Camelia, the Perl 6 bug

IRC log for #november-wiki, 2009-04-23

| Channels | #november-wiki index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:02 sri_kraih_ joined #november-wiki
01:06 wayland76 I don't understand the question "is there a precedent to the specification strategy of Perl 6?"  Is it rhetorical?
08:02 masak joined #november-wiki
08:31 masak still problems with getting the hitomi prototype script to DWIFM.
08:32 masak due to the nature of pride, I think there's a bug in Rakudo somewhere.
08:32 masak but I'd be happy to be proven wrong.
08:35 moritz_ can you paste your current version?
08:36 masak just about to. :)
08:36 masak http://gist.github.com/100394
08:39 moritz_ another bug in Match.perl, sigh
08:39 masak I'm having a hard time figuring out how to minimize it.
08:46 moritz_ somehow Match!_perl_quant gets a list of which one value is an undef
08:46 masak aye.
08:46 masak that's as far as I got too.
08:47 masak too bad we don't have a way to introspect a Match...
08:47 moritz_ sure we have
08:47 moritz_ %($/).pairs
08:47 moritz_ @($/).pairs
08:47 moritz_ stringification should still work
08:48 * masak tries
08:50 masak the former gives 'content', the latter ''.
08:50 masak this despite the fact that I do 'next if $chunk<content>' just before.
08:50 moritz_ can <content> match a zero-width string?
08:51 * masak checks
08:51 masak no.
08:51 moritz_ no.
08:51 masak it should.
08:52 moritz_ for ^self.list { ... } in Match.pm seems to be the problem
08:52 moritz_ no
08:53 masak hm, changing the rule 'content' from + to * causes the grammar to hang.
09:06 masak and yet, '<tag></tag>' should parse, and it doesn't now.
09:06 masak this might actually be a punishment for not writing tests first.
09:07 masak maybe I should start over...
09:07 eiro :)
09:07 eiro hello all
09:07 zarah ni hao eiro
09:09 masak ni hao eiro
09:11 eiro i wonder to myself where 'ni hao' comes from
09:11 wayland76 Chinese
09:12 wayland76 They all have bad knees, and inquire after each others'
09:12 masak eiro: it's really written 你好.
09:12 eiro it means 'hello' ?
09:12 masak eiro: yes.
09:12 wayland76 It does indeed
09:12 masak but literally 'you good'.
09:12 wayland76 As a statement, or a question?
09:12 masak as a statement, not a question.
09:12 wayland76 And good morally, or healthwise?
09:12 masak a question would be 'ni hao ma?'
09:13 masak wayland76: I don't know, actually.
09:13 wayland76 Ah, yes, that vaguely rings a bell
09:13 wayland76 probably healthwise
09:13 wayland76 Like "Was hael", 'be well'
09:13 masak as in 'you look good'. aye, probably.
09:13 wayland76 Or "I wish you good health"
09:13 masak sounds more like a goodbye.
09:14 masak anyway, the 'ni hao' greeting is so lexicalized that it doesn't mean anything like 'you good' anymore.
09:14 wayland76 Well, if the Jews (?) can greet with "Peace be upon you", or whatever it is...
09:15 masak http://en.wikipedia.org/wiki/Shalom_aleichem
09:15 zarah masak's link is also http://tinyurl.com/dh3r52
09:15 masak google++ wikipedia++
09:23 wayland76 Yah, I know, and Arabic is Salaam Alekoom (sp?).  But my point is, it sounds like "I wish you good health"
09:24 wayland76 So if the Hebrew one can be a greeting, then so can "I wish you good health"
09:24 masak aye.
09:24 wayland76 Wes þu Theoden hal!
09:24 wayland76 :)
09:36 spinclad and Ferþu hal al so!
09:37 masak is that Icelandic?
09:37 spinclad anglosaxon roughly
09:37 masak ah.
09:37 * spinclad imagines there's a question buried in 'ni hao', that's how i hear it anyhow
09:38 spinclad 'hao are you?'
09:38 * spinclad *dux*
09:39 masak :)
09:40 masak part of successfully learning new languages is successfully turning off the part of the brain that does that. at least for me.
09:40 wayland76 It could be like French.  "Ca va?  Ca va."
09:40 wayland76 It's a question, and its own response
09:41 masak well, the question has to have a 'ma' at the end.
09:41 masak that's how it works.
09:41 spinclad i heard 'ni hao ma'/'ni hao a' as variants when i was dabbling
09:41 masak aye.
