duktape

Commit Graph

Author	SHA1	Message	Date
Sami Vaarala	42f4c6c043	Guide fix for new DUK_CMDLINE_BAREBONES default	10 years ago
Sami Vaarala	11b54b64a0	Make DUK_CMDLINE_BAREBONES the default This will make it more probable that the initial compilation works without investigating a linker error. Ideally the Makefile would detect readline availability, but that would require `uname` or some other platform hackery, which is probably too heavyweight for a simple example Makefile.	10 years ago
Sami Vaarala	0f764ed5dd	Merge branch 'makefile-cygwin-note'	10 years ago
Sami Vaarala	23898c3b56	Cygwin note to Makefile Cygwin 'make dist' now works, and 'make dist-src' works as long as you have 'zip' and 'mkisofs' installed.	10 years ago
Sami Vaarala	63664907ce	Ditz issue update	10 years ago
Sami Vaarala	9db544a1b1	Merge branch 'date-timeval-edge-fix' Fix a few edge cases related to local time handling near the valid Ecmascript time value limits. Two test262 test cases failed due to recently added sanity checks (which were necessary to avoid integer overflows in extreme cases), and one assert was broken. There's no need for a release note because this merge fixes issues introduced by other changes in the release.	10 years ago
Sami Vaarala	b70df0ff6f	Fix Date corner case bugs, tzoffset/local time Tzoffset calculations would be skipped if a temporary time value was just outside the Ecmascript range. This can happen in some legitimate cases, so a +/- 24h leeway was added to the check. This fixes a few test262 cases documented in test-bug-date-timeval-edges.js. Similarly, add a +/- 24h leeway for an assert in local-time-to-parts conversion. When converting a time value to parts and the parts are requested in local time, a temporary time value (simulating local time) can be just outside the Ecmascript range. Allow +/- 24h leeway in the assert to avoid triggering it in valid edge cases. See test-bug-date-timeval-edges.js for demonstration of both issues.	10 years ago
Sami Vaarala	aba9d02641	Bug testcase for Date time value edge case bugs Two test262 cases failed. The tests use Date constructor to create timestamps near the Ecmascript valid range edge. The Date constructor operates with local time values, which caused some previous sanity checks to trigger: temporary time values which have a local tzoffset can be just outside the valid Ecmascript range which trigger some invalid behavior and one assert failure.	10 years ago
Sami Vaarala	1a318bc0be	Ditz issue update	10 years ago
Sami Vaarala	d029bfa7e0	Ditz issue update	10 years ago
Sami Vaarala	0f8433ecaf	Merge branch 'reorg-dist-files' Reorganize Makefiles and a few other files both in the repo itself and in the resulting dist package. This should make it clearer which makefiles are intended to work from the dist directory only (as opposed to the Duktape repo itself).	10 years ago
Sami Vaarala	644551a7e8	Release checklist trivia: new dist-files location	10 years ago
Sami Vaarala	2e6a94bbe2	README fix: refer to Makefile.hello	10 years ago
Sami Vaarala	83a8ff9d19	Fix make_dist script to use dist-files	10 years ago
Sami Vaarala	04aa6831dd	Fix Makefiles to work from dist root	10 years ago
Sami Vaarala	10f94b554d	Move files from examples to dist-files/ The coffeescript makefile is a special case, as it doesn't rely on Duktape.	10 years ago
Sami Vaarala	fcb80e1826	Ditz issue update	10 years ago
Sami Vaarala	710c7c59c9	Merge branch 'valstack-checked-size-restore' Fix value stack resize handling after a function call. Call handling must ensure that whatever value stack size has been requested by the application (with duk_{check,require}_stack()) is guaranteed also after a function call returns. Previously the guaranteed size might have been reduced (there's a testcase demonstrating this in the commit).	10 years ago
Sami Vaarala	56dc9ed9f7	Release note: stack reserve handling	10 years ago
Sami Vaarala	b681f26514	Fix Array.prototype.join() valstack spare handling Previously join() didn't require enough valstack space. The existing built-in spare hid this bug, so this had no concrete impact.	10 years ago
Sami Vaarala	3c7efebee2	Rework valstack resize to use a flags field Three separate int flags slightly bloat call sites for basically no gain. There may be a need for further flags later too.	10 years ago
Sami Vaarala	a8aae84753	Fix valstack restore: use compact flag Without compact flag, valstack resize helper would always add more spare, which would accumulate. This luckily broke test-bi-array-proto-join.js. The commit also removes one helper layer from valstack resizing.	10 years ago
