android: don't try to handle suppress_wildcard.

android: don't try to handle suppress_wildcard.

When deciding whether to include a target in the generated makefiles or
not, don't consider the suppress_wildcard flag as this just means that
any target which depends directly on the target in question will crash
gyp. The exact contents of the "all" target on the android backend are
not really relevant, so this is the simplest fix.

BUG=chromium:435772
R=thakis@chromium.org

Committed: https://code.google.com/p/gyp/source/detail?r=2010

[Torne, 2014-11-25 15:59:32]

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

[Torne, 2014-11-25 15:59:32]

Well, this is the fix for the bug, but apparently the gyp-android bot has been
broken for ages and nobody mentioned this to me or refrained from submitting
their changes anyway :)

[Nico, 2014-11-25 16:44:53]

lgtm to get things unblocked. Test, and a working bot would be nice :-)

[Torne, 2014-11-25 16:50:54]

On 2014/11/25 16:44:53, Nico wrote:
> lgtm to get things unblocked. Test, and a working bot would be nice :-)

We're looking at why the bot is broken, but it looks environmental - it started
failing at a point where the VM was rebuilt by the look of it, nothing has
changed in gyp or android. From our understanding of the failure it's impossible
for it to fail there, so that's nice :)

[Torne, 2014-11-27 16:16:43]

Having one more go at trying this (with a clobber) and if the bot still fails
because of the emulator being missing I'll land/roll this CL anyway to unblock
stuff and investigate the bot later.

[Torne, 2014-11-27 16:56:59]

Committed patchset #1 (id:1) manually as 2010 (presubmit successful).

[Torne, 2014-11-27 16:59:36]

Bot still broken. We'll investigate separately.