Perl 6 macros -- about github.com/masak/alma/ and how macros fit into Perl 6 -- irclog: colabti.org/irclogger/irclogger_log...rl6-macros
Set by masak on 26 August 2019.
00:38 lucasb left
masak will do :) 11:08
masak prefers to have No Known Bugs in the language... but not at the expense of refusing to reproduce them 11:09
12:52 Ven_de_Thiel joined
Ven_de_Thiel
.oO( I removed the runner script for my language, and since then, it has received exactly 0 bug reports )
12:52
13:42 lucasb joined
masak Ven_de_Thiel: converting to variables works: github.com/masak/alma/commit/21556...e57fb7?w=1 14:40
(at a guess, maybe you forgot to make it `BEGIN my`?)
Ven_de_Thiel yay! I guess it was because that was prior to my work on findAllAttributes() :)
oh, maybe I did, since it was used by subs... but of course the subs are invoked by a macro :) 14:41
masak right 14:42
the timing is what matters
I will keep massaging that code; it's fun 14:43
Ven_de_Thiel ...I'm glad :-) 14:47
3 things I was gonna do 1) API for inheritList 2) API for attrList 3) `rootType.copyWithAttributes(root, ...);` 14:50
masak the API things I would be much happier doing after the object system refactor 15:00
but I guess we can shove them in before 15:01
Qtypes are long overdue for some API methods
Ven_de_Thiel yay :) 15:39
masak hm, that branch triggers a MoarVM bug: travis-ci.org/masak/alma/builds/585625456 15:43
can't reproduce it locally, though
Ven_de_Thiel that's a lot of bytes 15:47
can you re-run the tests? I want to know if it's flapping 15:48
masak ok 15:50
I dislike the term "flapping". it makes light of essentially a real problem: an unreliable test. 15:51
when tests are unreliable, they decrease confidence and ultimately velocity/value 15:52
note though that the branch has been failing for a few commits already -- that's also data, I guess
Ven_de_Thiel I'm just cargo-culting the term 15:55
masak yeah, I know. not your fault :) and I'm just tilting at another windmill 16:01
well, it failed again. fails fairly reliably for a "flapper" 16:02
masak .oO( reliably unreliable )
Ven_de_Thiel hahaha 16:08
masak logic puzzles like "if you try to fail, and succeed, what have you really done?" kinda rub me the wrong way. someone can't keep track of their semantic levels, that's all. nothing to write home about 16:11
Ven_de_Thiel
.oO( we need more parentheses. Also lisp macro phases )
16:13
masak re "# XXX maybe we need `append` in Alma?" -- we could go the Perl 6 and JavaScript route and make .push variadic 17:20
I'm just a little bit concerned that it'd be one of those abstractions that aren't entirely free 17:21
interestingly, what you're really using `append` for here is a `gather`... :) 17:24
github.com/masak/alma/commit/7f3d9...6ffd41350f 17:30
:)
lucasb Wouldn't be nice to have Geth announce Alma commits here? :) 18:24
20:54 Ven`` joined
Ven`` masak: that's very cute :) 20:55
21:22 Ven`` left