Perl 6 - the future is here, just unevenly distributed

IRC log for #darcs, 2014-07-25

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

All times shown according to UTC.

Time Nick Message
00:29 Igloo joined #darcs
00:35 Igloo joined #darcs
00:38 favonia joined #darcs
01:25 mizu_no_oto joined #darcs
01:50 mizu_no_oto joined #darcs
02:12 Igloo joined #darcs
02:22 Igloo joined #darcs
05:40 dolio joined #darcs
06:59 lelit joined #darcs
07:17 edwardk_ joined #darcs
07:45 gh__ joined #darcs
07:46 gh__ hi
07:51 gh__ Heffalump, seems like the next haskell platform will provide the right version of containers for us: https://github.com/haskell/haskell-platform/blob/new-build/hptool/src/Releases2014.hs#L17
08:31 Heffalump gh__: cool
08:31 edwardk joined #darcs
09:27 lelf joined #darcs
11:05 lelf joined #darcs
11:21 mizu_no_oto joined #darcs
12:12 mizu_no_oto joined #darcs
13:19 lelf joined #darcs
13:20 lelf left #darcs
14:44 mizu_no_oto joined #darcs
15:29 amgarchIn9 joined #darcs
15:37 edwardk joined #darcs
15:51 edwardk joined #darcs
16:45 mdiaz joined #darcs
17:08 siel joined #darcs
17:44 edwardk joined #darcs
18:10 amgarchIn9 joined #darcs
18:11 Heffalump mdiaz: hi
18:14 mdiaz hi Heffalump
18:16 mdiaz I haven't finished yet, :-/.
18:19 Heffalump how far have you got?
18:26 mdiaz I'm writing to a file the list of commands that can be undone. In another file I write the index of the list.
18:28 lelf joined #darcs
18:28 mdiaz I added two arguments to the finalize... function , the flag Undoable/NoUndoable and a string with the name of the command who called it.
18:29 mdiaz I made a small interface to display the list of commands to undo.
18:30 Heffalump ok, that sounds like decent progress
18:31 Heffalump do you think we should also store the arguments to the commands as well as the names? This is all for the user's information only, right?
18:33 mdiaz I'm still not sure how to do that. The information passed to the functions who call finalize*  is incomplete.
18:34 mdiaz Also, Im' not sure if I can get the name of the command in all cases.
18:36 mdiaz Unless I hard code the names.
18:36 mizu_no_oto joined #darcs
18:40 Heffalump I think just hard coding the names at each call site is fine
18:40 Heffalump if the arguments aren't easily available then just forget them
18:40 mdiaz Ok.
18:42 mdiaz That is going to look weird in the case of aliases and commands that call the same function.
18:52 mdiaz s/weird/imprecise/
18:59 Heffalump hmm, true. Never mind though.
18:59 Heffalump It'll do for a first pass, we can think about how to improve it later.
19:01 Heffalump if you can easily access/pass in better information, then do, but otherwise just hard-code
19:01 mdiaz Oki.
19:05 lelit joined #darcs
19:13 lelf is it me or rollback is broken (does nothing)?
19:14 edwardk joined #darcs
19:35 amgarchIn9 joined #darcs
19:42 Heffalump lelf: sounds like just you
19:42 Heffalump I use it quite a lot
19:42 Heffalump oh, there is one known confusing case, where if you say "don't rollback this tag" then it doesn't offer you anything else because the tag depends on everthing else
19:43 lelf For me 2.8.4 works, fresh HEAD is not (just quits after selecting hunks)
19:48 Heffalump hmm, that's strange
19:48 Heffalump if it got as far as selecting hunks then that's not the problem I described above
19:48 Heffalump I'll give it a try..
20:10 Heffalump seems to work for me. I just remembered that it now doesn't create a "rollback" patch, it just rolls back the changes in working, as that's what 90% of uses want. Were you expecting a patch too?
20:11 lelf Ehmmmm. Yes? :)
20:13 Heffalump :-)
20:13 Heffalump you can just use darv record afer if you do want the pach
20:13 lelf And it even allows you to rollback it over and over again without saying anything
20:17 Heffalump hmm, that's not great
20:18 Heffalump it ought to probably print "Changes rolled back in working directory" or something
20:32 Rastus_Vernon joined #darcs
20:49 Heffalump joined #darcs
20:55 Heffalump mdiaz: one thing that occurred to me is to check what happens to pending when you undo - e.g. test your existing case of amend-record  with something like a previous "darcs add"
20:56 mdiaz Ok, I'll check it.
21:06 amgarchIn9 joined #darcs
21:08 mizu_no_oto joined #darcs
21:09 Heffalump lelf: I'll send in a patch to print out that message (it's written locally, just needs a bit of fiddling around to upload it)
21:10 lelf Heffalump: thx!
21:12 Heffalump silently doing the same thing again is because of the way darcs 2 handles duplicates silently
21:13 Heffalump you get a real conflict if you do have conflicting local changes, so I think the behaviour is reasonable
22:16 mizu_no_oto joined #darcs
22:23 alegadea joined #darcs
22:30 mizu_no_oto joined #darcs
22:51 mdiaz_ joined #darcs
23:24 lispy_ joined #darcs
23:29 edwardk_ joined #darcs
23:32 lambdabot joined #darcs

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