| samcv | does that work for you guys? | 00:00 | |
| ugh starting to drive myself crazy with trying to debug collation | 01:37 | ||
|
01:52
ilbot3 joined
|
|||
| Geth | MoarVM/collation-arrays: 79 commits pushed by (Daniel Green)++, (Jonathan Worthington)++, (Samantha McVey)++, (Stefan Seifert)++, MasterDuke17++, (Steve Mynott)++, lizmat++, (Lucas Buchala)++, (Timo Paulssen)++ review: github.com/MoarVM/MoarVM/compare/2...e9ef652eea |
02:34 | |
|
07:00
vendethiel joined
07:38
evalable6 joined,
committable6 joined,
quotable6 joined
07:39
unicodable6 joined,
bisectable6 joined,
bloatable6 joined,
greppable6 joined,
benchable6 joined,
statisfiable6 joined
07:41
coverable6 joined
08:08
domidumont joined
08:13
domidumont joined
09:16
nebuchadnezzar joined
10:30
lizmat joined
10:56
lizmat_ joined
13:29
zakharyas joined
14:58
zakharyas joined
16:06
zakharyas joined
17:03
zakharyas joined
17:16
domidumont joined
|
|||
| MasterDuke | is there a potential bug here? github.com/MoarVM/MoarVM/blob/mast...#L426-L432 | 18:23 | |
| i is an MVMint32, but an MVMString's num_graphs is an MVMuint32 | 18:24 | ||
| geekosaur | there's so many such mismatches in the string/char code in moarvm that I have to assume it's de facto safe | 18:34 | |
| you mostly can't trigger the edge conditions anyway :) | |||
| MasterDuke | i think `perl6 -e 'my $a = "a" x 2**32-1; say $a.chars; my $b = $a.substr(1); say $b.chars'` will trigger it, but it needs 17gb ram (hence my question in #perl6) | 18:39 | |
| samcv | MasterDuke, ideally we will use collapse_strands() instead of repeating the word inside all the different functions that need to collapse strands | 20:20 | |
| so looks like that is a place we can remove that code and just call collapse_strands() | 20:21 | ||
| Geth | MoarVM/collation-arrays: 201720a973 | (Samantha McVey)++ | src/strings/unicode_ops.c Generate collation values for Tangut, Nushu, and Unified_Ideograph's Add one to each of the collation values so each of them match Update some comments and fix a (maybe) bug in value separators |
21:19 | |
| samcv | MasterDuke, i'm working on making it so that and collapse_strands can both share the same loop | 21:55 | |
| since all they do is iterate a grapheme iterator until the end. even though one is copying the whole string and one a substring | 21:56 | ||
| Geth | MoarVM: 1e92fc9689 | (Samantha McVey)++ | src/strings/ops.c Refactor code into iterate_gi_into_string() to reduce code duplication Move the code from collapse_strands() into iterate_gi_into_string(). This allows us to use it in MVM_string_substring() and both reduce code duplication and also use more optimized code. |
22:27 | |
|
22:42
travis-ci joined
|
|||
| travis-ci | MoarVM build failed. Samantha McVey 'Refactor code into iterate_gi_into_string() to reduce code duplication | 22:42 | |
| travis-ci.org/MoarVM/MoarVM/builds/256696927 github.com/MoarVM/MoarVM/compare/0...92fc9689ab | |||
|
22:42
travis-ci left
|
|||
| samcv | oops i broke everything | 22:42 | |
| Geth | MoarVM: 1c23d9b703 | (Samantha McVey)++ | src/strings/ops.c Move collapse_strands() after iterate_gi_into_string() |
22:43 | |
|
23:05
travis-ci joined
|
|||
| travis-ci | MoarVM build passed. Samantha McVey 'Move collapse_strands() after iterate_gi_into_string()' | 23:05 | |
| travis-ci.org/MoarVM/MoarVM/builds/256699573 github.com/MoarVM/MoarVM/compare/1...23d9b70342 | |||
|
23:05
travis-ci left
|
|||
| samcv | \o/ now passing tests 1 through 18888 with my collation_arrays branch | 23:27 | |
| (unicode collation test) | |||
| ok out of all the tests (190,412 total tests) failing only 699 now \o/ | 23:30 | ||