On Wed, May 18, 2005 at 01:39:07PM -0700, Tupshin Harper wrote: > David Roundy wrote: > >I don't think this'll make a big difference efficiency-wise (once we > >introduce the new hunk formatting plan), but it does look perhaps a bit > >cleaner, it fixes a UI issue. It also might simplify conflict marking a > >bit. > > Could you or somebody else summarize the new hunk formatting plan? I've > been trying to follow all the planned changes but I don't have a good > mental picture of the scope of this change.
It's something that I think Juliusz and Ian and I talked about a while back on darcs-devel. The realization is that all the +'s and -'s in hunk patches really are irrelevant. If we instead just wrote something like hunk ./foo 1 - (1 lines, 9 bytes) + (2 lines, 9 bytes) old hunk: 123456789 new hunk: abcd efgh we could just copy the contents of the hunks verbatim into the patch (and read them from the patch). This would speed up formatting, and would mean that when we parse the patch we don't have to break the hunks into lines. Please note that the actual syntax hasn't been hammered out, just the information that needs to be conveyed--which is the number of lines and number of bytes. The idea is to introduce this patch format after we've introduced the repository formatting code, and darcs will only write the new patch type if it's supported in the repository format file. And any older darcs that supports repository formatting will die with a nice error message when it tries to read a repository containing new-format hunks. So the key here is that we need to get the repo format stuff in soon, since we'll only be backward-compatible as far as the first version of darcs that supports repository formatting. But by default we will create old-style hunks, so by default you'll have real backwards-compatibility. -- David Roundy http://www.darcs.net _______________________________________________ darcs-users mailing list [email protected] http://www.abridgegame.org/mailman/listinfo/darcs-users
