Summer of Perl | 6.2.12 released! | paste: sial.org/pbot/perl6 | pugs.blogs.com | pugscode.org | pugs.kwiki.org | www.treehugger.com/files/th_images/paradigm.jpg
Set by nothingmuch on 12 July 2006.
TimToady note, we're also thinking about playing similar games with postfix (), which would force for() to be considered a function rather than a keywords. 00:00
salty-horse TimToady, unrelated, I scanned the synopsis and didn't find a mention of foo$x being treated as foo $x
therefore giving room to for overloading, and much obfuscuation? 00:01
00:01 rashakil joined
TimToady I see that S03 is unclear on the subject. I will look at how to remedy that. 00:04
I was trying to say that you have to use parens to add a postfix: foo().[] and such, but what I said seems to disallow foo$bar. 00:05
mind you, there's part of me that thinks foo$bar *ought* to be disallowed...
nothingmuch quick off the top of your heads, i need a module to make '--from="some_date_spec"' parsing easy 00:11
e.g. today, tomorrow, sunday, 1/2 (locale sensitive)
00:12 mauke_ is now known as mauke
salty-horse google cpan fuzzy date ? 00:12
00:12 ruoso joined
nothingmuch there's a bajillion hits 00:13
i need the ones that are known to be good, hence the human factor =)
silug Date::Manip works nicely, but it is a bit slow 00:14
salty-horse no experience with it, but i also notice search.cpan.org/~muir/Time-modules-...rseDate.pm 00:15
lambdabot Title: Time::ParseDate -- date parsing both relative and absolute - search.cpan.org 00:16
salty-horse gnight 00:17
silug perl -MDate::Manip -le 'for $n qw(today tomorrow sunday) { print &ParseDate($n) }' 00:19
2006071719:18:37
2006071819:18:37
2006072300:00:00
things like "last sunday" work too
nothingmuch awesome 00:20
00:20 FurnaceBoy joined
silug "last week" and "last month" work too, although strangely "last year" doesn't 00:20
nothingmuch i doubt a mailing list summarizer will do a last year =) 00:21
silug true 00:22
00:22 SubStack joined 00:23 daxim_ joined
silug i love that things like "third monday in august" work. that would have saved me a lot of coding when i was setting up my LUG sites a few years back. 00:23
00:25 unobe is now known as unobe_away
robkinyon seen alias 00:28
jabbot robkinyon: I havn't seen alias, robkinyon 00:29
00:43 frederico joined
Jedai seen ingy 00:43
jabbot Jedai: ingy was seen 3 hours 10 seconds ago
00:54 hikozaemon joined
TreyHarris seen audreyt 01:04
jabbot TreyHarris: audreyt was seen 1 hours 14 minutes 37 seconds ago
TreyHarris ?eval $_ = 3; if .defined -> $id { say "yes: $id" } 01:05
01:05 evalbot_11605 is now known as evalbot_11606
evalbot_11606 Error: unexpected ">" expecting comment, operator or block 01:05
TreyHarris audreyt: if you see this when you get back, i'm trying to figure out this comment:
audreyt TreyHarris: the "if ... ->{}" form doesn't yet work 01:06
TreyHarris 09:47 < gaal> er. okay. I'm just wondering
oops :-)
09:39 < audreyt> if .defined -> $id { ... }
oh, you replied, and that was what messed up my copy :-)
ok, thanks 01:07
audreyt np :) 01:08
TreyHarris ?eval my $x = 3; undef $x; 01:09
evalbot_11606 Error: unexpected "$" expecting comment, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
TreyHarris is undef() gone?
