Sync: Generalize entity injection in Android tests

chrome/chrome_tests.gypi

Index: chrome/chrome_tests.gypi
diff --git a/chrome/chrome_tests.gypi b/chrome/chrome_tests.gypi
index ec2bd4646d152c4db515cce14f41b01c527a08ba..954a56c4c8b323cfb3e8f880b0cea31cbc359b9a 100644
--- a/chrome/chrome_tests.gypi
+++ b/chrome/chrome_tests.gypi
@@ -2965,6 +2965,7 @@
             '../base/base.gyp:base',
             '../base/base.gyp:base_java_test_support',
             '../content/content_shell_and_tests.gyp:content_java_test_support',
+            '../sync/sync.gyp:test_support_sync_proto_java',

[Nicolas Zea, 2015/04/09 20:38:16]

what's the typical naming scheme here? it seems like base_java_test_support is
using a different convention than test_support_sync_proto_java. Should we match
them?

[pval...(no longer on Chromium), 2015/04/09 21:28:05]

Most test infra targets I've encountered are test_support_FOO and all Java proto
targets are FOO_proto_java, so my creation here is a Frankenstein-like result.

I think base_java_test_support may be the outlier here, although there are some
other FOO_java_test_support targets.

All things considered, I'm fine with the current name as it describes the
important parts: it's for tests, it's for sync, and it contains protos. Another
idea: sync_test_proto_java?

[Nicolas Zea, 2015/04/09 21:40:42]

I'm fine with this, as long as it's mostly consistent. 

[nyquist, 2015/04/09 21:53:52]

To be honest, I think I like the sync_proto_java_test_support better. But not
important though, since like Patrick said, it's a Frankestein thing with two
naming concepts mixed.

           ],
           'variables': {
             'apk_name': 'ChromeSyncShellTest',