Hi Myrna, Thanks for pointing me to that issue :) I didn't know there was work in progress for it already, so I'll just assign that issue to me and try to pick up on Junjie's patch.
Tiago On Sun, Mar 22, 2009 at 10:18 PM, Myrna van Lunteren <[email protected] > wrote: > On Sun, Mar 22, 2009 at 2:53 PM, Tiago Espinha <[email protected]> wrote: > > Hey guys, > > I am working on DERBY-4090 and I started with the > > holdCursorExternalSortJDBC30.sql . The thing is that on the file's > overview, > > 10 tests are reported as follows: > > -- overview: > > -- TEST 0: test hold cursor with external sort (order by). > > -- TEST 1: basic heap scan tests (multiple rows) > > -- TEST 2: basic btree scan tests (zero rows/update nonkey field) > > -- TEST 3: basic btree scan tests (multiple rows/update nonkey field) > > -- TEST 4: basic btree scan tests (zero rows/read only/no group > fetch) > > -- TEST 5: basic btree scan tests (multiple rows/read only/no group > > fetch) > > -- TEST 6: basic tests for cursors with order by > > -- TEST 7: test of hold cursor code in > > DistinctScalarAggregateResultSet.java > > -- TEST 8: test of hold cursor code in GroupedAggregateResultSet.java > > -- TEST 9: test scan positioned on a row which has been purged. > > -- TEST 10: test scan positioned on a page which has been purged > > However, only two actual tests exist in that file. Is this normal or am I > > missing something? Why would the remaining tests have been removed? > > Tiago > > Hi Tiago, > > Converting this test is actually DERBY-3842 - currently owned by > Junjie, but I think you can just assign yourself and link it to > DERBY-4090. Junjie made a patch but there were some comments that > never got addressed. > > But to get to the point you were asking - looks to me like the top of > comments at the top of the test are somewhat bogus - looks like > someone might have done a cut-and-paste from holdCursorJDBC30.sql and > never adjusted the comments. > > HTH. > Myrna >
