Work in progress to port remoting_me2me_host_archive

Work in progress to port remoting_me2me_host_archive

R=rsesek@chromium.org, nicholss@chromium.org, sergeyu@chromium.org
BUG=622415
Committed: https://crrev.com/90f80f7c7b6dfa69822a68e6c48656809a4fd4e0
Cr-Commit-Position: refs/heads/master@{#403022}

[Dirk Pranke, 2016-06-27 00:01:55]

This is at least a start at what needs to be done, I think, and is closer than
what we had before. 

There's a lot of <noideadog> going on in this patch, though.

[nicholss, 2016-06-27 18:34:06]

On 2016/06/27 00:01:55, Dirk Pranke wrote:
> This is at least a start at what needs to be done, I think, and is closer than
> what we had before. 
> 
> There's a lot of <noideadog> going on in this patch, though.

Thanks! I am back today so I will start looking at this. I did the first pass
port to GN and thought I had it but did not know the mac side also produced a
zip...

[Dirk Pranke, 2016-06-27 18:45:51]

On 2016/06/27 18:34:06, nicholss wrote:
> On 2016/06/27 00:01:55, Dirk Pranke wrote:
> > This is at least a start at what needs to be done, I think, and is closer
than
> > what we had before. 
> > 
> > There's a lot of <noideadog> going on in this patch, though.
> 
> Thanks! I am back today so I will start looking at this. I did the first pass
> port to GN and thought I had it but did not know the mac side also produced a
> zip...

Yup, as you can see there was a bunch of other stuff missing or not done right
as well
(raw executables but no app bundles, etc.).

[nicholss, 2016-06-27 20:36:53]

On 2016/06/27 18:45:51, Dirk Pranke wrote:
> On 2016/06/27 18:34:06, nicholss wrote:
> > On 2016/06/27 00:01:55, Dirk Pranke wrote:
> > > This is at least a start at what needs to be done, I think, and is closer
> than
> > > what we had before. 
> > > 
> > > There's a lot of <noideadog> going on in this patch, though.
> > 
> > Thanks! I am back today so I will start looking at this. I did the first
pass
> > port to GN and thought I had it but did not know the mac side also produced
a
> > zip...
> 
> Yup, as you can see there was a bunch of other stuff missing or not done right
> as well
> (raw executables but no app bundles, etc.).


Ok I just compared the output from this update from GN and what GYP currently
produces and it looks like the zip has all the components and they look like
they are functional but I did find one oddity:

GYP:

  33M	./PreferencePanes/Chromoting.prefPane
259M	./PrivilegedHelperTools/ChromotingHost.bundle

GN:

137M	./PreferencePanes//Chromoting.prefPane
415M	./PrivilegedHelperTools//ChromotingHost.bundle



I am not sure what to make of this size difference. Maybe Sergey might have some
insights?

[Robert Sesek, 2016-06-27 20:42:28]

Re: size...

https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn
File remoting/host/BUILD.gn (right):

https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn#...
remoting/host/BUILD.gn:893: target(app_target_type, "remoting_me2me_host") {
This and :native_messaging_host may be missing some bundle_data resources.

[nicholss, 2016-06-27 20:43:42]

On 2016/06/27 20:36:53, nicholss wrote:
> On 2016/06/27 18:45:51, Dirk Pranke wrote:
> > On 2016/06/27 18:34:06, nicholss wrote:
> > > On 2016/06/27 00:01:55, Dirk Pranke wrote:
> > > > This is at least a start at what needs to be done, I think, and is
closer
> > than
> > > > what we had before. 
> > > > 
> > > > There's a lot of <noideadog> going on in this patch, though.
> > > 
> > > Thanks! I am back today so I will start looking at this. I did the first
> pass
> > > port to GN and thought I had it but did not know the mac side also
produced
> a
> > > zip...
> > 
> > Yup, as you can see there was a bunch of other stuff missing or not done
right
> > as well
> > (raw executables but no app bundles, etc.).
> 
> 
> Ok I just compared the output from this update from GN and what GYP currently
> produces and it looks like the zip has all the components and they look like
> they are functional but I did find one oddity:
> 
> GYP:
> 
>   33M	./PreferencePanes/Chromoting.prefPane
> 259M	./PrivilegedHelperTools/ChromotingHost.bundle
> 
> GN:
> 
> 137M	./PreferencePanes//Chromoting.prefPane
> 415M	./PrivilegedHelperTools//ChromotingHost.bundle
> 
> 
> 
> I am not sure what to make of this size difference. Maybe Sergey might have
some
> insights?

