Turns on use_mojo=1 for windows.

Turns on use_mojo=1 for windows.

BUG=353602
TEST=none
R=bradnelson@chromium.org, csharp@chromium.org, grt@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=258288

[sky, 2014-03-18 21:37:12]

bradnelson: chrome/installer/mini_installer.gyp
csharp: *isolate

Brad, if you know of other files that need to be updated for the new dll, let me
know. Or if you think there is a better reviewer.

Thanks,

  -Scott

[bradnelson, 2014-03-18 22:20:50]

lgtm, nothing occurs offhand.

[csharp, 2014-03-18 22:32:02]

lgtm

[sky, 2014-03-18 22:37:31]

+robertshield for chrome/installer/mini_installer.gyp

[grt (UTC plus 2), 2014-03-19 00:59:02]

https://codereview.chromium.org/204023002/diff/1/chrome/installer/mini_instal...
File chrome/installer/mini_installer.gyp (right):

https://codereview.chromium.org/204023002/diff/1/chrome/installer/mini_instal...
chrome/installer/mini_installer.gyp:231: '<(PRODUCT_DIR)/mojo_system.dll',
is the intent here for mojo_system.dll to be included in the installer archive? 
should this be conditional on use_mojo==1? if not, will things break if someone
sets use_mojo=0?
you may need to make the same change in mini_installer.gypi, which i think it
used by syzygy toolchain.

[sky, 2014-03-19 15:04:09]

Updated.

https://codereview.chromium.org/204023002/diff/1/chrome/installer/mini_instal...
File chrome/installer/mini_installer.gyp (right):

https://codereview.chromium.org/204023002/diff/1/chrome/installer/mini_instal...
chrome/installer/mini_installer.gyp:231: '<(PRODUCT_DIR)/mojo_system.dll',
On 2014/03/19 00:59:03, grt wrote:
> is the intent here for mojo_system.dll to be included in the installer
archive? 

Yes, mojo_system.dll needs to be part of the install.

> should this be conditional on use_mojo==1? if not, will things break if
someone
> sets use_mojo=0?

Good point. I had envisioned use_mojo as temporary until I straighten this out,
but you're right. While we have use_mojo I should honor it.

> you may need to make the same change in mini_installer.gypi, which i think it
> used by syzygy toolchain.

Done.

[sky, 2014-03-19 18:04:57]

grt: I suspect the mini_installer failure is real (
http://build.chromium.org/p/tryserver.chromium/builders/win_rel/builds/284341...
). I tried to run locally, but I don't seem to get very far:

$ python ./chrome/test/mini_installer/test_installer.py --target=Release --buil
d-dir=d:\\src\\build2\\src\\out ./chrome/test/mini_installer/config/config.conf
ig
Traceback (most recent call last):
  File "./chrome/test/mini_installer/test_installer.py", line 271, in <module>
    sys.exit(main())
  File "./chrome/test/mini_installer/test_installer.py", line 265, in main
    if not RunTests(mini_installer_path, config, options.force_clean):
  File "./chrome/test/mini_installer/test_installer.py", line 237, in RunTests
    variable_expander = VariableExpander(mini_installer_path)
  File "d:\src\build2\src\chrome\test\mini_installer\variable_expander.py", line
 112, in __init__
    mini_installer_product_name)
KeyError: u"Unknown mini_installer product name 'Google Chrome Installer'"

If you have any suggestions as to what I may have done wrong it would be much
appreciated.

Thanks

[grt (UTC plus 2), 2014-03-19 19:46:35]

On 2014/03/19 18:04:57, sky wrote:
> grt: I suspect the mini_installer failure is real (
>
http://build.chromium.org/p/tryserver.chromium/builders/win_rel/builds/284341...
> ). I tried to run locally, but I don't seem to get very far:
> 
> $ python ./chrome/test/mini_installer/test_installer.py --target=Release
--buil
> d-dir=d:\\src\\build2\\src\\out
./chrome/test/mini_installer/config/config.conf
> ig
> Traceback (most recent call last):
>   File "./chrome/test/mini_installer/test_installer.py", line 271, in <module>
>     sys.exit(main())
>   File "./chrome/test/mini_installer/test_installer.py", line 265, in main
>     if not RunTests(mini_installer_path, config, options.force_clean):
>   File "./chrome/test/mini_installer/test_installer.py", line 237, in RunTests
>     variable_expander = VariableExpander(mini_installer_path)
>   File "d:\src\build2\src\chrome\test\mini_installer\variable_expander.py",
line
>  112, in __init__
>     mini_installer_product_name)
> KeyError: u"Unknown mini_installer product name 'Google Chrome Installer'"
> 
> If you have any suggestions as to what I may have done wrong it would be much
> appreciated.

The failure you cite above was fixed in r257782. You probably just need to
rebase to get past that to see the underlying problem. The underlying problem is
likely that chrome.release doesn't list mojo_system.dll so it isn't included in
the archive. You also need to add it and its pdb to
chrome/tools/build/win/FILES.cfg. Sorry I didn't mention these before. It wasn't
clear to me that you really needed mojo_system.dll in the install.

[sky, 2014-03-19 20:15:53]

On 2014/03/19 19:46:35, grt wrote:
> On 2014/03/19 18:04:57, sky wrote:
> > grt: I suspect the mini_installer failure is real (
> >
>
http://build.chromium.org/p/tryserver.chromium/builders/win_rel/builds/284341...
> > ). I tried to run locally, but I don't seem to get very far:
> > 
> > $ python ./chrome/test/mini_installer/test_installer.py --target=Release
> --buil
> > d-dir=d:\\src\\build2\\src\\out
> ./chrome/test/mini_installer/config/config.conf
> > ig
> > Traceback (most recent call last):
> >   File "./chrome/test/mini_installer/test_installer.py", line 271, in
<module>
> >     sys.exit(main())
> >   File "./chrome/test/mini_installer/test_installer.py", line 265, in main
> >     if not RunTests(mini_installer_path, config, options.force_clean):
> >   File "./chrome/test/mini_installer/test_installer.py", line 237, in
RunTests
> >     variable_expander = VariableExpander(mini_installer_path)
> >   File "d:\src\build2\src\chrome\test\mini_installer\variable_expander.py",
> line
> >  112, in __init__
> >     mini_installer_product_name)
> > KeyError: u"Unknown mini_installer product name 'Google Chrome Installer'"
> > 
> > If you have any suggestions as to what I may have done wrong it would be
much
> > appreciated.
> 
> The failure you cite above was fixed in r257782. You probably just need to
> rebase to get past that to see the underlying problem. The underlying problem
is
> likely that chrome.release doesn't list mojo_system.dll so it isn't included
in
> the archive. You also need to add it and its pdb to
> chrome/tools/build/win/FILES.cfg. Sorry I didn't mention these before. It
wasn't
> clear to me that you really needed mojo_system.dll in the install.

Thanks for pointing me at the right places. New patch uploaded.

  -Scott

[grt (UTC plus 2), 2014-03-19 20:32:25]

lgtm

[sky, 2014-03-20 14:50:13]

Committed patchset #4 manually as r258288 (presubmit successful).