix warning issue (compile errors inside require()'d files were incorrectly supressed) 2000-08-15 16:44:40 +00:00
Zeev Suraski
636d126e16 - Fix leak and some logic 2000-08-14 20:16:27 +00:00
Andi Gutmans
e46ea8864c - This patch should hopefully fix situations where a constructor uses
- the $this pointer as a reference.
2000-08-14 19:17:26 +00:00
Stanislav Malyshev
9e750e64b9 Fix crash 2000-08-14 09:06:56 +00:00
Stanislav Malyshev
5090b1e8d5 Fix zend_fiel_handle handling. Should fix URL include
and various opened_path inconsistencies.
2000-08-13 18:00:50 +00:00
Andi Gutmans
e5015ed581 @ Fix problem with nested foreach()'s (Andi, Zend Engine) 2000-08-10 22:08:02 +00:00
Zeev Suraski
c06692e9ec The patch we promised - redesigned the compilation/execution API:
Advantages:
- Smaller memory footprint for the op arrays
- Slightly faster compilation times (due to saved erealloc() calls and faster zend_op
  initialization)
- include_once() & require_once() share the same file list
- Consistency between include() and require() - this mostly means that return()
  works inside require()'d files just as it does in include() files (it used to
  be meaningless in require()'d files, most of the time (see below))
- Made require() consistent with itself.  Before, if the argument was not a constant
  string, require() took the include() behavior (with return()).
- Removed lots of duplicate code.
Bottom line - require() and include() are very similar now;  require() is simply an include()
which isn't allowed to fail.  Due to the erealloc() calls for large op arrays, require()
didn't end up being any faster than include() in the Zend engine.
2000-08-09 19:22:35 +00:00
Andi Gutmans
50ddf9e3d5 - Use some more SEPARATE_ZVAL macros instead of replicated code. 2000-08-05 13:54:07 +00:00
Stanislav Malyshev
8b9ecac2dc Fix memory leak 2000-08-05 11:06:00 +00:00
Andi Gutmans
fcdc025d11 - Beautify code. Try and use more macros for splitting instead of
- replicating the code everywhere.
2000-08-04 08:11:54 +00:00
Andi Gutmans
5ae343b752 - Remove commented code 2000-08-02 17:51:42 +00:00
Zeev Suraski
735a0d148e Fix filename issues 2000-07-29 10:44:13 +00:00
Andi Gutmans
290525fdcd Always store full filename as compiled file name 2000-07-27 15:36:54 +00:00
Stanislav Malyshev
148317f979 Fix a bug in passing second parameter of RECV_INIT with is_ref set 2000-07-11 13:23:48 +00:00
Andi Gutmans
c1ed6ef9cc - Fix memory leak. 2000-07-09 20:21:20 +00:00
Andi Gutmans
40b31d7487 - Need to seperate if the hash isn't a reference 2000-07-09 17:12:37 +00:00
Stanislav Malyshev
be59613ddf Remove C++ commennts. 2000-07-07 09:50:50 +00:00
Andi Gutmans
58d1e0a2cc - Yet another fix... 2000-07-06 23:00:46 +00:00
Andi Gutmans
db8734c82f - One more... 2000-07-06 22:38:01 +00:00
Andi Gutmans
5f893e68ff - Complex fix for solving a problem with objects & method calls.
- Previous version is tagged PRE_METHOD_CALL_SEPERATE_FIX_PATCH.
- I need to check this fix on a server so if it doesn't work I will revert
- it.
2000-07-06 22:08:22 +00:00
Zeev Suraski
84eb48f0a1 Ok, this time here's some real Win32 system programming :)
Redesigned the timeout system using a single timeout thread and a single window,
and used a much quicker check.
2000-06-16 14:27:28 +00:00
Zeev Suraski
a6d9b5ed18 Macro it up the right way 2000-06-16 02:48:35 +00:00
Zeev Suraski
b791918443 Macro this up, so it can be moved to other places 2000-06-16 02:47:00 +00:00
Zeev Suraski
ef0bd3d102 - Move timeout code to Zend
- Implement timeouts in Win32
2000-06-16 01:54:56 +00:00
Zeev Suraski
53bdf0b5cc I can't think of a reason of why it should just be a notice... Make it a warning, like it was in PHP 3. 2000-06-10 01:12:12 +00:00
Zeev Suraski
03e842ccb8 Fix bug #4933 2000-06-09 15:42:49 +00:00
Andi Gutmans
edd7025645 - Remove old obsolete code. 2000-06-09 13:08:40 +00:00
Andi Gutmans
a712da5cee - Make unset consistent with the way array offsets work 2000-06-09 13:07:26 +00:00
Stanislav Malyshev
e346a6b538 Handle unset with empty key 2000-06-09 10:59:21 +00:00
Zeev Suraski
13b67621cf Fix a bug in static initializers/default values/class member variables that contained
array values
2000-05-31 19:07:09 +00:00
Andi Gutmans
715ff77344 - Fix include() when used on resources (shouldn't work but shouldn't crash
either).
2000-05-18 10:21:24 +00:00
Andi Gutmans
686bd5724b - Make $obj->test = 5; work again (assigning to uninitialized objects) 2000-05-06 18:40:48 +00:00
Andi Gutmans
f2471fd50a - Change the place CAST uses for the op_type 2000-05-03 17:56:44 +00:00
Zeev Suraski
974671eced *** empty log message *** 2000-04-27 21:48:47 +00:00
Zeev Suraski
4b6594c5e9 Change to using the #define's 2000-04-27 21:26:17 +00:00
Andi Gutmans
4f15966d1b - Fix order of JMPZNZ arguments 2000-04-17 17:16:47 +00:00
Zeev Suraski
021631d5eb *** empty log message *** 2000-04-12 00:50:21 +00:00
Zeev Suraski
d17e3d6b18 Fix warnings 2000-04-11 17:38:19 +00:00
Zeev Suraski
97c9603b02 Fix fd leak in include_once() 2000-04-11 17:30:37 +00:00
Andi Gutmans
73c6fbff7e - 2000-04-10 21:12:33 +00:00
Zeev Suraski
e663d856b3 Fix object overloading support 2000-04-10 20:21:13 +00:00
Andi Gutmans
c5d5c69bfe - Add warnings 2000-04-10 19:54:45 +00:00
Zeev Suraski
1cf5c33627 Thoroughly fix include_once() 2000-04-07 22:03:46 +00:00
Zeev Suraski
ed6e734047 Fix include_once() 2000-04-07 21:16:19 +00:00
Andi Gutmans
a9159c2eec Initial preparation for OO overloading patch 2000-04-06 16:34:55 +00:00