733260 - Remove jsuint

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

[David Mandelin [:dmandelin]]

I think this the last one (other than JSBool).

Comment 1

[David Mandelin [:dmandelin]]

Attachment: Patch

Most |jsuint| is for array indexes and lengths, so I used uint32_t per the new APIs. I also followed that for string lengths. I used |unsigned| at a few places case-by-case where it seemed that was wanted.

Comment 2

[Luke Wagner [:luke]]

Comment on attachment 603135 [details] [diff] [review]
Patch

It would be good to put something like this at the top of jsapi.h:

  /* Basic JSAPI type assumptions: */
  JS_STATIC_ASSERT(sizeof(unsigned) >= 4);
  JS_STATIC_ASSERT(sizeof(int) >= 4);

including any others that you can think of from the recent de-typedef-ification.

Comment 3

[David Mandelin [:dmandelin]]

(In reply to Luke Wagner [:luke] from comment #2)
> Comment on attachment 603135 [details] [diff] [review]
> Patch
> 
> It would be good to put something like this at the top of jsapi.h:
> 
>   /* Basic JSAPI type assumptions: */
>   JS_STATIC_ASSERT(sizeof(unsigned) >= 4);
>   JS_STATIC_ASSERT(sizeof(int) >= 4);

Good idea. I'm pretty sure those two being less than 32 bits is the major breakage to worry about. Added to the landed patch.

http://hg.mozilla.org/integration/mozilla-inbound/rev/ae005ec67376

Comment 4

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

(In reply to David Mandelin from comment #1)
> Created attachment 603135 [details] [diff] [review]
> Patch
> 
> Most |jsuint| is for array indexes and lengths, so I used uint32_t per the
> new APIs. I also followed that for string lengths. I used |unsigned| at a
> few places case-by-case where it seemed that was wanted.

You seem to have used |unsigned| everywhere.

Comment 5

[David Mandelin [:dmandelin]]

(In reply to Ms2ger from comment #4)
> (In reply to David Mandelin from comment #1)
> > Created attachment 603135 [details] [diff] [review]
> > Patch
> > 
> > Most |jsuint| is for array indexes and lengths, so I used uint32_t per the
> > new APIs. I also followed that for string lengths. I used |unsigned| at a
> > few places case-by-case where it seemed that was wanted.
> 
> You seem to have used |unsigned| everywhere.

Thanks for catching! I brain-farted when writing my replace script. I'll post a followup that puts it how I wanted momentarily.

Comment 6

[David Mandelin [:dmandelin]]

Attachment: Followup patch

Switches from unsigned to uint32_t all the places I originally intended.

Comment 7

[David Mandelin [:dmandelin]]

Followup landed:

http://hg.mozilla.org/integration/mozilla-inbound/rev/5ca23d2045b6

Comment 8

[Marco Bonardo [:mak]]

https://hg.mozilla.org/mozilla-central/rev/ae005ec67376
https://hg.mozilla.org/mozilla-central/rev/5ca23d2045b6

Comment 9

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Thanks, that makes a lot more sense.