A bunch of remaining parts of extension content verification

A bunch of remaining parts of extension content verification

-The real guts of content_hash_fetcher.cc, which fetches the
 verified_contents.json file from the webstore if needed and also runs
 tasks to compute and cache the block-level hashes of all files in an
 extension.

-The real guts of content_hash_reader.cc, which uses the work done by
 the content_hash_fetcher during validation of extension file content as
 it's read off of disk at time of use.

-Code to avoid verifying transcoded files (images used in browser
 process, and message catalogs).

-Don't allow downgrade of mode via kForceFieldTrials command line switch

-Various bits of plumbing to support all of the above

BUG=369895
R=rockot@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=270694

[asargent_no_longer_on_chrome, 2014-05-14 23:55:40]

Sorry for churn in middle of review, but I uploaded a new patch with a small fix
in one file. 

In testing against prod I realized the way we were calculating the url for
fetching verified contents wasn't working (I could have sworn it was for
staging, so maybe the omaha server configs are a little different or something
and the staging ones ignore some extra unneeded parameters we were sending that
the prod ones don't).

[Ken Rockot(use gerrit already), 2014-05-15 00:51:25]

lgtm

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/comp...
File extensions/browser/computed_hashes.h (right):

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/comp...
extensions/browser/computed_hashes.h:24: class Reader {
nit: I guess I'd rather see separate classes in separate source modules for
ComputedHashesReader and ComputedHashesWriter, but I don't feel strongly about
it.

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
File extensions/browser/content_hash_fetcher.cc (right):

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
extensions/browser/content_hash_fetcher.cc:417: jobs_.erase(key);
nit: jobs_.erase(found) instead

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
File extensions/browser/content_hash_reader.cc (right):

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
extensions/browser/content_hash_reader.cc:98: 
nit: Maybe worth a DCHECK(block_index >= 0)

[asargent_no_longer_on_chrome, 2014-05-15 04:02:48]

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/comp...
File extensions/browser/computed_hashes.h (right):

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/comp...
extensions/browser/computed_hashes.h:24: class Reader {
On 2014/05/15 00:51:25, Ken Rockot wrote:
> nit: I guess I'd rather see separate classes in separate source modules for
> ComputedHashesReader and ComputedHashesWriter, but I don't feel strongly about
> it.

I filed crbug.com/373597 for looking into doing this as a follow-up. One
advantage I saw of doing it this way was that I could have the key name
constants in one anonymous namespace in the shared .cc file. OTOH, most
consumers are typically only using one or the other, so having them declared in
the same header causes unnecessary code inclusion.

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
File extensions/browser/content_hash_fetcher.cc (right):

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
extensions/browser/content_hash_fetcher.cc:417: jobs_.erase(key);
On 2014/05/15 00:51:25, Ken Rockot wrote:
> nit: jobs_.erase(found) instead

Ooh, nice catch. Done.

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
File extensions/browser/content_hash_reader.cc (right):

https://codereview.chromium.org/289533003/diff/120001/extensions/browser/cont...
extensions/browser/content_hash_reader.cc:98: 
On 2014/05/15 00:51:25, Ken Rockot wrote:
> nit: Maybe worth a DCHECK(block_index >= 0)

Done.

[asargent_no_longer_on_chrome, 2014-05-15 04:03:04]

The CQ bit was checked by asargent@chromium.org

[commit-bot: I haz the power, 2014-05-15 04:03:59]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/asargent@chromium.org/289533003/130017

[commit-bot: I haz the power, 2014-05-15 05:44:20]

FYI, CQ is re-trying this CL (attempt #1).
Please consider checking whether the failures are real,
and report flakes to chrome-troopers@google.com.
The failing builders are:
  ios_dbg_simulator on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/ios_dbg_simulator/bui...)

[commit-bot: I haz the power, 2014-05-15 05:47:45]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-05-15 05:47:46]

Try jobs failed on following builders:
  ios_dbg_simulator on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/ios_dbg_simulator/bui...)

[USE_CHROMIUM_ACCT_INSTEAD, 2014-05-15 13:04:24]

The CQ bit was checked by asargent@google.com

[commit-bot: I haz the power, 2014-05-15 13:05:10]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/asargent@chromium.org/289533003/130017

[asargent_no_longer_on_chrome, 2014-05-15 17:02:44]

Committed patchset #3 manually as r270694 (presubmit successful).