Camelia, the Perl 6 bug

IRC log for #kaizendo, 2011-07-29

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

All times shown according to UTC.

Time Nick Message
11:05 sveinns sjn: Hooray! Have you looked at my pull request?
11:09 sjn not really (distracted)
11:10 sjn lemme check :)
11:15 sjn sveinns: what was your thoughts behind changing the old README?
11:15 sjn I mean replacing it?
11:16 sveinns urk. How do I check the diff?
11:17 sveinns (I assume it must be a mistake to change the README, if t it isn't just a change in the rewuired packages.)
11:18 sveinns I'm checking the diff now.
11:20 sveinns It was replaced by the automatic package setup. Just ignore that one, if you can.
11:20 sveinns Can you?
11:20 sjn git diff $branch-or-revision $file
11:20 sveinns https://github.com/sjn/Kaizendo/pull/2#diff-1
11:20 sjn no hurries
11:20 sveinns More gentle to the eye.
11:21 sveinns no, just take the old README. We don�'t need the new one.
11:22 sjn add some info into the old one, describing what's going on there? :)
11:23 sjn also, if you can, add a few words in scrubberTest.pl describing the purpose of that program
11:23 sjn POD is probably a good format for that :)
11:24 sjn also, you checked in an ignore.txt file
11:24 sjn that's supposed to be a .gitignore or .svnignore or .cvsignore
11:25 sjn but we already have one of those, so update that one instead :)
11:25 sveinns Ahyup.
11:25 sjn do you know what's supposed to be in that file?
11:25 sveinns Ahyup.
11:26 sjn also, update the Changes file so it says something meaningful about what you've done :)
11:26 sveinns OK.
11:27 sjn next issue; how do you run the tests?
11:27 sveinns prove t/
11:27 sveinns right?
11:27 sjn sure
11:28 sjn if that's the way, then write it in the README file
11:28 sveinns :-D
11:28 sjn the usual way is to let a Makefile.PL handle that (so you can just write "make test")
11:28 sjn but you don't have to do that here
11:29 sjn these books are not exactly "regular Perl modules"
11:29 sveinns nope.
11:29 sjn although, having a "make install" that would install a book into some sane location for the app to use, would be really cool
11:30 sveinns Please just keep posting issues, I just have to go put a layer of paint on my Gran's carport.
11:30 sjn let's wait with that though
11:30 sjn ook
11:30 sveinns make install we'll think about later, yeah.
11:30 sveinns She REALLY wants it to get done today.
11:31 sjn also, why did you call the "cleaned up" book "plain"?
11:31 sjn and why did you check it in?
11:33 sjn yeah, you should have a Makefile.PL, at least to be able to specify what dependencies you have
11:34 sjn then just make the "install" feature there a no-op
11:39 sjn <- afk
18:37 sveinns sjn: afk still?
18:40 sjn nah
18:40 sjn preparing a talk for yapc
18:53 sveinns ok
18:57 sveinns ok, working on fixing the things you mentioned. Thanks for the feedback!
18:59 sveinns sjn: btw, I think we need the useful files, and not the source files for putting in a sub-repository. What we ideally need is just a representation of the original, not the original itself.
19:00 sveinns This is an argument for having the plain/ files.
19:13 * sjn is a _big_ fan of keeping source files
19:13 sjn and then always generating the final product from those
19:14 sjn meaning, one is forced to think about how to create the correct result with code, instead of just "fixing the source"
19:14 sjn fixing the source is meaningful if we have full control of the original source
19:15 sjn as in, we're the upstream for everyone else
19:15 sjn in this case, we're not, so we have to work on what upstream provides
19:16 sjn the principle at work here, is "keep the repository as authoritative as possible"
19:16 sjn no committing of generated stuff
19:17 sjn if we're going into the "let's just fix the data on behalf of upstream" trap, we'll never get out of it
19:18 sjn (does this make sense to you?
19:18 sjn )

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