Mixed Content: Treat some Chrome schemes as restricting mixed content.

Mixed Content: Treat some Chrome schemes as restricting mixed content.

We should not load insecure content into documents under the following
schemes:

* chrome:
* chrome-search:
* chrome-extension-resource:
* chrome-devtools:

This patch registers those schemes as restricting mixed content, which
means the renderer will treat them similarly to 'https:' URLs with
regard to insecure subresource requests.

BUG=443597

[Mike West, 2014-12-19 10:24:06]

mkwst@chromium.org changed reviewers:
+ jochen@chromium.org

[Mike West, 2014-12-19 10:24:06]

Jochen, WDYT? I don't think we can write layout tests for this... is there a
reasonable way to test on the Chromium side?

[jochen (gone - plz use gerrit), 2014-12-19 10:31:29]

not sure what chrome-devtools: is used for?

can chrome-extension-resource: serve entire iframes? I'd expect that restricting
this will break stuff, no?

is chrome-search: only used for google, or also other search engines that might
be http?

[Mike West, 2014-12-19 10:35:59]

On 2014/12/19 10:31:29, jochen (slow) wrote:
> not sure what chrome-devtools: is used for?

The devtools interface. All that stuff down at the bottom of the window when you
open the inspector is served from this scheme. (+pfeldman)

> can chrome-extension-resource: serve entire iframes? I'd expect that
restricting
> this will break stuff, no?

It doesn't look like we serve anything other than *.js over this scheme. I don't
know if we have any resources meant to be served as full frame documents, but if
we do, restricting it doesn't seem unreasonable.

> is chrome-search: only used for google, or also other search engines that
might
> be http?

I really hope we don't allow the NTP to serve http from any search engine. :(

[Mike West, 2014-12-19 10:36:11]

mkwst@chromium.org changed reviewers:
+ pfeldman@chromium.org

[Mike West, 2014-12-19 10:36:12]

er. +pfeldman for real.

[Mike West, 2014-12-19 10:38:34]

On 2014/12/19 10:31:29, jochen (slow) wrote:
> is chrome-search: only used for google, or also other search engines that
might
> be http?

The only search engine other than Google that takes over the NTP is Bing
(according to
https://code.google.com/p/chromium/codesearch#chromium/src/components/search_...),
and its URLs are HTTPS. Given that we enable some APIs there that we don't have
elsewhere, it seems reasonable to lock it down a bit.

[Mike West, 2014-12-19 10:39:15]

mkwst@chromium.org changed reviewers:
+ kalman@chromium.org

[Mike West, 2014-12-19 10:39:15]

On 2014/12/19 10:35:59, Mike West wrote:
> > can chrome-extension-resource: serve entire iframes? I'd expect that
> restricting
> > this will break stuff, no?
> 
> It doesn't look like we serve anything other than *.js over this scheme. I
don't
> know if we have any resources meant to be served as full frame documents, but
if
> we do, restricting it doesn't seem unreasonable.

+kalman

[Mike West, 2014-12-19 13:42:19]

Lots of ChromeOS test failures, loading `http://127.0.0.1` inside `chrome://`
for login. I hope that's just a problem with the tests spinning up a fake gaia
server locally, and not representative of how we do login.

[Mike West, 2014-12-19 13:43:09]

+bartfab: Can you point me to someone to chat with about the login tests?

[bartfab (slow), 2014-12-19 13:49:25]

On 2014/12/19 13:43:09, Mike West wrote:
> +bartfab: Can you point me to someone to chat with about the login tests?

rogerta@ and guohui@ are good contacts for this code.

[bartfab (slow), 2014-12-19 13:53:00]

On 2014/12/19 13:49:25, bartfab wrote:
> On 2014/12/19 13:43:09, Mike West wrote:
> > +bartfab: Can you point me to someone to chat with about the login tests?
> 
> rogerta@ and guohui@ are good contacts for this code.

By the way: I wrote an HTTPSForwarder class
(chrome/browser/chromeos/login/test/https_forwarder.h) that I use in
saml_browsertest.cc. It serves as a wrapper around FakeGaia, allowing Chrome to
access FakeGaia over HTTPS in tests.

[not at google - send to devlin, 2014-12-19 20:39:22]

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

[pfeldman, 2015-05-15 23:44:16]

pfeldman@chromium.org changed reviewers:
- pfeldman@chromium.org

[jochen (gone - plz use gerrit), 2015-05-18 22:48:41]

https://codereview.chromium.org/809153004/diff/20001/chrome/renderer/chrome_c...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/809153004/diff/20001/chrome/renderer/chrome_c...
chrome/renderer/chrome_content_renderer_client.cc:422: // pages should not
directly embed insecure resources.
chrome: and chrome-devtools: are defined in content. I wonder whether we should
register them there?

[not at google - send to devlin, 2015-05-18 23:08:32]

kalman@chromium.org changed reviewers:
+ elijahtaylor@chromium.org

[not at google - send to devlin, 2015-05-18 23:08:33]

lgtm, FYI +elijahtaylor for the ARC question.

https://codereview.chromium.org/809153004/diff/20001/chrome/renderer/chrome_c...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/809153004/diff/20001/chrome/renderer/chrome_c...
chrome/renderer/chrome_content_renderer_client.cc:430:
extension_resource_scheme);
Double-check with the ARC people about this, they're the main clients of
chrome-extension-resource.

[not at google - send to devlin, 2015-05-18 23:09:09]

(lg for chrome-extension-resource that is)

[elijahtaylor1, 2015-05-18 23:46:33]

https://codereview.chromium.org/809153004/diff/20001/chrome/renderer/chrome_c...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/809153004/diff/20001/chrome/renderer/chrome_c...
chrome/renderer/chrome_content_renderer_client.cc:430:
extension_resource_scheme);
On 2015/05/18 23:08:33, kalman wrote:
> Double-check with the ARC people about this, they're the main clients of
> chrome-extension-resource.

I don't think this should be a problem for ARC. We actually don't use
chrome-extension-resource, shared module resources are delivered via reserved
_modules/ paths from the normal chrome-extension:// scheme.

[Mike West, 2015-05-19 08:12:06]

Patchset #3 (id:40001) has been deleted

[Mike West, 2015-05-20 05:12:34]

The CQ bit was checked by mkwst@chromium.org to run a CQ dry run

[Mike West, 2015-05-20 05:12:34]

The patchset sent to the CQ was uploaded after l-g-t-m from kalman@chromium.org
Link to the patchset: https://codereview.chromium.org/809153004/#ps60001 (title:
"Moving.")

[commit-bot: I haz the power, 2015-05-20 05:12:41]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/809153004/60001

[commit-bot: I haz the power, 2015-05-20 06:31:22]

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

[commit-bot: I haz the power, 2015-05-20 06:31:23]

Dry run: Try jobs failed on following builders:
  linux_chromium_chromeos_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[jochen (gone - plz use gerrit), 2015-05-21 07:54:16]

lgtm