On Fri, 2008-09-26 at 00:25 -0600, Shawn M Emery wrote: > There seems to be different standards in how changesets are commented > using mecurial. In specific if there are multiple CRs and multiple > modules being modified there is a disconnect in the changeset and the > associated comments. This makes it difficult to correlate specific > changes in a module to the corresponding CR. Could we have a process > that requires separate changesets corresponding to individual CR comments?
Won't this be very difficult with the current situation of per-repo changesets and thus the requirement to do a "hg reci" before pushing? I agree that it would certainly make more sense to have multiple changesets for multiple CRs. -M
