Do Regression Testing

[taywhited]

Is this done?

[SaajidM]

The feedback from is up. I'm not sure about the output.

From: taywhited Sent: Thursday, April 3, 2014 12:25 PM To: kubasub/chinese-checkers Reply To: kubasub/chinese-checkers Cc: SaajidM Subject: Re: [chinese-checkers] Do Regression Testing (#211)

Is this done?

---

Reply to this email directly or view it on GitHub: https://github.com/kubasub/chinese-checkers/issues/211#issuecomment-39472568

[kubasub]

@ChrisKdon will run the tests as mine freezes during the first test (if you recall that).

@SaajidM, can you confirm that the tests can be ran at any point now (as in there is no automated testing for the online features)?

[SaajidM]

Technically it should be run after all the changes are made but yeah I suppose at any point now should do.

From: Jakub Subczynski Sent: Thursday, April 3, 2014 12:40 PM To: kubasub/chinese-checkers Reply To: kubasub/chinese-checkers Cc: SaajidM Subject: Re: [chinese-checkers] Do Regression Testing (#211)

@ChrisKdon will run the tests as mine freezes during the first test (if you recall that).

@SaajidM, can you confirm that the tests can be ran at any point now (as in there is no automated testing for the online features)?

---

Reply to this email directly or view it on GitHub: https://github.com/kubasub/chinese-checkers/issues/211#issuecomment-39474384

[chriskdon]

@taywhited when do you need this by?

[taywhited]

Earlier the better, but 5:30 at the latest.

[kubasub]

@SaajidM @taywhited, I ran the tests, and there are two failures, but I do not understand why they occurred. Could you look at the tests on your comp, @SaajidM?

You can find the test results here.

[taywhited]

@kubasub @SaajidM, I have those types of tests listed under "Unit testing" in past binders. Are these results different than unit tests? If so, I'll make a new category under regression testing.

[SaajidM]

These are the automated ui tests from the last phase. So they are regression tests for this phase.

-------- Original message -------- From: taywhited notifications@github.com Date: 03/04/2014 4:33 PM (GMT-05:00) To: kubasub/chinese-checkers chinese-checkers@noreply.github.com Cc: SaajidM saajid.mohammed@hotmail.com Subject: Re: [chinese-checkers] Do Regression Testing (#211)

@kubasub @SaajidM, I have those types of tests listed under "Unit testing" in past binders. Are these results different than unit tests? If so, I'll make a new category under regression testing.

— Reply to this email directly or view it on GitHub.

[SaajidM]

It looks as though these are errors in the AI unit test suite. Thats james' area, I'm not sure why there is an error there.

Sent from Samsung tablet

-------- Original message -------- From: Jakub Subczynski notifications@github.com Date: 03/04/2014 4:16 PM (GMT-05:00) To: kubasub/chinese-checkers chinese-checkers@noreply.github.com Cc: SaajidM saajid.mohammed@hotmail.com Subject: Re: [chinese-checkers] Do Regression Testing (#211)

@SaajidM @taywhited, I ran the tests, and there are two failures, but I do not understand why they occurred.

You can find the test results here.

— Reply to this email directly or view it on GitHub.

[kubasub]

@SaajidM, could you fill out the regression form? I'm not sure what to write since I don't understand the test just based on its label

[taywhited]

@SaajidM, got it, thanks!

@kubasub, should I add these as they are to the binder even with the failures or will they be run again?

[kubasub]

@taywhited, I would just add them. It's a normal part of development (and I wouldn't know where to start with fixing any AI stuff hah)

[taywhited]

Done~

[JamesKostiuk]

If there is an error with AI it would be one of two things: either the error I mentioned during the last iteration which is a failure during the TestChainHops (or something like that) function, or an error in the Heuristic section. If it's the heuristic section it's because of one line I changed which will give slightly different values. If it's the TestChainHops it can be fixed by changing Easy to Medium

[kubasub]

It's the TestChainHops, but I think at this point I think we should just get err printed

[JamesKostiuk]

It's fixed. Because Easy was putting a random restriction on the maximum length it can hop, sometimes it wouldn't hit the max distance, so the test would fail. This has been corrected