Alan Barrett wrote: > On Tue, 17 Nov 2009, Julian Foad wrote: > > Stefan Sperling wrote: > > > 3) Hunks matching at or near their original offset in the target > > > file are always preferred over hunks matching at the same > > > location but with a greater offset relative to their original > > > offset. > > > > Would I be right in guessing that "matching at the same location > > but with a greater offset relative to their original offset" means > > "matching further away"? > > Point 3 is talking about how to choose between multiple hunks that > could all match at the same location in the input file, not about > how to choose between multiple locations in the input file that > could all be matched by the same hunk.
Oh yes, I see now. Thanks. Alan Barrett wrote: > By the way, patch(1) is described in > <http://www.opengroup.org/onlinepubs/009695399/utilities/patch.html>. The algorithm described under "Patch Application" there is the minimal requirement, and is really simple compared to Stefan's. We don't have to try to be too clever. A simple algorithm can work reasonably well. - Julian
