On Monday, August 10, 2015, Peter Kelly <pmke...@apache.org> wrote: > Also one other point I forgot to mention about history for releases - this > is available from the git repository for anyone who wants it, by going to > the tag/branch for that particular release. > > Many projects include a ChangeLog file in their release with a summary of > what’s changed (though not as detailed as individual files or a complete > git log). I think ChangeLog files make sense on a per-release basis, then > people can use that as a guide if they want to go into the git log and see > all the details.
+1 a ChangeLog file as a simple git log extract makes sense. > > — > Dr Peter M. Kelly > pmke...@apache.org <javascript:;> > > PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key> > (fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966) > > > On 10 Aug 2015, at 1:55 pm, Peter Kelly <pmke...@apache.org > <javascript:;>> wrote: > > > >> On 10 Aug 2015, at 11:51 am, Dennis E. Hamilton < > dennis.hamil...@acm.org <javascript:;>> wrote: > >> > >> I do that because source tarballs of release code don't have history. > Just a historical habit. > >> > >> Also, when I crib a file to a different project, likewise. > >> > >> Does it bother you so much that you don't want to see that kind of > thing from me? > > > > It’s not a major problem if it’s only a few config files here and there > like .gitignore. But I would not want to see it in source files. Looking at > the git log, we have a very extensive history of changes, and including the > commit messages (or even summaries) in every file would make them > considerably larger. It also complicates merging. I agree with you it would blow up the source for no reason, good argument. > > > > My preference is that we use git in the way it’s designed to be used, > instead of a roll-your-own log system in individual files (esp. when many > changes span multiple files). I don’t think I’ve ever seen anyone else do > this in a project which already uses a version control system. +1 it is mandatory that we keep the info in the version control system, the other can only be extra. I have seen a couple of projects that (somehow) get the git message automatically inserted in the source, not something I like. rgds jan i > > > > — > > Dr Peter M. Kelly > > pmke...@apache.org <javascript:;> > > > > PGP key: http://www.kellypmk.net/pgp-key < > http://www.kellypmk.net/pgp-key> > > (fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966) > > > > -- Sent from My iPad, sorry for any misspellings.