Ah, one more thing, does GN change how languages work? Looks like GN does not
contain the <language>.lproj inside of the Resources anymore. 

use to look like this: 


20K	./PrivilegedHelperTools//ChromotingHost.bundle/Contents/MacOS/RemoteAssistanceHost.bundle/Contents/Resources/sw.lproj

36K	./PrivilegedHelperTools//ChromotingHost.bundle/Contents/MacOS/RemoteAssistanceHost.bundle/Contents/Resources/ta.lproj

36K	./PrivilegedHelperTools//ChromotingHost.bundle/Contents/MacOS/RemoteAssistanceHost.bundle/Contents/Resources/te.lproj

32K	./PrivilegedHelperTools//ChromotingHost.bundle/Contents/MacOS/RemoteAssistanceHost.bundle/Contents/Resources/th.lproj

20K	./PrivilegedHelperTools//ChromotingHost.bundle/Contents/MacOS/RemoteAssistanceHost.bundle/Contents/Resources/tr.lproj

28K	./PrivilegedHelperTools//ChromotingHost.bundle/Contents/MacOS/RemoteAssistanceHost.bundle/Contents/Resources/uk.lproj

20K	./PrivilegedHelperTools//ChromotingHost.bundle/Contents/MacOS/RemoteAssistanceHost.bundle/Contents/Resources/vi.lproj

[nicholss, 2016-06-27 20:47:45]

https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn
File remoting/host/BUILD.gn (right):

https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn#...
remoting/host/BUILD.gn:893: target(app_target_type, "remoting_me2me_host") {
On 2016/06/27 20:42:28, Robert Sesek wrote:
> This and :native_messaging_host may be missing some bundle_data resources.

yes, I verified they are, working on adding localization. But that does not
explain why GN is 4x as large as GYP also given GYP has the loc files.

[Robert Sesek, 2016-06-27 21:00:13]

On 2016/06/27 20:47:45, nicholss wrote:
> https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn
> File remoting/host/BUILD.gn (right):
> 
>
https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn#...
> remoting/host/BUILD.gn:893: target(app_target_type, "remoting_me2me_host") {
> On 2016/06/27 20:42:28, Robert Sesek wrote:
> > This and :native_messaging_host may be missing some bundle_data resources.
> 
> yes, I verified they are, working on adding localization. But that does not
> explain why GN is 4x as large as GYP also given GYP has the loc files.

Does the GN version have more things in its bundle? Can you run diff -r between
the two?

[nicholss, 2016-06-27 21:06:29]

On 2016/06/27 21:00:13, Robert Sesek wrote:
> On 2016/06/27 20:47:45, nicholss wrote:
> > https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn
> > File remoting/host/BUILD.gn (right):
> > 
> >
>
https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn#...
> > remoting/host/BUILD.gn:893: target(app_target_type, "remoting_me2me_host") {
> > On 2016/06/27 20:42:28, Robert Sesek wrote:
> > > This and :native_messaging_host may be missing some bundle_data resources.
> > 
> > yes, I verified they are, working on adding localization. But that does not
> > explain why GN is 4x as large as GYP also given GYP has the loc files.
> 
> Does the GN version have more things in its bundle? Can you run diff -r
between
> the two?

Yes, GYP has loc, (about 1MB worth) and GN does not (I am working to add it at
the moment.)

But the size difference is in the binary:

GN - 137M Jun 27 13:08 remoting_host_prefname
GYP - 32M Jun 27 13:24 remoting_host_prefpane

Both are debug builds so maybe this does not matter until I do a release build
on each and compare. Let me try that...

[Dirk Pranke, 2016-06-27 21:46:55]

On 2016/06/27 21:06:29, nicholss wrote:
> On 2016/06/27 21:00:13, Robert Sesek wrote:
> > On 2016/06/27 20:47:45, nicholss wrote:
> > >
https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn
> > > File remoting/host/BUILD.gn (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn#...
> > > remoting/host/BUILD.gn:893: target(app_target_type, "remoting_me2me_host")
{
> > > On 2016/06/27 20:42:28, Robert Sesek wrote:
> > > > This and :native_messaging_host may be missing some bundle_data
resources.
> > > 
> > > yes, I verified they are, working on adding localization. But that does
not
> > > explain why GN is 4x as large as GYP also given GYP has the loc files.
> > 
> > Does the GN version have more things in its bundle? Can you run diff -r
> between
> > the two?
> 
> Yes, GYP has loc, (about 1MB worth) and GN does not (I am working to add it at
> the moment.)
> 
> But the size difference is in the binary:
> 
> GN - 137M Jun 27 13:08 remoting_host_prefname
> GYP - 32M Jun 27 13:24 remoting_host_prefpane
> 
> Both are debug builds so maybe this does not matter until I do a release build
> on each and compare. Let me try that...

