On Wed, Jan 24, 2018 at 6:49 PM Branko Čibej <br...@apache.org> wrote:
> On 24.01.2018 22:32, Julian Foad wrote: > > When 'svn patch' applies an 'add file' patch onto a WC path whose > > local schedule is 'delete', it changes the schedule to 'replace'. > > > > Sequences such as this... > > > > > svn rm foo > > > svn diff > rm-foo.diff > > > svn patch --reverse-diff rm-foo.diff > > > > or this... > > > > > svn add foo > > > svn diff > add-foo.diff > > > svn commit > > > svn patch --reverse-diff add-foo.diff > > > svn patch add-foo.diff > > > > ... result in 'foo' being scheduled as 'replace'. > > > > stsp and I discussed on IRC ( > > http://colabti.org/irclogger/irclogger_log/svn-dev?date=2017-12-15#l19 > > ) and agreed that this is not what users would generally want or expect. > > > > I propose to make 'patch' always generate a 'modified' (or unmodified) > > schedule when it applies an 'add file' diff (or reverse-applies a > > 'delete file' diff) onto a schedule 'deleted' working copy file. > > > > It is not inherently necessary that shelving and manual use of 'svn > > patch' should share the same solution, but in this case I think that > > is best. > > Why is this not what users would expect? "Delete" + "add" has always > been "replace" in Subversion. The only other reasonable option I can > think of would be to generate a delete/add tree conflict and let the > user decide what to do about it. Silently undoing an "svn rm" in the > working copy is exactly what I would _not_ expect. Both 'svn rm' and > 'svn patch' are explicit user operations and we can't just assume that > one or the other were mistakes. > > FWIW, I agree with Branko. By starting a tree conflict the user can chose a path to go. If svn marked the file modified, they would not actually have a way to accomplish a "replace" if that was the desired behavior (though I suppose a delete, commit, then 'svn patch' could be used in that scenario). Troy -- Brane > >
