Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

isPotentiallyTrustworthy() previously took an |errorMessage| argument
that it always set to the same string when it returned false. Now there
is just a static method to get that string, and callers, if they so
choose, can call the static method to populate an error message.

This is in preparation for adapting Document::isSecureContext() for
--site-per-process. In an OOPIF world, we need to replicate a flag
indicating whether a sandboxed frame's origin should be considered
potentially trustworthy, but we don't want to also replicate an error
string as well when it isn't potentially trustworthy.

BUG=571079
Committed: https://crrev.com/64b1060f5898227113951469bd35100389ad26ee
Cr-Commit-Position: refs/heads/master@{#377002}

[estark, 2016-02-22 21:50:38]

Description was changed from

==========
Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

isPotentiallyTrustworthy() previously took an |errorMessage| argument
that it always set to the same string when it returned false. Now there
is just an exported constant set to that string, and callers, if they so
choose, can use the string in an error message.

This is in preparation for adapting Document::isSecureContext() for
--site-per-process. In an OOPIF world, we need to replicate a flag
indicating whether a sandboxed frame's origin should be considered
potentially trustworthy, but we don't want to also replicate an error
string as well when it isn't potentially trustworthy.

BUG=571079
==========

to

==========
Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

isPotentiallyTrustworthy() previously took an |errorMessage| argument
that it always set to the same string when it returned false. Now there
is just a static method to get that string, and callers, if they so
choose, can call the static method to populate an error message.

This is in preparation for adapting Document::isSecureContext() for
--site-per-process. In an OOPIF world, we need to replicate a flag
indicating whether a sandboxed frame's origin should be considered
potentially trustworthy, but we don't want to also replicate an error
string as well when it isn't potentially trustworthy.

BUG=571079
==========

[estark, 2016-02-23 06:01:46]

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

[estark, 2016-02-23 06:01:47]

Hi Mike -- what do you think of removing the |errorMessage| argument to
SecurityOrigin::isPotentiallyTrustworthy() and replacing it with a static method
that callers can use to get the error message? See the CL description for why I
want to do this; it simplifies the OOPIF-ification of
Document::isSecureContext(). The follow-up CL
https://codereview.chromium.org/1723753002/ might also help make it clear.

(If you're not on board with this change, though, I think that's okay and I can
still make isSecureContext() OOPIF-friendly... it might just be a little
clunkier.)

Thanks!

[Mike West, 2016-02-23 09:10:37]

On 2016/02/23 at 06:01:47, estark wrote:
> Hi Mike -- what do you think of removing the |errorMessage| argument to
SecurityOrigin::isPotentiallyTrustworthy() and replacing it with a static method
that callers can use to get the error message? See the CL description for why I
want to do this; it simplifies the OOPIF-ification of
Document::isSecureContext(). The follow-up CL
https://codereview.chromium.org/1723753002/ might also help make it clear.
> 
> (If you're not on board with this change, though, I think that's okay and I
can still make isSecureContext() OOPIF-friendly... it might just be a little
clunkier.)
> 
> Thanks!

I'm fine with this: LGTM.

The original thought was that we'd have more specific error messages to explain
_why_ the context wasn't secure, but that never happened (and doesn't look like
it needs to happen (and if it does need to happen, we could do it with the
static method)).

[estark, 2016-02-23 16:15:27]

The CQ bit was checked by estark@chromium.org

[commit-bot: I haz the power, 2016-02-23 16:16:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1716303002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1716303002/40001

[commit-bot: I haz the power, 2016-02-23 16:25:02]

Description was changed from

==========
Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

isPotentiallyTrustworthy() previously took an |errorMessage| argument
that it always set to the same string when it returned false. Now there
is just a static method to get that string, and callers, if they so
choose, can call the static method to populate an error message.

This is in preparation for adapting Document::isSecureContext() for
--site-per-process. In an OOPIF world, we need to replicate a flag
indicating whether a sandboxed frame's origin should be considered
potentially trustworthy, but we don't want to also replicate an error
string as well when it isn't potentially trustworthy.

BUG=571079
==========

to

==========
Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

isPotentiallyTrustworthy() previously took an |errorMessage| argument
that it always set to the same string when it returned false. Now there
is just a static method to get that string, and callers, if they so
choose, can call the static method to populate an error message.

This is in preparation for adapting Document::isSecureContext() for
--site-per-process. In an OOPIF world, we need to replicate a flag
indicating whether a sandboxed frame's origin should be considered
potentially trustworthy, but we don't want to also replicate an error
string as well when it isn't potentially trustworthy.

BUG=571079
==========

[commit-bot: I haz the power, 2016-02-23 16:25:03]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-02-23 16:25:49]

Description was changed from

==========
Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

isPotentiallyTrustworthy() previously took an |errorMessage| argument
that it always set to the same string when it returned false. Now there
is just a static method to get that string, and callers, if they so
choose, can call the static method to populate an error message.

This is in preparation for adapting Document::isSecureContext() for
--site-per-process. In an OOPIF world, we need to replicate a flag
indicating whether a sandboxed frame's origin should be considered
potentially trustworthy, but we don't want to also replicate an error
string as well when it isn't potentially trustworthy.

BUG=571079
==========

to

==========
Remove SecurityOrigin::isPotentiallyTrustworthy() |errorMessage|

isPotentiallyTrustworthy() previously took an |errorMessage| argument
that it always set to the same string when it returned false. Now there
is just a static method to get that string, and callers, if they so
choose, can call the static method to populate an error message.

This is in preparation for adapting Document::isSecureContext() for
--site-per-process. In an OOPIF world, we need to replicate a flag
indicating whether a sandboxed frame's origin should be considered
potentially trustworthy, but we don't want to also replicate an error
string as well when it isn't potentially trustworthy.

BUG=571079

Committed: https://crrev.com/64b1060f5898227113951469bd35100389ad26ee
Cr-Commit-Position: refs/heads/master@{#377002}
==========

[commit-bot: I haz the power, 2016-02-23 16:25:49]

Patchset 3 (id:??) landed as
https://crrev.com/64b1060f5898227113951469bd35100389ad26ee
Cr-Commit-Position: refs/heads/master@{#377002}