Sami Vaarala	05e8c20dd8	Fix valstack guaranteed size after function call	10 years ago
Sami Vaarala	1173953348	Testcase trivia for Array.prototype.join()	10 years ago
Sami Vaarala	9759c74f90	API testcase for stack reserve vs. function calls This documents a value stack reserve handling bug in Duktape 0.11.0 and documents the desired behavior.	10 years ago
Sami Vaarala	eba565613c	Ditz issue update	10 years ago
Sami Vaarala	5aa528eae2	Ditz issue update	10 years ago
Sami Vaarala	a27472e5a0	Ditz issue update	10 years ago
Sami Vaarala	aa28c89c08	Merge branch 'doc-compilation-options' Add an overview section on how Duktape is compiled, with specific emphasis on: - no official Makefile or build script - Duktape can be compiled as a static or dynamic library too - application MUST include duktape.h; all API calls can be macros C++ discussion is moved into Compiling section.	10 years ago
Sami Vaarala	7fd9a1d287	Add compiling overview to guide	10 years ago
Sami Vaarala	6049fa6797	Relocate C++ section in Guide	10 years ago
Sami Vaarala	82c251ac23	Switch section order in guide	10 years ago
Sami Vaarala	36bc2a6d3c	Ditz issue update	10 years ago
Sami Vaarala	7481df185a	Merge branch 'emscripten-memory-init-file' Add --memory-init-file 0 option to avoid generating dukweb.js.mem which is not correctly passed around, breaking the test REPL page.	10 years ago
Sami Vaarala	c8ec5723e9	Add --memory-init-file 0 to Emscripten options Without this, newer Emscripten will emit a memory init file which is more awkward than passing around a single file. This change fixes the current Dukweb REPL page. Using an explicit memory init file would of course also be possible and a better fix might be to use the Emscripten default.	10 years ago
Sami Vaarala	a2dd8db962	Ditz issue update	10 years ago
Sami Vaarala	1c6548439a	Merge branch 'date-overflow-fixes' Fix a few overflow issues in Date handling. This caused an assert failure and some incorrect behavior when operating with years outside the valid Ecmascript range.	10 years ago
Sami Vaarala	7b05fe8256	Release note: Date fixes	10 years ago
Sami Vaarala	e3c7d0d4a0	Fix Date integer overflow bugs Avoid doing tzoffset computations when timevalue is outside Ecmascript range. If we don't, some integer overflows happen and some internal asserts will fail. This fixes test-bug-date-insane-setyear.js. Avoid integer overflow in duk__make_day() when year is outside valid Ecmascript range. Return NaN directly instead. This fixes test-bug-date-setyear-overflow.js.	10 years ago
Sami Vaarala	b61f20d639	Some 'e.stack' to 'e.stack \|\| e' fixes in tests	10 years ago
Sami Vaarala	fc60360fdd	Add test cases for Date out-of-range handling	10 years ago
Sami Vaarala	6146bdbbc0	Ditz issue update	10 years ago
Sami Vaarala	8c740bc2dd	Ditz issue update	10 years ago
Sami Vaarala	3037bca796	Ditz issue update	10 years ago
Sami Vaarala	ec85b856a3	Merge branch 'api-setglobal-fixes' Fix bug in duk_set_global_object(): must create a new global lexical env object instead of just overwriting _target property of the current one. Otherwise the _target change will affect other contexts with the same (initial) global object incorrectly. Add a lot more tests to duk duk_set_global_object() testcase. Add a note on the API being experimental for now. The detailed semantics can change with other sandboxing changes.	10 years ago
Sami Vaarala	9b89c8c32e	API experimental note for duk_set_global_object()	10 years ago
Sami Vaarala	d0344a6de8	Improvements to duk_set_global_object() testcase	10 years ago
Sami Vaarala	2c3108d92a	Fix duk_set_global() global scope replacement Must create a new scope instead of just setting _target of the existing one: if just _target is set, it will affect other threads which share the same (initial) built-ins. Instead, we must create a new scope object and replace it entirely.	10 years ago
Sami Vaarala	e9cfec20fd	Merge branch 'bug-proxy-finalize-twice' Fix bug for Proxy finalizer handling at heap destruction. If a Proxy was reachable and its target had a finalizer, the finalizer would get executed both for the target and for the Proxy object.	10 years ago
Sami Vaarala	54e60f62d9	Release note: Proxy finalization fix	10 years ago

1 2 3 4 5 ...

3842 Commits (3358df2f8b76a7743016bfa5c77c2657e6ae9e84) All Branches Search

3842 Commits (3358df2f8b76a7743016bfa5c77c2657e6ae9e84)

All Branches