On 16.10.2009 14:16, Sune Foldager wrote: > On Friday, October 16, 2009, at 02:02PM, "Adrian Buehlmann" > <adr...@cadifra.com> wrote: >> Ah sorry. >> I didn't know that. > > Np; the feature is not quite done yet either, as it doesn't work with 'hgtk > vdiff' and consequently with diffing from the shell context menu. I have been > meaning to do that sometime. > >> What I was trying to solve was that there was no way to see the diff >> to the second parent with "Visualize Change", since it always showed >> only the diff to the first parent, no matter whether "other >> parent" was activated or not (the list of files window always >> showed only the files that changed compared to the first parent). > > Yes, and that's a very good thing, I agree :-). Even better is to see both at > once, of course, if the differ supports it. > >> Should we add an extra test for when vdiffnowin is set then? >> I.e. always only define >> >> opts['change'] = str(rev) >> >> for merge csets if vdiffnowin is set? > > Yeah, that's another possiblity that would do it. Maybe it's a bit 'ugly' to > have two checks along the code path, but on the other hand... well... what > does Steve think? :-p. I think I am +1 on it. > >> What 3-way diff tool are you using and how have you configured it in .hgrc/ >> mercurial.ini? (If I knew that I could test your setup?) >> I tested with kdiff3. > > I also use kdiff3, with the following configuration in mercurial.ini: > > [extdiff] > cmd.vdiff = kdiff3 > opts.vdiff = $parent1 $child $parent2 > > And then > > [tortoisehg] > vdiff = vdiff > > as well. For non-merge changesets, $parent2 is simply left out by extdiff. > 3-way mode is only used if the $parent2 token is found in the opts line. > > When I said that vdiffnowin doesn't work from hgtk vdiff and shell > extensions, it's actually a bit worse, since it doesn't understand the new > tokens when parsing the extdiff config, and so can't launch the differ > correctly. Something should be done about that as well, but perhaps Steve > would like to directly support 3-way diff in thg now that my patch is in hg. > We talked about that a while ago. >
I can reproduce the problem with 23e60a084681 now. Thanks for posting your config settings. Using revision 74061568b17f (the one before my patch) I get a three-way diff for merge changesets in kdiff3, with 23e60a084681 (my patch) kdiff3 shows only two way diffs for merge csets. The three-way diffs are a really interesting feature. ------------------------------------------------------------------------------ Come build with us! The BlackBerry(R) Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9 - 12, 2009. Register now! http://p.sf.net/sfu/devconference _______________________________________________ Tortoisehg-develop mailing list Tortoisehg-develop@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/tortoisehg-develop
