html_viewer: Move webcrypto to a place where html_viewer can use it.

html_viewer: Move webcrypto to a place where html_viewer can use it.

A good chunk of the web (Facebook and YouTube) require webcrypto. This
takes the code in the content/ directory, puts it in a new top level
component directory, and uses it from html_viewer.

BUG=476568
TBR=jochen

Committed: https://crrev.com/56f123297ad298113d9a2dc6ed4b7de37dad9e3b
Cr-Commit-Position: refs/heads/master@{#325571}

[Elliot Glaysher, 2015-04-10 22:12:04]

erg@chromium.org changed reviewers:
+ ben@chromium.org, jam@chromium.org, sky@chromium.org

[Elliot Glaysher, 2015-04-10 22:12:05]

Let's have a fight about naming.

AFAIK, we don't want to put content/ stuff in components/. We also don't want to
have html_viewer depend on content/. We need a toplevel directory to hold
componentized code from content/ in a place where html_viewer can access it.

I asked Ben for a name, he said "things," probably as a joke. Thus, I have used
"things" as a placeholder to get this discussion going.

This patch works as is, but requires more love before it is in a commitable
state, but I need y'all to agree on a name before I move forward.

[Ben Goodger (Google), 2015-04-10 22:26:26]

Here is my backstory for src/things. I'm only half joking. 

Consider src/components. It is a directory that:

- has a name that sound like a generic storage place for random crap
- has specific rules about what can actually go in it

In this case, we have some random crap that doesn't fit these rules. What does
one do?

#1 create a toplevel specifically for this, src/webcrypto. Seems undesirable as
in time the number of such directories is likely to be large.
#2 create a toplevel that specifically describes the nature of random crap like
this src/content_can_depend_on_it_but_cant_depend_on_content o_o;;;
#3 create a toplevel with another generic name and rely on DEPS/yelling at
people to enforce. => hence src/things.

[jam, 2015-04-13 14:34:26]

On 2015/04/10 22:26:26, Ben Goodger (Google) wrote:
> Here is my backstory for src/things. I'm only half joking. 
> 
> Consider src/components. It is a directory that:
> 
> - has a name that sound like a generic storage place for random crap
> - has specific rules about what can actually go in it
> 
> In this case, we have some random crap that doesn't fit these rules. What does
> one do?
> 
> #1 create a toplevel specifically for this, src/webcrypto. Seems undesirable
as
> in time the number of such directories is likely to be large.
> #2 create a toplevel that specifically describes the nature of random crap
like
> this src/content_can_depend_on_it_but_cant_depend_on_content o_o;;;
> #3 create a toplevel with another generic name and rely on DEPS/yelling at
> people to enforce. => hence src/things.

I wonder if mandoline is a good reason to consider redefining what "components"
is? i.e. we punted on that when we first created src/components and said that
they all depend on content, and ignored components that might only depend on
base or net. what if we had content depends on some modules in src/components,
and used DEPS rules to enforces that these directories don't depend on content?
the motivation would be that we wouldn't have two top level directories with
similar purposes.

[Elliot Glaysher, 2015-04-14 01:06:26]

I have moved things to components/ and have gotten the gyp build working, too.

[jam, 2015-04-14 14:47:12]

Can you send a PSA to chromium-dev to update the policy on src\components? i.e.
something to the effect that we have changed the rule that content can't depend
on components because of the new html_viewer work. only shared code between
these two will go in components\ and content will depend on it.

