Add GN output prefix override and allow empty output extensions.

Add GN output prefix override and allow empty output extensions.

Previously setting the output extension to an empty string would give the
default extension. This patch differentiates unset (default extension) from
set-to-empty-string (which can now mean "no extension").

A flag is added to out-out of the target output prefix application for targets
that don't want to have "lib" at the beginning on Posix systems. Even though
the flag might more naturally be called "override_output_prefix" I called it
"output_prefix_override" so that all of the output-name-related variables are
called "output_....".

Committed: https://crrev.com/2a642a10300c49bcd33b6246e16650cf61a42d3b
Cr-Commit-Position: refs/heads/master@{#385927}

[brettw, 2016-04-07 21:08:27]

Description was changed from

==========
Add GN output prefix override and allow empty output extensions.

Previously setting the output extension to an empty string would give the
default extension. This patch differentiates unset (default extension) from
set-to-empty-string (which can now mean "no extension").

A flag is added to out-out of the target output prefix application for targets
that don't want to have "lib" at the beginnin on Posix systems.
==========

to

==========
Add GN output prefix override and allow empty output extensions.

Previously setting the output extension to an empty string would give the
default extension. This patch differentiates unset (default extension) from
set-to-empty-string (which can now mean "no extension").

A flag is added to out-out of the target output prefix application for targets
that don't want to have "lib" at the beginning on Posix systems. Even though
the flag might more naturally be called "override_output_prefix" I called it
"output_prefix_override" so that all of the output-name-related variables are
called "output_....".
==========

[brettw, 2016-04-07 21:08:51]

brettw@chromium.org changed reviewers:
+ rsesek@chromium.org

[brettw, 2016-04-07 21:08:56]

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

[commit-bot: I haz the power, 2016-04-07 21:10:02]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1868023002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1868023002/1

[Robert Sesek, 2016-04-07 21:34:47]

LGTM. Thanks for doing this!

https://codereview.chromium.org/1868023002/diff/1/tools/gn/target.cc
File tools/gn/target.cc (right):

https://codereview.chromium.org/1868023002/diff/1/tools/gn/target.cc#newcode363
tools/gn/target.cc:363: std::string Target::GetComputedOutputName(bool
include_prefix) const {
This method is only ever called by non-test code with include_prefix=true. With
the introduction of output_prefix_override_ it may be a little confusing to call
this with true and get no prefix. Consider removing it?

https://code.google.com/p/chromium/codesearch#search/&q=GetComputedOutputName...

[Dirk Pranke, 2016-04-07 21:43:54]

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

[Dirk Pranke, 2016-04-07 21:43:54]

Do we want to explicitly allow (and test for) output_names that contain
directories as part of this?

IIUC we currently support directories in names kinda unintentionally -- and
they're used -- and it might be good to be explicit about it.

Or that could be a separate CL.

https://codereview.chromium.org/1868023002/diff/1/tools/gn/target_unittest.cc
File tools/gn/target_unittest.cc (right):

https://codereview.chromium.org/1868023002/diff/1/tools/gn/target_unittest.cc...
tools/gn/target_unittest.cc:298: // Target with an output prefix override should
not a prefix.
nit: s/should not a/should not have a/

[commit-bot: I haz the power, 2016-04-07 21:50:00]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-07 21:50:00]

Dry run: This issue passed the CQ dry run.

[brettw, 2016-04-07 23:12:29]

If we want to change the path thing, we would need to first update the build and
I don't want to block this patch on that. Disallowing paths is pretty low on my
priority list.

https://codereview.chromium.org/1868023002/diff/1/tools/gn/target.cc
File tools/gn/target.cc (right):

https://codereview.chromium.org/1868023002/diff/1/tools/gn/target.cc#newcode363
tools/gn/target.cc:363: std::string Target::GetComputedOutputName(bool
include_prefix) const {
Good point.

[brettw, 2016-04-07 23:12:33]

The CQ bit was checked by brettw@chromium.org

[brettw, 2016-04-07 23:12:33]

The patchset sent to the CQ was uploaded after l-g-t-m from rsesek@chromium.org
Link to the patchset: https://codereview.chromium.org/1868023002/#ps20001
(title: " ")

[commit-bot: I haz the power, 2016-04-07 23:13:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1868023002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1868023002/20001

[commit-bot: I haz the power, 2016-04-07 23:51:29]

Description was changed from

==========
Add GN output prefix override and allow empty output extensions.

Previously setting the output extension to an empty string would give the
default extension. This patch differentiates unset (default extension) from
set-to-empty-string (which can now mean "no extension").

A flag is added to out-out of the target output prefix application for targets
that don't want to have "lib" at the beginning on Posix systems. Even though
the flag might more naturally be called "override_output_prefix" I called it
"output_prefix_override" so that all of the output-name-related variables are
called "output_....".
==========

to

==========
Add GN output prefix override and allow empty output extensions.

Previously setting the output extension to an empty string would give the
default extension. This patch differentiates unset (default extension) from
set-to-empty-string (which can now mean "no extension").

A flag is added to out-out of the target output prefix application for targets
that don't want to have "lib" at the beginning on Posix systems. Even though
the flag might more naturally be called "override_output_prefix" I called it
"output_prefix_override" so that all of the output-name-related variables are
called "output_....".
==========

[commit-bot: I haz the power, 2016-04-07 23:51:29]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-04-07 23:52:57]

Description was changed from

==========
Add GN output prefix override and allow empty output extensions.

Previously setting the output extension to an empty string would give the
default extension. This patch differentiates unset (default extension) from
set-to-empty-string (which can now mean "no extension").

A flag is added to out-out of the target output prefix application for targets
that don't want to have "lib" at the beginning on Posix systems. Even though
the flag might more naturally be called "override_output_prefix" I called it
"output_prefix_override" so that all of the output-name-related variables are
called "output_....".
==========

to

==========
Add GN output prefix override and allow empty output extensions.

Previously setting the output extension to an empty string would give the
default extension. This patch differentiates unset (default extension) from
set-to-empty-string (which can now mean "no extension").

A flag is added to out-out of the target output prefix application for targets
that don't want to have "lib" at the beginning on Posix systems. Even though
the flag might more naturally be called "override_output_prefix" I called it
"output_prefix_override" so that all of the output-name-related variables are
called "output_....".

Committed: https://crrev.com/2a642a10300c49bcd33b6246e16650cf61a42d3b
Cr-Commit-Position: refs/heads/master@{#385927}
==========

[commit-bot: I haz the power, 2016-04-07 23:52:58]

Patchset 2 (id:??) landed as
https://crrev.com/2a642a10300c49bcd33b6246e16650cf61a42d3b
Cr-Commit-Position: refs/heads/master@{#385927}