Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-soc, 2008-05-28

| Channels | #perl6-soc index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
02:15 Auzon joined #perl6-soc
17:02 [particle1 joined #perl6-soc
17:58 Auzon we have a meeting in 1.5 hours, correct?
17:58 [particle] yes
17:58 Auzon Thanks.
18:02 moritz_ all those time zones are confusing ;)
18:02 Auzon Yep. That's why I used a relative time to make sure :)
18:43 pmichaud joined #perl6-soc
19:28 TimToady joined #perl6-soc
19:30 moritz_ ok, everybody here?
19:31 TimToady I'm just lurking...
19:31 moritz_ TimToady: you're the authorotative lurker ;)
19:31 [particle] hi-ho
19:31 Auzon Hello
19:31 moritz_ great, let's start
19:31 moritz_ Auzon: you first
19:32 Auzon OK. So I've added a few test cases this week and last Friday so far, as well as some reorganization.
19:32 pmichaud here.
19:32 Auzon I posted my spec schedule to the planet just a bit ago, since not everyone can see my application.
19:32 Auzon actual link: http://auzon.blogspot.com/2008/05/test-case-schedule-for-gsoc2008.html
19:33 Auzon I'm trying to constrain myself to those areas in the given week, since the spec is huge and I could waste a lot of time jumping between different synopses
19:34 moritz_ what takes the most time? reading S\d\d ?
19:35 Auzon Reading the spec takes a lot of the time. I started with fairly small ones this week, I think, so it's not horrible.
19:36 Auzon I spent a fair bit of time on Monday verifying the correctness of existing test cases that I wanted to add to. They weren't in t/spec, so I also moved them to t/spec and added smartlinks. That took a lot longer than I thought it would
19:38 [particle] i'd like to do this meeting like parrotsketch. if you're not familiar, first we do a round of status reports (from auzon, moritz, me, others?) followed by questions/comments.
19:38 moritz_ ok, that should be more productive
19:38 [particle] seems like auzon is done, so moritz, what have you been up to?
19:38 moritz_ * monitored Auzon's commits, seeemed good
19:39 moritz_ * fudged a few specs myself
19:39 moritz_ * sent patch to improve rakudo's testing
19:39 moritz_ * answered questions
19:39 moritz_ * tried to be available
19:39 moritz_ EOR
19:40 [particle] ok. this week has been mostly setup for gsoc
19:40 [particle] i've applied some of moritz++'s rakudo patches
19:40 [particle] and answered some questions on #parrot and #perl6
19:40 [particle] done a few spectest changes myself, and i'm working on something to make test management easier
19:41 [particle] i hope to have that finished today or tomorrow, and blog about it
19:41 [particle] i have five questions for auzon, and a few for moritz.
19:41 [particle] anyone else want to report related activity, or should we move to question time?
19:41 pmichaud * queue two questions
19:42 moritz_ * queue oone question, one comment
19:42 [particle] ok. let's move on to questions, then. i'll save mine for last. pmichaud?
19:42 pmichaud comment:  I like the use of a weekly schedule -- smart plan
19:43 pmichaud Q1:  would it be possible for the developers to queue specific tests we'd like to have fleshed out?
19:43 pmichaud (or specific areas of testing)
19:43 pmichaud for example...
19:43 pmichaud the question arose today:   does   @a.push(@b)   flatten @b ?
19:44 pmichaud we could ask this on p6l, but it might be nice if we could point to a test (or generate a test) that answers the question.
19:44 pmichaud or, speaking as a developer, I'd like to be able to ask Auzon the above question and have him point to the test that demonstrates the answer (possibly doing some research to find the answer if it's not obvious)
19:45 moritz_ I fear the latter will distract too much, because it requires jumping through the various synopsis
19:46 moritz_ it's mostly the question if you want to support immediate development, or instead refactor/extend the test suite faster
19:46 moritz_ what do you think is more important?
19:47 moritz_ I vote for a compromise, which is "write the test if it's in of the synopsis that you already covered"
19:47 pmichaud I'm very focused on immediate development, so I'm obviously biased that way.  But I see the other point as well.  My question was more out of curiosity than anything else -- perhaps we just find another route to do that.
19:47 pmichaud "write the test if in an already covered synopsis" works for me
19:47 pmichaud especially since S29 is week #2, and that's where I expect most of the questions to come from :-)
19:48 moritz_ Auzon: would that that be acceptable for you?
19:48 Auzon That's similar to what I've been doing so far, so yes.
19:48 [particle] (focused testing)++
19:48 pmichaud that would eliminate my second question, which was "should S29 come after S02 and S03 in sequence?"  But I think I'd prefer to have S29 right where it is now.  :-)
19:49 [particle] ok, looks like we're on to moritz now
19:49 pmichaud I don't expect it to be a huge number of questions, fwiw.
19:49 moritz_ pmichaud: I think your first question is only answerd half-way
19:49 moritz_ how should Auzon report features that would enable a big number of tests?
19:50 moritz_ for example if he discovers that most parse failures are due to $x not being implemented?
19:50 moritz_ ticket?
19:50 moritz_ irc?
19:50 moritz_ mail?
19:50 pmichaud either file a ticket (rakudobug) or contact me directly.
