[darcs-users] Eclipse integration, refactoring, and combinato rial explosion?

[Kevin Smith]

David Roundy wrote:
> On Fri, Dec 12, 2003 at 04:05:35PM +0100, Marnix Klooster wrote:
>>So this really is language-dependent, and not simple search-and-replace.
> 
> Yes, but it can at least be approximated by a set of existing darcs
> patches.

(snip)

Yes. It can already be approximated with hunk patches, and perhaps more 
closely with replace patches. However, the example given is one of the 
simpler cases to work around.

Consider the case of renaming a member method from "start" to "begin", 
where it is called from twenty files, and there are several other 
classes with both "begin" and "start" methods, both defined in and 
called from those same twenty files.

As far as I can tell, replace cannot help with this. Even if we did 
decide to use it, any overlapping replace patches from other branches 
would be more likely to be wrong than right.

Now, darcs is already no worse than any other RCS in this case. So we're 
not trying to "fix" it. We're just trying to figure out if there is some 
way that darcs can jump from being a next-generation RCS to being the 
first next-next generation of RCS, leaving all the others way back in 
the dust :-)

I still stand by my claim that the replace patch is more often dangerous 
than helpful, by the way.

Kevin