Add cdm::ContentDecryptionModule_8

Add cdm::ContentDecryptionModule_8

Changes:
- Add new Initialize() method
- Add additional KeyStatus values
- Add new InitDataType enum and use in CreateSessionAndGenerateRequest()

BUG=450861
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=294503

[jrummell, 2015-03-18 21:46:06]

jrummell@chromium.org changed reviewers:
+ ddorwin@chromium.org, sandersd@chromium.org, xhwang@chromium.org

[jrummell, 2015-03-18 21:46:06]

PTAL. PS#1 is a clone of CDM_7 renamed, PS#2 is the new stuff.

[ddorwin, 2015-03-19 02:42:00]

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
File content_decryption_module.h (right):

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:828: // Lets the CDM know whether distinctive
identifiers are allowed or should
If that's what it did, we would call it Set...(). We could do that, but perhaps
there are other reasons a CDM might want to be initialized beyond the instance
creation. (Note that this cannot fail as defined.)

If we keep this, it should be something like:
// Initializes the CDM instance, providing information about permitted
functionalities.

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:830: virtual void Initialize(bool
allow_distinctive_identifier) = 0;
Should we have allow_persistent_state rather than relying on CdmFileIo failing?

[xhwang, 2015-03-19 17:16:23]

just nits....

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
File content_decryption_module.h (right):

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:311: kStatusPending = 5
nit: The order of these values are different from the spec. Is this for
historical reasons? Not a big deal though.

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:356: kWebm = 2
s/Webm/WebM/

[jrummell, 2015-03-19 19:08:50]

Updated.

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
File content_decryption_module.h (right):

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:311: kStatusPending = 5
On 2015/03/19 17:16:22, xhwang wrote:
> nit: The order of these values are different from the spec. Is this for
> historical reasons? Not a big deal though.

I don't want to change the existing 0..3 due to existing CDMs using them.

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:356: kWebm = 2
On 2015/03/19 17:16:22, xhwang wrote:
> s/Webm/WebM/

Done.

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:828: // Lets the CDM know whether distinctive
identifiers are allowed or should
On 2015/03/19 02:42:00, ddorwin wrote:
> If that's what it did, we would call it Set...(). We could do that, but
perhaps
> there are other reasons a CDM might want to be initialized beyond the instance
> creation. (Note that this cannot fail as defined.)
> 
> If we keep this, it should be something like:
> // Initializes the CDM instance, providing information about permitted
> functionalities.

Comment updated.

https://codereview.chromium.org/1021603002/diff/20001/content_decryption_modu...
content_decryption_module.h:830: virtual void Initialize(bool
allow_distinctive_identifier) = 0;
On 2015/03/19 02:42:00, ddorwin wrote:
> Should we have allow_persistent_state rather than relying on CdmFileIo
failing?

Done. The CDM may want to know this rather than waiting for the FileIO to fail.

[xhwang, 2015-03-19 19:15:38]

looking good. Shall we drop CDM6 support as well?

[ddorwin, 2015-03-19 19:16:18]

LG

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
File content_decryption_module.h (right):

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
content_decryption_module.h:465: class ContentDecryptionModule_6 {
As discussed, we can remove this one.

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
content_decryption_module.h:833: // If |allow_persistent_state| is false, only
"temporary" sessions may be
FYI, Chrome should be enforcing the session type requirements, but it's fine for
the CDM to check too I guess.

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
content_decryption_module.h:834: // created. Attempts by the CDM to use
CreateFileIO() will always fail.
// If |allow_persistent_state| is false, the CDM must not attempt to persist
state. Calls to CreateFileIO() will fail.

[jrummell, 2015-03-19 20:05:50]

Updated.

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
File content_decryption_module.h (right):

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
content_decryption_module.h:465: class ContentDecryptionModule_6 {
On 2015/03/19 19:16:18, ddorwin wrote:
> As discussed, we can remove this one.

Will do as a separate CL.

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
content_decryption_module.h:833: // If |allow_persistent_state| is false, only
"temporary" sessions may be
On 2015/03/19 19:16:18, ddorwin wrote:
> FYI, Chrome should be enforcing the session type requirements, but it's fine
for
> the CDM to check too I guess.

Acknowledged.

https://codereview.chromium.org/1021603002/diff/40001/content_decryption_modu...
content_decryption_module.h:834: // created. Attempts by the CDM to use
CreateFileIO() will always fail.
On 2015/03/19 19:16:18, ddorwin wrote:
> // If |allow_persistent_state| is false, the CDM must not attempt to persist
> state. Calls to CreateFileIO() will fail.

Done.

[ddorwin, 2015-03-19 21:27:28]

lgtm

[jrummell, 2015-03-19 21:28:02]

The CQ bit was checked by jrummell@chromium.org

[commit-bot: I haz the power, 2015-03-19 21:28:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1021603002/60001

[commit-bot: I haz the power, 2015-03-19 21:29:09]

Committed patchset #4 (id:60001) as
http://src.chromium.org/viewvc/chrome?view=rev&revision=294503