On May 29, 2009, at 2:28 PM, C. Titus Brown wrote:
> Marek, note that I pulled apart some of the functions in blast_test so > that I could get separate failures for blastall problems vs other > issues. Having tests that may fail at multiple points makes it > possible > to run the test, have it fail at the first point, and then not run the > remaining parts of the test. I see one minor problem with commit 207547, which is heading the opposite direction vs. what we are now doing to the "live" (as opposed to "canned") blast tests. We recently added a test that checks that the queryDB option produces the exact same results as running the sequences one by one. You changed that test to test against a fixed set of "correct results". But blast produces slightly different results on different platforms; that is why we get errors in blast_test. We coded our queryDB-matches-single test in such a way that it is insensitive to blast differences on different platforms; you've changed it back into breaking on any platform where blast produces slightly different results. If you don't mind I'll convert that test back by hand... -- Chris --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "pygr-dev" group. To post to this group, send email to pygr-dev@googlegroups.com To unsubscribe from this group, send email to pygr-dev+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/pygr-dev?hl=en -~----------~----~----~----~------~----~------~--~---
