Re: commit: Fix specifier inheritance behavior

Michael Sperber writes: We *are* talking about the commit 'bot. But you're not helping at all---you've made zero progress toward getting in my shoes (which I suspect are the same style as those worn by Didier and Jerry). So I'm thinking out loud in hope that you'll get the picture. Can we agree on the following? It is *not* a good thing when almost all merge notices simply replicate an earlier (series of) patch(es). It is *not* a good thing when there's no easy way to identify nontrivial merge notices (where a human had to sort out a conflict). It is *not* a good thing when every merge notice is preceded by a patch notice that appears to remove all work done concurrently with and committed before the patch in question. These all have been problems with the commit 'bot in the past. These are a natural consequence of generating commit notices with "hg diff -r (N-1) -f N". I'm not sure what the 'bot does now, since we haven't seen any merges recently. _______________________________________________ XEmacs-Patches mailing list XEmacs-Patches(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches