Chromium implementation of WebPublicSufixList.

The WebPublicSuffixList API allows embedders to tell Blink about what constitutes a top-level domain.  It is the source of truth that says that for google.co.uk, co.uk is the top-level domain, and not just .uk. Blink has, until now, been completely ignorant of this complication.

This is the third part of a two-sided patch. The tests will be on the Blink side, landing as the fourth part of the two-sided patch (once this has landed on the chrome side).


BUG=307407
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=235828

[Tom Sepez, 2013-11-05 21:03:29]

Requires webkit side changes first.  Published for the sake of completeness.

[Tom Sepez, 2013-11-15 19:49:09]

James, please review. Thanks.

[jamesr, 2013-11-15 23:13:46]

https://codereview.chromium.org/60893002/diff/130001/content/renderer/rendere...
File content/renderer/renderer_webkitplatformsupport_impl.cc (right):

https://codereview.chromium.org/60893002/diff/130001/content/renderer/rendere...
content/renderer/renderer_webkitplatformsupport_impl.cc:210:
public_suffix_list_(new WebPublicSuffixListImpl()),
why does this need heap allocation?

https://codereview.chromium.org/60893002/diff/130001/content/renderer/webpubl...
File content/renderer/webpublicsuffixlist_impl.h (right):

https://codereview.chromium.org/60893002/diff/130001/content/renderer/webpubl...
content/renderer/webpublicsuffixlist_impl.h:16: virtual size_t
getPublicSuffixLength(const blink::WebString&) OVERRIDE;
this is chromium, all parameters get names

do not use OVERRIDE when implementing an interface in a different repository.
this makes it too hard to do two-sided patches.

https://codereview.chromium.org/60893002/diff/130001/content/renderer/webpubl...
content/renderer/webpublicsuffixlist_impl.h:17: virtual
~WebPublicSuffixListImpl() OVERRIDE;
never use OVERRIDE for d'tors - it doesn't make sense

[Tom Sepez, 2013-11-15 23:22:58]

https://codereview.chromium.org/60893002/diff/130001/content/renderer/rendere...
File content/renderer/renderer_webkitplatformsupport_impl.cc (right):

https://codereview.chromium.org/60893002/diff/130001/content/renderer/rendere...
content/renderer/renderer_webkitplatformsupport_impl.cc:210:
public_suffix_list_(new WebPublicSuffixListImpl()),
On 2013/11/15 23:13:47, jamesr wrote:
> why does this need heap allocation?
fixed.

https://codereview.chromium.org/60893002/diff/130001/content/renderer/webpubl...
File content/renderer/webpublicsuffixlist_impl.h (right):

https://codereview.chromium.org/60893002/diff/130001/content/renderer/webpubl...
content/renderer/webpublicsuffixlist_impl.h:16: virtual size_t
getPublicSuffixLength(const blink::WebString&) OVERRIDE;
On 2013/11/15 23:13:47, jamesr wrote:
> this is chromium, all parameters get names
> 
> do not use OVERRIDE when implementing an interface in a different repository.
> this makes it too hard to do two-sided patches.

Done.

https://codereview.chromium.org/60893002/diff/130001/content/renderer/webpubl...
content/renderer/webpublicsuffixlist_impl.h:17: virtual
~WebPublicSuffixListImpl() OVERRIDE;
On 2013/11/15 23:13:47, jamesr wrote:
> never use OVERRIDE for d'tors - it doesn't make sense

Done.

[jamesr, 2013-11-15 23:25:33]

ok - owners LGTM but please let Adam look at this as well before landing.

[abarth-chromium, 2013-11-16 00:49:44]

LGTM

[commit-bot: I haz the power, 2013-11-18 19:08:35]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/tsepez@chromium.org/60893002/170002

[commit-bot: I haz the power, 2013-11-18 22:03:58]

Change committed as 235828