diff --git a/bugs/issue-39f8f5bfd85083fe4e3af1bea400cff87127f459.yaml b/bugs/issue-39f8f5bfd85083fe4e3af1bea400cff87127f459.yaml new file mode 100644 index 00000000..22ab2495 --- /dev/null +++ b/bugs/issue-39f8f5bfd85083fe4e3af1bea400cff87127f459.yaml @@ -0,0 +1,34 @@ +--- !ditz.rubyforge.org,2008-03-06/issue +title: add finalizer support for plain buffers +desc: |- + Buffers very often need finalizer support because they can contain native + structures pointing to malloc()'d memory (or socket handles, etc). Currently + user code must use an actual object and attach a finalizer for that object. + + Adding a fixed finalizer field would add buffer overhead for those buffers + that don't need finalization which is not ideal (but certainly a reasonable + approach anyway). + + Another approach would be to support finalizers only for dynamic buffers: add + a flag for "finalizer present" and if the flag is set, store a finalizer + pointer in the dynamic data area (before or after actual data). + + A third approach would be to similarly use dynamic buffers but put the pointer + in the duk_hbuffer_dynamic struct (depending on whether or not a finalizer is + needed). However, this would require user to know beforehand when a finalizer + is needed because it affects the duk_hbuffer alloc size. +type: :feature +component: duk +release: v1.1 +reporter: Sami Vaarala +status: :unstarted +disposition: +creation_time: 2014-10-06 08:59:55.754356 Z +references: [] + +id: 39f8f5bfd85083fe4e3af1bea400cff87127f459 +log_events: +- - 2014-10-06 08:59:56.282297 Z + - Sami Vaarala + - created + - "" diff --git a/bugs/issue-49cce34deaa9df3b034e55d5f4a67de9c1ca2c69.yaml b/bugs/issue-49cce34deaa9df3b034e55d5f4a67de9c1ca2c69.yaml index d8ed6603..1b56b451 100644 --- a/bugs/issue-49cce34deaa9df3b034e55d5f4a67de9c1ca2c69.yaml +++ b/bugs/issue-49cce34deaa9df3b034e55d5f4a67de9c1ca2c69.yaml @@ -3,9 +3,9 @@ title: go through FIXMEs in ecmascript-testcases and fix the worst coverage issu desc: "" type: :task component: duk -release: v1.0 +release: v1.1 reporter: Sami Vaarala -status: :in_progress +status: :unstarted disposition: creation_time: 2014-09-01 20:26:21.841131 Z references: [] @@ -20,3 +20,7 @@ log_events: - Sami Vaarala - changed status from unstarted to in_progress - "" +- - 2014-10-07 11:24:50.853508 Z + - Sami Vaarala + - assigned to release v1.1 from v1.0 + - "" diff --git a/bugs/issue-7514d4ae323366dd2243d8e8823d70c0063a041f.yaml b/bugs/issue-7514d4ae323366dd2243d8e8823d70c0063a041f.yaml new file mode 100644 index 00000000..6b77374d --- /dev/null +++ b/bugs/issue-7514d4ae323366dd2243d8e8823d70c0063a041f.yaml @@ -0,0 +1,18 @@ +--- !ditz.rubyforge.org,2008-03-06/issue +title: windows dllexport / dllimport symbol macros +desc: Need to use different declspec when compiling Duktape and when compiling an application. +type: :task +component: duk +release: v1.0 +reporter: Sami Vaarala +status: :unstarted +disposition: +creation_time: 2014-10-07 11:14:26.567577 Z +references: [] + +id: 7514d4ae323366dd2243d8e8823d70c0063a041f +log_events: +- - 2014-10-07 11:14:26.743566 Z + - Sami Vaarala + - created + - "" diff --git a/bugs/issue-afdd37e32e5abfc07c1ff7bb8c8086c509847979.yaml b/bugs/issue-afdd37e32e5abfc07c1ff7bb8c8086c509847979.yaml new file mode 100644 index 00000000..711b3528 --- /dev/null +++ b/bugs/issue-afdd37e32e5abfc07c1ff7bb8c8086c509847979.yaml @@ -0,0 +1,24 @@ +--- !ditz.rubyforge.org,2008-03-06/issue +title: document version compatibility expectations +desc: |- + Most important is expectations for a maintained release (like 1.0), such