audreyt it's now undefine()
TreyHarris ahh, i missed that. S29 needs to be tracked in email like the official synopses 01:10
01:44 unobe joined 01:45 unobe_away is now known as unobe 01:50 unobe joined 02:11 lisheeghan joined 02:15 Kattana joined 02:25 weinig is now known as weinig|zzz 02:27 miyagawa joined 02:31 xdg joined 03:04 Khisanth joined 03:31 scw joined 03:59 amnesiac joined 04:09 amnesiac joined, rlb3 joined 04:20 Khisanth joined 04:40 SSelva joined
svnbot6 r11607 | audreyt++ | * Pugs::Compiler::Rule (fglock++'s branch): more fix on getting ::Ratchet initialization working 04:48
05:07 SSelva left, SSelva joined 05:09 szbalint_ joined
svnbot6 r11608 | audreyt++ | * Implement $$match for real. 05:19
05:21 SSelva left 05:30 scook0 joined 05:37 szbalint_ joined, Kattana joined, hikozaemon joined, daxim_ joined, rashakil joined, esammer joined, lumi joined, silug joined, typester joined, xern_ joined, drbean joined, Jedai joined, lollan joined, theorbtwo joined, thepler joined, svnbot6 joined, dvorak joined, broquaint joined, kcwu joined, perlbot joined, buubot joined, Shabble joined, Grrrr joined, bennymack joined, Yaakov joined, revdiablo joined, monkey_ joined, LCamel joined, tcliou joined, _calvin_ joined, geoffb joined, trym joined, meppl joined, ajs joined, PerlJam joined, Odin- joined, cwest joined, xerox joined, hcchien joined, nelhage joined, webmind joined, nnunley joined, audreyt joined, ingy joined, cls_bsd joined, obra joined, amv joined, pjcj joined, jiing joined, clkao joined, nperez joined, rafl joined, notsri_ joined
gaal Should we make U+nnnn a string literal? (equivalent to chr(0xnnnn) but folded)? 06:00
06:04 justatheory joined
TimToady do you mean in Perl or Haskell? 06:04
TreyHarris anyone know of a free fixed-width font containing that newline character? My graphic-designer spouse's Mac, loaded with hundreds of fonts, has only two with that glyph, and both are variable-width :-( 06:12
gaal TimToady: in Perl. By "folded" I mean constant-folded 06:22
"\x[nnnn]" 06:23
TimToady It seems rather special-casey for "\xnnnn"
gaal just seems useful when copy-pasting from reference tables :) 06:24
TimToady I'm using Sazanami Gothic at the moment, but then I'm studying Japanese... 06:25
gaal has been looking for a monospaced font with Hebrew, Arabic, and Cyrillic to replace Courier New... 06:27
TimToady I think asking for a monospaced Arabic font is kind of strange... 06:30
gaal hey, they use xterms too...
Hebrew Yod is ugliest in fixed-width: it's just about two pixels wide. 06:31
×Øואים?
TimToady doesn't look too bad in my font. 06:32
gaal Maybe I should just create my own frankenfont.
06:32 kanru joined
TimToady The mem is a bit squished against the question mark though. 06:33
gaal TimToady: it's *reasonable*, but can't be pretty. It's like seeing unkerned text after reading a quality typeset book.
TreyHarris TimToady: you may find this of interest. A couple of computing professors at Middlesex University say they've devised a test that can predict, before any programming instruction whatsoever, whether someone will be able to learn to program or not:
www.codinghorror.com/blog/archives/000635.html
lambdabot Title: Coding Horror: Separating Programming Sheep from Non-Programming Goats
gaal it would be, because the question mark is from the ASCII range, optimized for English glyphs no doubt. 06:34
does Japanese use a Western question mark? 06:35
miyagawa gall: we use both
TimToady yes, but it has double-wide punctuations when it makes sense
TreyHarris the test is snippets of code that just show imperative variable assignment, and ask what the values assigned are. people who are consistent in their guesses of how assignment might work (whether they get the answers right or not) are much more likely to learn to program than people who are inconsistent or who refuse to answer
TimToady "?" vs "ļ¼Ÿ"
miyagawa full-width question mark is U+FF1F btw 06:36
gaal TreyHarris: I'd argue that indirection is an important major hurdle for beginners -> intermediate programmers
TreyHarris miyagawa: what is traditional? "ka" (sorry, I can't type hiragana in this window) followed by a circle?
gaal like barre on guitar 06:37
TreyHarris gaal: what do you mean by "indirection"? abstraction? references?
gaal miyagawa, TimToady: thanks
miyagawa TreyHarris: sorry I don't get your question ...
gaal TreyHarris: pointers/references
TimToady It sounds like the test is trying to determine whether you can run a model in your head. 06:38
gaal lambdaheads would object to that :)
TreyHarris miyagawa: before the question mark was imported, what if any mark was used to end questions?
