Fix content verification failures on unusual content script paths

Fix content verification failures on unusual content script paths

If content script paths had a leading "./" in them, we'd fail to find
expected hashes for them because we weren't properly stripping the
leading junk from the relative path.

BUG=437675
Committed: https://crrev.com/73d92c7c8328979eb489a8af59b8d61940a296c6
Cr-Commit-Position: refs/heads/master@{#348207}

[asargent_no_longer_on_chrome, 2015-09-08 23:21:42]

asargent@chromium.org changed reviewers:
+ rockot@chromium.org

[asargent_no_longer_on_chrome, 2015-09-08 23:21:42]

https://codereview.chromium.org/1315713009/diff/1/chrome/test/data/extensions...
File
chrome/test/data/extensions/content_verifier/dot_slash_paths/_metadata/verified_contents.json
(left):

https://codereview.chromium.org/1315713009/diff/1/chrome/test/data/extensions...
chrome/test/data/extensions/content_verifier/dot_slash_paths/_metadata/verified_contents.json:1:
[{"description":"treehash per
file","signed_content":{"payload":"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","signatures":[{"header":{"kid":"publisher"},"protected":"eyJhbGciOiJSUzI1NiJ9","signature":"nsuqgedzxEY0hfCqN9xHmKrIhXgIFljzTg8hFTiOnh2HhKr3I1HwhcBDbVSQ9Q0sYXW-lQTXHWR8cfcwABk4UbSQfhBa2Z2y9qoe-nyApf49anPVPG2SHbtBujYPF9qjSL0uGQpOFUSvICTiaEHl5pTUv1DvMOyiQ17JytqHaHF_towTgbSq-_b_D_8sK755S58L0BuJUhc31EJZ5llIwA1BlC8omQDpbch8QSzUQ6E0lLzK7MVkQRhVeK4bo5iMkbkQhPoTJDDFzAYwN2eafucfaZSWAADzhEp5oZbof-2QPzsuBkdOPD3F-LJ0cZ7JnW8shm1d6J9kD9PKxcreFw"},{"header":{"kid":"webstore"},"protected":"eyJhbGciOiJSUzI1NiJ9","signature":"CTCIYz86Sh043zJuIpW246A1IH64nZ2mc8zFZk4JqlWNBrkzQi38QRmSM5fKtwGPRjxLKXLsJ1LMV4HRIQHbAAJ3r4DoGF85vpJ3YgnAmFLt3uEfKC4qu8SWDVY5Ad5vZbGf1BjvTljqZJVqPjEvSQmWaNYqTKQs_3Sviil994Nq6M9tChkcQzUjZOBQSz0MTBXeYtSkla80V13VvUJ204vv_hIg0GW3x0opQ70r8qXa25aEB1zPNUavi844IgUM5fdLirMprYCjxn47Qby7MxEIchX3b9cQDiOhqFcqluD5b_s3e3fXTkAaHnc4e5GSv2SPZb10GQH97U-_5gC8tw"}]}}]
FYI, I removed this file because I had never really intended to check it in in
the first place - when I initially added this code I uploaded the extension to
the webstore, then downloaded it and just unzipped it to add to the repository.
(These files are added during the webstore's .crx packaging step, and contain
the signed expected hashes)

[Ken Rockot(use gerrit already), 2015-09-09 16:56:10]

lgtm

https://codereview.chromium.org/1315713009/diff/1/extensions/browser/content_...
File extensions/browser/content_verifier.cc (right):

https://codereview.chromium.org/1315713009/diff/1/extensions/browser/content_...
extensions/browser/content_verifier.cc:65: static base::FilePath
MakePathRelative(const base::FilePath& path) {
nit since you're moving this, might as well nix the "static" and move it to an
anonymous namespace instead

https://codereview.chromium.org/1315713009/diff/1/extensions/browser/content_...
extensions/browser/content_verifier.cc:90: const base::FilePath& relative_path)
{
nit: Maybe instead of calling this |relative_path| and the real one
|actual_relative_path|, the real one could be changed to |relative_path| and
this could be something like |path_spec|? Since it's not really the relative
path, right?

[asargent_no_longer_on_chrome, 2015-09-09 17:42:39]

https://codereview.chromium.org/1315713009/diff/1/extensions/browser/content_...
File extensions/browser/content_verifier.cc (right):

https://codereview.chromium.org/1315713009/diff/1/extensions/browser/content_...
extensions/browser/content_verifier.cc:65: static base::FilePath
MakePathRelative(const base::FilePath& path) {
On 2015/09/09 16:56:10, Ken Rockot wrote:
> nit since you're moving this, might as well nix the "static" and move it to an
> anonymous namespace instead

Done. Renamed to "NormalizeRelativePath" also.

https://codereview.chromium.org/1315713009/diff/1/extensions/browser/content_...
extensions/browser/content_verifier.cc:90: const base::FilePath& relative_path)
{
On 2015/09/09 16:56:10, Ken Rockot wrote:
> nit: Maybe instead of calling this |relative_path| and the real one
> |actual_relative_path|, the real one could be changed to |relative_path| and
> this could be something like |path_spec|? Since it's not really the relative
> path, right?

I sort of like calling the argument relative_path since that provides a clue
that the data supplied will be interpreted as a relative path, and "path_spec"
makes me think of absolute paths since GURL::spec() gives you the text of the
full URL. 

I renamed "actual_relative_path" to "normalized_path" instead.

[asargent_no_longer_on_chrome, 2015-09-09 17:42:45]

The CQ bit was checked by asargent@chromium.org

[asargent_no_longer_on_chrome, 2015-09-09 17:42:45]

The patchset sent to the CQ was uploaded after l-g-t-m from rockot@chromium.org
Link to the patchset: https://codereview.chromium.org/1315713009/#ps20001
(title: "address nits, autoformat")

[commit-bot: I haz the power, 2015-09-09 17:43:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1315713009/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1315713009/20001

[commit-bot: I haz the power, 2015-09-09 18:09:00]

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

[commit-bot: I haz the power, 2015-09-09 18:09:01]

Try jobs failed on following builders:
  win8_chromium_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_ng/...)

[asargent_no_longer_on_chrome, 2015-09-10 18:03:45]

The CQ bit was checked by asargent@chromium.org

[asargent_no_longer_on_chrome, 2015-09-10 18:03:46]

The patchset sent to the CQ was uploaded after l-g-t-m from rockot@chromium.org
Link to the patchset: https://codereview.chromium.org/1315713009/#ps40001
(title: "fix windows compile error")

[commit-bot: I haz the power, 2015-09-10 18:03:56]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1315713009/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1315713009/40001

[commit-bot: I haz the power, 2015-09-10 19:20:20]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-09-10 19:22:26]

Patchset 3 (id:??) landed as
https://crrev.com/73d92c7c8328979eb489a8af59b8d61940a296c6
Cr-Commit-Position: refs/heads/master@{#348207}

[commit-bot: I haz the power, 2015-09-23 12:13:01]

Patchset 3 (id:??) landed as
https://crrev.com/73d92c7c8328979eb489a8af59b8d61940a296c6
Cr-Commit-Position: refs/heads/master@{#348207}