releasable6 | Next release in ≈2 days and ≈11 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft | 07:00 | |
patrickb | timo: My feeling has always been that there has to be some place where all the meta fields are nicely specced / documented. But up until now I did not find that place. Sounds a bit like El Dorado... | 08:32 | |
09:39
gfldex left
09:40
gfldex joined
09:58
sena_kun joined
11:16
lizmat left,
lizmat_ left,
lizmat joined
11:26
Altai-man joined
11:29
sena_kun left
12:01
Geth joined
|
|||
Geth | nqp/main: eac3ca0130 | (Elizabeth Mattijsen)++ | tools/templates/MOAR_REVISION Bump MoarVM to get 64bit native opts, MasterDuke++ |
12:02 | |
rakudo/main: 61fa0c85a6 | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION Bump NQP to get 64bit native opts, MasterDuke++ |
|||
nine | Anyone know how the LTM implementation actually works? | 12:03 | |
lizmat | I think at this point, there's no active memory of that with the active Raku core devs | 12:04 | |
I hope to be mistaken | |||
timo | i know a little bit about longlit in the NFA | 12:08 | |
well, LTM is really about the whole NFA, longlit is just to prefer branches that are literal matches over branches that match based on character class or similar | |||
12:26
Guest93 joined
|
|||
Guest93 | nine: 6guts.wordpress.com/2012/06/07/ltm...ernations/ | 12:26 | |
12:44
Guest93 left
|
|||
Geth | rakudo/main: 5807369f78 | (Stefan Seifert)++ | src/Raku/ast/regex.rakumod RakuAST: store regex NFAs in their code object This is required for LTM to do its thing. |
12:56 | |
nine | My that was a hard find | ||
1184 now | 12:57 | ||
lizmat | hhmmm t/spec/S02-types/multi_dimensional_array.t failed in the run for me, but succeeds when run alone: so a flapper :-( | 13:07 | |
nine | Failed in what way? | 13:11 | |
lizmat | no idea, it just showed in the diff with a - | 13:12 | |
this is a general beef I have with "make spectest", that I may look at soo | |||
ner rather than later | |||
nine | lizmat: you may be interested in my test-log patch that's in use when building rakudo rpm packages: build.opensuse.org/projects/devel:...f?expand=1 | 13:18 | |
It stores the full test output in separate files on disk and when make test fails dumps that to the console, so I can look up the actual error in the build log | 13:19 | ||
lizmat | that's what actually inspired me to create my own "run-tests" script in the modules that I maintain :-) | ||
so we agree we need something like this for "make spectest" and "make stresstest" and "make test"... right ? | 13:20 | ||
nine | sure | 13:27 | |
15:15
librasteve_ joined
|
|||
timo | yeah | 17:18 | |
ugexe | we already do that for CI via test harnesses dont we? | 17:40 | |
timo | we spit out ... i think junitxml? with a plugin for TAP::Harness or something in that space | 17:56 | |
21:15
Altai-man left
21:16
Altai-man joined
|
|||
Geth | rakudo/main: 8b60dc2608 | (Stefan Seifert)++ | src/Raku/Grammar.nqp RakuAST: reinstate parsing of numeric args to internal regex modifiers Fixes regression caused by commit 7fcbfd657cdbde316c1fced6c0947c83f83f9153 |
21:41 | |
nine | 1183 | ||
Geth | rakudo/main: 0bfbf5fa8b | (Stefan Seifert)++ | src/Raku/ast/regex.rakumod Revert "Don't allow \c[] with multiple graphemes in regex" This reverts commit ccccfb7240a5228d17c8ec4c0cfa46635a6f04c6. The change broke a lot of spec tests. |
21:50 | |
22:44
Altai-man left
|
|||
lizmat | wow, 2 years ago ? | 23:08 | |
nine: did we lose t/spec/S05-modifier/ratchet ? | 23:39 | ||
and t/12-rakuast/regex.rakutest ? | 23:40 | ||
afk& |