TimToady TreyHarris: the ka is often used in conversation with a rising intonation, so is often written with a ? anyway.
gaal they'd say it's Von Neuman chauvinism to make that test
TimToady And ka can men "or"ish things too.
*mean
or "some" as in "something". 06:39
so the ? would not be entirely redundant.
miyagawa TreyHarris: Ah, good question. Yeah, if the sentence itself composes the questional verb, the question mark is optional
i wasn't good for Japanese old literature in school and can't tell what's the traditional before the question mark was imported :/
TimToady and in spoken Japanese the ka is often omitted when a rising tone on the final syllable can convey a question.
miyagawa eight. 06:40
TimToady You'll often hear: "nani?"
miyagawa right.
TimToady from the women, or "nanda?" from the men
gaal TimToady: anyway, re: arabic and fixed-width: a lot of the characters are in fact reasonably close in width, if they are connectives. Alif for example is thin, but if it's connected on the right it's still reasonable to give it a cell width: ļŗ„
TreyHarris gaal: my first CS course was taught with Haskell. I really think it made a big difference in my abilities (but permanently ruined my code's readability for complete novices who can't handle anything vaguely functional). but i still "thought" imperatively and translated into Haskell
TimToady some of the ligatures look to be at least triple width though... 06:41
TreyHarris TimToady: my teachers were all women and didn't even attempt to teach male forms, so I say "nani" and get laughed at :-)
gaal lim-alif is kinda wide, but can be squished in...
it's sin and final 'sad that are really cramped. 06:42
miyagawa TrryHarris: Male using "nani" is nothing hillarious. it'd be okay
06:43 iblechbot joined
TimToady ļ·† 06:43
now, you probably better use "kana..." rather than "kashira..."
and don't say "atashi" instead of "watashi"...
miyagawa right. "kana" can be used by both male/female but "kashira" implies female 06:44
TreyHarris what about "o-<noun>"? that's the one that I really don't get when I'm supposed to use.
TimToady and don't use the "wa" particle (not to be confused with the "ha" particle that is pronounced "wa")
depends on how respectful you're wanting to be. 06:45
gaal goes bike shopping &
TimToady women are just in general expected to be more respectful by and large
many of the o-nouns you can't really be understood without the o in any case
it's sorta become part of the noun. 06:46
but the o-verb suru kinds of things are more productive, I think.
TreyHarris TimToady: ah, that's the crucial bit i was missing. i was trying to form a hypothesis by exclusion, and didn't realize the particle was attaching at the lexical level
TimToady it varies, of course, but if you just say "cha" they might just wonder if you said "ja", whereas "ocha" is much less ambiguous to the ear. 06:49
it really does tend to be driven a lot by what will seem ambiguous--particle omission works the same way. 06:50
TreyHarris right
TimToady and unfortunately there's no way to know what'll be ambiguous without knowing the majority of the language. 06:51
I don't know how many times I've struggle to parse a bit of writing, and finally figured out that it was a proper noun.
miyagawa some historical nouns have "omi" as a prefix, which is a doubled version of "o"
TimToady interesting. 06:52
TreyHarris ?eval sub foo (Int $x) { say $x }; foo("hello")
06:52 evalbot_11606 is now known as evalbot_11609
evalbot_11609 OUTPUT[hello ] Bool::True 06:52
TreyHarris is that always supposed ot work, or is that a TBD with pugs?
miyagawa "omiotsuke" is a tripple "o"-ed version of "tsuke" == picke
pickle
TimToady that's a TBD, I'm sure. 06:53
I only know "tsukemono"... 06:54
but if I ever wish to be very respectful to a pickle... :) 06:55
TreyHarris ok--so there's no catch-all "you tried to call a sub, and I don't have a matching one, but I do have one by that name, so I'll just use it" DWIM
lumi Isn't this for being respectful while mentioning a pickle? 06:56
TimToady I could argue that Int is just a role that is required of $x, and that say is not required to access $x via the Int role, I suppose...
but it seems a stretch.
lumi: hai, joudan desu... 06:57
miyagawa-san: is omi spelled å¾”ē„ž? 07:00
miyagawa TimToady: it comes with triple U+5FA1's: gogen-allguide.com/o/omiotsuke.html 07:01
lambdabot Title: ōæ½x8Cōæ½xE4ōæ½x8Cōæ½xE4ōæ½x8Cōæ½xE4ōæ½x95tōæ½x82ōæ½xAFōæ½x81iōæ½x82ōæ½xA8ōæ½x82Ż‚ōæ½xA8ōæ½x82ōæ½xAFōæ½x81j - ōæ½x8CźŒ¹ōæ½x97Rōæ½x97ōæ½x88ōæ½x8Eōæ½xABōæ½x93T
miyagawa oh, and I was wrong. omiotsuke stands for miso-soup, not pickle (tsukemono). 07:02
TimToady sugoi. 07:06
I think I'll have to bookmark that site...
miyagawa :)
07:12 Aankhen`` joined 07:18 TimToady_ joined 07:46 hugues23 joined 08:02 buetow joined 08:08 kolibrie joined 08:09 drrho joined 08:20 justatheory joined 08:25 nothingmuch joined 09:04 Revision17 joined 09:29 szbalint joined 09:33 bsb left 09:44 ruoso joined 09:54 avar joined 09:56 froh-doh joined 10:04 justatheory joined 10:08 justatheory joined 10:09 elmex joined
gaal audreyt: ping 10:17
audreyt pong 10:18
gaal hey, stylistic question. I've always been a little unhappy with the deep patmatching in emitNode 10:21
what do you say about changing
emitNode _ e n@(MkYamlNode{nodeElem = YamlStr str}) = ...
to
emitNode _ e n | YamlStr str <- nodeElem (MkYamlNode n) = ...
audreyt that works fine
gaal I don't know if it's better :)
audreyt it is
though
YamlStr str <- nodeElem n 10:22
suffices
audreyt is still playing this commit/revert/commit game with allison on parrot-porters
gaal also, we could follow the new naming conventions and get even shorter identifiers :) but that'd break compatibility and we're a Packaged Library now :)
audreyt nah. please break compatibility. 10:23
before cabal-install gets deployed, Packaged Libraries means very little
gaal whoo, you're right about not needing the record tag there! yay
s/tag/constructor/ 10:24
audreyt should sleep, too
*wave*
10:24 ludan joined
xerox Seeya! 10:25
gaal night :)
clkao: svk sucks because it doesn't have a "svk ci --ephemeral" option to help audreyt in commit-revert games :) 10:27
audreyt svk ci -P -
is that :)
gaal ah, obviously I should revert my last statement :)
er, no, that's just spelled "diff", no? :) 10:28
diff with the option to interactively select files to include. 10:29
webmind s/is/ie/ 10:31
nothingmuch moose 10:44
11:12 wesle1 joined 11:17 iblechbot joined
gaal nothingmuch: moose! 11:20
11:28 nothingmuch joined
gaal audreyt: one place our convention comes out a little clunky is in variants that have a sort of catch-all case. So data ValUndef = UndefUnit | ..., or, in the present case, data YamlAnchor = AnchorUnit | ... (used to be MkYamlAnchor, I'm thinking "unit" is misleading here) 11:33
maybe AnchorSimple
er, actually that should be UUnit and ASimple 11:34
11:37 wesleyyaya joined
gaal maybe in that case UUndef and AAnchor look less ridiculous :) 11:40
11:57 wesleyyaya is now known as wesle1
svnbot6 r11610 | gaal++ | * util/claim_drift.pl: claim files more aggressively (works better for 12:12
r11610 | gaal++ | first claim)
12:26 frederico joined 12:56 amnesiac_ joined, amnesiac_ is now known as amnesiac, Termy joined 12:59 bpphillips joined 13:00 vel joined 13:03 tifo joined 13:05 tup joined 13:15 Qiang joined
gaal fglock: is it ok to svn rm perl5/Pugs-Compiler-Rule/Makefile.old ? 13:20
13:23 FurnaceBoy joined 13:25 fglock joined 13:28 weinig|zzz is now known as weinig 13:30 jkanter joined
svnbot6 r11611 | fglock++ | pX/fglock/PCR - '$match->perl' 13:40
13:44 buubot joined 13:50 vel joined
svnbot6 r11612 | fglock++ | pX/fglock/PCR - some tests pass 13:58
14:01 justatheory joined 14:05 justatheory joined 14:08 justatheory joined 14:11 chris2 joined 14:12 orafu joined 14:13 christopher joined 14:16 Jedai joined, justatheory joined 14:18 justatheory joined
svnbot6 r11614 | gaal++ | * HsSyck: 14:19
r11614 | gaal++ | - use Pugs naming convention and other misc. cleanups
r11614 | gaal++ | - this change breaks the API (only by name), so bump version to 0.2.
r11614 | gaal++ | * Pugs:
r11614 | gaal++ | - chase API changes and re-DrIFT.
r11614 | gaal++ | You may need to 'ghc-pkg unregister pugs-HsSyck' or even
r11614 | gaal++ | 'make clean' after this update.
14:23 justatheory joined 14:24 justatheory joined 14:25 justatheory joined 14:26 cjeris joined 14:37 justatheory joined 14:39 justatheory joined 14:41 iblechbot joined 14:42 justatheory joined 14:53 justatheory joined
svnbot6 r11615 | fglock++ | pX/fglock/PCR - 03-match.t passes 38/46 14:55
14:59 esammer_ joined 15:00 weinig is now known as weinig|work
svnbot6 r11616 | fglock++ | pX/fglock/PCR - refactored non-anchored match in Regex 15:16
15:24 weinig|work is now known as weinig 15:35 putter joined 15:41 amnesiac joined 15:44 weinig joined, lambdabot joined 15:53 frederico joined 16:28 mj41_ joined 16:29 ruoso joined
svnbot6 r11618 | fglock++ | pX/fglock/PCR - 03-match.t all tests pass, including TODOs 16:32
16:36 Aankh|Clone joined
ruoso fglock, hi... 16:45
fglock, did the Makefile work on Windows?
I didn't have how to test it 16:46
16:46 weinig is now known as weinig|bbl
fglock ruoso: I didn't test - I'm currently working in PCR TODOs 16:46
I'll merge later, then I can fix any Makefile problems too 16:47
ruoso ok... 16:49
you can blame me for any problem :)
fglock ok :) 16:50
16:58 chris2 joined 17:00 buetow joined 17:07 hcarty joined 17:32 Aankh|Clone is now known as Aankhen`` 17:37 jkanter joined 17:44 hcarty left 17:53 buetow joined, lollan joined 17:57 orafu joined
svnbot6 r11621 | fglock++ | pX/fglock/PCR - docs update; passes 199/228 17:59
18:02 vel joined 18:23 scsibug_ joined 19:07 larsen joined 19:09 bpphillips left 19:18 DaGo joined 19:42 pochi joined 20:01 rodi joined, rodi is now known as baaaaazzz 20:02 baaaaazzz is now known as rodi 20:21 [particle] joined 20:27 justatheory joined
svnbot6 r11627 | daxim++ | check if parrot is in PATH 20:36
20:42 hcart1 joined 20:50 fglock left 21:06 SageLT joined 21:09 SubStack joined 21:33 SubStack joined 21:41 weinig|bbl is now known as weinig
svnbot6 r11629 | Sage++ | More regex support. Almost finished internal regex translations. 22:07
22:21 qubar2000 joined 22:27 qubar2000 is now known as Arador 22:58 cjeris left 23:15 bsb joined 23:36 furrybeagle joined
pasteling0 "Dr_Pi" at 72.136.69.198 pasted "Build Problems in Mac OS 10.3.9 and Fink. (It worked before)" (8 lines, 702B) at sial.org/pbot/18486 23:47
daxim_ install IPC::Cmd 23:50
23:59 mauke_ joined

Powered by Cro and the RakuĀ® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!