Adds AndroidWebView and ChromiumTestShell tests to the Android FYI bot.

build/android/buildbot/buildbot_functions.sh

Index: build/android/buildbot/buildbot_functions.sh
diff --git a/build/android/buildbot/buildbot_functions.sh b/build/android/buildbot/buildbot_functions.sh
index 25e89c40a82f98eaf69a4cf47b791dd935c712bd..ba563b79a785b2568694f77f92347775d0a54264 100755
--- a/build/android/buildbot/buildbot_functions.sh
+++ b/build/android/buildbot/buildbot_functions.sh
@@ -212,18 +212,36 @@ function bb_run_unit_tests {
 #   $1: TEST_APK.
 #   $2: EXTRA_FLAGS to be passed to run_instrumentation_tests.py.
 function bb_run_instrumentation_test {
+  (
   local TEST_APK=${1}
   local EXTRA_FLAGS=${2}
   local INSTRUMENTATION_FLAGS="-vvv"
   INSTRUMENTATION_FLAGS+=" --test-apk ${TEST_APK}"
   INSTRUMENTATION_FLAGS+=" ${EXTRA_FLAGS}"
-  build/android/run_instrumentation_tests.py ${INSTRUMENTATION_FLAGS}
+  set +e

[bulach, 2012/10/02 10:40:46]

I'm not terribly familiar with bash, but is a "set -e" needed for symmetry?

[klundberg, 2012/10/02 18:40:06]

I discussed this with Isaac and we don't need "set -e" since this is now a
separate process and "set +e" only affects what is inside the parentheses.

+  python build/android/run_instrumentation_tests.py ${INSTRUMENTATION_FLAGS}
+  local TEST_RUN_EXIT_CODE=$?
+  if (( $TEST_RUN_EXIT_CODE != 0 )); then

[Isaac (away), 2012/10/02 05:13:37]

use a [[ ]] test here

[klundberg, 2012/10/02 18:40:06]

Done.

+    echo "@@@STEP_FAILURE@@@"
+  fi
+  # Regardless of whether tests failed or crashed we don't want to stop the
+  # entire buildbot flow.
+  return 0

[Isaac (away), 2012/10/02 05:13:37]

return is not technically necessary here but it's fine for clarity.

[klundberg, 2012/10/02 18:40:06]

Done.

+  )
 }
 
-# Run content shell instrumentation test on device.
-function bb_run_instrumentation_tests {
-  build/android/adb_install_content_shell
-  local TEST_APK="ContentShellTest"
+# Run instrumentation test.
+# Args:
+#   $1: TEST_APK.
+#   $2: EXTRA_FLAGS to be passed to run_instrumentation_tests.py.

[bulach, 2012/10/02 10:40:46]

nit: $3: ... ? :)

[klundberg, 2012/10/02 18:40:06]

Done.

+function bb_run_instrumentation_tests_apk {
+  local APK=${1}
+  local TEST_APK=${2}
+  local APK_PACKAGE=${3}
+  echo ${1} ${2} ${3}
+  python build/android/adb_install_apk.py --apk ${APK} \
+      --apk_package ${APK_PACKAGE}
+
   # Use -I to install the test apk only on the first run.
   # TODO(bulach): remove the second once we have a Smoke test.
   bb_run_instrumentation_test ${TEST_APK} "-I -A Smoke"
@@ -232,6 +250,16 @@ function bb_run_instrumentation_tests {
   bb_run_instrumentation_test ${TEST_APK} "-A LargeTest"
 }
 
+# Run content shell instrumentation test on device.
+function bb_run_instrumentation_tests {

[bulach, 2012/10/02 10:40:46]

hmm, these three methods sound too similar..
would these be any clearer?
bb_run_instrumentation_test_internal
bb_run_all_instrumentation_tests_for_apk
bb_run_all_instrumentation_tests

[klundberg, 2012/10/02 18:40:06]

I discussed this with Isaac and since we don't care about test sizes anymore, I
have simplified the scripts.

+  bb_run_instrumentation_tests_apk "ContentShell-debug.apk" \
+      "ContentShellTest" "org.chromium.content_shell"
+  bb_run_instrumentation_tests_apk "ChromiumTestShell-debug.apk" \
+      "ChromiumTestShellTest" "org.chromium.chrome.browser"
+  bb_run_instrumentation_tests_apk "AndroidWebView-debug.apk" \
+      "AndroidWebViewTest" "org.chromium.android_webview"
+}
+
 # Zip and archive a build.
 function bb_zip_build {
   echo "@@@BUILD_STEP Zip build@@@"