[Webkit-unassigned] [Bug 189947] [JSC] Merge node-jsc Webkit changes upstream

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Sep 27 08:52:23 PDT 2018


https://bugs.webkit.org/show_bug.cgi?id=189947

--- Comment #1 from Yusuke Suzuki <yusukesuzuki at slowstart.org> ---
(In reply to Koby from comment #0)
> Hi, I'm Koby Boyango, the authro of node-jsc, a node.js fork to
> JavaScriptCore and iOS.
> Currently, as I needed to make some changes in WebKit, I have my own fork
> (https://github.com/mceSystems/webkit).
> 
> I would love to offer the changes I've made to WebKit, and hopefully merge
> as much as possible upstream. The changes are mainly around JavaScriptCore,
> WTF and building the JSCOnly port with cmake. Following the discussion on
> the webkit-dev mailiing list:
> https://lists.webkit.org/pipermail/webkit-dev/2018-September/030140.html
> I'm creating this "umbrella" bug, and I'll create a new (blocking) bug for
> each patch.
> 
> This is the first time I'm working with WebKit and try to send patches, so
> please let me know if i'm not doing something correctly with the patches.
> 
> Koby

Reading https://github.com/mceSystems/node-jsc/blob/master/deps/jscshim/docs/webkit_fork_and_compilation.md document, and I have one suggestion.

> Added CustomAPIValue: Similar in functionality to v8's accessor (and used to implement them), allowing object properties that appear like data properties, but invoke native callbacks when accessed.

JavaScriptCore has a similar functionality, called, CustomValue. While CustomAccessor in JSC looks as an accessor, CustomValue looks like a value, but it invokes a function.
Let's look at RegExpConstructor.cpp's regExpConstructorInput etc. They are values. But they invoke corresponding functions to produce returned values.
You can leverage CustomValue mechanism to implement it I think.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20180927/63e84a46/attachment.html>


More information about the webkit-unassigned mailing list