[SOL-19] Removing the archetypes (eventually they'll go elsewhere)

[mmerrell-sauce]

I'm working on an archetype for the more up-to-date samples, and they belong with the sample frameworks, not in this project

[enriquegh]

@mmerrell-sauce this might've gotten a bit stale.
Are we still planning on removing them?

We've seen people trying to use the TestNG archetype and by extension the testng artifact (link) which iirc is deprecated.

[mmerrell]

We have a new Java framework that should drop sometime soon. I hope to include a new archetype with that. If the people using the current one are able to get it to run successfully, I think we should keep it alive--it's not the best way to get tests working, but it's not the worst. I think removing it before we have a replacement ready would create more frustration.

Let me know what you think.

[enriquegh]

@mmerrell main issue here is that the archetype uses the sauce_testng module that was deprecated a bit ago
https://github.com/saucelabs/sauce-java/blob/master/quickstart-archetype/quickstart-webdriver-testng/src/main/resources/archetype-resources/src/test/java/SampleSauceTest.java#L34
https://github.com/saucelabs/sauce-java/blob/master/quickstart-archetype/quickstart-webdriver-testng/src/main/resources/archetype-resources/pom.xml#L35

I don't think we should be encouraging a sample archetype that uses code we no longer consider valid.
While it might "just work" now, things like updating SauceREST (which we do still maintain as a company) might break it at any moment's notice.

[nadvolod]

@enriquegh did this ever get resolved? If not, what do you propose that we do?

[enriquegh]

@nadvolod don't believe so.
I'd propose that we either:

• completely remove the archetypes since (at least the testNG one) uses deprecated dependencies
• update them to not use deprecated dependencies so it's up to date for whenever we move them to their new home

I'm good with either and happy to make a PR for either if needed.

[nadvolod]

@enriquegh I like the removal. Might be the fastest way to solve the problem. If you can send a PR, would be great 👍

[enriquegh]

@nikolay-advolodkin @mmerrell-sauce I've rebased this PR with master and did a small addition to Marcus' original work.

The Travis' tests seem to have all failed but based on Travis history these haven't been working for a while? Best I can tell they're not failing because of the changes done here.

If you'd like me to look into them let me know but I think those might be a PR on its own.