https://codereview.chromium.org/1077273002/diff/50001/components/webcrypto/al...
File components/webcrypto/algorithm_implementation.h (right):

https://codereview.chromium.org/1077273002/diff/50001/components/webcrypto/al...
components/webcrypto/algorithm_implementation.h:5: #ifndef
CONTENT_CHILD_WEBCRYPTO_ALGORITHM_IMPLEMENTATION_H_
nit: please update header guards

https://codereview.chromium.org/1077273002/diff/50001/components/webcrypto/al...
components/webcrypto/algorithm_implementation.h:14: namespace components {
the convention for code in components\foo\ is that the "foo" namespace is use
(not nested in "components")

https://codereview.chromium.org/1077273002/diff/50001/content/DEPS
File content/DEPS (right):

https://codereview.chromium.org/1077273002/diff/50001/content/DEPS#newcode58
content/DEPS:58: # Allow inclusion of specific components that we depend on.
we need a comment saying that content can only depend on components that are
shared with html_viewer

https://codereview.chromium.org/1077273002/diff/50001/content/DEPS#newcode59
content/DEPS:59: "+components/webcrypto",
since only content/child depends on this, better to have this be in
content/child/DEPS

[Elliot Glaysher, 2015-04-14 23:44:49]

https://codereview.chromium.org/1077273002/diff/60001/content/child/DEPS
File content/child/DEPS (right):

https://codereview.chromium.org/1077273002/diff/60001/content/child/DEPS#newc...
content/child/DEPS:6: "+components/tracing",
It looks like someone already beat us to depending on components/ from content/.
Should I remove the comment?

[jam, 2015-04-15 17:15:09]

lgtm

https://codereview.chromium.org/1077273002/diff/60001/content/child/DEPS
File content/child/DEPS (right):

https://codereview.chromium.org/1077273002/diff/60001/content/child/DEPS#newc...
content/child/DEPS:6: "+components/tracing",
On 2015/04/14 23:44:49, Elliot Glaysher wrote:
> It looks like someone already beat us to depending on components/ from
content/.
> Should I remove the comment?

yeah that's been the case for a while (we changed our mind about what components
 is after this landed and we never had a chance to fix this).

keep the comment as is though

[Elliot Glaysher, 2015-04-15 17:18:42]

The CQ bit was checked by erg@chromium.org

[commit-bot: I haz the power, 2015-04-15 17:19:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1077273002/70001

[Elliot Glaysher, 2015-04-15 17:19:59]

The CQ bit was unchecked by erg@chromium.org

[Elliot Glaysher, 2015-04-15 17:24:41]

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

[Elliot Glaysher, 2015-04-15 17:24:42]

+jochen for third_party/Webkit/public/platform DEPS stamp.

[Elliot Glaysher, 2015-04-16 17:06:20]

jam told me to tbr jochen

[Elliot Glaysher, 2015-04-16 17:06:35]

The CQ bit was checked by erg@chromium.org

[Elliot Glaysher, 2015-04-16 17:06:36]

The patchset sent to the CQ was uploaded after l-g-t-m from jam@chromium.org
Link to the patchset: https://codereview.chromium.org/1077273002/#ps80001
(title: "missed one.")

[commit-bot: I haz the power, 2015-04-16 17:07:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1077273002/80001

[commit-bot: I haz the power, 2015-04-16 17:18:32]

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

[commit-bot: I haz the power, 2015-04-16 17:18:34]

Try jobs failed on following builders:
  cast_shell_android on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/cast_shell_andr...)
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[jochen (gone - plz use gerrit), 2015-04-16 17:35:13]

Stampety stamp lgtm

[Elliot Glaysher, 2015-04-16 18:04:06]

The CQ bit was checked by erg@chromium.org

[Elliot Glaysher, 2015-04-16 18:04:07]

The patchset sent to the CQ was uploaded after l-g-t-m from jam@chromium.org,
jochen@chromium.org
Link to the patchset: https://codereview.chromium.org/1077273002/#ps90001
(title: "Rebase to ToT")

[commit-bot: I haz the power, 2015-04-16 18:04:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1077273002/90001

[commit-bot: I haz the power, 2015-04-17 00:52:00]

Committed patchset #10 (id:90001)

[commit-bot: I haz the power, 2015-04-17 00:53:03]

Patchset 10 (id:??) landed as
https://crrev.com/56f123297ad298113d9a2dc6ed4b7de37dad9e3b
Cr-Commit-Position: refs/heads/master@{#325571}