Parrot 3.5.0 "Menelaus" released | parrot.org | Log: irclog.perlgeek.de/parrot/today
Set by moderator on 27 June 2011.
cotto soh_cah_toa, looking now 00:13
00:17 lucian left 00:19 whiteknight joined
nopaste "cotto" at 192.168.1.3 pasted "hbdb build fix" (23 lines) at nopaste.snit.ch/56980 00:26
cotto soh_cah_toa, this approach fixes the build.
still segfaults when I try the "help" command
00:26 woosley joined
whiteknight too...many...trac...emails.... 00:29
cotto a few tickets got knocked off though
00:36 eternaleye left 00:37 eternaleye joined 00:40 PacoLinux left
whiteknight yes, can't argue with that 00:46
kid51_at_dinner If I don't ask about old tickets, the cage doesn't get cleaned :-) 00:50
00:50 kid51_at_dinner is now known as kid51 00:52 benabik joined
benabik ~~ 00:54
Finally made it home.
cotto benabik, wb 00:55
01:06 daniel-s joined
whiteknight I find myself spending a lot more time on github reading code and commits than I spend on trac. When we were on SVN I was spending the time on trac and was looking at tickets and stuff at the same time 01:07
but now that the fun stuff is happening on github, I'm not going to trac unless I'm specifically reminded about tickets that need attention
cotto same here
kid51 Cage cleaning is a dirty job, but somebody's got to do it. 01:08
whiteknight kid51: you're a gentleman and a scholar, no doubt.
I clearly don't have the attention span for that kind of work 01:09
along with myriad other personality flaws
cotto yet here you are 01:10
I have to say that I strongly prefer to have hackathons before YAPC now. I felt pretty useless after 3 days of yapc-ing. 01:14
whiteknight speaking of which, I think that we are ripe to have a parrot-related hackathon sometime soon 01:16
cotto whiteknight, virtual or irl? 01:17
whiteknight virtual
cotto good, because I don't want to travel at all soon
whiteknight well, irl too, but travel is a pain
cotto whiteknight, do you have any intention of going to the GSoC mentor's summit? dukeleto has me sold on it. 01:18
whiteknight we do see nice productivity boosts for a hackathon, if we don't do it too often
no, I can't go
cotto er, dukeleto++ has me sold
whiteknight I basically am not doing any travel this year
unfortunately
cotto weren't you going to hit fosscon?
aloha, going to fosscon? 01:19
aloha cotto: Search me, bub.
whiteknight I don't know if I can go to fosscon or not. I'm not signed up yet 01:21
Of course, that doesn't involve traveling. It's happening only a few blocks from where I work
cotto no excuses for you 01:22
kid51 aloha, going to fosscon is kid51 01:24
aloha kid51: Okay.
kid51 aloha, going to fosscon?
aloha kid51: going to fosscon is kid51
kid51 aloha, going to fosscon is also soh_cah_toa
aloha kid51: Okay.
kid51 aloha, going to fosscon?
aloha kid51: going to fosscon is kid51 or soh_cah_toa
kid51 aloha, going to fosscon is also possibly_pmichaud
aloha kid51: Okay. 01:25
cotto ponders the quantum superposition of pmichaud and not pmichaud
kid51 msg soh_cah_toa lib/Parrot/Test/HBDB.pm line 118 no need for 'require'; see perldoc.perl.org/base.html 01:26
aloha OK. I'll deliver the message.
01:35 contingencyplan left
bubaflub hey whiteknight, can i ask you a few questions about VTABLE overrides? 01:36
whiteknight bubaflub: you need to learn! You don't need to ask to ask
just ask
bubaflub just knocking before entering 01:37
cotto barge it. It'll be fun.
