gm: add --ignoreTests flag

gm: add --ignoreTests flag
BUG=skia:1600
R=scroggo@google.com

Committed: https://code.google.com/p/skia/source/detail?r=11374

[epoger, 2013-09-18 18:49:29]

https://codereview.chromium.org/23526065/diff/1/gm/tests/outputs/ignoring-one...
File gm/tests/outputs/ignoring-one-test/output-expected/json-summary.txt
(right):

https://codereview.chromium.org/23526065/diff/1/gm/tests/outputs/ignoring-one...
gm/tests/outputs/ignoring-one-test/output-expected/json-summary.txt:1: {
Patchset 1: Just copies the boilerplate from an existing gm self-test to create
a new one from, using the commands below.  Nothing substantive in this patchset.

cd gm/tests/outputs
OLD=compared-against-different-pixels-json
NEW=ignoring-one-test

 DIRS=$(cd $OLD ; find output-expected -not -path "*/.svn*" -type d)
FILES=$(cd $OLD ; find output-expected -not -path "*/.svn*" -type f)
for DIR in $DIRS; do
  mkdir -p $NEW/$DIR
done
svn add $NEW
for FILE in $FILES; do
  svn cp $OLD/$FILE $NEW/$FILE
done

[epoger, 2013-09-18 18:57:41]

https://codereview.chromium.org/23526065/diff/2001/gm/gmmain.cpp
File gm/gmmain.cpp (right):

https://codereview.chromium.org/23526065/diff/2001/gm/gmmain.cpp#newcode1383
gm/gmmain.cpp:1383: DEFINE_string(ignoreTests, "", "Space delimited list of
tests for which we should ignore "
Patchset 2: Adding the ignoreTests flag, but it doesn't actually DO anything
yet.

https://codereview.chromium.org/23526065/diff/2001/gm/tests/outputs/ignoring-...
File gm/tests/outputs/ignoring-one-test/output-expected/stdout (right):

https://codereview.chromium.org/23526065/diff/2001/gm/tests/outputs/ignoring-...
gm/tests/outputs/ignoring-one-test/output-expected/stdout:3: GM: writing to
gm/tests/outputs/ignoring-one-test/output-actual/writePath
Patchset 2 also updates the output for the new gm self-test to pass (that output
will change again, once I actually implement the flag in gmmain)

[epoger, 2013-09-18 20:10:35]

Ready to review at patchset 3.

I recommend that you ignore patchset 1 and instead just review the changes after
that point, since patchset 1 was a bunch of boilerplate svn copies.

https://codereview.chromium.org/23526065/diff/18001/gm/gmmain.cpp
File gm/gmmain.cpp (right):

https://codereview.chromium.org/23526065/diff/18001/gm/gmmain.cpp#newcode2171
gm/gmmain.cpp:2171: if (FLAGS_ignoreTests.contains(shortName)) {
Patchset 3: if the --ignoreTests flag has been passed, use it.

https://codereview.chromium.org/23526065/diff/18001/gm/tests/outputs/ignoring...
File gm/tests/outputs/ignoring-one-test/output-expected/json-summary.txt
(right):

https://codereview.chromium.org/23526065/diff/18001/gm/tests/outputs/ignoring...
gm/tests/outputs/ignoring-one-test/output-expected/json-summary.txt:4:
"failure-ignored" : {
The diff between patchsets 2 and 3 in this file shows us that the new flag is
being observed as expected.

[scroggo, 2013-09-18 20:16:54]

lgtm

https://codereview.chromium.org/23526065/diff/18001/gm/gmmain.cpp
File gm/gmmain.cpp (right):

https://codereview.chromium.org/23526065/diff/18001/gm/gmmain.cpp#newcode2171
gm/gmmain.cpp:2171: if (FLAGS_ignoreTests.contains(shortName)) {
On 2013/09/18 20:10:35, epoger wrote:
> Patchset 3: if the --ignoreTests flag has been passed, use it.

Might it make sense to allow pattern matching? (If so, that's probably a
separate CL, but just wanted to throw it out there.)

[epoger, 2013-09-19 06:16:36]

https://codereview.chromium.org/23526065/diff/18001/gm/gmmain.cpp
File gm/gmmain.cpp (right):

https://codereview.chromium.org/23526065/diff/18001/gm/gmmain.cpp#newcode2171
gm/gmmain.cpp:2171: if (FLAGS_ignoreTests.contains(shortName)) {
On 2013/09/18 20:16:54, scroggo wrote:
> On 2013/09/18 20:10:35, epoger wrote:
> > Patchset 3: if the --ignoreTests flag has been passed, use it.
> 
> Might it make sense to allow pattern matching? (If so, that's probably a
> separate CL, but just wanted to throw it out there.)

That's a worthy suggestion, thanks.  As per
https://code.google.com/p/skia/issues/detail?id=1600#c7 , I think I'll keep it
simple (as-is) for now, but we may want to re-evaluate later on...

[epoger, 2013-09-19 06:18:47]

Committed patchset #3 manually as r11374 (presubmit successful).