as + backporting of fixes (what kind?) and features (what kind?). + + API change compatibility: outline the basic expectation that APIs will either + remain compatible or be removed entirely so that compile fails reliably. Same + for feature options. +type: :task +component: duk +release: v1.0 +reporter: Sami Vaarala +status: :unstarted +disposition: +creation_time: 2014-10-06 21:24:19.887598 Z +references: [] + +id: afdd37e32e5abfc07c1ff7bb8c8086c509847979 +log_events: +- - 2014-10-06 21:24:20.092617 Z + - Sami Vaarala + - created + - "" diff --git a/bugs/issue-b8d66be953b48a7975b2f12dbc0dd8cdda9103bf.yaml b/bugs/issue-b8d66be953b48a7975b2f12dbc0dd8cdda9103bf.yaml new file mode 100644 index 00000000..44dfab63 --- /dev/null +++ b/bugs/issue-b8d66be953b48a7975b2f12dbc0dd8cdda9103bf.yaml @@ -0,0 +1,34 @@ +--- !ditz.rubyforge.org,2008-03-06/issue +title: "platform header possibility: explicit duk_config.h which is built by a config script" +desc: |- + The current model for feature options, compiler detection, etc is that a + single header file does all the necessary detection on the fly. There is + no need to change Duktape sources or introduce new headers for this to work. + + Another model would be to expect a user to provide a ``duk_config.h`` header + in their include path. For major platforms, a config script could generate + the duk_config.h. For other platforms, a copied and manually edited + duk_config.h could be used instead. + + The upside of this would be that all the platform/compiler detection stuff + would become external. Adding new platforms would have no impact on Duktape + sources at all. + + Even with this approach there could be a default duk_config.h which would + actually do feature detection for common platforms (Linux, Windows, OSX, + mobile phones). +type: :task +component: duk +release: v1.1 +reporter: Sami Vaarala +status: :unstarted +disposition: +creation_time: 2014-10-07 10:32:37.312764 Z +references: [] + +id: b8d66be953b48a7975b2f12dbc0dd8cdda9103bf +log_events: +- - 2014-10-07 10:32:37.600706 Z + - Sami Vaarala + - created + - "" diff --git a/bugs/issue-bf5ea58aef8e046806a2c71f1cf7ce301aa9c422.yaml b/bugs/issue-bf5ea58aef8e046806a2c71f1cf7ce301aa9c422.yaml new file mode 100644 index 00000000..3146ad89 --- /dev/null +++ b/bugs/issue-bf5ea58aef8e046806a2c71f1cf7ce301aa9c422.yaml @@ -0,0 +1,18 @@ +--- !ditz.rubyforge.org,2008-03-06/issue +title: "unify bitops tests: use shared test values and large matrix type tests for better coverage" +desc: Add also coercions with side effects to the test value set. +type: :task +component: duk +release: v1.1 +reporter: Sami Vaarala +status: :unstarted +disposition: +creation_time: 2014-10-06 20:39:51.311452 Z +references: [] + +id: bf5ea58aef8e046806a2c71f1cf7ce301aa9c422 +log_events: +- - 2014-10-06 20:39:51.492438 Z + - Sami Vaarala + - created + - "" diff --git a/bugs/issue-f3d60a9ea765a47372cb6ae0dbeb84d9a6d20444.yaml b/bugs/issue-f3d60a9ea765a47372cb6ae0dbeb84d9a6d20444.yaml new file mode 100644 index 00000000..38c1907c --- /dev/null +++ b/bugs/issue-f3d60a9ea765a47372cb6ae0dbeb84d9a6d20444.yaml @@ -0,0 +1,18 @@ +--- !ditz.rubyforge.org,2008-03-06/issue +title: "expression evaluation order test case: duktape may differ from standard now (e.g. bitops)" +desc: "" +type: :task +component: duk +release: v1.1 +reporter: Sami Vaarala +status: :unstarted +disposition: +creation_time: 2014-10-06 20:32:56.215116 Z +references: [] + +id: f3d60a9ea765a47372cb6ae0dbeb84d9a6d20444 +log_events: +- - 2014-10-06 20:32:56.412102 Z + - Sami Vaarala + - created + - ""