Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2017-04-30

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

All times shown according to UTC.

Time Nick Message
00:15 geekosaur joined #moarvm
01:19 ggoebel joined #moarvm
02:24 MasterDuke_ timotimo: TUI?
05:17 samcv guys what's the file open limit for MVM?
05:17 samcv is it mostly just the os?
08:21 AlexDani` joined #moarvm
09:50 timotimo terminal UI
09:50 timotimo samcv: yeah, ulimit tells you what it is
09:50 timotimo i'm not sure what the maximum maximum is
09:50 samcv yeah got it figured out
09:51 samcv it was wayyyyyyy lower than the system limit
09:51 samcv was like 1600
09:51 timotimo it's usually 1024 i believe
09:51 samcv maybe was that
09:51 samcv sounds abotu right
13:15 Geth ¦ MoarVM: pmurias++ created pull request #588: Coerce s n
13:15 Geth ¦ MoarVM: review: https://github.com/MoarVM/MoarVM/pull/588
14:20 zakharyas joined #moarvm
15:29 stmuk joined #moarvm
15:37 stmuk_ joined #moarvm
15:58 stmuk joined #moarvm
16:07 stmuk joined #moarvm
16:16 stmuk_ joined #moarvm
16:21 stmuk joined #moarvm
16:31 stmuk_ joined #moarvm
16:34 zakharyas1 joined #moarvm
16:34 stmuk joined #moarvm
16:35 zakharyas1 joined #moarvm
16:37 stmuk__ joined #moarvm
16:39 stmuk_ joined #moarvm
17:16 domidumont joined #moarvm
17:22 domidumont joined #moarvm
18:07 timotimo i think in general you don't want to try to parse something like the spdx grammar with perl6 grammars
19:07 zakharyas joined #moarvm
19:13 harrow joined #moarvm
20:28 stmuk joined #moarvm
21:09 samcv yeah idk lol. uguxe wanted to do it |:
21:09 samcv it seems hard. but. i think it must be possible
21:09 samcv timotimo, cause you can have parenthesis and precedence
21:10 samcv (MIT AND GPL-1.0) OR BSD-2-Clause
21:10 samcv for example
21:10 samcv arbitrary parens and stuff
21:12 stmuk_ joined #moarvm
21:16 stmuk joined #moarvm
21:24 stmuk_ joined #moarvm
21:26 timotimo well, parents are okay for recursive descent
21:26 yoleaux 20:24Z <japhb> timotimo: I noticed yesterday you were discussing raw and processed keyboard input in terminal UIs.  See https://github.com/ab5tract/Terminal-Print/blob/raw-input/lib/Terminal/Print/RawInput.pm6 for early work on this, with https://github.com/ab5tract/Terminal-Print/blob/raw-input/examples/tris.p6 for a test program.  Last I tried it works fine right up until it stops accepting keyboard input; see https://github.com/ab5tract/
21:26 yoleaux Terminal-Print/issues/19 for the sa
21:26 timotimo but other than that you have an operator precedence parser really
21:26 stmuk joined #moarvm
21:26 timotimo that should totally have a module in the ecosystem
21:26 timotimo operator precedence parsers in general
21:31 samcv yeah
21:31 stmuk__ joined #moarvm
21:39 samcv timotimo, i think (haven't done precedence yet. more important for it to parse)
21:39 samcv but i think i maybe have it working the best i've had it
21:40 timotimo i've been able to trace it btw
21:40 timotimo it's "succeeding" at parsing THIS as a or-thingie over and over and over again
21:41 timotimo i wonder if it'd help if you make it ** 2..* instead of just +
21:41 timotimo because it isn't supposed to think "THIS" is an or-statement
21:42 samcv ok pushing latest changes
21:42 samcv https://github.com/samcv/SPDX-Parser/blob/next/lib/SPDX/Parser.pm6 it's not dying anymore since i moved paren-expression into multiple syms
21:42 samcv so compound expression can be either one with parens or not
21:44 samcv it only fails parsing one of my things in the test file. not checked accuracy yet
21:44 timotimo we have only WITH, AND, and OR?
21:44 samcv yes
21:44 samcv and parens
21:47 timotimo right
21:47 timotimo and WITH is tightest, then AND, then OR, and parens do whatever you want
21:47 samcv yeah
21:48 timotimo that's not even complex enough for the shunting yard algorithm
21:48 samcv this is not that bad a parse https://gist.github.com/samcv/cdc10399a866be0b329fab4451ddf07b
21:49 samcv i will look into that
21:50 samcv i think it'll work for simple ones at least
21:50 samcv going to try some things
21:51 timotimo cool, you made it works
21:55 samcv what i'm thinking of. is storing the licenses in an array. where each element of the array is another array
21:56 timotimo like a tree
21:57 samcv so '(MIT AND FOO) OR GPL' would be ['MIT', 'FOO'], ['GPL']
21:57 samcv so you need to accept all of the licenses from one of those two, so you can have MIT AND FOO or GPL
21:58 samcv and then we can check to make sure that at least one element matches
21:58 samcv this may be a little simplistic but it should work for all the cases i can think of right now
21:58 timotimo but what if you have (MIT OR FOO) AND GPL?
21:59 samcv then you have ['GPL', 'FOO'] ['GPL', 'MIT']
21:59 timotimo oooh
21:59 timotimo so you "multiply" it out
21:59 samcv yeah
22:00 timotimo and when you have a "WITH" it'll be part of the string itself?
22:00 samcv yeah with applies to that one license
22:00 samcv it's basically a license variation
22:02 samcv my $match = 0; for ^@list.elems { $match++ if @list[$_].all ne any(@blacklist) }
22:03 timotimo (MIT OR ARTISTIC-2.0 OR GPL-2 OR GPL-3 OR AGPL) AND (X11 OR FREEBSD OR NASA) AND (ARTISTIC-1.0 OR ME OR YOU)
22:03 samcv that's a lot of licenses
22:04 samcv timotimo, see this https://github.com/Juerd/p6-mqtt/issues/3
22:04 timotimo yikes!
22:04 samcv though tbh some of those identifiers are deprecated it found
22:04 samcv and are replaced with FOO WITH exceptionname
22:04 timotimo interesting
22:04 samcv but regardless
22:05 samcv was an interesting experiment
22:05 timotimo yeah i agree
22:27 brrt joined #moarvm
22:28 brrt http://brrt-to-the-future.blogspot.nl/2017/04/letting-templates-do-what-you-mean.html
22:30 timotimo oh!
23:24 geekosaur joined #moarvm
23:29 geekosaur joined #moarvm
23:42 Geth joined #moarvm
23:51 Geth joined #moarvm

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