Move MessageLocations constructor out of header

Move MessageLocations constructor out of header

This is a bit unfortunate, but otherwise we would have to include
objects.h before message.h, since for the initialization of a Handle<T>,
the compiler checks that Object* can be assigned to T*. So it would
need to know about the inheritance for initializing Handle<Script> and
Handle<JSFunction>.

R=mstarzinger@chromium.org, titzer@chromium.org
BUG=

Committed: https://crrev.com/e2ec04d2ab020d5f70f7ca5861ca50f6b86a8a80
Cr-Commit-Position: refs/heads/master@{#35626}

[Michael Starzinger, 2016-04-11 13:58:19]

https://codereview.chromium.org/1872373002/diff/1/src/messages.h
File src/messages.h (right):

https://codereview.chromium.org/1872373002/diff/1/src/messages.h#newcode28
src/messages.h:28: MessageLocation(Handle<Script> script, int start_pos, int
end_pos,
If it is just this constructor that requires the include, then I would vote for
moving the constructor implementation into message.cc and forward declare the
necessary "class Script" and "class JSFunction" instead.

[Clemens Hammacher, 2016-04-11 14:08:47]

https://codereview.chromium.org/1872373002/diff/1/src/messages.h
File src/messages.h (right):

https://codereview.chromium.org/1872373002/diff/1/src/messages.h#newcode28
src/messages.h:28: MessageLocation(Handle<Script> script, int start_pos, int
end_pos,
On 2016/04/11 13:58:18, Michael Starzinger wrote:
> If it is just this constructor that requires the include, then I would vote
for
> moving the constructor implementation into message.cc and forward declare the
> necessary "class Script" and "class JSFunction" instead.

It's about the default constructor below, but yes, it probably makes more sense
to just move this constructor out of the header. I'll update the patch in a
second.

[Clemens Hammacher, 2016-04-11 14:10:18]

Description was changed from

==========
Include objects.h from messages.h

Having to include the whole objects header here is seems bad, but it is
necessary for the initializiation of the Handle<Script> and
Handle<JSFunction> in MessageLocation. The compiler checks that Object*
can be assigned to T* when initializing Handle<T>.
This would trigger an error whenever messages.h is included before any
other header.

R=mstarzinger@chromium.org, titzer@chromium.org
BUG=
==========

to

==========
Move MessageLocation constructor out of header

This is a bit unfortunate, but otherwise we would have to include
objects.h before message.h, since for the initialization of a Handle<T>,
the compiler checks that Object* can be assigned to T*. So it would
need to know about the inheritance for initializing Handle<Script> and
Handle<JSFunction>.


R=mstarzinger@chromium.org, titzer@chromium.org
BUG=
==========

[Michael Starzinger, 2016-04-11 14:49:46]

LGTM.

[Clemens Hammacher, 2016-04-11 15:28:20]

Description was changed from

==========
Move MessageLocation constructor out of header

This is a bit unfortunate, but otherwise we would have to include
objects.h before message.h, since for the initialization of a Handle<T>,
the compiler checks that Object* can be assigned to T*. So it would
need to know about the inheritance for initializing Handle<Script> and
Handle<JSFunction>.


R=mstarzinger@chromium.org, titzer@chromium.org
BUG=
==========

to

==========
Move MessageLocations constructor out of header

This is a bit unfortunate, but otherwise we would have to include
objects.h before message.h, since for the initialization of a Handle<T>,
the compiler checks that Object* can be assigned to T*. So it would
need to know about the inheritance for initializing Handle<Script> and
Handle<JSFunction>.


R=mstarzinger@chromium.org, titzer@chromium.org
BUG=
==========

[Clemens Hammacher, 2016-04-12 15:26:08]

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

[commit-bot: I haz the power, 2016-04-12 15:26:13]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1872373002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1872373002/100001

[Clemens Hammacher, 2016-04-12 15:28:11]

The CQ bit was unchecked by clemensh@chromium.org

[titzer, 2016-04-18 16:22:58]

On 2016/04/12 15:26:13, commit-bot: I haz the power wrote:
> Dry run: CQ is trying da patch. Follow status at
>  https://chromium-cq-status.appspot.com/patch-status/1872373002/100001
> View timeline at
>  https://chromium-cq-status.appspot.com/patch-timeline/1872373002/100001

lgtm

[Clemens Hammacher, 2016-04-19 14:48:08]

The CQ bit was checked by clemensh@chromium.org

[Clemens Hammacher, 2016-04-19 14:48:09]

The patchset sent to the CQ was uploaded after l-g-t-m from
mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1872373002/#ps100001
(title: "rebase")

[commit-bot: I haz the power, 2016-04-19 14:48:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1872373002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1872373002/100001

[commit-bot: I haz the power, 2016-04-19 15:13:15]

Committed patchset #6 (id:100001)

[nodir1, 2016-04-22 18:43:00]

Description was changed from

==========
Move MessageLocations constructor out of header

This is a bit unfortunate, but otherwise we would have to include
objects.h before message.h, since for the initialization of a Handle<T>,
the compiler checks that Object* can be assigned to T*. So it would
need to know about the inheritance for initializing Handle<Script> and
Handle<JSFunction>.


R=mstarzinger@chromium.org, titzer@chromium.org
BUG=
==========

to

==========
Move MessageLocations constructor out of header

This is a bit unfortunate, but otherwise we would have to include
objects.h before message.h, since for the initialization of a Handle<T>,
the compiler checks that Object* can be assigned to T*. So it would
need to know about the inheritance for initializing Handle<Script> and
Handle<JSFunction>.

R=mstarzinger@chromium.org, titzer@chromium.org
BUG=

Committed: https://crrev.com/e2ec04d2ab020d5f70f7ca5861ca50f6b86a8a80
Cr-Commit-Position: refs/heads/master@{#35626}
==========

[nodir1, 2016-04-22 18:43:01]

Patchset 6 (id:??) landed as
https://crrev.com/e2ec04d2ab020d5f70f7ca5861ca50f6b86a8a80
Cr-Commit-Position: refs/heads/master@{#35626}