Camelia, the Perl 6 bug

IRC log for #parrot, 2012-03-25

Parrot | source cross referenced

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:31 whiteknight joined #parrot
00:32 whiteknight good evening, #parrot
00:37 wagle joined #parrot
00:48 benabik o/ whiteknight
01:07 whiteknight hello benabik
01:55 alester joined #parrot
01:59 alester Are other folks having miniparrot sadness?
02:02 whiteknight alester: I haven't tried recently
02:02 dalek Rosella: cd82644 | Whiteknight++ | src/ (5 files):
02:02 dalek Rosella: [Net] Implement URI encoding, based on URI::Escape from the parrot library. Add builtins for the various required string operations. Use the same basic algorithm to implement an encoding routine for form data
02:02 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/cd82644bb6
02:02 dalek Rosella: af076fc | Whiteknight++ | src/unstable/net/ (2 files):
02:02 dalek Rosella: [Net] Make sure to set the Content-Length for post data
02:02 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/af076fc9c4
02:02 dalek Rosella: 333ccdd | Whiteknight++ | src/unstable/net/ (2 files):
02:02 dalek Rosella: [Net] fix a few small problems I added when I refactored Protocol
02:02 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/333ccdd46f
02:10 benabik joined #parrot
02:23 dalek Rosella: 3607d2e | Whiteknight++ | s (8 files):
02:23 dalek Rosella: [Net] Break the protocol classes out into separate files. Misc cleanups
02:23 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/3607d2e878
03:04 Psyche^ joined #parrot
06:20 dalek parrot: d1142ed | dukeleto++ | config/gen/makefiles/root.in:
06:20 dalek parrot: Unbreak the parrallel build by making prt0* depend on miniparrot
06:20 dalek parrot: review: https://github.com/parrot/parrot/commit/d1142ed3c2
06:24 alester dukeleto++
06:30 taryk left #parrot
06:39 nbezzala joined #parrot
09:08 dalek parrot: 003b2b8 | nine++ | src/pmc/resizablepmcarray.pmc:
09:08 dalek parrot: Fix documentation to match the VTABLE function names they document
09:08 dalek parrot: review: https://github.com/parrot/parrot/commit/003b2b8060
09:08 dalek parrot/threads: 87c6d52 | nine++ | t/pmc/task.t:
09:08 dalek parrot/threads: Parrot's Test::More is not ready for threads (writes to shared data)
09:08 dalek parrot/threads: review: https://github.com/parrot/parrot/commit/87c6d52c32
09:08 dalek parrot/threads: cd053dc | nine++ | / (2 files):
09:08 dalek parrot/threads: Fix Parrot_pmc_get_type_str when the class hash is a Proxy
09:08 dalek parrot/threads: review: https://github.com/parrot/parrot/commit/cd053dcaa8
09:09 dalek parrot/threads: fdb60a5 | nine++ | src/thread.c:
09:09 dalek parrot/threads: Don't create proxies for PMCNULL
09:09 dalek parrot/threads: review: https://github.com/parrot/parrot/commit/fdb60a5f50
09:09 dalek parrot/threads: 65e2704 | nine++ | / (8 files):
09:09 dalek parrot/threads: Fix the wait op for threads
09:09 dalek parrot/threads:
09:09 dalek parrot/threads: A task scheduled on another thread will get cloned to be local to this
09:09 dalek parrot/threads: thread. The two task objects are then linked by a "partner" pointer. To
09:09 dalek parrot/threads: wait for a task of another thread one now only has to add the current
09:09 dalek parrot/threads: task to the waiter list of the original task object.
09:09 dalek parrot/threads:
09:09 dalek parrot/threads: A task will schedule its waiters when it is done and now will also see
09:09 dalek parrot/threads: if the partner task has any waiters and schedule them on the partner's
09:09 dalek parrot/threads: interp.
09:09 dalek parrot/threads:
09:09 dalek parrot/threads: In addition the main thread will now no longer exit as long as tasks on
09:09 dalek parrot/threads: other threads are running (an explicit exit op does still work however).
09:09 dalek parrot/threads: The main thread's tasks could still be waiting for other tasks to
09:09 dalek parrot/threads: finish.
09:09 dalek parrot/threads: review: https://github.com/parrot/parrot/commit/65e27042d9
10:05 mj41 joined #parrot
10:07 nine msg whiteknight: increased performance of chameneos.pir by another 3x by pushing the interp lock down into the GC. So it's now 11 times faster than yesterday this time. But this comes at the cost of the single threaded load being now 3.72 % slower than on master with signficance 17.0σ
