Comment 1 Boris Zbarsky [:bz] (still a bit busy) 2012-04-27 23:07:15 PDT

Created attachment 619254 [details] [diff] [review]
Codegen for typed array and arraybuffer arguments.

I think this depends on the infrastructure bug 748267.  Also, I didn't do typed array return values yet, because I have no real way to test them.

Comment 2 :Ms2ger (⌚ UTC+1/+2) 2012-04-28 00:04:37 PDT

Comment on attachment 619254 [details] [diff] [review]
Codegen for typed array and arraybuffer arguments.

Review of attachment 619254 [details] [diff] [review]:
-----------------------------------------------------------------

::: content/base/src/nsXMLHttpRequest.cpp
@@ +2633,5 @@
>    aContentType.SetIsVoid(true);
>    aCharset.Truncate();
>  
> +  PRInt32 length = aArrayBuffer->mLength;
> +  char* data = reinterpret_cast<char*>(aArrayBuffer->mData);

Can this be a static_cast?

Comment 3 Boris Zbarsky [:bz] (still a bit busy) 2012-04-28 00:12:45 PDT

No.  Can't static_cast from uint8_t* to char*.  At least not on my compiler.  ;)

Comment 4 Boris Zbarsky [:bz] (still a bit busy) 2012-05-22 19:20:23 PDT

Created attachment 626287 [details] [diff] [review]
Updated to be a little simpler

Comment 5 Peter Van der Beken [:peterv] 2012-05-23 14:25:18 PDT

Comment on attachment 626287 [details] [diff] [review]
Updated to be a little simpler

Review of attachment 626287 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/bindings/Codegen.py
@@ +1414,5 @@
> +        elif type.isArrayBufferView():
> +            name = "ArrayBufferView"
> +            jsname = "TypedArrayObject"
> +        else:
> +            name = type.name

Can't you always set name to type.name, even for ArrayBuffer and ArrayBufferView?

Comment 6 Boris Zbarsky [:bz] (still a bit busy) 2012-05-23 14:29:17 PDT

> Can't you always set name to type.name, even for ArrayBuffer and ArrayBufferView?

I think so, yes.  Will do.

Comment 7 Boris Zbarsky [:bz] (still a bit busy) 2012-05-24 22:10:11 PDT

http://hg.mozilla.org/mozilla-central/rev/faa1929dc247

Comment 8 Boris Zbarsky [:bz] (still a bit busy) 2012-05-29 20:57:22 PDT

Added decent tests in https://hg.mozilla.org/integration/mozilla-inbound/rev/e0d8fa7fe174