Register O3D/GTalk PPAPI plugins if available

Register O3D/GTalk PPAPI plugins if available


BUG=97848
TEST=manual. Check chrome:plugins, run gtalk if plugins available


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=114487

[Josh Horwich, 2011-12-12 18:35:52]

Please review - I picked you as a reviewer based on git blame output, so if
someone else is best, please let me know.

Background on this change is to try to register PPAPI plugins for GTalk and O3D
on ChromeOS and Linux - this is needed for a sans-GTK browser such as when
running under Aura.

[jam, 2011-12-12 23:09:51]

i'm ooo. can you please send this to someone else?

[Josh Horwich, 2011-12-13 01:18:24]

Antoine,
  Can you please review my change, or recommend me the correct reviewer if you
would like someone else to do it? This is to register my PPAPI plugins for Aura
builds (linux/ChromeOS only for now). Thank you!

- Josh

[piman, 2011-12-13 01:48:02]

If possible I would really try to avoid the ifdef madness - especially between
linux and Chrome OS, where I don't think a difference is warranted.

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_content_cl...
File chrome/common/chrome_content_client.cc (right):

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_content_cl...
chrome/common/chrome_content_client.cc:58: const char kGTO3DPluginMimeType[] =
"application/vnd.gtpo3d.auto";
Do we really need different mime types? Can't we just pick one that's always the
right one for the pepper plugin?
vnd.o3d.auto vs vnd.gtpo3d.auto is historical to avoid a conflict between the
old O3D plugin and the new GTalk-special O3D plugin. But in Pepper land, there
will never be the former.

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_paths.cc
File chrome/common/chrome_paths.cc (right):

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_paths.cc#n...
chrome/common/chrome_paths.cc:69:
FILE_PATH_LITERAL("/opt/google/o3d/libppo3dautoplugin.so");
Do we need a different path on Chrome OS vs non-Chrome OS?
Do we need the path for these plugins to be different than the other pepper
plugins?
Why not simply always install them in /opt/google/chrome/pepper with the rest,
and remove these ifdefs?

[Josh Horwich, 2011-12-13 22:06:30]

Please review - I simplified the registration of O3D/GTalk plugins as per your
feedback.

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_content_cl...
File chrome/common/chrome_content_client.cc (right):

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_content_cl...
chrome/common/chrome_content_client.cc:58: const char kGTO3DPluginMimeType[] =
"application/vnd.gtpo3d.auto";
On 2011/12/13 01:48:02, piman wrote:
> Do we really need different mime types? Can't we just pick one that's always
the
> right one for the pepper plugin?
> vnd.o3d.auto vs vnd.gtpo3d.auto is historical to avoid a conflict between the
> old O3D plugin and the new GTalk-special O3D plugin. But in Pepper land, there
> will never be the former.

Done. I've consolidated it to one MIME type now.

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_paths.cc
File chrome/common/chrome_paths.cc (right):

http://codereview.chromium.org/8899014/diff/1/chrome/common/chrome_paths.cc#n...
chrome/common/chrome_paths.cc:69:
FILE_PATH_LITERAL("/opt/google/o3d/libppo3dautoplugin.so");
On 2011/12/13 01:48:02, piman wrote:
> Do we need a different path on Chrome OS vs non-Chrome OS?
> Do we need the path for these plugins to be different than the other pepper
> plugins?
> Why not simply always install them in /opt/google/chrome/pepper with the rest,
> and remove these ifdefs?

Done - I moved it to the pepper directory on all Linux flavors. I also kept it
inside of the POSIX && !MACOSX conditional so we don't waste time during startup
calling PathExists on those platforms.

[piman, 2011-12-14 00:55:21]

lgtm

[commit-bot: I haz the power, 2011-12-14 18:25:07]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jhorwich@chromium.org/8899014/11001

[commit-bot: I haz the power, 2011-12-14 20:36:14]

Change committed as 114487