[SUREFIRE-2006] Don't use Services Transformer in shadefire

[slawekjaranowski]

Following this checklist to help us incorporate your
contribution quickly and easily:

• Make sure there is a JIRA issue filed
  for the change (usually before you start working on it). Trivial changes like typos do not
  require a JIRA issue. Your pull request should address just this issue, without
  pulling in other changes.
• Each commit in the pull request should have a meaningful subject line and body.
• Format the pull request title like [SUREFIRE-XXX] - Fixes bug in ApproximateQuantiles,
  where you replace SUREFIRE-XXX with the appropriate JIRA issue. Best practice
  is to use the JIRA issue title in the pull request title and in the first line of the
  commit message.
• Write a pull request description that is detailed enough to understand what the pull request does, how, and why.
• Run mvn clean install to make sure basic checks pass. A more thorough check will
  be performed on your pull request automatically.
• You have run the integration tests successfully (mvn -Prun-its clean install).

If your pull request is about ~20 lines of code you don't need to sign an
Individual Contributor License Agreement if you are unsure
please ask on the developers list.

To make clear that you license your contribution under
the Apache License Version 2.0, January 2004
you have to acknowledge this by using the following check-box.

• I hereby declare this contribution to be licenced under the Apache License Version 2.0, January 2004

• In any other case, please file an Apache Individual Contributor License Agreement.

[Tibor17]

It's good that you have looked at it.
After checking the original content of the JAR, the SPI was there 4 times and the sources too.
Now it is without duplicates of *.class and SPI?

The POJO test should not be modified like this because the users will not use the ShadeFire provider. This provider was developed mainly for our internal purposes. The POJO does not need to have any JUnit dependency and no provider because it assumes that the tests are so simple where the public test methods start with name test.
Anyway, if you like to test the ShadeFire, it's good but we should also keep the original IT and this can be accomplished by adding a new Maven profile for ShadeFire in the IT's POM which is maybe the most simple solution.

[Tibor17]

Maybe it is a bug in the maven-shade-plugin, WDYT?

[slawekjaranowski]

https://issues.apache.org/jira/browse/MSHADE-182

[slawekjaranowski]

• *.class are not duplicated
• we have one additional service - from original jar - MasterProcessChannelProcessorFactory

shade/META-INF/services/
├── org.apache.maven.shadefire.surefire.spi.MasterProcessChannelProcessorFactory
├── org.apache.maven.surefire.api.provider.SurefireProvider
└── org.apache.maven.surefire.spi.MasterProcessChannelProcessorFactory

I hope it is not issue

[Tibor17]

Anyway, if you like to test the ShadeFire, it's good but we should also keep the original IT and this can be accomplished by adding a new Maven profile for ShadeFire in the IT's POM which is maybe the most simple solution.

[slawekjaranowski]

I will fix duplicate services and prepare separate IT test

[slawekjaranowski]

Fixed:

• there are no duplicated services
• new IT test

[Tibor17]

Here we can open more alternatives for test errors if we build the sources against ${java.specificaation.version}.

[slawekjaranowski]

What is your proposition?
Most of IT tests have the same, even with 1.7

[Tibor17]

The old tests with j7 are old tests, that's the history.
Once, Robert Scholte provided an integration test with

<maven.compiler.source>${java.specificaation.version}</maven.compiler.source>
<maven.compiler.target>${java.specificaation.version}</maven.compiler.target>

It was very interesting configuration. We know that the bytecode is not always the same beween versions which is good for us because we can find more bugs in these tests.