| timo | i think i misread the code and it's fine the way it is, oops | 08:14 | |
| bartolin_ | hi timo, I wanted to say thousand thanks for github.com/rakudo/rakudo/commit/81d43af66e. Compiling rakudo-j became so incredible slow without that. | 08:17 | |
| timo | YW! i totally lucked into that one :) | 08:19 | |
|
09:44
sena_kun joined
|
|||
| Geth | rakudo/main: 0db7149f4f | (Elizabeth Mattijsen)++ | src/core.c/OperatorProperties.rakumod Fix docu typos |
12:34 | |
| rakudo/main: 485474df40 | (Elizabeth Mattijsen)++ | 2 files RakuAST: fix handling of markup in =alias As described in github.com/Raku/RakuDoc-GAMMA/issues/53 The fix is a bit of a hack: take the contents of the =alias specification minus the lemma, wrap that in a =begin fake / =end fake block and run .AST on it. Then remove the outer block and process ... (6 more lines) |
12:39 | ||
| rakudo: usev6++ created pull request #5691: Restore old logic in find_best_dispatchee |
13:17 | ||
| rakudo/main: 65a98ce22a | (Christian Bartolomäus)++ (committed using GitHub Web editor) | src/Perl6/bootstrap.c/BOOTSTRAP.nqp Restore old logic in find_best_dispatchee (#5691) The refactoring from 70b8b229fa changed the logic to detect a mismatch between the expected type and the passed in value when doing the type check for native parameters. But if I'm not mistaken the new version got some cases wrong: ... (10 more lines) |
13:59 | ||
| timo | would love to see tests that systematically try to cover this space, but honestly that sounds difficult | 14:16 | |
| bartolin_ | yeah, that would be nice. I tried to come up with some simple examples that would show the changed behaviour, but I failed. And even the problem with variable interpolation (github.com/rakudo/rakudo/issues/5688) which seemed to hit this code did only show up on the JVM backend. | 14:31 | |
| timo | are you already aware of the coverage log that moar offers? could help figure out what's going on on the moar side, even if not on the jvm side | 15:06 | |
| bartolin_ | no, I have to admit that I'm not aware of that. Do you have a pointer ready? | 15:08 | |
| timo | there's two env vars, one sets the filename and the other the mode, where you can decide if you want lines to only report being hit once, or every time (so kinda like a tracing mode) | 15:21 | |
| and there's also nqp::coveragecontrol to switch modes while it's running | |||
| comma also has support for coverage built in, integrated with the test running stuff | 15:22 | ||
| bartolin_ | thanks, will have a look into that | 15:30 | |
| Geth | rakudo/fix-negated-junctions: bdf1e2bbd3 | ab5tract++ | src/Perl6/Optimizer.nqp Scope to symmetric comparison ops |
16:16 | |
| ab5tract | `ne` was being a total jerk, so that's still NYI | 16:17 | |
| timo | desu ne | 16:28 | |
| ab5tract learns some Japanese | 16:40 | ||
| Geth | nqp/main: 4651b898bf | (Christian Bartolomäus)++ (committed using GitHub Web editor) | 4 files [JVM] Deserialize VMNull as such The old deserialization code gave back a Java null when it encountered a VMNull. This led to problems when compiling Rakudo's settings: Class attributes that were initialized to nqp::null ended up as NQPMu. Fixes github.com/Raku/nqp/issues/828. |
17:23 | |
| rakudo/main: ba84cd998d | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION Bump NQP to get bartolin's JVM fixes |
17:33 | ||
| rakudo/main: 3f0f974e4b | (Elizabeth Mattijsen)++ | src/Raku/Grammar.nqp RakuAST: allow * is =config identifier |
17:36 | ||
|
18:29
codesections joined
18:40
librasteve_ left
18:56
codesections left,
codesections1 joined
18:58
codesections1 is now known as codesections
19:21
codesections left
19:22
librasteve_ joined,
codesections joined,
codesections left
19:23
codesections joined
19:28
codesections left
|
|||
| lizmat detects an off-by-one | 19:43 | ||
| m: dd "foo doo".rindex(" ",7) | 19:44 | ||
| camelia | index start offset (7) out of range (0..7) in block <unit> at <tmp> line 1 |
||
| Geth | rakudo/fail-gracefully-outside-of-supported-terminal-emulator: 3c01c08aab | ab5tract++ | 5 files Fail gracefully when REPL is initialized without TTY This patch was originally intended to address R#4378 (#4378)[^1], but that problem is actually unresolvable as long as starting `raku` outside of a TTY automatically engages the wait-until-EOF mode (the same behavior as calling `perl` without `-e`). ... (11 more lines) |
22:04 | |
| linkable6 | R#4378 [open]: github.com/rakudo/rakudo/issues/4378 [Windows] $*IN.t is False on Windows, causing issues with the REPL | ||
| nqp/add-is-interactive: 2e8924f9e8 | ab5tract++ | src/HLL/Compiler.nqp Add is-interactive to HLL::Compiler This commit nukes the hard-to-grok distinction of `interactive` vs `non-interactive` repl-modes and replaces it with the following options: - `tty` ... (21 more lines) |
22:14 | ||
|
22:30
sena_kun left
|
|||
| Geth | rakudo/fix-negated-junctions: 808e5e48d0 | ab5tract++ | 3 files Use OperatorProperties |
23:54 | |
| rakudo/fix-negated-junctions: 5f2b8c1550 | ab5tract++ | 3 files Use OperatorProperties |
23:56 | ||
rakudo/fix-negated-junctions: 9 commits pushed by (Elizabeth Mattijsen)++, (Christian Bartolomäus)++, ab5tract++
|
23:57 | ||