Pierangelo Masarati writes: > test040 is currently specific to back-hdb because only back-hdb > supports subtree rename (it could be moved to a backend-specific > add-on to test005)
That's true. Though if back-newWonderfulDatabase is implemented someday and also supports subtree rename, test005 would have to be renamed back. I wasn't even thinking of such cases, I was looking at the sql tests and passwd-search - these backends cannot use the general tests. Different config files, different database contents. Same for back-dnssrv, back-null, and others. For back-perl and back-shell one would have to implement a lot of OpenLDAP as perl or shell scripts in order to run the full test suite, I doubt anyone would bother:-) Yet now that I think of it, there ought to be some general tests which both back-sql and these other backends can run. > test035 and test036 are specific to back-meta, but use any of > bdb,hdb,ldbm as remote storage > > test028, test029 and test039 are specific to back-ldap; same as above So we don't rename them. If they are made backend-specific we get into trouble, but if we leave them as they are they work fine. Same for test020 and test030 below. > test020 is specific to slapo-pcache, and uses back-ldap as proxy and > any of bdb, hdb or ldbm as storages; it should be modified to run with > back-meta (and possibly with back-sql, as it may benefit from local > caching) as proxies Hmm. My scheme still wouldn't run this test for back-sql. Maybe tests named "scripts/all-test*" should be run for any backend. Then add a more convenient scheme for specifying which backends support which features, and filter out which tests to run that way. Could handle at least test005 above the same way. One important feature would be whether or not slapadd is supported. Come to think of it, which backends can run a test chould be seen as just another feature, maybe (Or most other tests could be renamed dbtest* or something which would only be run for bdb, ldbm and hdb, but that seems more invasive...) > test030 can test back-relay, back-ldap and back-meta, using any of bdb, > hdb or ldbm as storages; this currently tests all 9 combinations, so ot's > pretty exaustive; maybe too much, in some cases. > I mean: when only one backend type is involved, everything is fine; when > more are involved, we should allow the full test matrix as well as partial > ones, trading testing time versus exaustion. In principle, since the test > suite is mainly intended as a means to trace regressions, exaustion should > be the preferred choice. Yes. -- Hallvard
