677329 - Implement WebIDL array types

Status: RESOLVED WONTFIX

(Core :: XPCOM, defect)

Description

[(not currently active) Ted Mielczarek]

These would be useful. I'd like to use them. Once we kill off the C implementation of xpidl this won't be too hard.

Comment 1

[(not currently active) Ted Mielczarek]

Attachment: Support WebIDL array types in xpidl.py parser

This patch adds the necessary parsing changes to xpidl.py to handle WebIDL array types. I also added a small test suite for the parser. It's nowhere near comprehensive, but it tests around the areas I was changing, so it made me feel better.

This isn't actually useful yet, as it needs to be hooked up to header.py and the patch that hasn't been finished yet for bug 458936. We'll want to handle Array types by synthesizing an additional length parameter, presumably. (Unless we want to go crazy and make xptcall handle nsTArray or something neat like that.)

Comment 2

[(not currently active) Ted Mielczarek]

I split the tests off into their own patch.

Comment 3

[(not currently active) Ted Mielczarek]

Attachment: Support WebIDL array types in xpidl.py parser

On top of the patch from the other bug. We might not actually want to take this. khuey and I have been talking, and we may fork the xpidl.py parser to make a webidl.py parser, so as not to intermingle all the legacy xpidl crap in with WebIDL.

Comment 4

[Jeff Muizelaar [:jrmuizel]]

(In reply to Ted Mielczarek [:ted, :luser] from comment #3)
> Created attachment 551877 [details] [diff] [review]
> Support WebIDL array types in xpidl.py parser
> 
> On top of the patch from the other bug. We might not actually want to take
> this. khuey and I have been talking, and we may fork the xpidl.py parser to
> make a webidl.py parser, so as not to intermingle all the legacy xpidl crap
> in with WebIDL.

That sounds like a decent idea.

Comment 5

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Is there a timeline for the WebIDL parser?

Comment 6

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

No.

Comment 7

[(not currently active) Ted Mielczarek]

We totally have a separate WebIDL parser now, so use that.