19:50 pmichaud any of those work.  ticket is nice because multiple people see it (tickets are automatically forwarded to p6c)
19:50 moritz_ ok
19:50 [particle] i'd prefer to track auzon's work via tickets, too
19:51 moritz_ ok, then it's ticket
19:51 [particle] easier to generate a report from rt than from irc log
19:51 moritz_ Auzon: rakudobug@perl.org (if you don't know already)
19:51 Auzon Tickets it is. I'll also make noise on IRC for the benefit of mncharity and other implementers.
19:51 pmichaud outstanding.
19:51 moritz_ very good
19:51 moritz_ ok, next
19:52 moritz_ I noticed that the non- t/spec/ part of the synopsis contains several artifacts
19:52 moritz_ mostly outdated syntax
19:52 moritz_ should we have a place were we collect the most frequent now-errors?
19:52 moritz_ like @list[-1] (instead of @list[*-1]) and such things
19:52 pmichaud ("of the synopsis"  ==>  "of the tests"?)
19:52 moritz_ if yes, where?
19:52 moritz_ pmichaud: yes, thanks
19:53 moritz_ perhaps some file under t/ ?
19:53 [particle] t/ bitrot is a problem
19:53 moritz_ t/COMMON-MISTAKES ?
19:53 [particle] t/deprecated-syntax.pod?
19:53 moritz_ sounds better
19:53 [particle] whatever the name, i think a file would be helpful during the conversion to t/spec/
19:54 moritz_ ok, I'll create one today or tomorrow
19:54 [particle] moritz_++
19:54 [particle] ok, auzon, my questions. the first two are related
19:54 [particle] are you comfortable with the tools you have?
19:54 [particle] are all (or any) of the implementations of perl 6 working well enough for you now?
19:55 Auzon What tools do you mean? Just everything in general?
19:55 [particle] yes, os, editor, compiler, etc
19:55 moritz_ stuff like svn, smartlinks.pl
19:56 moritz_ whatever you use for coding
19:56 Auzon Yeah, I'm happy. I'm still considering using SVK instead of SVN, but I don't see the benefit at present.
19:56 pmichaud (fwiw, svn works just peachy for me.)
19:57 [particle] fyi: i'm going to create a docs/dev/toolchain.pod file in the parrot repo, that describes what tools i use, and has room for others to do the same
19:57 [particle] svn works for me, too
19:57 Auzon I'm relying on Pugs on feather since the current revision hasn't built in a while. Other than that, everything works well.
19:57 [particle] ok, good. the next two questions are also related:
19:57 [particle] are you getting responses to your questions/comments in a timely manner?
19:57 [particle] is there anything more we can be doing to help you?
19:59 Auzon For the most part, questions are being handled fine via IRC. Any time issues are the result of me preferring to be awake while most people are asleep ;)
19:59 [particle] :)
20:00 Auzon As for helping me, just make note of any things that are obviously lacking in tests that you encounter. mncharity mentioned that the hash tests are very lacking, since elf passed all/most of the existing ones.
20:01 moritz_ ok
20:01 Auzon That way, I know what will help the most and I don't need to spend time looking through S\d\d for things that aren't tested well.
20:01 [particle] ok, i'd like to note that via a ticketing system or list mail, so it's somewhat formalized
20:01 [particle] if we don't have something that works today, we'll set it up soon
20:01 [particle] one last question, that i'll ask every week: do you feel you're on track?
20:02 TimToady I told the phone call you might be a little late
20:02 moritz_ ;)
20:02 Auzon Currently, I feel I'm a bit behind in numbers, but not coverage. The first specs in my list have the easier/implemented things covered
20:03 [particle] ok. we'll keep an eye on that
20:03 [particle] moritz: do you feel Auzon is on track?
20:03 moritz_ [particle]: I share his opinion
20:03 moritz_ and he's done good work so fwar
20:03 moritz_ s/w/
20:04 [particle] great. i need to run to another meeting. keep up the good work!
20:04 moritz_ btw one announce, I'll be offline from thursday to sunday, so you'll need to get your feedback from @other_channel_members
20:04 obra joined #perl6-soc
20:04 moritz_ anything else?
20:04 rdice joined #perl6-soc
20:04 Auzon What questions, if any, should I ask in #parrot?
20:05 [particle] any you need answered.
20:05 Auzon or would be better asked in #parrot
20:05 pmichaud #perl6 is fine, for the most part.
20:05 [particle] i'd prefer to keep perl6 questions to #perl6
20:05 Auzon Alright, sounds good.
20:05 moritz_ let's call it a week, and hope that the future meetings will be shorter
20:06 pmichaud _maybe_ if you have a very rakudo-specific implementation question they could go in #parrot
20:06 pmichaud otherwise I think #perl6 is better
20:07 moritz_ ok, thank you all for attending
20:07 pmichaud until next week!
20:07 pmichaud Auzon++ particle++ moritz_++
20:07 moritz_ and pmichaud++
20:07 moritz_ (btw we don't track karma here ;)
20:09 pmichaud ++ is my generic "excellent job" postfix sigil
20:09 moritz_ I know ;)
23:40 Juerd Auzon: Two benefits of svk: 1. it's faster; 2. on feather, pugs is local so it's even faster-er. :)

| Channels | #perl6-soc index | Today | | Search | Google Search | Plain-Text | summary