Re: PHPUnit bugs/fixes

May be we should add a "strong-typed" and "typeless" mode to PHPUnit??? Lukas Smith wrote:

From: Wolfram Kriesing [mailto:lists@kriesing.de] I am getting back to this. who uses PHPUnit (extensively), please jump in and lets solve this! I think a good test platform should be tested too :-) and we need a good test platform for all the packages (BTW i think all the packages (including those i am maintaining) should have a PHPUnit test, so verification would be an easy task)

I dont know PHPUnit all that well. Anyways I know that Paul Cooper started on PHPUnit tests for MDB. Working on MDB I cannot say how important Metabase testing suite currently is to ensure that the (new) drivers are doing what they are supposed to. But then again running your tests against the Metabase Wrapper of MDB is not cool, nor does it make the bugtracking very transparent :-)

can u go more into detail? what do you really mean by "running your tests against the Metabase"?

So I really want to have a native test suite for MDB.

why that? if PHPUnit was a unified testing platform, which strictly adheres to a spec, which in my eyes basically is JUnit with the necessary modifications for PHP, then it should be the best testing tool. (online there are some articles about how to test DB-applications using JUnit, this should be a start ,only diff is that we use PHPUnit :-) ) spec with modifications to the JUnit? i think they are necessary simply because PHP as opposed to java is i.e. typeless. in php you can rely on the result of

    $count = $db->getOne("SELECT count(*) FROM myTable");

to be the same as

    7
Which would not be the fact according to PHPUnit test as they currently work, since $count is a string and 7 an integer.

-- Wolfram ... translating template engine .... http://sf.net/projects/simpletpl ... authentication system .... http://sf.net/projects/auth