*in
bubaflub whiteknight: i've got a few basic overrides for GMP Integers, but i wanna add add_* and subtract_* and all those goodies.
whiteknight ok
bubaflub but i'm having trouble with set_[integer|float|string]_native 01:38
kid51 msg soh_cah_toa lib/Parrot/Test/HBDB.pm lines 204 and 243: You should not place your test function calls inside an 'if' block. If the conditions for the 'if' to be true are not met, the tests won't be run and it will screw up your plan().
aloha OK. I'll deliver the message.
whiteknight you're having problems with our problems.
bubaflub are these called when i do something like: var x = new GMP.Integer(32); x = 33;
cotto kid51, thanks for reviewing that code
bubaflub well, misery doesn't need company but sure enjoys it
whiteknight bubaflub: definitive resource is the Parrot source. Take a look at src/ops/set.ops to see where the various vtables are being called, and also look at src/pmc/default.pmc to see what the default fallback behavior is 01:39
bubaflub whiteknight: ok.
whiteknight set_integer_native may fall back to set_number_native or something stupid like that
kid51 msg soh_cah_toa Also at lines 204 and 243, you need to account for possibility that @lines is empty and that $lines[0] is undefined. That's why you're getting: Use of uninitialized value $this in pattern match (m//) at lib/Parrot/Test/HBDB.pm line 243.
aloha OK. I'll deliver the message.
whiteknight set_p_i seems to call VTABLE_set_integer_native 01:40
set_p_n and set_p_s are set_number_native and set_string_native respectively 01:41
bubaflub whiteknight: ok, i think it's just Winxed doing some magic, because the following works
var x = new GMP.Integer(32); x.set_native_integer(35); 01:42
then the assert that checks x is 35 passes
if i do
x = new GMP.Integer(32); x = 35; when i try to check the value x is not a GMP.Integer
whiteknight Get the generated PIR and see what op it is generated. I suspect x = 35 may be a box op, not a set op 01:43
or, box then set
To be certain, generate an explicit op: ${ set x, 35 };
bubaflub whiteknight: you're absolutely right, i'm getting a box op 01:45
i'll do it with the $
well
hmm
kid51 msg soh_cah_toa lib/Parrot/Test/HBDB.pm lines 166 and 175: You should modify each assignment to end in || ''; to avoid possible uninitialized values.
aloha OK. I'll deliver the message.
bubaflub how would i box first?
whiteknight We'll have to ask NotFound if there is any way in Winxed to generate a set_p_i op
bubaflub: what do you mean?
bubaflub i mean, i know in Winxed i can do things like int(x) and what not; can i coerce it to be a GMP Integer in the same way? 01:46
whiteknight no
at least, I don't think so
bubaflub ok
that's fine
kid51 msg soh_cah_toa lib/Parrot/Test/HBDB.pm sub _generate_pbc(): sub is not completely encapsulated; $parrot and $builder are defined outside scope of subroutine.
aloha OK. I'll deliver the message.
cotto kid51++ 01:47
kid51 msg soh_cah_toa lib/Parrot/Test/HBDB.pm END {}. Since you create hello.pbc in the test file (not in the module) why not unlink it in the test file? 01:50
aloha OK. I'll deliver the message.
whiteknight bubaflub: anything else you need? I 01:53
kid51 msg soh_cah_toa t/tools/hbdb/cmds.t line 39 test_bad_cmd: Recommend passing strings assigned to $bad_cmd and $err_msg as arguments to the function itself rather than hard-coding them inside the function. Increases flexibility in testing.
aloha OK. I'll deliver the message.
whiteknight m going to sign off soon
bubaflub whiteknight: nah, i'm good for now. imma blog post and push some commits. thanks for all your help.
i'm sure i'll be harassing you again soon enough.
whiteknight bubaflub: I look forward to it!
later
01:54 whiteknight left
soh_cah_toa kid51++: thanks a bunch for the code review. i'll take a look at those things now 01:56
dalek rrot/soh-cah-toa/hbdb: 92fbcf4 | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm:
Removed "require Exporter" since "use base qw(Exporter)" loads it implicitly
 02:00
rrot/hbdb: ca41f3e | jkeenan++ | include/parrot/hbdb.h:
Correct typo in inline comment.
