Trent W. Buck wrote: > Ganesh Sittampalam <[email protected]> writes: > >> On Tue, 17 Nov 2009, Trent W. Buck wrote: >> >>> Ganesh Sittampalam <[email protected]> writes: >>> >>>> Generally when reviewing patches, I read them directly, perhaps >>>> have a look at the source they apply to, and then decide that >>>> they're fine so long as the tests pass. Actually keeping track of >>>> them between that point and running the tests and pushing is a bit >>>> of a pain, and there's a cognitive shift that feels quite >>>> disruptive. >>> >>> Let me turn this around: how hard would it be to automate this, so >>> that patches that don't apply cleanly are automatically set to >>> needs-amend? >>> >>> Is it as simple as "apply --dry-run $attachment --repodir ..." >>> within >>> the procmail script? >> >> I think the difficulty here is uniquely identifying the correct >> attached bundle to apply, once a patch has been amended or similar. > > I was assuming that this was happening as the mail came in, before it > got injected into roundup. It would use the first MIME attachment of > type text/x-darcs-patch (or whatever "darcs send" uses). It would > simply ignore any mail that was sent without such an attachment (i.e. > people who send -O and then manually attach without setting the > appropriate MIME type).
But the end result of testing the patch needs to affect roundup somehow - I guess you are implying that it should always be the latest patch that controls the roundup status? Also, this idea doesn't work so well for doing something long-running or possibly unreliable like building and running the tests. >> This has been a bit of a source of concern to me from the point of >> view of human confusion too. Perhaps amendments should become fresh >> items in the tracker? > > I think that problem is orthogonal to the OP's problem :-) I freely admit to having two orthogonal problems ;-) Ganesh =============================================================================== Please access the attached hyperlink for an important electronic communications disclaimer: http://www.credit-suisse.com/legal/en/disclaimer_email_ib.html =============================================================================== _______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
