On Fri, Mar 26, 2010 at 10:41 AM, Martin Bligh <[email protected]> wrote:
> On Fri, Mar 26, 2010 at 10:09 AM, John Admanski <[email protected]> > wrote: > > I don't suppose I could convince you to use a character other than '.' to > > identify hierarchal tests. I know it's the more natural way to do it in > > python, but it now means that if I see a result for X.Y I can't tell if > it's > > test X with tag Y or test X.Y, and if you have multiple tags it's even > > messier (is X.Y.Z test X or X.Y or X.Y.Z?) > > In practice it probably won't be much of a problem for a person, there > > probably won't be much of a naming conflict, but when trying to write any > > kind of generic queries or scripts it'll be more of a pain. > > Hmm. fair point. I'll see if I can change it to / ... I ran into problems > doing > that the first time, can't recall why. Will do it as a separate patch > though. > I doubt / can be made to work without a lot of...work. We use the test name to construct result paths all over the place, so using a character which the filesystem will interpret is just going to be painful. Really, we should actually add an explicit check to run_test to reject any test name or tag with a / in it. I recommend using some other glyph (colon?) that means nothing to both python and the filesystem.
_______________________________________________ Autotest mailing list [email protected] http://test.kernel.org/cgi-bin/mailman/listinfo/autotest