 02:06
kid51 Hmm, that's not waht I wanted to do.
bubaflub incoming 02:07
dalek rrot-gmp: 2ab89a4 | bubaflub++ | / (8 files):
vtable overrides for everyone!

this includes get_bool, get_number, set_integer_native, set_number_native, and set_string_native with tests. the rest are commented out in the main source		 02:08
cotto dukeleto, ping
soh_cah_toa kid51: wow, i'm surprised you caught that. you've got quite an eye :)
kid51 soh_cah_toa: Spelling error at line 23 of include/parrot/hbdb.h
dalek rrot-gmp: 8ad02d7 | bubaflub++ | src/GMP/Integer.winxed.template:
fix typo
 02:11
cotto kid51, do you have an outline or notes from the BoF at YAPC::NA? 02:12
dalek rrot/soh-cah-toa/hbdb: 7ead923 | soh_cah_toa++ | include/parrot/hbdb.h:
Fixed typo in inline comment thanks to kid51++
kid51 soh_cah_toa: I'm trying to diagnose those warnings during the build of src/hbdb.c -- but your C foo is far in advance of mine!
cotto mine were very minimal 02:13
soh_cah_toa kid51: you will go insane looking at those ;)
kid51 Each of them is googleable. 02:14
nopaste "kid51" at 192.168.1.3 pasted "hbdb branch build errors" (60 lines) at nopaste.snit.ch/56984 02:15
kid51 Most of the *other* warnings have already been cleared up by mikehh++ in master
soh_cah_toa whoa, you have a lot more and different ones than i do
cotto soh_cah_toa, that reminds me, you need to make sure to sync with master every few weeks. 02:16
kid51 soh_cah_toa: Do you have warnings from * src/hbdb.c* above and beyond what I pasted?
soh_cah_toa let me nopaste my build warnings
benabik home again again. I was planning on blogging something useful today but put up a placeholder instead. DID: Drove. WILL DO: Blog. BLOCKERS: Lack of sleep. 02:17
kid51 benabik: Sleep tonight. Enjoy hacking on holiday weekend tomorrow and Monday ;-) 02:18
dalek website: benabik++ | GSoC 6: Unexpected YAPC 02:19
website: www.parrot.org/content/gsoc-6-unexpected-yapc
cotto a wild yapc appeared
nopaste "soh_cah_toa" at 192.168.1.3 pasted "HBDB Build Warnings" (27 lines) at nopaste.snit.ch/56985 02:20
dalek kudo/nom: ebe1051 | jonathan++ | NOMMAP.markdown:
Remove a couple of dealt with nommap items.
kid51 soh_cah_toa: The only warnings you need to be concerned with are those in src/hbdb.c. All the others are long-standing warnings in other source code files. 02:21
soh_cah_toa ok good
kid51 But you/we do need to be concerned about the warnings in hbdb.c. They ought to be gone by the point where you merge into master. 02:22
cotto kid51, no worries. They will be.
kid51 soh_cah_toa: Are you configuring with gcc or g++? 02:25
benabik cotto: YAPC used "Drive to far away hotel". It was super effective! 02:26
soh_cah_toa kid51: gcc, though cotto mentioned i should also use g++ today. how do i change that w/ make w/o editing the Makefile?
kid51 soh_cah_toa: Yes, I'm puzzled by why I'm getting all those errors in src/pmc/structview.c and you're not -- especially since I just pulled your branch fresh
cotto soh_cah_toa, you have to rerun Configure.pl with the right arguments 02:27
In general, you shouldn't edit the makefile directly for that.
kid51 perl Configure.pl --cc=g++ --link=g++ --ld=g++ --test && make && make test 02:28
soh_cah_toa ok thanks
kid51 You will also have to fix 2 instances of unused assert macros: prove t/codingstd/c_arg_assert.t (or: make codetest) 02:29
cotto --test isn't generally necessary (but won't hurt) 02:30
benabik So how is everyone? I assume y'all got home safe.
