On 1/15/18, Karel Gardas <[email protected]> wrote: > > I've been always wondering how serious are various failures man gets > from building/testing fossil release.
Probably I should pay more attention to the tests. But the truth is, I never run them. My way of testing Fossil is to put the latest copy on my development machine, on https://fossil-scm.org/ and on https://sqlite.org/ and run it for a few days. If you look at the very faint line at the bottom of any screen on the Fossil self-hosting website (ex: https://www.fossil-scm.org/fossil/timeline) it tells you exactly which version of Fossil generated that screen. The version is usually the latest trunk check-in, or if not the latest, not more than one or two versions behind. There is a similar line at the bottom of the page on https://www.sqlite.org/src/timeline and on various other sites I host. So, in other words, testing consists mostly of running Fossil in the wild. On the SQLite and Fossil sites alone, Fossil generates about one page-view per second. All day. Every day. Maybe I'm being naive, but I feel like that is plenty of testing. Any problems that appear when running "make test" are very likely in the test logic, and not in Fossil. The tests in Fossil have always been a bit dodgy. We'd welcome your help in making the Fossil tests better! :-) -- D. Richard Hipp [email protected] _______________________________________________ fossil-users mailing list [email protected] http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
