Граф коммитов

9718 Коммитов

Автор SHA1 Сообщение Дата
igor%mir2.org f390d556b2 Bug 357016: Patch from Edward Lee to restore ability to compile SpiderMonkey with C++ compiler. It was broken after my changes. r=myself 2007-07-12 07:48:50 +00:00
jwalden%mit.edu f51f364d16 Back out bug 387390; it causes the marquee binding to throw security exceptions. 2007-07-12 03:14:27 +00:00
mrbkap%gmail.com 7ba5e08f88 Don't trust the filename of cloned functions. bug 387390, r=jst sr=brendan 2007-07-12 01:27:03 +00:00
mrbkap%gmail.com 91be19e9bc Always return XPCSafeJSObjectWrappers from Components.utils.Sandbox and evalInSandbox. This makes this interface much, much easier to use safely, as the wrapper takes care of several common problems that occur when touching regular JS objects directly. bug 386635, r=jst sr=brendan 2007-07-11 22:52:29 +00:00
mrbkap%gmail.com 8d7965a0f9 Clear dormantNext when we're done using it. bug 387725, r=igor 2007-07-11 21:57:05 +00:00
nboyd%atg.com a83b7e8719 New skips of new tests. 2007-07-11 16:53:38 +00:00
nboyd%atg.com 7389c08ecb Implement block scope (let) and array comprehensions.
This required significant changes to symbol table management.
2007-07-11 16:00:03 +00:00
nboyd%atg.com 29b1032a5e Switch to 1.4 as the souce and target. 2007-07-11 13:55:09 +00:00
igor%mir2.org 740c486981 Bug 386885: JSAtom.number is removed in favour of using atoms itself for hashing. r=brendan 2007-07-11 09:25:45 +00:00
neil%parkwaycc.co.uk 11d0936e48 Followup to bug 348748 to fix the alignment of trailing backslashes in macros
[xxx_cast<foo*>(bar) is three characters shorter than NS_XXX_CAST(foo*, bar)]
2007-07-11 08:46:49 +00:00
bzbarsky%mit.edu ef8b8db3a9 Clean up the XDR properly even if JS_XDRScript fails. In particular, don't
double-free things.  Bug 387572, r+sr=brendan
2007-07-11 03:10:43 +00:00
bzbarsky%mit.edu 9bba06ca93 Relanding bug 326777, r=bsmedberg, sr=jst 2007-07-11 00:57:17 +00:00
mrbkap%gmail.com 64ad36be39 Create exceptions with the right parent. bug 387390, r=jst sr=brendan 2007-07-11 00:37:03 +00:00
mrbkap%gmail.com cee94558fe Ensure that we have a valid scope chain at all times so C++ callers can create objects in the right scope. bug 387591, r=brendan 2007-07-11 00:29:25 +00:00
mrbkap%gmail.com 0996fe32e7 Avoid asserting for user-controlled thing (through __proto__ setting). bug 387182, r+sr=jst 2007-07-11 00:27:57 +00:00
brendan%mozilla.org c422a3ae05 to{Source,{,Locale}String} are not generic (387501, r=mrbkap). 2007-07-10 05:07:20 +00:00
bzbarsky%mit.edu 4b9e120afb Backing out to fix red 2007-07-10 04:07:45 +00:00
bzbarsky%mit.edu af53b414ca Make sure that we push a null JSContext on the current thread's XPConnect stack
before processing events.  Bug 326777, r=bsmedberg, sr=jst
2007-07-10 03:48:07 +00:00
mrbkap%gmail.com 5a0bb7f9f2 Attempt to fix leaks. 2007-07-10 00:24:16 +00:00
mrbkap%gmail.com ec88adb127 Back out this change to hopefully fix the rest of the regressions test failures (from bug 386695). 2007-07-09 23:34:58 +00:00
mrbkap%gmail.com 52995f0ccd Associate XPCWrappedNativeScopes with XPCContexts, like the DOM does with windows and contexts, so that we can push a context that's allowed to actually call XPCWrappedJS functions. bug 352791, r=jst sr=brendan 2007-07-09 21:51:57 +00:00
mrbkap%gmail.com 52bf2b3523 Don't create exceptions using the global object as the parent, since that's bogus, use the scope chain instead. bug 386695, r+sr=jst 2007-07-09 21:46:18 +00:00
igor%mir2.org ffb0bddb6a Bug 385729: objects and regexps are stored in separated tables in JSScript. r=brendan,mrbkap 2007-07-08 09:03:37 +00:00