kid51 Configuring with g++, I get this:
src/hbdb.c:118: error: braces around scalar initializer for type 'hbdb_cmd_t*'
... in 2 groups of 3 ... 02:31
cotto benabik, safe, yes. sick, also yes.
kid51 ... illustrating the fact that gcc and g++ throw different build warnings
benabik cotto: Oh, bleck. Get better?
cotto benabik, it's either that or get worse. ;]
benabik cotto: ... True? 02:32
soh_cah_toa kid51: yes, i think that's the one i'm getting
kid51 Oh wait: With an all g++ build, the build fails in that branch
src/hbdb.c:118: error: braces around scalar initializer for type 'hbdb_cmd_t*' 02:33
make: *** [src/hbdb.o] Error 1
So it warns in gcc but FAILs in g++
soh_cah_toa interesting
kid51 Hypothesis: Once you fix this in the g++ build, you will pass on gcc without warnings. Proceed to test hypothesis while ... 02:34
kid51 goes for ice cream
soh_cah_toa good idea
benabik (ice cream)++
benabik eats his ice cream and drinks his hard cider. 02:35
02:41 cosimo left, rurban_ joined 02:44 rurban left, rurban_ is now known as rurban
dalek website: bubaflub++ | Parrot-GMP: Random Numbers, Distutils, Plumage, VTABLE overrides 02:50
website: www.parrot.org/content/parrot-gmp-r...-overrides
02:51 kurahaupo joined
bubaflub msg dukeleto bloggity blog blog blah blah www.parrot.org/content/parrot-gmp-r...-overrides 02:53
aloha OK. I'll deliver the message.
03:06 theory joined 03:11 woosley left
nopaste "kid51" at 192.168.1.3 pasted "This enables g++ build to complete -- but causes gcc build to fail" (13 lines) at nopaste.snit.ch/56992 03:13
03:55 cosimo joined 04:00 kid51 left 04:25 daniel-s left
dukeleto bubaflub++ # blag toast 04:40
bubaflub hola dukeleto
dukeleto: wanna skype it up sometime?
dukeleto bubaflub: sure, but I just got back from hiking, and will likely pass out soon
bubaflub: perhaps tomorrow
bubaflub dukeleto: yeah. i'm watching some anime with the wife. 04:41
04:41 whiteknight joined
dukeleto bubaflub: nice work recently! Very nice to see you close lots of github issues 04:41
bubaflub dukeleto: thanks! 04:44
dalek rrot/soh-cah-toa/hbdb: 525a293 | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm:
Assignment of @lines now defaults to an empty string on error (kid51++)
 04:46
rrot/soh-cah-toa/hbdb: 77167c2 | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm:
Began refactoring duplicate code into _enter_cmd() and _close_fh()
05:02 zby_home_ joined 05:03 zby_home left
dalek rrot/soh-cah-toa/hbdb: 64fea2a | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm:
Moved call to waitpid() to inside _enter_cmd()
 05:05
kudo/nom: f37d574 | moritz++ | src/core/ (4 files):
log()
 05:19
kudo/nom: a08f69e | moritz++ | / (6 files):
exp, infix ** for Complex
05:20 fperrad joined
dalek rrot/soh-cah-toa/hbdb: 74662fd | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm:
Moved definition of to inside _generate_pbc() to increase encapsulation
 05:29
soh_cah_toa ugh, stupid bash variable interpolation. should've used single quotes 05:31
05:31 soh_cah_toa left
dalek kudo/nom: a930204 | moritz++ | / (2 files):
allow underscores in rat literals. Tests.
 05:33
kudo/nom: 6f27881 | moritz++ | / (3 files):
two more passing test files, add List.fmt, grep() accepts non-Code matchers
 05:50
05:52 NotFound_b joined
NotFound_b bubaflub: you need to use the operator =: 05:53
x = new GMP.Integer(32); x =: 35;
06:09 redicaps joined
NotFound_b msg bubaflub you need to use the operator =: ---> x = new GMP.Integer(32); x =: 35; 06:14
aloha OK. I'll deliver the message.
06:29 kurahaupo left
dalek sella/gh-pages: c1e32d9 | Whiteknight++ | libraries/string.md:
update string docs with concepts
 06:30
sella/gh-pages: 28c7b1a | Whiteknight++ | _layouts/rosella.html:
Add link to String docs
sella: 418a2a2 | Whiteknight++ | / (2 files):
Update DelimiterRegion to throw an error if we find an unclosed region. Add a test for this behavior.
sella: 55eef91 | Whiteknight++ | t/string/tokenizer/Delimiter.t:
test for what happens when we have no text between or after delimiters in Delimiter.t
sella: f225561 | Whiteknight++ | / (2 files):
Add tests for Tokenizer. Small fixes
sella: 4ebbdc9 | Whiteknight++ | t/string/ (2 files):
+tests for Token
sella: f078d32 | Whiteknight++ | / (17 files):
String is now stable
sella: 6c3ca27 | Whiteknight++ | README.md:
add Path and String to README
 06:31
06:49 benabik left, benabik joined 06:51 benabik left, benabik joined 07:06 dodathome joined
dalek sella: 08afc84 | Whiteknight++ | s (5 files):
Fix Template to work with recent changes to String. Add a Context object to hold temporaries with clobbering user data. Fix for loop to work with hashes.
 07:16
07:16 whiteknight left, particle left 07:18 particle joined 07:27 mj41 joined
dalek rrot/m0-prototype: eb17d98 | chromatic++ | src/m0/c/ (4 files):
Moved run_ops() to m0_ops.c and added m0_ops.h.
 07:59
rrot/m0-prototype: 5ec429a | chromatic++ | src/m0/c/ (7 files):
Added skeleton for "Hello, world!" handling.

This adds m0_ops.c with its header.
rrot/m0-prototype: f9c6a20 | chromatic++ | src/m0/c/ (4 files):
Made "Hello, world!" example work.

Little else will work, and there are some definite hacks here, but progress is progress.
rrot/m0-prototype: 57ad086 | chromatic++ | src/m0/c/ROADMAP:
Updated ROADMAP.
benabik Woo! More chromatic work!
08:06 theory left 08:19 redicaps left 08:52 mikehh left 08:54 daniel-s joined 09:32 ligne left 10:15 lichtkind joined 10:19 PacoLinux joined
dalek kudo/nom: 430e461 | pmichaud++ | src/core/Any.pm:
Any.grep doesn't need a gather/take, .map is already lazy.
 10:19
10:25 mj41 left 10:33 mj41 joined 10:41 rurban_ joined 10:44 rurban left, rurban_ is now known as rurban 10:51 Eclesia joined
Eclesia hi 10:52
11:37 ligne joined
Eclesia those 'using ...' are painful ... 11:46
11:46 ligne left
Eclesia NotFound : how do I use a function located in a sub-namespace ? 11:47
in winxed
11:51 whiteknight joined
Eclesia hi whiteknight 11:52
whiteknight good morning Eclesia. How are you doing today?
Eclesia more or less fine, fighting with winxed 11:53
and you ?
11:53 rohit_nsit08 joined 11:55 zby_home_ left
Eclesia whiteknight: perhaps you know, but how can I call a method in a sub-namespace ? 11:55
function*
whiteknight you have to look it up
Eclesia means ?
