Switch CDM.h to use uint32_t for size types per style guide.

Switch CDM.h to use uint32_t for size types per style guide.

Style guide now clearly says to use unsigned types for sizes
and counts:
http://dev.chromium.org/developers/coding-style#TOC-Types

Since CDM.h communicates over PPAPI, we use a fixed uint32_t
instead of size_t.

BUG=none
TEST=compiles
R=ddorwin@chromium.org, xhwang@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=230180

[DaleCurtis, 2013-10-08 23:05:06]

Not sure we can change all these after all, but here's what a clean world would
look like.

[DaleCurtis, 2013-10-08 23:05:30]

Also, here's what CdmWrapper would look like:

https://codereview.chromium.org/26592003/

[ddorwin, 2013-10-08 23:53:10]

LG. Will wait to see what issues we might encounter.

[DaleCurtis, 2013-10-10 21:22:43]

It seems to work just fine with an older WV CDM, a little hard to tell
completely since half the WV tests are failing right now though...

[ddorwin, 2013-10-11 18:57:33]

LGTM, but let's wait until the tests are passing again before landing this.

[DaleCurtis, 2013-10-21 22:31:51]

xhwang: You want to weigh in here?

[xhwang, 2013-10-21 23:30:10]

lgtm % nits

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h
File content_decryption_module.h (right):

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h#n...
content_decryption_module.h:177: int32_t samples_per_second;
hmm, are these also "counts"?

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h#n...
content_decryption_module.h:271: uint32_t platform_key_certificate_length;
nit: we use "size" for data size in this file. This is consistent with what we
use in /media.

[DaleCurtis, 2013-10-21 23:40:55]

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h
File content_decryption_module.h (right):

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h#n...
content_decryption_module.h:177: int32_t samples_per_second;
On 2013/10/21 23:30:11, xhwang wrote:
> hmm, are these also "counts"?

The line is "Use unsigned integer types (preferably size_t) for object and
allocation sizes, object counts, array and pointer offsets, vector indices, and
so on."

I tried to only convert values which were allocation sizes or vector indices,
since you can stretch the definition of "object counts" pretty far.

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h#n...
content_decryption_module.h:271: uint32_t platform_key_certificate_length;
On 2013/10/21 23:30:11, xhwang wrote:
> nit: we use "size" for data size in this file. This is consistent with what we
> use in /media.

Well, kind of, this is consistent with the names used in the function prototypes
(always _length).  If you want to change all the prototypes, I'm happy to change
this.  Otherwise I think it should stay as length for consistency.

[xhwang, 2013-10-21 23:51:11]

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h
File content_decryption_module.h (right):

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h#n...
content_decryption_module.h:177: int32_t samples_per_second;
On 2013/10/21 23:40:55, DaleCurtis wrote:
> On 2013/10/21 23:30:11, xhwang wrote:
> > hmm, are these also "counts"?
> 
> The line is "Use unsigned integer types (preferably size_t) for object and
> allocation sizes, object counts, array and pointer offsets, vector indices,
and
> so on."
> 
> I tried to only convert values which were allocation sizes or vector indices,
> since you can stretch the definition of "object counts" pretty far.

agreed :)

https://codereview.chromium.org/26592004/diff/1/content_decryption_module.h#n...
content_decryption_module.h:271: uint32_t platform_key_certificate_length;
On 2013/10/21 23:40:55, DaleCurtis wrote:
> On 2013/10/21 23:30:11, xhwang wrote:
> > nit: we use "size" for data size in this file. This is consistent with what
we
> > use in /media.
> 
> Well, kind of, this is consistent with the names used in the function
prototypes
> (always _length).  If you want to change all the prototypes, I'm happy to
change
> this.  Otherwise I think it should stay as length for consistency.

sgtm, I just found that we also use "length" in SendKeyMessage :(

[DaleCurtis, 2013-10-22 19:35:08]

Committed patchset #2 manually as r230180 (presubmit successful).