Re: [General] Language-independent test representation

I haven't been following this thread closely enough, but just wanted
to quickly say that the approach I'll probably take to certifying how
the source tests will be translatable into Perl will be to make sure I
can translate a small subset (one node type, for example).  With that
info, I can definitely feed back any gotchas doing the translation for
Perl.

Re. whether it's in XML or Java, I see the value and ease of writing
the tests in an existing language (with constraints) and then
translating to other languages.  Again, by translating a subset of the
tests, I can feedback what constructs can't easily be automatically
converted.  I also agree with the point about making several
assertions in one test case (a la xUnit).

But note that these are just gut feelings, I won't be able to get to
this for a few weeks.

  -- Ken

Received on Monday, 23 April 2001 18:49:24 UTC