Make sure you're building both with the same levels of symbols ... I will do
some
builds locally to double-check this.

(and yes, the GN version does not have all of the localized strings).

[nicholss, 2016-06-27 23:21:04]

On 2016/06/27 21:46:55, Dirk Pranke wrote:
> On 2016/06/27 21:06:29, nicholss wrote:
> > On 2016/06/27 21:00:13, Robert Sesek wrote:
> > > On 2016/06/27 20:47:45, nicholss wrote:
> > > >
> https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn
> > > > File remoting/host/BUILD.gn (right):
> > > > 
> > > >
> > >
> >
>
https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn#...
> > > > remoting/host/BUILD.gn:893: target(app_target_type,
"remoting_me2me_host")
> {
> > > > On 2016/06/27 20:42:28, Robert Sesek wrote:
> > > > > This and :native_messaging_host may be missing some bundle_data
> resources.
> > > > 
> > > > yes, I verified they are, working on adding localization. But that does
> not
> > > > explain why GN is 4x as large as GYP also given GYP has the loc files.
> > > 
> > > Does the GN version have more things in its bundle? Can you run diff -r
> > between
> > > the two?
> > 
> > Yes, GYP has loc, (about 1MB worth) and GN does not (I am working to add it
at
> > the moment.)
> > 
> > But the size difference is in the binary:
> > 
> > GN - 137M Jun 27 13:08 remoting_host_prefname
> > GYP - 32M Jun 27 13:24 remoting_host_prefpane
> > 
> > Both are debug builds so maybe this does not matter until I do a release
build
> > on each and compare. Let me try that...
> 
> Make sure you're building both with the same levels of symbols ... I will do
> some
> builds locally to double-check this.
> 
> (and yes, the GN version does not have all of the localized strings).

Ok ignore the size comments. It seems the default level of symbols is more
verbose out of the box for GN is all and the release versions are very close. So
the outstanding thing I am looking at is how to get the loc files generated.
Still working that out.

[Robert Sesek, 2016-06-27 23:22:11]

On 2016/06/27 23:21:04, nicholss wrote:
> On 2016/06/27 21:46:55, Dirk Pranke wrote:
> > On 2016/06/27 21:06:29, nicholss wrote:
> > > On 2016/06/27 21:00:13, Robert Sesek wrote:
> > > > On 2016/06/27 20:47:45, nicholss wrote:
> > > > >
> > https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn
> > > > > File remoting/host/BUILD.gn (right):
> > > > > 
> > > > >
> > > >
> > >
> >
>
https://codereview.chromium.org/2093393002/diff/20001/remoting/host/BUILD.gn#...
> > > > > remoting/host/BUILD.gn:893: target(app_target_type,
> "remoting_me2me_host")
> > {
> > > > > On 2016/06/27 20:42:28, Robert Sesek wrote:
> > > > > > This and :native_messaging_host may be missing some bundle_data
> > resources.
> > > > > 
> > > > > yes, I verified they are, working on adding localization. But that
does
> > not
> > > > > explain why GN is 4x as large as GYP also given GYP has the loc files.
> > > > 
> > > > Does the GN version have more things in its bundle? Can you run diff -r
> > > between
> > > > the two?
> > > 
> > > Yes, GYP has loc, (about 1MB worth) and GN does not (I am working to add
it
> at
> > > the moment.)
> > > 
> > > But the size difference is in the binary:
> > > 
> > > GN - 137M Jun 27 13:08 remoting_host_prefname
> > > GYP - 32M Jun 27 13:24 remoting_host_prefpane
> > > 
> > > Both are debug builds so maybe this does not matter until I do a release
> build
> > > on each and compare. Let me try that...
> > 
> > Make sure you're building both with the same levels of symbols ... I will do
> > some
> > builds locally to double-check this.
> > 
> > (and yes, the GN version does not have all of the localized strings).
> 
> Ok ignore the size comments. It seems the default level of symbols is more
> verbose out of the box for GN is all and the release versions are very close.
So
> the outstanding thing I am looking at is how to get the loc files generated.
> Still working that out.

This is how I set it up for Chrome's lprojs:
https://cs.chromium.org/chromium/src/chrome/BUILD.gn?q=chrome/BUILD.gn&sq=pac...