ha found it ... 11:57
12:19 mj41 left 12:33 Eclesia left 12:58 mj41 joined 13:17 mj41 left 13:21 kid51 joined 13:24 mj41 joined 13:26 rohit_nsit08 left 13:32 mj41 left 13:36 rohit_nsit08 joined 13:38 lucian joined
rohit_nsit08 whiteknight: there? 13:46
13:46 ambs joined, bluescreen left
rohit_nsit08 Coke: ping 13:46
13:46 bluescreen joined
whiteknight hello rohit_nsit08 13:57
rohit_nsit08 whiteknight: hello 13:59
i was trying to access a namespace's property using its pmc object
codepad.org/1pu2ItmE
line 9
but getting error , while same thing with $P2.'find_var'() is working 14:00
any mistake i'm doing? 14:01
whiteknight I think get_global expects a Key, not a NameSpace
use 'find_var' instead
rohit_nsit08 I expected this, but $P2 is the namespace object , documentation says that I can use a namespace object to search into the properties 14:04
here, docs.parrot.org/parrot/latest/html/...s.pod.html 14:05
search for this, $P1 = get_global $P0, "bill" 14:06
allison This is pretty cool www.stargit.net/#github (look up github/parrot) 14:12
whiteknight rohit_nsit08: that's either a bug or a problem with documentation. Open a ticket 14:27
dalek sella: e3534e3 | Whiteknight++ | src/unstable/ (3 files):
Add a new test_template utility program to create new stub test files using the Template library.
 14:29
sella: 0054712 | Whiteknight++ | src/unstable/ (3 files):
update the test_template utility to generate harnesses too. Add two templates for harnesses in winxed and nqp
rohit_nsit08 whiteknight: ya seems to some documentation error. ok I'll open a ticket for that 14:31
whiteknight allison: yes, that is very cool
kid51 I can't get that stargit thing to do anything under my direction 14:46
14:51 fperrad left
dalek rrot: 1ee5ec8 | jkeenan++ | src/packfile/pf_items.c:
Add number of Trac ticket for inline comment.
 15:08
kid51 whiteknight: ping 15:13
whiteknight kid51: pong
kid51 ready for phone? 15:14
whiteknight yes 15:15
15:20 mj41 joined 15:25 arnsholt left 15:27 lichtkind left
bubaflub ~ 15:30
dalek rrot-gmp: cd43896 | bubaflub++ | t/integer/vtable/set_ (3 files):
use Winxed operator instead of dropping down to PIR, NotFound++
 15:33
15:38 bluescreen left 15:39 bluescreen_ left 15:45 arnsholt joined 15:46 zby_home joined
lucian wishes for better distribution/testing utils 15:58
whiteknight: hi. i'm trying to figure out what rosella's setup.winxed is doing 16:09
whiteknight lucian: hopefully, "building things" 16:10
lucian whiteknight: yes, that much i gathered :)
i've monkey'd up a setup.winxed for puffin, based on yours, and I get this "Reading version information...Unable to open filehandle from path 'src/_build/_versions.winxed' " 16:11
lucian is a little confused
whiteknight oh, Rosella has a VERSIONS file. setup.winxed reads that file and outputs a .winxed file to add all those numbers into a hash 16:13
so you probably want to rip out all that crap
16:15 theory joined
lucian whiteknight: right. good, i was wondering if it's necessary 16:20
16:20 contingencyplan joined
whiteknight no, not necessary. 16:21
I should document better
16:21 NotFound_b left
lucian whiteknight: it's still very useful, thanks 16:22
16:24 mj41 left 16:30 kid51_ joined 16:34 kid51 left 16:35 kid51_ left 16:36 kid51 joined 16:43 daniel-s left 16:44 mj41 joined 16:58 lichtkind joined
cotto ~~ 17:03
17:37 fperrad joined
whiteknight cotto: I don't have time now, we have to go out to the inlaws' soon, but later we should get together and talk profiling 17:55
cotto whiteknight, ok. It's on the list of 3 +/- Parroty things I want to get done today. 17:57
whiteknight okay, awesome. I'm on vacation this week so I am going to have more than average time to focus on parrot
cotto good deal 17:58
whiteknight I want to get my packfiles branch mergable, and I want to get a plan written down for profiling. All my Rosella-related todo items for the week are already done as of this morning
it helps when I don't sleep all night
cotto sleep is great, but if you're stuck with insomnia hacking is a good second choice 17:59
bubaflub lucian: this is a bit late, but i've got a distutils setup.winxed at github.com/bubaflub/parrot-gmp/blo...tup.winxed - the only parts that are specific to my project are the check_dependencies() and clean_build_dir() functions 18:00
cotto I'm excited that chromatic has started hacking on a C m0 interp.