jwalden%mit.edu ef68fcf595 Bug 348748 - Replace all instances of NS_STATIC_CAST and friends with C++ casts (and simultaneously bitrot nearly every patch in existence). r=bsmedberg on the script that did this. Tune in next time for Macro Wars: Episode II: Attack on the LL_* Macros. 2007-07-08 07:08:56 +00:00
igor%mir2.org 36ec5a5746 Bug 387229: now DUMP_SCOPE_STATS includes js_scope_stats into the log. r=brendan 2007-07-07 21:55:29 +00:00
dbaron%dbaron.org b6ea4ec030 Fix bogus trace names. b=387223 r=brendan 2007-07-07 07:30:58 +00:00
mrbkap%gmail.com 80bf1a96e6 Allow chrome files to touch implicit XPCNativeWrappers. bug 387084, r+sr=jst 2007-07-07 00:52:44 +00:00
jwatt%jwatt.org e5919b741b Fixing bug 317304. Implement weak references for JavaScript. Patch by Alex Fritze <alex@croczilla.com>. r+sr=shaver@mozilla.org 2007-07-06 20:13:14 +00:00
brendan%mozilla.org 5f235ddce2 Fix degenerate unsigned (enum) comparisons (357016, r=igor). 2007-07-05 23:25:27 +00:00
crowder%fiverocks.com 3d18858ebb Bug 357016: changing this back to (void *) for now to fix burning tree 2007-07-05 20:54:44 +00:00
crowder%fiverocks.com 43f7310bf8 Bug 357016: fixing brendan's nit from comment 23 2007-07-05 20:41:59 +00:00
crowder%fiverocks.com 0c198a012a Bug 357016: spidermonkey should be buildable by a C++ compiler, patch by
crowder/Jason Orendorff, r=brendan
2007-07-05 20:37:48 +00:00
mrbkap%gmail.com 807678c499 Ensure that accesses to implicit XPCNativeWrappers are actually legal. bug 384750, r+sr=jst 2007-07-05 20:13:43 +00:00
mrbkap%gmail.com 0609e26629 Ensure that accesses to implicit XPCNativeWrappers are actually legal. bug 384750, r+sr=jst 2007-07-05 20:01:12 +00:00
dbaron%dbaron.org acfb4f5b38 Put comment in the right place. b=180380 2007-07-03 21:31:02 +00:00
mrbkap%gmail.com aaa3a85b3c Backing out these changes to fix tinderbox orange. 2007-07-03 01:12:41 +00:00
mrbkap%gmail.com cb63d56a5f Ensure that accesses to implicit XPCNativeWrappers are actually legal. bug 384750, r+sr=jst 2007-07-02 23:39:41 +00:00
brendan%mozilla.org d45f131a3a Fix redundant statement and out-of-date comment. 2007-07-02 21:06:52 +00:00
igor%mir2.org b3db2ca9a8 Bug 349326: for-in loop now always closes iterator objects. r=brendan 2007-07-02 12:13:27 +00:00
timeless%mozdev.org b2a20d910c Bug 385765 [@ jsdASObserver::Observe] You can't dereference a NULL nsCOMPtr with operator->()
r=dveditz sr=dveditz
2007-07-01 19:25:10 +00:00
sayrer%gmail.com a07b272bd1 Bug 180380. nsXPCComponents object and its wrapper leaked at shutdown. Patch by David Baron, David Bradley, and Robert Sayre. r=jst/bzbarsky, sr=dbaron 2007-06-30 23:00:09 +00:00
dbaron%dbaron.org 324cd46264 Suspect all native wrappers during cycle collection (last part, rest already done by Graydon). b=368869 r=graydon sr=jst 2007-06-30 03:03:30 +00:00
dbaron%dbaron.org fe691f69ab Make wrapped native debug dump log the native that is wrapped. b=385549 r+sr=jst 2007-06-30 03:02:15 +00:00
nboyd%atg.com e3954fc48f See bug 382340. Additional patch from Hannes Wallnoefer:
The varargs patch does not handle the case where a varargs method is called
with a java array. For example, the following code returns a List containing a
string array as single element, instead of a List containing all the strings in
the array:

java.util.Arrays.asList(java.util.TimeZone.getAvailableIDs());
2007-06-28 12:52:40 +00:00
jst%mozilla.org 0c323997cb Fixing bug 380474. Use the inner object when looking up methods and adding event listeners. r=mrbkap@mozilla.org, sr=brendan@mozilla.org 2007-06-28 00:28:02 +00:00
mrbkap%gmail.com ebe9f0c093 Skip initial holes when computing the start value of reduce(Right). bug 386030, r=brendan 2007-06-28 00:04:49 +00:00
brendan%mozilla.org 7707ad29a1 Added push benchmark from Gabriel Sjoberg <gabrielsjoberg@gmail.com>. 2007-06-27 00:11:11 +00:00
