80% code coverage for StorIO-Common

[artem-zinnatullin]

Part of #450.

This PR adds tests for StorIO-Common, now coverage of this module is about 99%.

@nikitin-da PTAL

[artem-zinnatullin]

YES, I've finally fixed the problem with test coverage of OnSubscribeExecuteAsBlocking.

It was race condition which never happens on build machine with more than one Processor Core but on Travis CI and on our Jenkins we have only one core and test execution became different.

@nikitin-da PTAL!

[nikitin-da]

Why should it be faster withfinal? As for me it's not uniquely (http://stackoverflow.com/a/4279442)

[artem-zinnatullin]

Hm, I think it will be faster because of JIT.
27 июля 2015 г. 8:59 AM пользователь "Dmitrii Nikitin" <
notifications@github.com> написал:

In
storio-common/src/test/java/com/pushtorefresh/storio/internal/EnvironmentTest.java
#461 (comment):

•    field.set(null, false); // No Environment will think that RxJava is not in the ClassPath
  

•    try {
  

•        Environment.throwExceptionIfRxJavaIsNotAvailable("yolo");
  

•        fail();
  

•    } catch (IllegalStateException expected) {
  

•        assertEquals("yolo requires RxJava in classpath," +
  

•                        " please add it as compile dependency to the application",
  

•                expected.getMessage()
  

•        );
  

•    } finally {
  

•        // Return previous value of the field
  

•        field.set(null, prevValue);
  

•        // Restoring FINAL modifier (for better tests performance)
  

•        modifiersFieldOfTheField.setInt(field, field.getModifiers() & FINAL);
  

Why should it be faster withfinal? As for me it's not uniquely (
http://stackoverflow.com/a/4279442)

—
Reply to this email directly or view it on GitHub
https://github.com/pushtorefresh/storio/pull/461/files#r35507387.

[nikitin-da]

Well done!