It's a little gross since GN can't express this very easily.

[nicholss, 2016-06-27 23:59:44]

Attempting to add the loc generation code, I am running into some issues, it
seems the original script is unable to load the localization lib, it claims to
not be able to find keys but if I search in the directly supplied in the command
line python is running I do find the keys so I am a little confused. 

https://paste.googleplex.com/6117143695327232

[nicholss, 2016-06-28 18:48:07]

Localization is added in this CL:

 https://codereview.chromium.org/2110593002

[Dirk Pranke, 2016-06-28 18:49:56]

On 2016/06/27 23:59:44, nicholss wrote:
> Attempting to add the loc generation code, I am running into some issues, it
> seems the original script is unable to load the localization lib, it claims to
> not be able to find keys but if I search in the directly supplied in the
command
> line python is running I do find the keys so I am a little confused. 
> 
> https://paste.googleplex.com/6117143695327232

Did you resolve the issue you were hitting? Is your CL ready for review, or do
you
need help w/ that (or anything else)?

[nicholss, 2016-06-29 16:55:43]

On 2016/06/28 18:49:56, Dirk Pranke wrote:
> On 2016/06/27 23:59:44, nicholss wrote:
> > Attempting to add the loc generation code, I am running into some issues, it
> > seems the original script is unable to load the localization lib, it claims
to
> > not be able to find keys but if I search in the directly supplied in the
> command
> > line python is running I do find the keys so I am a little confused. 
> > 
> > https://paste.googleplex.com/6117143695327232
> 
> Did you resolve the issue you were hitting? Is your CL ready for review, or do
> you
> need help w/ that (or anything else)?

The new update looks good. Good call on not trying to use the remoting_localize
template. I would like to merge this CL and then I will use it from trunk to add
the other locs to the other apps. Does that sound ok to everyone?

[nicholss, 2016-06-29 16:57:33]

LGTM

[Dirk Pranke, 2016-06-29 18:02:09]

The CQ bit was checked by dpranke@chromium.org

[Dirk Pranke, 2016-06-29 18:02:22]

The CQ bit was unchecked by dpranke@chromium.org

[Dirk Pranke, 2016-06-29 18:02:41]

Description was changed from

==========
Work in progress to port remoting_me2me_host_archive

R=rsesek@chromium.org, nicholss@chromium.org
BUG=622415
==========

to

==========
Work in progress to port remoting_me2me_host_archive

R=rsesek@chromium.org, nicholss@chromium.org, sergeyu@chromium.org
BUG=622415
==========

[Dirk Pranke, 2016-06-29 18:02:41]

dpranke@chromium.org changed reviewers:
+ sergeyu@chromium.org

[Dirk Pranke, 2016-06-29 18:02:59]

@sergeyu - can I get an OWNERS approval?

[Sergey Ulanov, 2016-06-29 23:09:45]

lgtm

[Dirk Pranke, 2016-06-29 23:11:03]

The CQ bit was checked by dpranke@chromium.org

[commit-bot: I haz the power, 2016-06-29 23:11:53]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-06-30 00:24:50]

Description was changed from

==========
Work in progress to port remoting_me2me_host_archive

R=rsesek@chromium.org, nicholss@chromium.org, sergeyu@chromium.org
BUG=622415
==========

to

==========
Work in progress to port remoting_me2me_host_archive

R=rsesek@chromium.org, nicholss@chromium.org, sergeyu@chromium.org
BUG=622415
==========

[commit-bot: I haz the power, 2016-06-30 00:24:52]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-06-30 00:25:00]

CQ bit was unchecked.

[commit-bot: I haz the power, 2016-06-30 00:27:39]

Description was changed from

==========
Work in progress to port remoting_me2me_host_archive

R=rsesek@chromium.org, nicholss@chromium.org, sergeyu@chromium.org
BUG=622415
==========

to

==========
Work in progress to port remoting_me2me_host_archive

R=rsesek@chromium.org, nicholss@chromium.org, sergeyu@chromium.org
BUG=622415

Committed: https://crrev.com/90f80f7c7b6dfa69822a68e6c48656809a4fd4e0
Cr-Commit-Position: refs/heads/master@{#403022}
==========

[commit-bot: I haz the power, 2016-06-30 00:27:41]

Patchset 4 (id:??) landed as
https://crrev.com/90f80f7c7b6dfa69822a68e6c48656809a4fd4e0
Cr-Commit-Position: refs/heads/master@{#403022}