You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
ATM we test a little bit of code for compliance with direct mode, using hand-picked tests. IMHO this should change. There is quite a chunk of usable functionality in direct mode, but from the test perspective it doesn't look like it. I think we should switch to a negative list, disabling only test that fail (possibly for a known reason). The current approach will or will not allow for the status quo to get worse -- we just won't know -- hence it isn't suitable for making progress towards direct mode compatibility.
The text was updated successfully, but these errors were encountered:
This is a bit more of an effort than I expected it to be.
Apparently nose can't easily exclude a list of tests out of the box, but only entire files/directories or via regexes.
Current approach: Have a "dry-run" of nose to discover all tests (sadly by name), do a lot of grep'ing and sed'ing to get a unique list of tests adressed via /path/to/file.py:test, build difference to exclusion list and build a nose.cfg from the result to run those tests only.
The exclusion list itself (actually there are two - DM and V6) is build similarly but in an one time effort and is supposed to be maintained manually.
ATM we test a little bit of code for compliance with direct mode, using hand-picked tests. IMHO this should change. There is quite a chunk of usable functionality in direct mode, but from the test perspective it doesn't look like it. I think we should switch to a negative list, disabling only test that fail (possibly for a known reason). The current approach will or will not allow for the status quo to get worse -- we just won't know -- hence it isn't suitable for making progress towards direct mode compatibility.
The text was updated successfully, but these errors were encountered: