Parrot 3.5.0 "Menelaus" released | parrot.org | Log: irclog.perlgeek.de/parrot/today
Set by moderator on 27 June 2011.
whiteknight wonders why P6 has a Cool type, but not Radical, Tubular, or Awesome ones 00:09
sub foo(Awesome $x)
or why not go all Paris Hilton with a Hot type?
sub bar(Hot $ornot)
00:23 Coke left 00:24 Coke joined 00:29 kid51 left 00:34 Coke left, Coke joined 00:40 theory left 01:08 zby_home left 01:15 kid51 joined
dalek sella: 5ef0dae | Whiteknight++ | t/string/ (2 files):
Implement several tests for String
01:19
sella: 1aa7f86 | Whiteknight++ | src/ (2 files):
Mark not_implemented the functions I'm not ready to implement yet
sella: 34fa915 | Whiteknight++ | / (6 files):
Update tests for CClass, Delimiter, and DelimiterRegion tokenizers
01:37 JimmyZ joined
dukeleto ~~ 01:45
whiteknight hello dukeleto 01:47
you home now?
01:49 dukelet0 joined
dukelet0 installed parrot-gmp via plumage just now, and it detected rosella as a dependency and installed that too. Spiffyy 01:49
01:51 theory joined
whiteknight nice 01:52
kid51 Does anyone have a link to the slides cotto presented at YAPC? (or at OSBridge, as a fallback)?
01:56 Drossel joined
dukelet0 looks like he hasn't put them on www.yapc2011.us/yn2011/talk/3311 yet 01:57
cotto++ is currently wandering around Chicago during a long layover
dukelet0 sat next to him on the first flight that we shared this morning
01:58 Kulag left
dukelet0 kid51: i don't have a summary of the hackathon yet. Going to a wedding in a few mins. 01:59
02:00 ligne joined
kid51 Thanks, I will post as is. 02:01
(A Friday night wedding?)
dalek website: jkeenan++ | Parrot at YAPC::NA::2011, Asheville NC
website: www.parrot.org/content/parrot-yapcn...heville-nc
02:04 benabik joined 02:05 dukelet0 left
dalek sella/gh-pages: 41e9f74 | Whiteknight++ | libraries/string.md:
Add in draft docs for the String library
02:07
02:18 kid51 left 02:19 Drossel left, Kulag joined
dalek sella: 67d90fb | Whiteknight++ | / (3 files):
small fixes. Add FileSystem to list of libraries in README. string, path, and prototype library tests get run together as the 'unstable' set. Move the testing logic out of the main handler in setup.winxed, to prevent weird backtraces from appearing.
02:24
02:24 whiteknight left
dalek website: soh_cah_toa++ | Battery Low, No Time for Title 02:33
website: www.parrot.org/content/battery-low-no-time-title
02:41 rurban_ joined 02:44 rurban left, rurban_ is now known as rurban, Kulag left 02:50 Kulag joined 02:57 fantazo_ joined 03:01 JimmyZ left 03:02 fantazo left
cotto ~~ 03:05
dukeleto, I caught an earlier flight back home. I got back around 12:00 local time. 03:06
much better than previous eta of 19:30 03:08
03:13 JimmyZ joined
bubaflub seen colomon 03:27
aloha colomon was last seen in #perl6 3 hours 50 mins ago saying "o/, #perl6".
benabik ~~ 03:29
bubaflub msg colomon i've made a breaking change to parrot-gmp - .pbc files are now installed into GMP/*.pbc instead of src/GMP/*.pbc 03:31
aloha OK. I'll deliver the message.
cotto msg kid51 My slides are at mksig.org/talks/yapcna11/slides/ . I also added them to the talk's page on the yapc site. 03:53
aloha OK. I'll deliver the message.
dalek kudo: 72d158a | pmichaud++ | build/Makefile.in:
Peg spectest to a specific revision, since they're starting to target the nom branch.
04:36
04:44 theory left
cotto sorear, ping 04:58
05:33 fperrad joined 05:49 Kulag left 05:50 Kulag joined 05:51 daniel-s joined
moritz cotto: sorear mentioned that he might be offline until Tuesday 06:24
cotto moritz, ok. 06:27
moritz, where does he live?
moritz cotto: dunno
06:54 zby_home joined
Util is back home, safe and sleepy. 06:58
cotto great conference, but it's really nice to be home again 07:01
Util ditto, cotto 07:04
EJOKESOURCEUNDERFLOW 07:05
dalek kudo/nom: 4434d03 | pmichaud++ | src/binder/types.c:
Change types.c so that RPAs come back as Parcel instead of Array.
