Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2017-11-20

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

All times shown according to UTC.

Time Nick Message
00:56 bisectable6 joined #moarvm
00:56 committable6 joined #moarvm
00:56 evalable6 joined #moarvm
00:57 unicodable6 joined #moarvm
00:57 releasable6 joined #moarvm
02:02 ilbot3 joined #moarvm
02:02 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
02:55 ilbot3 joined #moarvm
02:55 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
03:48 samcv joined #moarvm
06:16 ilbot3 joined #moarvm
06:16 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
06:20 brrt joined #moarvm
06:22 ugexe joined #moarvm
06:27 brrt good * #moarvm
06:31 ugexe joined #moarvm
07:05 domidumont joined #moarvm
07:10 domidumont joined #moarvm
07:37 samcv good * brrt
07:39 brrt ohai samcv
07:39 samcv i probably don't need to include NFG_CHECK changes to the changelog right
07:43 brrt oh, are you rolling out the release now?
07:44 brrt i was wondering whether we should include the labelling-issue-fix as a cherry-pick
07:44 samcv if you think that should be done i'd do it now
07:50 Ven joined #moarvm
07:52 Geth ¦ MoarVM: c0ac87af5f | (Bart Wiegmans)++ | src/jit/label.c
07:52 Geth ¦ MoarVM: Fix labeling off-by-one
07:52 Geth ¦ MoarVM:
07:52 Geth ¦ MoarVM: The last 'object' label would not be counted as a instruction
07:52 Geth ¦ MoarVM: label (and by implication treated as a basic block label). This would
07:52 Geth ¦ MoarVM: cause the required label not to be setup, which would make try/catch
07:52 Geth ¦ MoarVM: on some frames fail.
07:52 Geth ¦ MoarVM: review: https://github.com/MoarVM/MoarVM/commit/c0ac87af5f
08:08 Ven joined #moarvm
08:13 Ven joined #moarvm
08:17 zakharyas joined #moarvm
09:33 robertle joined #moarvm
09:41 Ven joined #moarvm
09:57 Ven joined #moarvm
09:58 Ven joined #moarvm
10:14 Geth_ joined #moarvm
10:55 Ven joined #moarvm
10:58 Ven_ joined #moarvm
11:00 brrt joined #moarvm
11:56 Ven joined #moarvm
12:05 geospeck joined #moarvm
13:46 zakharyas joined #moarvm
14:23 geospeck joined #moarvm
14:57 brrt joined #moarvm
15:11 geospeck joined #moarvm
15:13 geospeck joined #moarvm
15:18 AlexDaniel samcv: how is it going? :)
15:25 zakharyas joined #moarvm
16:01 Geth joined #moarvm
16:59 zakharyas joined #moarvm
17:29 timotimo joined #moarvm
17:36 dogbert17 joined #moarvm
17:42 ilmari joined #moarvm
17:42 ilmari[m] joined #moarvm
17:44 [Coke] joined #moarvm
17:47 domidumont joined #moarvm
18:21 robertle joined #moarvm
18:46 samcv back. good. should release in an hour or so
19:11 lizmat joined #moarvm
19:12 AlexDaniel alright
19:14 samcv AlexDaniel: the extra time was partially me checking some Unicode stuff to make sure no other regressions. i decided that we were good on that front, but I am planning on making some new tests after the release
19:15 AlexDaniel samcv: that's awesome, thanks
19:15 samcv so it seems we need to have at minimum tests for each property value. and i also discovered we had 0 tests for <:LC>
19:15 samcv which is Letter Cased, which combines like Ll Lu and a few other ones. but it does apparently work so that's nice
19:16 samcv i'm thinking we might want a test file that tests both regex and .uniprop as well. since they take different codepaths
19:16 samcv and there have been bugs in one but not the other before
19:17 samcv and we can ensure that the unions like <:L> work for all L* things and LC works for all casey things etc. as well as .uniprop all in one to make sure both are covered and allow us to extend it in the future
19:18 samcv also of note before my changes this last month we didn't return general category "Cn" for any codepoint which is for unassigned. so it's nice that we now respond that
19:18 samcv probably missed it during my grant because no codepoint is explicitly assigned that general category in UnicodeData.txt
19:25 brrt joined #moarvm
20:23 Geth ¦ MoarVM: 56ed28cbbc | (Samantha McVey)++ | docs/ChangeLog
20:23 Geth ¦ MoarVM: Update ChangeLog with commits since last release
20:23 Geth ¦ MoarVM: review: https://github.com/MoarVM/MoarVM/commit/56ed28cbbc
20:29 lizmat samcv++
20:36 samcv feel free for people do do better reorganization or make the ones that haven't been clairified (some of the ones with commit numbers) or better categorize
20:45 AlexDaniel samcv: it's WIP, right? looking at “Unsorted and Unmodified commit text” :)
20:45 AlexDaniel ah-ha, just re-read your last message
20:45 samcv yeah it's not final
20:45 samcv i've already gone through a lot of them looking at the commit message and adding to text and adding which function was modified etc
20:50 AlexDaniel I've been wondering if there's any way to make this work faster/easier, but couldn't come up with any solution
21:01 AlexDaniel maybe one way to do it is by asking committers to log their changes, and then just do some finishing touches
21:01 AlexDaniel but in that case it shouldn't be happening in the repo at all, but on the wiki or something
21:01 AlexDaniel and then it can be committed before the release
21:01 AlexDaniel otherwise it's too much pressure
21:02 AlexDaniel yeah, ok, we'll try that
21:04 AlexDaniel (after the release ofc)
21:40 geospeck joined #moarvm
22:06 timotimo samcv: the first entry in JIT has a surprising "op" in the "middle"
22:08 timotimo i'd write "1LL" instead of "1 LL" for the integer literal fix
22:28 lizmat and another Perl 6 Weekly hits the Net: https://p6weekly.wordpress.com/2017/11/20/2017-47-more-tpcia-videos/
22:40 AlexDaniel lizmat: “Open tickets overview” seems to be pointing nowhere
22:40 lizmat oops
22:40 AlexDaniel lizmat: and that's not related to core development I guess?
22:42 lizmat added link + moved to blog posts   # AlexDaniel++
23:08 brrt joined #moarvm
23:10 jnthn We could just try and encourage people to write really good commit subjects and produce the changelog from git log --oneline :)
23:10 jnthn And anybody who writes "fix bug" gets nominated to do the release :P
23:11 * jnthn wonders why the weekly is adding Rakudo to talk titles that didn't contain it
23:12 jnthn Even my grant name got twisted
23:12 japhb jnthn: Perl 5 added "Pumpking" at the front too.
23:14 japhb Attempt to prioritize clarity over literal quoting?
23:16 ilmari japhb: we didn't, actually. it was just a proposal by mst
23:16 ilmari http://blogs.perl.org/users/matt_s_trout_mst/2013/02/pumpkin-perl---redux.html and linked posts
23:16 japhb ilmari: jnthn and I meant in the Perl 6 Weekly -- talk and blog post titles were subtly mutated
23:18 ilmari japhb: oh, silly me. when you used "perl 5" as a subject in a sentence I thought you meant someone related to perl 5 did something
23:18 japhb ilmari: Apologies for the confusion.  :-)
23:41 lizmat joined #moarvm

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