Vision for 2.0: Production Users | Don't forget to post closed tickets in your report. | Note: This channel is only for our Tuesday status meetings; you probably want #parrot instead.
Set by moderator on 16 December 2009.
00:16 cotto joined 00:17 davidfetter left 11:53 bluescreen joined 13:00 kid51 joined, kid51 left, kid51 joined
kid51 kid51's report 13:09
In last several weeks have reviewed some older TT tickets. Resolved some.
In other cases, I cc-ed or msg-ed various Parrot devs for a consultation. If I poked you, please respond. 13:10
Blocking on:
trac.parrot.org/parrot/ticket/1393
Why does a particular test abort (SIGNAL 6) when run thru a test harness but not when run by itself? 13:11
And why only on one OS (Linux/i386) and not on another (Darwin/PPC)?
Being unable to get a clean PASS on 'make test' due to this prevents me from evaluating other TTs, e.g., JimmyZ's patches. 13:12
EOR
Question for discussion: What steps are we going to take between now and end of January to actualize goals set at recent online summit for preparation for Rakudo* release in 2Q? 13:28
kid51 goes to $job
14:27 bluescreen joined 16:36 PacoLinux joined
japhb Two week report ("Two, two, TWO weeks in one!"): 17:07
DONE:
* (Parrot) Fix TT #1326, data_json fail which broke crow.pir
* Add TT #1391 requesting per-parrot-build GUID that Plumage can inspect
* Add canned searches for 'component: plumage' tickets in Trac
* Update TASKS to point to Trac ticket searches
* Update Gitorious and README project overviews to reflect current status
* Convert all remaining bits of Glue.pir to Util.nqp
* Split off all Plumage-specific bits of Util.nqp into Plumage::Util
* Replace JSON hackery at startup with hash() and => NQP-rx syntax
* Stop using PGE; switch to native NQP-rx regex functionality everywhere
* Add 02-load-all.t to unify testing that all required modules load properly
* Add add-tests hacking doc
* Fix minor Win32 bug
* Create blog on blogs.perl.org, start posting there
WIP:
* Tickets in Parrot Trac for Plumage testing low hanging fruit
* TT #1389, t/03-util.t
* More to come when I'm unblocked again
MAD PROPZ:
* tene++ # Fat arrow (=>) syntax for NQP-rx 17:08
* cotto++ # Work on TT #1378
* lithos++ # Pimping Plumage developments in the weekly summary
* darbelo++ # Metadata updates, Makefile fixes, Getopt::Obj rename tracking
* fperrad++ # Metadata updates, testing
* dukeleto++ # Tapir replacement for Plumage's old test harness
NEXT UP:
* Not much this week, unless I get lucky.
BLOCKERS:
* $DAY_JOB, @holidays, @visitors
* Probably blocked for at least another week. :-(
EOR
17:42 mikehh joined 17:52 davidfetter joined
Util Extra $WORK this week. All Parrot work delayed 1 week. Cannot attend meeting today. Expect tuits to return this weekend. EOR 18:21
mikehh What I did since my last report: 18:26
* building and testing parrot on amd64/i386, with gcc/g++, with and without --optimize
* some branch testing
* some fixes
* working through the gc book
What I intend to do in the next week:
* testing and fixing 18:27
* investigate gc some more
.eor
anyone here? 18:37
davidfetter no 19:10
19:13 plobsing_ joined
plobsing_ sorry I'm late 19:14
What I Did:
* work on pmc_freeze_cleanup (fixed segfaults, more cleanups)
What I Plan:
* more of the same
* eliminate EXTRA_IS_PROP_HASH and visit_info.extra_flags
* fix segfaultiness in t/pmc/eval.t
EOR 19:15
mikehh you are not late - nothing has happened yet 19:16
19:28 Coke joined
Coke Done: (parrot) 19:38
- Makefile cleanup - trunk - break and reconstruct build deps using tools/dev/checkdepend.pl
- branches/one_make - working on eliminating recursive make
- branches/one_make - working on splitting out single makefile into
includes.
- PIRC
- branches/pirc_config - add a --pirc option to select between
pirc/imcc; it gets as far as #defining a variable that someone else
could use to determine which imcc to build with. Any takers?
- Ticket cleanup.
Closed TT#'s: 1343, 231, 982, 307, 1277
Opened TT#'s: 1398, 1387
Primary focus right now is makefile cleanup:
* eliminate recursive make, but also:
* split up makefile into more managable chunks
* improve and validate dependencies
* move build steps out of config/ and into the make process as possible
.
Also, JimmyZ sent in a CLA; did he get 2 votes and a mentor? 19:41
mikehh JimmyZ's had some interesting patches, although I think they needed to be broken down a bit (a bit too much in a given patch) 19:48
I am quite happy to be a secondary mentor - not nthe primary one though 19:49
Coke mikehh: I'd be happy to hear build feedback on the one_make branch; I am pretty sure there are build deps missing at the moment. 19:55
mikehh Coke: I ran some tests on it - make -j corevm has a couple of problems - looking at it 19:59
21:29 plobsing_ left