07:05 ligne left
moritz now digs up the first R* tarball and uses that for timing as well 07:11
07:30 Kulag left 07:32 Kulag joined 07:41 Kulag left, Kulag joined 07:45 Kulag left, Kulag joined 07:47 fperrad left 07:51 fperrad joined 07:52 Kulag left, Kulag joined 08:08 Kulag left, Drossel joined 08:17 Drossel left 08:18 Kulag joined 08:24 Kulag left 08:25 Kulag joined 08:29 mj41 joined 09:02 ligne joined 09:16 fantazo_ left 09:25 JimmyZ_ joined, JimmyZ left, JimmyZ_ is now known as JimmyZ 10:06 mj41 left 10:14 ambs joined 10:16 JimmyZ left 10:41 rurban_ joined 10:42 whiteknight joined 10:44 rurban left, rurban_ is now known as rurban
whiteknight good morning, #parrot 10:46
tadzik good morning whiteknight 10:47
whiteknight hello tadzik. How are you doing? 10:48
tadzik not too bad, enjoying my free time at last 10:49
10:53 ambs left
dalek sella/gh-pages: a28dbdc | Whiteknight++ | libraries/path.md:
Add in a quick stub for Path docs
11:02
11:21 JimmyZ joined 11:56 mj41 joined 12:10 mj41 left 12:22 zby_home left 12:48 lichtkind joined
cotto ~ 12:50
stupid brain is still on nc time
12:57 ambs joined
benabik I'm still on NC time. 12:58
That less of a problem for me, since I'm travelling north, not west 12:59
13:05 kid51 joined 13:31 lichtkind left 13:37 benabik left 13:44 dafrito joined 13:46 bluescreen joined
dalek kudo/nom-complex-insanity: a5bfdd0 | moritz++ | src/core/Complex.pm:
optimize and fix a few ops. Not correct yet, but compiles
14:09
kudo/nom-complex-insanity: 6fe7e6f | moritz++ | src/core/Complex.pm:
fix infix -
kudo/nom-complex-insanity: 316fcb9 | moritz++ | src/core/ (3 files):
more work to get complex.t passing

includes Complex to Num, Int, Rat coercion, Num -> Rat and some bridge methods in Real
kudo/nom-complex-insanity: 1e80b76 | moritz++ | src/core/Numeric.pm:
- should subtract, not add :-)
cotto "nomplex"? 14:11
;)
moritz, ping 14:19
aloha, clock?
aloha cotto: LAX: Sat, 07:19 PDT / CHI: Sat, 09:19 CDT / NYC: Sat, 10:19 EDT / UTC: Sat, 14:19 UTC / LON: Sat, 15:19 BST / BER: Sat, 16:19 CEST / TOK: Sat, 23:19 JST / SYD: Sun, 00:19 EST
moritz cotto: pong 14:21
dalek kudo/nom-complex-insanity: a425e09 | moritz++ | src/core/Real.pm:
get infix:<-> with two different Real types working
14:23
cotto moritz, I'd like ilbot to be able to star snippets of conversation, indicating that something important is being talked about. Which part of ilbot's code would be the best place to add something like that? 14:24
moritz cotto: 1) a new column in the the db schema
cotto so, non-trivial 14:25
moritz cotto: 2) I think the least intrusive approach would be JS overlay, ie if you enable starry mode, an ajax query asks for which columns are starred, and displays it 14:26
cotto: not all that non-trivial. It's a simple ALTER TABLE
cotto moritz, ok. Does ilbot have any built-in upgrading code or is it necessary to do that manually? 14:27
moritz cotto: manually
cotto: ie you change schema.sql, I'll update the DB on the server 14:28
cotto moritz, what about having a separate table for that? Is having a single table important?
