RE: early version of the XML Query Test Suite
"Michael Kay" <[email protected]> Sun, 14 Aug 2005 22:04:28 +0100
| Newsgroups | gmane.comp.web.query-languages |
|---|---|
| Message-ID | <[email protected]> |
You have to accept that it's intrinsically unlikely that a test suite produced by W3C, or by any other organization, will happen to be organized on the same lines as the one you have designed in-house. Michael Kay http://www.saxonica.com/ > -----Original Message----- > From: [email protected] [mailto:[email protected]] On > Behalf Of Per Bothner > Sent: 14 August 2005 07:55 > To: Liam Quin > Cc: [email protected] > Subject: Re: early version of the XML Query Test Suite > > > Liam Quin wrote: > > So the test cases aren't written in the eXtreme or agile programming > > sense of identifying bugs, > > I'm sure you'd agree that a regression testsuite is essential for any > serious program development (at least of a programming langauge > implementation), not just "eXtreme or agile programming". > > > but rather to give coverage of the > > specification, and this different goal leads to a different > approach. > > Right. However, since implementors *will* need their own regression > test suites, it would make things easier if we could have a framework > for multiple purposes. Implementing and maintaining multiple testing > frameworks is possible (and probably not very difficult), but it does > add an extra management burden and needless complexity. > -- > --Per Bothner > [email protected] http://per.bothner.com/ > >