On Sat, 2015-12-19 at 14:51 +0100, Dirk Bächle wrote: > […] > not with Mercurial so far, but here's what I tend to do at work > (Clearcase) when things get complicated: > > For every conflicting file I'll accept the "new" changes completely, > overwriting old stuff. This is just to get the basic "merge" > operation through. > Afterwards, I look at the single "diffs" and do the actual merge work > by trying to amend those places where some of the "old" stuff > is still needed.
I think I may go this route. If this were Git or Bazaar I can perform the full merge and then generate a diff to capture things that got deleted. Mercurial must have this capability somewhere. > Note how you can list the "to be resolved" files with > > hg resolve -la My difficulty just now is that my set up is to spawn a meld job for each (3-way) merge conflict resolution. For my normal workflow this works well as I get maybe two or three files in need of working on, and few changes to resolve. For this case, there are 30 or 40 files that have merge conflicts, many of which are trivial, but some of which need real thought. I am rapidly coming to the conclusion, that forcing the full set of default changesets overiding the supposed conflicts and then finding the problems and reusing the previous solutions at that point is the way forward. My reasoning here is that for a couple of the conflicts I saw, the new changesets were deleting code that the old python3 brnch had changes in. This is most easily handled by not handling and letting the default changesets win the conflict battle forcibly. Also doing this allows me to do "conflict resolution" file by file individually. The problem I was having earlier was that Mercurial was driving the use of meld to resolve conflicts automatically and not completing the merge quickly. > Hope this helps you further, you may want to combine the "amend" work > with using the MQ extension... I think this has been most useful, in that it has made me decisive. The 16 months of changesets will go in forcibly and then I will redo the python3 updates based on the overwritten changesets. I will not rely on history in the repository but have two clones one the past, one the future, then I can run scripts (probably written in D, as I am feeling perverse :-) comparing the old and new at my leisure. You have to love DVCS. :-) -- Russel. ============================================================================= Dr Russel Winder t: +44 20 7585 2200 voip: sip:[email protected] 41 Buckmaster Road m: +44 7770 465 077 xmpp: [email protected] London SW11 1EN, UK w: www.russel.org.uk skype: russel_winder
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Scons-dev mailing list [email protected] https://pairlist2.pair.net/mailman/listinfo/scons-dev