moritz cotto: that can work too. Just think about what queries will have to be done, and if separate tables will make that easier or harder 14:30
dalek kudo/nom-complex-insanity: 2c0fd1c | moritz++ | src/core/Num.pm:
fix Num.Rat
14:34
14:39 PacoLinux left 14:40 PacoLinux_ joined, PacoLinux_ is now known as PacoLinux
dalek kudo/nom: a5bfdd0 | moritz++ | src/core/Complex.pm:
optimize and fix a few ops. Not correct yet, but compiles
14:41
kudo/nom: 6fe7e6f | moritz++ | src/core/Complex.pm:
fix infix -
kudo/nom: 316fcb9 | moritz++ | src/core/ (3 files):
more work to get complex.t passing

includes Complex to Num, Int, Rat coercion, Num -> Rat and some bridge methods in Real
kudo/nom: 1e80b76 | moritz++ | src/core/Numeric.pm:
- should subtract, not add :-)
kudo/nom: a425e09 | moritz++ | src/core/Real.pm:
get infix:<-> with two different Real types working
kudo/nom: 2c0fd1c | moritz++ | src/core/Num.pm:
fix Num.Rat
kudo/nom: e51937b | moritz++ | src/core/ (4 files):
Merge branch 'nom-complex-insanity' into nom
kudo/nom: 6518be4 | moritz++ | t/spectest.data:
another passing test file
15:18 lucian joined 15:46 kid51 left 15:59 dafrito left
dalek sella/gh-pages: 8fea1fd | Whiteknight++ | / (2 files):
Add a link to Path
16:31
sella/gh-pages: 998bd29 | Whiteknight++ | libraries/path.md:
Flesh out Path library docs
sella/gh-pages: fdeba36 | Whiteknight++ | libraries/path.md:
More Path docs
sella: 6bc45f1 | Whiteknight++ | / (2 files):
Fix Path to be able to handle multi-character separator sequences. Add tests for this, for non-standard separators, and for attr-hash-attr searches
sella: 42996df | Whiteknight++ | / (10 files):
Path library is now stable
sella: 929084b | Whiteknight++ | / (4 files):
Update Path to use new-style constructors instead
16:31 daniel-s left 16:35 JimmyZ left 16:52 zby_home joined
dalek kudo/nom: 67fe2fd | jonathan++ | src/Perl6/SymbolTable.pm:
Fix $_ bug.
17:08
17:21 lichtkind joined 17:28 fperrad left
dalek kudo/nom: fa88a07 | (Solomon Foster)++ | t/spectest.data:
Turn on S32-str/lcfirst.t now that the { $_ = whatever } bug has been fixed.
17:28
17:31 fperrad joined 17:32 dodathome joined
bubaflub ping whiteknight 17:34
dalek kudo/nom: 16ccf25 | moritz++ | t/spectest.data:
four more passing test files
17:40
17:51 mj41 joined
bubaflub unping whiteknight 17:53
18:09 bluescreen_ joined
lichtkind haha 18:11
18:12 contingencyplan joined 18:33 whiteknight left 18:41 rurban_ joined 18:44 rurban left, rurban_ is now known as rurban 19:03 fperrad left 19:04 fperrad joined 19:40 kid51 joined 19:42 lucian left 19:50 preflex left 19:53 preflex joined
dalek TT #2141 created by jkeenan++: Test of parrotbug only 20:10
TT #2141: trac.parrot.org/parrot/ticket/2141
TT #2141 closed by jkeenan++: Test of parrotbug only
TT #2141: trac.parrot.org/parrot/ticket/2141
cotto I like where this is going. 20:24
kid51++
dalek TT #2142 created by jkeenan++: Yet another test of bug ticket creation 20:25
TT #2142: trac.parrot.org/parrot/ticket/2142
TT #2142 closed by jkeenan++: Yet another test of bug ticket creation
TT #2142: trac.parrot.org/parrot/ticket/2142
TT #2141 reopened by jkeenan++: Test of parrotbug only
TT #2141: trac.parrot.org/parrot/ticket/2141
TT #2141 closed by jkeenan++: Test of parrotbug only
TT #2141: trac.parrot.org/parrot/ticket/2141
rrot/m0-prototype: 2b67d33 | chromatic++ | src/m0/c/Makefile:
Added debugging binary build rule.
20:27
rrot/m0-prototype: 606aa97 | chromatic++ | src/m0/c/ (3 files):
Added m0b constant segment parsing.
rrot/m0-prototype: 84311ae | chromatic++ | src/m0/c/ (3 files):
Added m0b metadata segment parsing.
rrot/m0-prototype: 844b9a4 | chromatic++ | src/m0/c/ (3 files):
Added m0b bytecode segment parsing.
rrot/m0-prototype: 8b1430a | chromatic++ | src/m0/c/ROADMAP:
Added a ROADMAP for this C M0 implementation.