er, chromatic++
whiteknight I'm extremely sensitive to caffeine. Yesterday afternoon I had several glasses of iced tea. badda bing, I don't sleep all night
yes, chromatic++ indeed
cotto Some guys from thinkgeek were giving out a bunch of caffeinated stuff. I don't have the LD50, but it's getting there. 18:02
whiteknight okay, signing off now. I'll be back on later tonight. Maybe much much much later if I can't keep my damn hands off the caffein 18:03
later
18:03 whiteknight left
lucian bubaflub: ah, thanks. i'll have a look 18:13
bubaflub lucian: cool. i'm more than happy to answer any questions / try anything out on my machine. 18:17
dalek rrot/tt1047/type_sizes: b9c08dc | jkeenan++ | / (2 files):
Per discussion in trac.parrot.org/parrot/ticket/1047, constrain opcode type to have same size as integers.
 18:22
18:23 rurban left 18:29 kid51 left 18:38 rurban_ joined 18:41 rurban__ joined 18:44 rurban__ is now known as rurban 18:46 rurban_ left 18:55 kurahaupo joined 19:13 kurahaupo left
dukeleto ~~ 19:44
19:46 soh_cah_toa joined 20:00 Coke left 20:01 Coke joined 20:08 fperrad left, fperrad joined 20:23 fperrad left 20:30 dodathome left 20:43 PacoLinux left 20:48 mj41 left
nopaste "soh_cah_toa" at 192.168.1.3 pasted "Build Errors on Master Branch" (20 lines) at nopaste.snit.ch/57060 20:53
soh_cah_toa is anyone else getting those errors? 20:54
that's on the master branch w/ the most recent changes (commit 1ee5ec8)
cotto soh_cah_toa, looks fine to me. Did you try make reconfig && make ? 20:57
soh_cah_toa let me try
'reconfig' isn't a make target. are you talking about 'realclean'? 20:58
cotto soh_cah_toa, it should be
should be equivalent to realclean and Configure.pl with whatever arguments you originally passed to it
soh_cah_toa that's odd. now it is after running 'realclean'
cotto whee 20:59
soh_cah_toa hmm...now it's building. how odd...
dalek rrot/soh-cah-toa/hbdb: 5979d51 | soh_cah_toa++ | src/hbdb.c:
Fixed last of the build warnings. However, I don't think this can be a final solution
 21:01
kudo/nom: f131a60 | moritz++ | src/core/Numeric.pm:
sin, cos, tan as functions
 21:02
cotto soh_cah_toa, that won't work. It casts a function pointer to a struct. 21:04
soh_cah_toa i know
i was just trying solutions and that finally got the warnings to stop 21:05
cotto soh_cah_toa, what about the nopaste I msg'd you yesterday? 21:06
soh_cah_toa cotto: i don't remember it
cotto nopaste.snit.ch/56980 21:07
soh_cah_toa ah, that should work 21:08
wow, that's so obvious 21:09
21:11 ambs left
soh_cah_toa cotto: how do you feel about moving all those types into include/parrot/hbdb.h and making command_table extern? i think it makes sense to keep types in a separate header 21:13
also, i may need some of them in src/runcores/cores.c soon. that way, i could just include the header rather than compiling it together w/ src/hbdb.c
cotto soh_cah_toa, definitely. I was waiting until you'd gotten the build working to bring it up. 21:14
soh_cah_toa ok good
at first, i was trying to limit the scope of all my typedefs/structs but now their scope is increasing elsewhere. plus, it's standard practice 21:15
dalek rrot/soh-cah-toa/hbdb: 11e5be0 | soh_cah_toa++ | src/hbdb.c:
Used cotto's method to reduce build warnings. Almost works
 21:23
nopaste "soh_cah_toa" at 192.168.1.3 pasted "Build Warnings With cotto's Method" (13 lines) at nopaste.snit.ch/57094 21:25
soh_cah_toa cotto: do you get those warnings/errors too?