brendan%mozilla.org e18f187bdc Re-land fix for cloned function object prototyping (300079, r=igor/mrbkap). 2007-06-26 18:19:54 +00:00
enndeakin%sympatico.ca 288cb44ead Bug 382034, support optional args in idl, try again with fix for crash calling toString, r+sr=shaver 2007-06-25 14:23:26 +00:00
enndeakin%sympatico.ca 54d2bdefc8 Bug 382034, back out due to test content/base/test/test_bug352728.html failing 2007-06-22 15:21:14 +00:00
enndeakin%sympatico.ca 5daa0a155e Bug 382034, support optional arguments in idl, r+sr=shaver 2007-06-22 14:22:23 +00:00
dbaron%dbaron.org 8bdd21ccff Back out again. 2007-06-22 03:11:41 +00:00
mrbkap%gmail.com fb6c310783 Actually populate the string cache so we don't leak all deflated strings. bug 385378, r=brendan 2007-06-22 00:31:50 +00:00
dbaron%dbaron.org d12d968af6 Bug 368869, suspect all native wrappers for cycle collection. Relanding. Patch by graydon@mozilla.com. r=brendan, sr=jst 2007-06-21 22:25:51 +00:00
mrbkap%gmail.com 38bbdc645d Remove useless variable. bug 385368, patch from Gabriel Sjoberg <gabrielsjoberg@gmail.com>, r=mrbkap 2007-06-21 20:38:45 +00:00
sayrer%gmail.com 772193bcc2 Bug 381651. Tweak XPCOMUtils. Patch by Nickolay_Ponomarev <asqueella@gmail.com>. r=sayrer, sr=bsmedberg 2007-06-21 04:53:35 +00:00
mrbkap%gmail.com dff4d95df2 Implement a full nsIScriptSecurityManager in xpcshell. bug 239969, r+sr=jst 2007-06-21 00:10:48 +00:00
mrbkap%gmail.com 5a5684e2d3 Use the latest version of JS, so the shell always has the newest features. bug 385159, r=brendan 2007-06-20 23:53:55 +00:00
sayrer%gmail.com 0d5bf93622 back out patch for bug 300079 due to leak regressions, as the new textframe code needs to land 2007-06-20 18:53:28 +00:00
brendan%mozilla.org 09eb9ff53c Test a hypothesis about the shutdown leak that's biting 300079's patch (a=sayrer). 2007-06-20 17:03:15 +00:00
bclary%bclary.com 7b1bf215e5 JavaScript Tests - regression test for bug 300079, by Brendan Eich 2007-06-20 07:50:52 +00:00
brendan%mozilla.org 7798c7faf7 Fix violation of function prototyping due to cloned function object implementation (300079, r=mrbkap; expecting r=igor after the fact, want to get this in for widespread testing tomorrow). 2007-06-20 06:37:15 +00:00
sayrer%gmail.com 8237229b94 Bug 384168. C.u.import doesn't prevent recursion in the presence of circular dependencies. r=brendan, sr=bsmedberg 2007-06-20 06:29:49 +00:00
timeless%mozdev.org a168ae84b1 Bug 365363 Return value for GetJSDValue ignored leading to death
r=silver sr=biesi
2007-06-20 05:56:13 +00:00
brendan%mozilla.org 12922bb001 Function atom if non-null must be string-keyed (385134, r=mrbkap). 2007-06-20 05:11:38 +00:00
gavin%gavinsharp.com af4a85e03e Back out previous patch for bug 239969 (bustage, and tree is closed) 2007-06-20 02:30:54 +00:00
mrbkap%gmail.com d41dfa9270 Implement an nsIScriptSecurityManager for the xpcshell. bug 239969, r+sr=jst 2007-06-20 02:13:22 +00:00
mrbkap%gmail.com f2b945f6da Backing this patch out again to fix sessionstore. See bug 300079 and 385085 2007-06-20 02:06:00 +00:00
brendan%mozilla.org 667871f79f Fix violation of function prototyping due to cloned function object implementation (300079, r=igor). 2007-06-19 18:49:53 +00:00
crowder%fiverocks.com bd4ba9d8a8 Bug 384809 : Various JS engine crashes/leaks in OOM conditions, r=brendan, patch by Gavin Reaney, gavin@picsel.com 2007-06-19 04:10:03 +00:00
brendan%mozilla.org 166d29078b Clean up property attributes (384846, r=mrbkap). 2007-06-19 02:52:50 +00:00
crowder%fiverocks.com 7ecfec5d58 Bug 384901: leak in jsfile.c, patch by Robin Ehrlich, rehrlich@ubiqinc.com, r=mrbkap 2007-06-18 21:40:30 +00:00
mrbkap%gmail.com e94ecb5e66 Protect js_GetStringBytes from callers that happen during the last GC. Patch from brendan, bug 336907, r=daumling sr=shaver 2007-06-18 21:38:31 +00:00