20:27 dodathome left, fperrad left
cotto Oooh. I also like where that's going. 20:29
chromatic++
bubaflub incoming 20:31
dalek rrot-gmp: 5835862 | bubaflub++ | src/GMP/ (6 files):
update these from loading from src/GMP to just plain ole GMP/
rrot-gmp: 7f5adb8 | bubaflub++ | t/integer/vtable/get_string.t:
test for vtable override get_string
rrot-gmp: 03673fe | bubaflub++ | / (4 files):
add vtable override for get_integer and tests
rrot-gmp: f6e8aae | bubaflub++ | gmpdoc2winxed.pl:
some macros are actually functions in disguise
rrot-gmp: 4977b67 | bubaflub++ | gmpdoc2winxed.pl:
remove extra blank-line
rrot-gmp: 6ec8757 | bubaflub++ | src/GMP/Integer. (2 files):
regenerate sources from new script

this picks up mpz_cmp_si, mpz_cmp_ui, mpz_sgn, mpz_odd_p, mpz_even_p
rrot-gmp: 008f934 | bubaflub++ | t/integer/cmp/01-cmp.t:
add tests for mpz_cmp_ui and mpz_cmp_si
rrot-gmp: 6b28b83 | bubaflub++ | / (4 files):
fix mpz_sgn, mpz_odd_p, mpz_even_p macros -> write my own functions
rrot-gmp: c4830ab | bubaflub++ | t/integer/ (2 files):
add tests for mpz_sgn, mpz_odd_p and mpz_even_p
kid51 cotto: Please express opinion in TT #920 and/or TT #2108.
bubaflub can someone explain the difference between the get_string VTABLE and the get_repr VTABLE? 20:37
cotto kid51, replied to 920 20:38
20:41 mj41 left
cotto bubaflub, get_repr isn't extensively used. The intent may have been to provide a human-readable format similar to Perl 6's .perl method. 20:42
bubaflub cotto: ok, so no real need to provide an override for get_repr? 20:43
cotto bubaflub, not if you don't want one
get_string is the important one
bubaflub cotto: ok, that's what i thought.
cotto: also, there is get / set number which takes a float - but all of the GMP functions only handle doubles. won't the conversion between the two lose precision? 20:44
cotto bubaflub, FLOATVAL is double afaict 20:46
bubaflub cotto: oh, perfect.
cotto are you on x86? 20:47
bubaflub cotto: yep.
cotto: so double == FLOATVAL == N registers
cotto bubaflub, check include/parrot/config.h and look at what Parrot_Float is defined to
I think it's constant across platforms. 20:48
bubaflub cotto: yup, it's a double for me. i don't imagine it'll be that way for everyone, though
cotto: i stand corrected.
cotto bubaflub, so you're happy? 20:49
bubaflub cotto: i don't usually love being wrong, but in this case i do.
cotto The naming of FLOATVAL is lta
bubaflub another vtable question - can you explain the difference between add and i_add 20:51
cotto i == in-place
20:52 ambs left
nopaste "kid51" at 192.168.1.3 pasted "TT #756: Am I correctly running the code provided in this ticket?" (36 lines) at nopaste.snit.ch/56924 20:52
bubaflub cotto: ok, that makes sense. the non-i variants will return a new (in my case) GMP integer
cotto bubaflub, exactly 20:53
bubaflub cotto: ok. is there a way i can trigger divide_floor? i imagine the plain ole / just calls the divide vtable.
cotto kid51, StringArray isn't a valid core PMC type. We have ResizableStringArray and FixedStringArray 20:54
20:54 Eclesia joined
Eclesia hi 20:54
bubaflub hello Eclesia. how goes the project?
Eclesia did not work on it this week. 20:55
kid51 cotto: So what behavior would we want parrot to do in that case? 20:56
cotto bubaflub, look for mentions of "floor_divide" in t/ .
bubaflub cotto: of course; thanks a bunch.
kid51 trac.parrot.org/parrot/ticket/756
cotto kid51, ah. coke is complaining about a very wrong failure for that code. 20:58
kid51, I don't see a bus error, so the ticket looks closeable from my perspective. 20:59
kid51 Can you take it and provide a reason for closing it? 21:02
cotto kid51, done 21:07
dalek kudo/nom: 14ef250 | moritz++ | src/core/Real.pm:
cis, unpolar
21:08 soh_cah_toa joined
dalek TT #756 closed by cotto++: invalid PMC causes bus error with const. 21:12
TT #756: trac.parrot.org/parrot/ticket/756
21:20 Eclesia left 21:25 lichtkind left 21:26 lucian joined, lucian left 21:27 perlite left 21:29 perlite joined 21:40 lucian joined 21:42 Psyche^ joined 21:46 lucian left 21:47 Patterner left, Psyche^ is now known as Patterner
soh_cah_toa cotto: ping 22:34
cotto soh_cah_toa, back at you 22:36
soh_cah_toa cotto: i saw your "make flag macros less noisy" commit. is this what you were talking about at the hackathon that fixed the "c++ blah blah blah" compiler warnings?
