Re: Design query: Ev2 and multiple moves

2011-11-05 Thread Daniel Shahaf
On Saturday, November 05, 2011 8:21 PM, "Greg Stein" wrote: > [ how to move A1->A2->A3->...->An->A1 circularly in a single txn ] ... > I'm not a master of topography or graphs, but I suspect that any given > permutation of N nodes can be reduced to a set of rotations of subsets > of those N nodes.

Re: Design query: Ev2 and multiple moves

2011-11-05 Thread Bill Tutt
I just know folks are going to hate me for mentioning this, but. Tom Lord came up with a way to store/apply arbitrary rename applications against trees. Apparently my google-fu failed me tonight because I couldn't find the original reference/explanation. So bear with me as I recall the details

Re: [PATCH] configure: pre-create build directories for VPATH builds

2011-11-05 Thread Jonathan Nieder
Daniel Shahaf wrote: > And thirdly, with this fix 'make extraclean-swig; make extraclean-swig' > will still fail :-) Seemed to work for me. > Perhaps then you'd resend a patch that fixes the bug differently? Gladly --- the mkdir-init method does seem nicer. Thanks for your review and kind advi

Re: [PATCH] configure: pre-create build directories for VPATH builds

2011-11-05 Thread Jonathan Nieder
Greg Stein wrote: > On Nov 5, 2011 6:06 AM, "Philip Martin" wrote: >> We have a mkdir-init target in the Makefile that is supposed to create >> the build directories. > > Right. It creates what is needed rather than a scattershot hope. Thanks for the pointer. Running "make mkdir-init" works fin

Re: Design query: Ev2 and multiple moves

2011-11-05 Thread Hyrum K Wright
On Sat, Nov 5, 2011 at 7:21 PM, Greg Stein wrote: > Julian raised a question when I saw him in September: how does Ev2 > deal with a node swap? ie. swap the contents of A and B, retaining > metadata that they were moves [rather than copies from history]. > > In Ev2, we attempt to disallow "mv A B

Design query: Ev2 and multiple moves

2011-11-05 Thread Greg Stein
Julian raised a question when I saw him in September: how does Ev2 deal with a node swap? ie. swap the contents of A and B, retaining metadata that they were moves [rather than copies from history]. In Ev2, we attempt to disallow "mv A B ; mv B C". The semantics around the interface say you should

Re: Editor v2 - suggestions and queries

2011-11-05 Thread Greg Stein
On Fri, Nov 4, 2011 at 11:16, Julian Foad wrote: >... >>> + *        internally respond to cancellation. If the driver then calls >>> + *        'complete' without intercepting the cancellation, the user >>> + *        would not be happy to see that commit being completed. >> >> That would be a bu

Re: svn commit: r1197998 - in /subversion/trunk/subversion: libsvn_client/patch.c tests/cmdline/patch_tests.py

2011-11-05 Thread Konstantin Kolinko
2011/11/6 Daniel Shahaf : > > > On Saturday, November 05, 2011 5:54 PM, s...@apache.org wrote: >> Author: stsp >> Date: Sat Nov  5 17:54:31 2011 >> New Revision: 1197998 >> >> URL: http://svn.apache.org/viewvc?rev=1197998&view=rev >> Log: >> Make 'svn patch' ignore "/dev/null" patch target paths to

Re: svn commit: r1197998 - in /subversion/trunk/subversion: libsvn_client/patch.c tests/cmdline/patch_tests.py

2011-11-05 Thread Daniel Shahaf
On Saturday, November 05, 2011 5:54 PM, s...@apache.org wrote: > +++ subversion/trunk/subversion/tests/cmdline/patch_tests.py Sat Nov 5 > 17:54:31 2011 > @@ -3858,6 +3858,61 @@ def patch_reversed_add_with_props2(sbox) > 1, # dry-run >

Re: svn commit: r1197998 - in /subversion/trunk/subversion: libsvn_client/patch.c tests/cmdline/patch_tests.py

2011-11-05 Thread Daniel Shahaf
On Saturday, November 05, 2011 5:54 PM, s...@apache.org wrote: > Author: stsp > Date: Sat Nov 5 17:54:31 2011 > New Revision: 1197998 > > URL: http://svn.apache.org/viewvc?rev=1197998&view=rev > Log: > Make 'svn patch' ignore "/dev/null" patch target paths to improve > compatibility with patche

Re: svn commit: r1197961 - in /subversion/trunk/subversion/libsvn_client: client.h copy.c diff.c info.c merge.c ra.c

2011-11-05 Thread Greg Stein
Woah. This is even crazier than what you fixed. *nobody* asks for the end revision. In fact, the code would have crashed since END_REVISION was always NULL. Thus, ->kind must have never been unspecified. But no matter, as the param is useless. Also note that only one callsite asks for the start, an

Re: svn_wc_adm_probe_open3 fails on windows

2011-11-05 Thread Barry Scott
On 4 Nov 2011, at 20:21, Bert Huijben wrote: > > >> -Original Message- >> From: Barry Scott [mailto:ba...@barrys-emacs.org] >> Sent: vrijdag 4 november 2011 19:57 >> To: Bert Huijben >> Cc: 'Subversion Development' >> Subject: Re: svn_wc_adm_probe_open3 fails on windows >> >> >> On 4

Re: [PATCH] configure: pre-create build directories for VPATH builds

2011-11-05 Thread Greg Stein
On Nov 5, 2011 6:06 AM, "Philip Martin" wrote: > > Jonathan Nieder writes: > > > Index: configure.ac > > === > > --- configure.ac (revision 1197804) > > +++ configure.ac (working copy) > > @@ -1374,6 +1374,15 @@ > > > > AC

Re: [PATCH] configure: pre-create build directories for VPATH builds

2011-11-05 Thread Philip Martin
Jonathan Nieder writes: > Index: configure.ac > === > --- configure.ac (revision 1197804) > +++ configure.ac (working copy) > @@ -1374,6 +1374,15 @@ > > AC_OUTPUT > > +if test "$abs_srcdir" != "$abs_builddir" > +then >

Re: [PATCH] commit --include-externals (v2)

2011-11-05 Thread Stefan Sperling
On Sat, Nov 05, 2011 at 12:48:09AM +0100, Neels J Hofmeyr wrote: > Not having the option of choosing a default, I'm rethinking which behavior > should be the default. I think both are valid use cases: > > UC1) Prevent accidental commits to stuff coming from different areas, so > exclude externals

Re: [PATCH] configure: pre-create build directories for VPATH builds

2011-11-05 Thread Daniel Shahaf
On Friday, November 04, 2011 5:41 PM, "Jonathan Nieder" wrote: > With r1197804, I tried: > > svn export ~/src/subversion svn-test > cd svn-test > ./autogen.sh > mkdir BUILD > cd BUILD > ../configure --prefix=$HOME/opt/subversion > make extraclean-swig