crowder%fiverocks.com aba6d7eb3b Bug 383607: add ARM pure endian double support to JS engine, r=crowder, patch by Gavin Reaney gavin@picsel.com 2007-06-18 21:31:48 +00:00
crowder%fiverocks.com 2df69f0107 bug 379056: quieting the assertions here for all but myself and mrbkap, r=mrbkap 2007-06-18 20:24:00 +00:00
igor%mir2.org 48eb00a693 Bug 384478: Patch from Mike Moening to fix warning when compiling with VC. r=me 2007-06-18 19:33:21 +00:00
bclary%bclary.com 853b5a0c8b JavaScript Tests - update test range to 48.5-51.5%, bug 211590 2007-06-18 18:35:32 +00:00
jwatt%jwatt.org 61e0b6bdb6 Fixing bug 383917. The new Components.utils.import code shouldn't spam the console. r=sayrer@gmail.com, sr=brendan@mozilla.com 2007-06-18 15:36:16 +00:00
bzbarsky%mit.edu 73f31550fa Backing out bug 300079 to see whether that fixes tinderbox orange. 2007-06-18 14:32:35 +00:00
brendan%mozilla.org b343c26cfb Fix violation of function prototyping due to cloned function object implementation (300079, r=igor). 2007-06-18 05:33:36 +00:00
bclary%bclary.com 32685f9683 JavaScript Test - catch indirect eval exception on trunk, bug 352797 2007-06-18 05:20:11 +00:00
bclary%bclary.com 7d86e31bbf JavaScript Test - catch indirect eval exception on trunk, bug 361571 2007-06-18 05:17:54 +00:00
brendan%mozilla.org d1fcc58d6f Remove __callee__ property of Call prototypes (384642, r=igor). 2007-06-18 05:10:19 +00:00
brendan%mozilla.org 494b1e6457 Emit JSOP_GROUP when optimizing away delete (382981, r=mrbkap). 2007-06-17 18:05:37 +00:00
brendan%mozilla.org 1bd5fea126 Fix recent regression in CheckSideEffects for paren-expr case (384680, r=mrbkap). 2007-06-16 19:48:44 +00:00
bclary%bclary.com 5ba863fe52 JavaScript Tests - move test to extensions subsuite 2007-06-16 18:41:15 +00:00
bclary%bclary.com 8f1662d813 JavaScript Test - add strict warning tests for eval rename 2007-06-16 17:14:17 +00:00
brendan%mozilla.org 96e2ed07f2 Better entrainment avoidance for Call.arguments (383269, r=igor). 2007-06-16 00:29:21 +00:00
mrbkap%gmail.com 2be57a7534 Restrict who can claim to implement nsISecurityCheckedComponent. bug 352882, r=jst sr=bzbarsky 2007-06-15 21:44:51 +00:00
mrbkap%gmail.com 2631a0c112 Use a better filename when eval is used across scopes. bug 369211, r=brendan 2007-06-15 21:00:26 +00:00
bclary%bclary.com d31f84b537 JavaScript Test - regression test for bug 380831, by Jesse Ruderman 2007-06-15 18:46:59 +00:00
bclary%bclary.com c3e3b692df JavaScript Tests - update test to ignore 1ms differences, bug 354998 2007-06-15 17:16:08 +00:00
bclary%bclary.com e32ce1c95f JavaScript Test - update test to reflect new uneval(undefined) on trunk, bug 356238 2007-06-15 17:00:46 +00:00
nboyd%atg.com ac1427d243 Fix boolean value passed to __iterator__ when called by runtime as a result of
for..in loop.
Also fix formatting.
2007-06-15 15:39:11 +00:00
nboyd%atg.com 85813cad67 Fix bug 381180 - __lookupGetter__ finds a function even when it should not
(similarly for __lookupSetter__)
Patch from Roshan James <roshanj@google.com>.
2007-06-15 12:40:53 +00:00
igor%mir2.org 7b840a3384 Bug 378793: Patch from Mike Moening and me to implement per-context debug hooks. r=brendan 2007-06-15 06:44:20 +00:00
mrbkap%gmail.com b31ccc1a27 Allow C.u.Sandbox to take a principal or an nsIScriptObjectPrincipal. bug 353732, r+sr=jst 2007-06-14 23:50:28 +00:00
bclary%bclary.com 59ff04c95a JavaScript Test - regression test for bug 369666, by Brendan Eich 2007-06-14 22:43:09 +00:00
bclary%bclary.com e6b34fd7c1 JavaScript Tests - regression tests for bug 351102, by Jesse Ruderman, Igor Bukanov 2007-06-14 22:39:03 +00:00
bclary%bclary.com ac87924a6e JavaScript Tests - regression test for bug 375183, by Igor Bukanov 2007-06-14 22:34:09 +00:00