cotto soh_cah_toa, it fixes a few of them. In some cases, I think it's just gcc being too picky. 22:37
soh_cah_toa ok. how did you discover that that was the problem? how does casting it as it's own type and explicitly writing out the assignment instead of the |= shorthand fix it? 22:38
cotto soh_cah_toa, do you have any pending changes? I have another similar patch but I don't want to clobber anything you're working on.
soh_cah_toa, I don't honestly know. 22:39
though one of the macros was broken
the TEST macro assigned instead of just testing (iirc)
soh_cah_toa interesting. i didn't realize that 22:40
yeah, i think i do have changes. it's half-finished though. let me get to my laptop and push them
cotto soh_cah_toa, no rush.
soh_cah_toa, I also highly recommend avoiding single-letter variable names except in loops. They generally don't make code easier to understand except when there's a known convention they follow. 22:43
dalek TT #2143 created by jkeenan++: examples/benchmarks/overload.pl: Can we remove this file? 22:46
TT #2143: trac.parrot.org/parrot/ticket/2143
cotto soh_cah_toa, do you know for sure that the short version of commands work in hbdb? I think there's a bug in src/hbdb.c +718 22:47
dalek rrot/soh-cah-toa/hbdb: b0430db | soh_cah_toa++ | src/hbdb.c:
Added 'quit' command
22:50
rrot/soh-cah-toa/hbdb: 9dc4dbc | soh_cah_toa++ | src/ (2 files):
Some of the work on breakpoints I started at the YAPC hackathon. It's incomplete though
cotto soh_cah_toa, I'd recommend building with g++. It's much better about catching some kinds of errors. 22:52
soh_cah_toa sure
cotto: actually, now that i think about it, i haven't even tried the short commands
dalek rrot/soh-cah-toa/hbdb: 4b8f590 | cotto++ | src/hbdb.c:
add some more consts to sate gcc
22:53
rrot/soh-cah-toa/hbdb: cc7ee99 | cotto++ | / (2 files):
a few pointer and const fixes
soh_cah_toa cotto: i removed those const's b/c i thought you wanted me to avoid them for now 22:54
cotto soh_cah_toa, only where they're not required for the build
soh_cah_toa what do you mean? i don't think a const is ever "required" 22:55
cotto C++ is more picky than gcc about such things. 22:56
soh_cah_toa alright 22:57
is your's running? mine keeps segfaulting. i think i remember this happening at the hackathon after changing the way i initialized the command table 22:59
cotto same here
soh_cah_toa wonderful
cotto I was about to mention it
at least it's consistent
inconsistent segfaults are not fun 23:00
soh_cah_toa true
cotto I suspect that this one will be pretty easy to debug. 23:01
and you get to run your debugger in a debugger, which is a plus
dalek rrot: f69ef9a | jkeenan++ | t/pmc/io_iterator.t:
Add number of Trac ticket tracking 'todo' item
23:02 sjn left
soh_cah_toa yeah, it's definitely b/c of the way i had to explicitly define each field of the command table 23:04
when i changed it back to the old way, it runs
23:25 sjn joined
dalek TT #956 closed by cotto++: Singleton PMCs with the active_destroy flag set cause segfaults when ... 23:32
TT #956: trac.parrot.org/parrot/ticket/956
TT #1023 closed by cotto++: [TODO] Explore using Apple's new Blocks and Grand Central Dispatch
TT #1023: trac.parrot.org/parrot/ticket/1023
cotto kid51, when grepping the source for functions, make sure to run make realclean first to avoid getting results from (most) generated files. 23:34
dalek rrot/soh-cah-toa/hbdb: e53f5b4 | soh_cah_toa++ | / (2 files):
Fixed merge conflict
soh_cah_toa i have serious git problems. ugh
dalek rrot/soh-cah-toa/hbdb: 4fdb6bc | soh_cah_toa++ | src/hbdb.c:
This is a test to see if my git issues are gone
23:36
23:41 kid51 is now known as kid51_at_dinner 23:50 lucian joined
dalek rrot/soh-cah-toa/hbdb: c7d969e | soh_cah_toa++ | src/hbdb.c:
Reformatted a few comments
23:52
rrot/soh-cah-toa/hbdb: dc05051 | soh_cah_toa++ | src/hbdb.c:
Fixed "comparison between pointer and integer" build warning in parse_command()
rrot/soh-cah-toa/hbdb: ba32b81 | soh_cah_toa++ | src/hbdb.c:
Fixed two "assignment makes integer from pointer without a cast" build warnings in hbdb_cmd_break()
soh_cah_toa cotto: great, now the only build warnings that remain are the "braces around scalar initializer" ones regarding the command table 23:56
i'll definitely need your help on that though b/c i haven't got a clue as to how i can solve that problem