cotto soh_cah_toa, I didn't intend to suggest that you use those struct names, just that that was a way to to fix the code. 21:27
same thing with gcc
soh_cah_toa yeah, i know
21:28 rohit_nsit08 left
soh_cah_toa cotto: \\o/ i have a possible solution 21:37
21:37 kurahaupo joined
soh_cah_toa will push in a second 21:37
dalek rrot/soh-cah-toa/hbdb: 4bd6a6a | soh_cah_toa++ | src/hbdb.c:
Used array notation instead of char * to fix last of the build warnings
 21:40
soh_cah_toa uh oh, segfaults :( 21:41
dalek kudo/nom: 4d0dabf | moritz++ | src/core/Numeric.pm:
add sqrt function
soh_cah_toa so close 21:42
21:42 Psyche^ joined, rohit_nsit08 joined 21:47 Patterner left, Psyche^ is now known as Patterner
dalek kudo/nom: bf3eccf | moritz++ | src/Perl6/Actions.pm:
fix rationals with zeros after the decimal point
 21:50
kudo/nom: e2a6299 | moritz++ | t/spectest.data:
run simplified trig tests
dukeleto nom nom nom 21:51
makes me hungry
bubaflub hola dukeleto 22:10
22:10 lichtkind left 22:20 Coke left, luben joined, Coke joined
dukeleto bubaflub: werd 22:29
dukeleto .nom($food)
22:30 sjn left
bubaflub dukeleto: i'm heading out in 10 minutes - got some time to skype later tonight? 22:30
22:36 mikehh joined 22:37 lucian_ joined, contingencyplan left
cotto since dalek is ignoring my blog, I'll just leave this here: reparrot.blogspot.com/2011/07/parro...-2011.html 22:38
22:39 lucian left
dalek kudo/nom: dcf30ef | jonathan++ | src/Perl6/ (2 files):
Assorted fixes to placeholder parameters.
 22:44
kudo/nom: dcdd4b7 | jonathan++ | t/spectest.data:
Two more passing test files (placeholder related).
rrot/m0-spec: 2836026 | cotto++ | docs/pdds/draft/pdd32_m0.pod:
add a suggestion from particle++ for an M0 completeness test
 22:45
cotto dukeleto, I'd like your thoughts on that blog post. 22:56
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#16891), fulltest) at 3_5_0-55-g1ee5ec8 22:59
Ubuntu 11.04 i386 (g++)
cotto: interesting post - I worry at the word optimization - too many things don't work properly 23:07
it tends to come down to 'premature optimization' far to often 23:09
s/to often/too often/ 23:10
cotto mikehh, That's a good point. In the future, I'll consider adding some hard numbers to that section to demonstrate what kind of measurable improvements are likely. 23:16
but not all optimization is premature, even if much of it is
see also: chromatic 23:18
mikehh cotto: there is a lot of cruft that needs to be removed in parrot, we have mede a lot of improvements in a lot of areas 23:19
cotto mikehh, definitely. There's more to do still.
mikehh but far to often a lot of junk gets left behind and slows things down and also obfuscuates thing
things
cotto see also: imcc 23:20
mikehh definately :-}
I was following pirc a while back, but unfortunately we had a bus number problem there 23:21
I have been looking at gc quite a bit, but we have a similar bus number problem there 23:22
dalek kudo/nom: 62204fe | tadzik++ | NOMMAP.markdown:
Update NOMMAP with the Block in a package case
 23:32
23:53 bluescreen joined 23:54 rohit_nsit08 left
Coke pastebin.com/PZ5FCbTw - from rohit - he's trying to figure out why lines 14 & 17 don't generate the same output. 23:55
23:56 zby_home left
Coke now catches up to 10am... 23:57
23:59 bluescreen_ joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!