On Fri, 4 Jun 2004, Stas Bekman wrote:
Jeff Trawick wrote:
There is one more problem. Not the whole API is tested and many APIs are tested only partially. IMHO, you want to run coverage tests first and only after passing a sensible coverage threshold, should 1.0 be released.
This goal is in the STATUS file as a showstopper, but it has been decided that it isn't really a showstopper. While I agree with you (obviously, I'm the one who added it as a showstopper years ago), the reality is that very few people have been interested in writing tests for APR, which is why every new test generally finds more non-portable bugs in APR.
Please see my solution proposal in reply to Greg.
If we waited to have a full test suite, at the rate our test suite is growing, it would take 10 years to really be ready to release 1.0. :-(
We face the same problem with APR perl glue. Partly due to apr itself (not having API frozen and us finding problems in it), partly due to a lack of help writing tests. The happy compromise that we have found recently is that for mod_perl 2.0 we are going to release only those APIs that have been used and tested and fully documented. Other APIs will be there, but they will be a subject to change post 2.0 and therefore should be not relied upon. I'm half way through finishing the tests and documentation and predict having mod_perl 2.0 out somewhere this summer, using the new approach.
Also, IMHO, all those folks who are pushing for 1.0 should instead help writing tests. It's not fun, but asking to release 1.0 just because the sub-set of the apr API that you happen to use works for you, is just plain not so nice to others who rely on other parts of apr.
-- __________________________________________________________________ Stas Bekman JAm_pH ------> Just Another mod_perl Hacker http://stason.org/ mod_perl Guide ---> http://perl.apache.org mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com http://modperlbook.org http://apache.org http://ticketmaster.com
