Update legacy Tuple-using code.

Update legacy Tuple-using code.

BUG=440675
TEST=no change
R=mdempsky@chromium.org, thakis@chromium.org
TBR=ben@chromium.org

Committed: https://chromium.googlesource.com/chromium/src/+/95c2a1b7d5626b287be55e41a5cfc33c6880eb46

[Avi (use Gerrit), 2014-12-22 16:35:49]

avi@chromium.org changed reviewers:
+ mdempsky@chromium.org

[Avi (use Gerrit), 2014-12-22 16:35:51]

The rest...

[Nico, 2014-12-22 17:03:08]

thakis@chromium.org changed reviewers:
+ thakis@chromium.org

[Nico, 2014-12-22 17:03:13]

awesome, lgtm!

I left a few "can possibly be a variadic template" comments below for possible
follow-up work (not necessarily by you).

https://codereview.chromium.org/821453003/diff/180001/base/memory/raw_scoped_...
File base/memory/raw_scoped_refptr_mismatch_checker.h (right):

https://codereview.chromium.org/821453003/diff/180001/base/memory/raw_scoped_...
base/memory/raw_scoped_refptr_mismatch_checker.h:50: struct
ParamsUseScopedRefptrCorrectly<Tuple<>> {
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/base/observer_list_thre...
File base/observer_list_threadsafe.h (right):

https://codereview.chromium.org/821453003/diff/180001/base/observer_list_thre...
base/observer_list_threadsafe.h:171: void Notify(Method m) {
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/components/password_man...
File
components/password_manager/content/browser/content_credential_manager_dispatcher_unittest.cc
(left):

https://codereview.chromium.org/821453003/diff/180001/components/password_man...
components/password_manager/content/browser/content_credential_manager_dispatcher_unittest.cc:308:
CredentialManagerMsg_SendCredential::Read(message, &send_param);
Heh.

https://codereview.chromium.org/821453003/diff/180001/ipc/ipc_message_macros.h
File ipc/ipc_message_macros.h (right):

https://codereview.chromium.org/821453003/diff/180001/ipc/ipc_message_macros....
ipc/ipc_message_macros.h:456: static bool Dispatch(const Message* msg, T* obj,
S* sender, P* parameter,   \
Dispatch() could probably be a variadic template function.

https://codereview.chromium.org/821453003/diff/180001/ipc/ipc_message_macros....
ipc/ipc_message_macros.h:600: typedef IPC::MessageSchema<IPC_TUPLE_IN_##in_cnt
in_list> Schema;         \
This file used to use these fancy token pasting hacks to get the number right
– I think this can now probably all go away, possibly with some more variadic
templates in this file.

https://codereview.chromium.org/821453003/diff/180001/ipc/ipc_message_utils.h
File ipc/ipc_message_utils.h (right):

https://codereview.chromium.org/821453003/diff/180001/ipc/ipc_message_utils.h...
ipc/ipc_message_utils.h:523: struct ParamTraits<Tuple<>> {
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/ppapi/host/dispatch_hos...
File ppapi/host/dispatch_host_message.h (right):

https://codereview.chromium.org/821453003/diff/180001/ppapi/host/dispatch_hos...
ppapi/host/dispatch_host_message.h:25: Tuple<>& arg) {
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/ppapi/proxy/dispatch_re...
File ppapi/proxy/dispatch_reply_message.h (right):

https://codereview.chromium.org/821453003/diff/180001/ppapi/proxy/dispatch_re...
ppapi/proxy/dispatch_reply_message.h:23: inline void DispatchResourceReply(ObjT*
obj, Method method,
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/ppapi/proxy/nacl_messag...
File ppapi/proxy/nacl_message_scanner.cc (right):

https://codereview.chromium.org/821453003/diff/180001/ppapi/proxy/nacl_messag...
ppapi/proxy/nacl_message_scanner.cc:152: void ScanTuple(const Tuple<A>& t1,
ScanningResults* results) {
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/ppapi/proxy/ppapi_messa...
File ppapi/proxy/ppapi_message_utils.h (right):

https://codereview.chromium.org/821453003/diff/180001/ppapi/proxy/ppapi_messa...
ppapi/proxy/ppapi_message_utils.h:26: struct TupleTypeMatch1<Tuple<A>, A> {
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/printing/pdf_render_set...
File printing/pdf_render_settings.h (right):

https://codereview.chromium.org/821453003/diff/180001/printing/pdf_render_set...
printing/pdf_render_settings.h:28: const gfx::Rect& area() const { return
::get<0>(*this); }
Out of interest, why do you need the ::?

https://codereview.chromium.org/821453003/diff/180001/testing/generate_gmock_...
File testing/generate_gmock_mutant.py (right):

https://codereview.chromium.org/821453003/diff/180001/testing/generate_gmock_...
testing/generate_gmock_mutant.py:249: inline R DispatchToMethod(T* obj, Method
method,
can possibly be a variadic template

https://codereview.chromium.org/821453003/diff/180001/testing/generate_gmock_...
testing/generate_gmock_mutant.py:275: CreateFunctor(T* obj, R
(U::*method)(%(params)s), %(args)s) {
can possibly be a variadic template

[mdempsky, 2014-12-22 17:47:30]

lgtm

Out of curiosity, was this generated manually or with tooling?

Agreed with Nico that several of the spots could be further simplified with
variadic templates, but I'd vote for doing those in followup CLs.

[Nico, 2014-12-22 17:48:19]

On Mon, Dec 22, 2014 at 9:47 AM, <mdempsky@chromium.org> wrote:

> lgtm
>
> Out of curiosity, was this generated manually or with tooling?
>
> Agreed with Nico that several of the spots could be further simplified with
> variadic templates, but I'd vote for doing those in followup CLs.
>

I vote for that too.


>
> https://codereview.chromium.org/821453003/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Avi (use Gerrit), 2014-12-22 17:57:57]

Comments:

1. This was done manually.
2. Variadic templates make my brain hurt, plus I didn't want to mix a mechanical
change with a more complicated change.

[commit-bot: I haz the power, 2014-12-22 18:00:52]

Patchset 11 (id:??) landed as
https://crrev.com/95c2a1b7d5626b287be55e41a5cfc33c6880eb46
Cr-Commit-Position: refs/heads/master@{#309441}

[Avi (use Gerrit), 2014-12-22 18:00:55]

Committed patchset #11 (id:200001) manually as
95c2a1b7d5626b287be55e41a5cfc33c6880eb46 (presubmit successful).

[Avi (use Gerrit), 2014-12-22 18:03:03]

https://codereview.chromium.org/821453003/diff/180001/printing/pdf_render_set...
File printing/pdf_render_settings.h (right):

https://codereview.chromium.org/821453003/diff/180001/printing/pdf_render_set...
printing/pdf_render_settings.h:28: const gfx::Rect& area() const { return
::get<0>(*this); }
On 2014/12/22 17:03:12, Nico wrote:
> Out of interest, why do you need the ::?

Because "get" is used for the name of lots of member functions of the various
base classes of Tuple, and I got the most complicated error messages about how
it couldn't decide which base class's "get" I meant if I didn't qualify it with
the ::.