Camelia, the Perl 6 bug

IRC log for #darcs-theory, 2008-09-08

| Channels | #darcs-theory index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
10:09 kowey joined #darcs-theory
16:35 lispy Igloo: I read some emails and now I think it might be helpful if you responded to David's email and said that minidarcs is a darcs prototype, for demoing semantics, not a reimplementation that is trying to steal thunder.
16:35 Igloo lispy: I think it's working on a replacement theory that he is really objecting to
16:37 lispy hmmm, well what I'm saying would re-enforce it has a different patch theory
16:38 lispy The discussion seems to be delicate
16:40 Igloo I'm pretty sure David understand that it has a different theory
16:41 kowey he just doesn't think of darcs 2 theory as being broken
16:41 lispy Yeah, I wasn't arguing, just understanding what you said and agreeing
16:41 lispy kowey: and yet he's said darcs2 usage should be avoidid
16:41 lispy avoided*
16:41 lispy I'm really confused by that
16:41 kowey well "fundamentally broken"
16:41 kowey he thinks new repos should be in darcs 2
16:42 lispy Then why isn't that the default?
16:42 kowey old repos should probably think twice about converting (though hashed is a very good idea)
16:42 kowey probably because there aren't enough darcs 2 out there yet
16:42 kowey and that gcau needs to be re-written to support darcs 2 properly
16:42 lispy I wonder what that means
16:43 lispy BTW, shouldn't gcau be called unmerge?
16:43 kowey he says that it's based on assumptions that are no longer true under darcs 2
16:43 kowey my hand-written notes that are taking forever to get transcribed
16:43 lispy That's a very fundamental function, so that's a scary statment
16:44 Igloo gcau does exactly what the acronym says (I think; I don't have the code in front of me)
16:44 kowey i think get_common_and_uncommon is pretty explicit
16:45 kowey whereas unmerge might require some thought
16:45 kowey for example, the unmerge name would lead me to think that each one of the two results is a separate branch
16:46 Igloo Right, unmerge would take a merged thing
16:46 Igloo Not two things to be merged
20:47 ertai Igloo: I use darcs2 repo formats for all my new repositories without any problem
20:47 ertai Igloo: in fact this was for everyone sorry
20:48 ertai However I have a very disiplined usage of darcs w.r.t conflicts.
20:57 lispy Igloo: ah, it's because of the return type of gcau that I think unmerge is a nice name
20:58 lispy Igloo: you feed it two, possibily identical repositories, and it tells you which patches are unique to each
21:32 kowey joined #darcs-theory

| Channels | #darcs-theory index | Today | | Search | Google Search | Plain-Text | summary