10:07 aloha OK. I'll deliver the message.
11:06 whiteknight joined #parrot
11:20 contingencyplan joined #parrot
12:50 PacoAir joined #parrot
13:14 benabik ~~
13:23 dalek parrot/m0: 061d58e | dukeleto++ | src/m0/ (2 files):
13:23 dalek parrot/m0: [m0] Start adding convert_* ops to C implementation
13:23 dalek parrot/m0: review: https://github.com/parrot/parrot/commit/061d58e303
13:23 dalek parrot/m0: 2e8c33a | dukeleto++ | .travis.yml:
13:23 dalek parrot/m0: [ci] Update travis config to test m0 stuff
13:23 dalek parrot/m0: review: https://github.com/parrot/parrot/commit/2e8c33a051
13:23 dalek parrot/m0: b08ccd1 | dukeleto++ | .travis.yml:
13:23 dalek parrot/m0: [ci] Update travis config to use a perl worker
13:23 dalek parrot/m0: review: https://github.com/parrot/parrot/commit/b08ccd1bdb
13:28 taryk joined #parrot
13:29 taryk left #parrot
13:32 dalek parrot: 4c5d82d | dukeleto++ | docs/project/cage_cleaners_guide.pod:
13:32 dalek parrot: [doc][ci skip] Add a note to the cage cleaner guide about running parrot+valgrind
13:32 dalek parrot: review: https://github.com/parrot/parrot/commit/4c5d82deea
13:51 jashwanth joined #parrot
13:54 kid51 joined #parrot
14:05 taryk joined #parrot
14:06 jashwanth joined #parrot
14:09 szbalint joined #parrot
14:10 dalek parrot: 9dbe4f5 | jkeenan++ | / (2 files):
14:10 dalek parrot: Add counsel to test changes to Makefile templatest with 'make -jN' for N>1.
14:10 dalek parrot: review: https://github.com/parrot/parrot/commit/9dbe4f519b
14:10 dalek parrot: c29badb | jkeenan++ | docs/project/cage_cleaners_guide.pod:
14:10 dalek parrot: Merge branch 'master' of git@github.com:parrot/parrot
14:10 dalek parrot: review: https://github.com/parrot/parrot/commit/c29badb30a
14:35 jsut joined #parrot
15:16 mdupont joined #parrot
15:50 taryk left #parrot
15:51 Psyche^ joined #parrot
17:16 jashwanth joined #parrot
17:27 patspam joined #parrot
17:38 jashwanth joined #parrot
17:50 jashwanth joined #parrot
18:02 kid51 joined #parrot
18:34 wknight-phone joined #parrot
19:12 patspam joined #parrot
19:27 mj41 joined #parrot
19:29 dalek rakudo/typed-ex-hash: d34720f | moritz++ | src/core/Failure.pm:
19:29 dalek rakudo/typed-ex-hash: make Failure throw more consistently
19:29 dalek rakudo/typed-ex-hash: review: https://github.com/rakudo/rakudo/commit/d34720f641
19:29 dalek rakudo/nom: 06b4eca | moritz++ | src/core/Failure.pm:
19:29 dalek rakudo/nom: make Failure throw more consistently
19:29 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/06b4eca92a
19:59 dalek rakudo/nom: 008dd40 | moritz++ | src/core/ (3 files):
19:59 dalek rakudo/nom: start to fail() with typed exceptions from Str.Numeric
19:59 dalek rakudo/nom:
19:59 dalek rakudo/nom: somewhere along the way it still stringifies though :(
19:59 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/008dd40862
21:17 patspam joined #parrot
21:24 dalek rakudo/nom: 34c3e35 | moritz++ | src/core/Failure.pm:
21:24 dalek rakudo/nom: replace a wad of PIR code with Perl 6; also preserves the argument type of &fail
21:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/34c3e35285
21:48 jsut_ joined #parrot
21:49 patspam_ joined #parrot
21:50 patspam joined #parrot
21:58 dalek rakudo/nom: 57a6818 | pmichaud++ | src/core/ (3 files):
21:58 dalek rakudo/nom: Merge branch 'nom' of github.com:rakudo/rakudo into nom
21:58 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/57a6818729
22:54 whiteknight yay! I've *finally* been able to use my new Rosella library to post a file to smolder
22:55 benabik \o/
22:55 whiteknight I spent hours not realizing that say(http_header); say(http_content); would print different data to the console than the socket was seeing
22:55 benabik why?
22:55 whiteknight because the socket expects two newlines between header and content, where say() was automatically adding one to the console
22:56 whiteknight so it *looked* like I had two newlines, but I was only sending one over the socket
22:57 whiteknight it was only when I condensed to say(header + content) that I saw my mistake
22:57 benabik d'oh
22:57 whiteknight srsly

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

Parrot | source cross referenced