09:41 wayland76 Wes = be.  þu = thou.  Theoden is the guy's name.  hal = hale, whole healthy
09:41 masak nice.
09:42 wayland76 Is "al so" good Old English?  :)
09:42 masak ah, good old english. :)
09:42 spinclad dunno rilly, just a guess
09:43 spinclad 'and' is rightwise 'and', though, that much i know
09:43 wayland76 IIRC, it's a quote from LoTR.  The Rohirrim spoke Old English (=Anglo-Saxon), but with the Mercian dialect that Modern English is descended from, instead of the West Saxon that appears in most manuscripts
09:44 wayland76 "al so" sort of looks right and feels wrong or something :)
09:44 wayland76 (The Theoden line is the quote from LoTR, not the rest)
09:44 wayland76 All hail Duke spinclad!
09:44 * wayland76 also dux :)
09:46 * spinclad accepts the ill-gained honour
09:46 spinclad ill-claimed
09:47 wayland76 Who started Web.pm?  masak?
09:47 masak wayland76: depends what you mean.
09:48 masak wayland76: correct answers can indicate Juerd, me, viklund, ihrd and/or Tene.
09:48 wayland76 All hail kings Juerd, masak, viklund, ihrd and/or Tene
09:48 wayland76 :)
09:48 masak heh. let's reach a modest amount of success first. :P
09:49 masak help appreciated.
09:49 wayland76 Ok.  Some  hail kings Juerd, masak, viklund, ihrd and/or Tene
09:49 wayland76 :)
09:49 wayland76 Well, I need a working Rakudo first.  That's kinda waiting on a branch of Parrot to be merged, which will happen after the release sometime when kid51 gets around to it
09:50 wayland76 After that, Parrot may need another patch, but then hopefully I should be able to get things working more
09:50 wayland76 By which I mean, I'll probably be working on a "make install" target for Rakudo
09:51 masak ok.
09:51 wayland76 But once I have the working RPM, I'll be here :)
09:52 wayland76 Hmm.  Actually, maybe I might have a go at Tree first, but hopefully that won't take too long
09:52 * spinclad could try for a .deb
09:52 wayland76 Especially if the XML grammar is already done :)
09:52 wayland76 spinclad: Well, the process of someone working on a .deb would either help me or be helped by me, I think
09:53 wayland76 Basically, Rakudo has to be configurable about which directory it gets installed in, so that it goes where RPM wants it to
09:53 spinclad of course.  i'll try to be around to contribute again, after all these months
09:54 masak wayland76: krunen++ is also doing an XML parser. see his project at github.
09:54 wayland76 Speaking of which, pleased to meet you :)
09:55 wayland76 Oh, great.  I started playing with Moritz' ages ago, but Rakudo wasn't up to it at that point
09:56 wayland76 I should also point out that I'm intending that Tree will be able to input/output XML :)
09:56 wayland76 So it can certainly work in conjunction with Web.pm :)
09:56 wayland76 (XML among other things)
09:59 spinclad good to meet thee, wayland76, and greetings all around.  i love what's been shaping in November, Web.pm, rakudo, (elf, smop, std, ...)
10:07 masak me too.
10:17 masak lunch &
12:45 moritz_ masak: does the Match.perl "work" for you (ie not die) when you revert d0a5de55a1da8c7307e6c486fd9b548fcbc0b727?
12:45 * masak tries
12:49 moritz_ it's Match.pm line 39 where .value is undef, but I don't see how a legal match object should ever have that.
12:50 masak moritz_: nope, gives the same error: "Method '!_perl' not found for invocant of class 'Failure'"
12:51 moritz_ yes, verified... I inserted a say .value.PARROT, and it says Failure
12:52 moritz_ given that $_ comes from %(self).pairs, that should never happen.
12:52 masak agreed.
12:52 masak and yet she moves.
12:52 moritz_ => bug in Rakudo or PGE
12:52 masak ...as opposed to .perl, you mean?
12:53 moritz_ yes, a bug in non-setting code
12:53 masak maybe time to minimize that failure after all.
12:53 masak it shouldn't be impossible.
13:56 p6eval joined #november-wiki
17:32 ruoso joined #november-wiki
22:02 zarah joined #november-wiki
22:05 wayland76 joined #november-wiki

| Channels | #november-wiki index | Today | | Search | Google Search | Plain-Text | summary