> > On Jun 18, 10:44pm, Andrej Borsenkow wrote: > } > } I was unsure about added value of this patch and there was no comments. > > Certainly it would be nice to at least *explain* the test failures for > the Y* tests when --disable-dynamic ... either that, or we should add > zpty to the list of modules that are always linked by default. > > Probably comptestinit should not proceed if zmodload zpty fails, and > the %prep section should fail in that case, so that we don't report > failures on individual tests. > > If Andrej's patch -- or some combination of it with the below -- isn't > considered better, how about this? > I guess, we better leave completion tests as in Bart patch. It is impossible (at least, I have no idea how) to decide if zsh/zpty failed or just was missing, so skipping other tests is the most we can do. And here is for the rest. -andrej
Attachment:
zsh-unimplemented.diff
Description: Binary data