Use gtest_filter syntax in jstest filter

Use gtest_filter syntax in jstest filter

This is mainly to support negative pattern.
https://code.google.com/p/googletest/wiki/AdvancedGuide#Running_a_Subset_of_the_Tests

BUG=
R=nyquist@chromium.org

Committed: e3ef0c4c8e46ab6205a4682ce889a0b47aab0be3

[wychen, 2015-03-31 06:53:38]

wychen@chromium.org changed reviewers:
+ cjhopman@chromium.org, nyquist@chromium.org

[wychen, 2015-03-31 06:53:38]

PTAL

[cjhopman, 2015-03-31 17:55:41]

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBase.java (right):

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:141: for (int i = 0; i <
filter.length(); i++) {
how about just split on "-" and then split each side on ":" (and maybe have a
func that converts String[] to List<Regex> appropriately)? up to you.

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:144: Boolean isNeg =
false;
Doesn't this need to be initialized outside the loop?

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:110: results =
suite.run(new TestLogger.NullLogger(), "TestClass*:*test1-*test2");
Should add a test with multiple negative patterns.

[wychen, 2015-03-31 18:26:49]

PTAL.

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBase.java (right):

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:141: for (int i = 0; i <
filter.length(); i++) {
On 2015/03/31 17:55:40, cjhopman wrote:
> how about just split on "-" and then split each side on ":" (and maybe have a
> func that converts String[] to List<Regex> appropriately)? up to you.

Done.

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:144: Boolean isNeg =
false;
On 2015/03/31 17:55:40, cjhopman wrote:
> Doesn't this need to be initialized outside the loop?

isNeg is used to tell whether the current segment begins with "-", so this
should stay here. However, I see the current implementation looks messy and
harder to understand.

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/20001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:110: results =
suite.run(new TestLogger.NullLogger(), "TestClass*:*test1-*test2");
On 2015/03/31 17:55:41, cjhopman wrote:
> Should add a test with multiple negative patterns.

Done.

[cjhopman, 2015-03-31 20:27:54]

https://codereview.chromium.org/1047853004/diff/40001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/40001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:125: results =
suite.run(new TestLogger.NullLogger(), "-TestClass.test1-TestClass.test2");
Ah, here's a bit of the confusion. From my reading of
https://code.google.com/p/googletest/wiki/AdvancedGuide#Running_a_Subset_of_t...,
the format for the filter is:

pos1:pos2:pos3-neg1:neg2:neg3

[nyquist, 2015-03-31 23:03:05]

https://codereview.chromium.org/1047853004/diff/40001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/40001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:125: results =
suite.run(new TestLogger.NullLogger(), "-TestClass.test1-TestClass.test2");
On 2015/03/31 20:27:54, cjhopman wrote:
> Ah, here's a bit of the confusion. From my reading of
>
https://code.google.com/p/googletest/wiki/AdvancedGuide#Running_a_Subset_of_t...,
> the format for the filter is:
> 
> pos1:pos2:pos3-neg1:neg2:neg3

Yeah, that's how I understand it as well.

[wychen, 2015-04-01 15:44:07]

PTAL.

https://codereview.chromium.org/1047853004/diff/40001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/40001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:125: results =
suite.run(new TestLogger.NullLogger(), "-TestClass.test1-TestClass.test2");
On 2015/03/31 23:03:05, nyquist wrote:
> On 2015/03/31 20:27:54, cjhopman wrote:
> > Ah, here's a bit of the confusion. From my reading of
> >
>
https://code.google.com/p/googletest/wiki/AdvancedGuide#Running_a_Subset_of_t...,
> > the format for the filter is:
> > 
> > pos1:pos2:pos3-neg1:neg2:neg3
> 
> Yeah, that's how I understand it as well.

Oh! I misread it. Fixed.
I thought it was something like this: "pos1-neg1:pos2:pos3-neg2-neg3".

[nyquist, 2015-04-01 18:27:07]

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBase.java (right):

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:130: List<RegExp> pos;
Nit: private final?

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:157: if (i > 0) {
Could we reorder the if/else and do |if (i == 0)| instead? It feels like they
kind of should be in pos...neg order.

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:125: results =
suite.run(new TestLogger.NullLogger(), "-TestClass.test1:TestClass.test2");
Nit: If you have time and the will to do it, feel free to add tests for
- single positive and multiple negatives
- multiple positives and multiple negatives

[wychen, 2015-04-01 19:05:53]

PTAL

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBase.java (right):

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:130: List<RegExp> pos;
On 2015/04/01 18:27:07, nyquist wrote:
> Nit: private final?

Done.

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBase.java:157: if (i > 0) {
On 2015/04/01 18:27:07, nyquist wrote:
> Could we reorder the if/else and do |if (i == 0)| instead? It feels like they
> kind of should be in pos...neg order.

Done.

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:125: results =
suite.run(new TestLogger.NullLogger(), "-TestClass.test1:TestClass.test2");
On 2015/04/01 18:27:07, nyquist wrote:
> Nit: If you have time and the will to do it, feel free to add tests for
> - single positive and multiple negatives
> - multiple positives and multiple negatives

Done.

[nyquist, 2015-04-01 20:41:14]

lgtm

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
File javatests/org/chromium/distiller/JsTestSuiteBaseTest.java (right):

https://codereview.chromium.org/1047853004/diff/60001/javatests/org/chromium/...
javatests/org/chromium/distiller/JsTestSuiteBaseTest.java:125: results =
suite.run(new TestLogger.NullLogger(), "-TestClass.test1:TestClass.test2");
On 2015/04/01 19:05:53, Wei-Yin Chen wrote:
> On 2015/04/01 18:27:07, nyquist wrote:
> > Nit: If you have time and the will to do it, feel free to add tests for
> > - single positive and multiple negatives
> > - multiple positives and multiple negatives
> 
> Done.

thanks!

[wychen, 2015-04-02 01:13:22]

Committed patchset #5 (id:80001) manually as
e3ef0c4c8e46ab6205a4682ce889a0b47aab0be3 (presubmit successful).