|
6macros: discussing the finer points of Perl 6 macros, Qtrees, and how to stay sane | irclog: irclog.perlgeek.de/6macros/today Set by moderator on 28 July 2015. |
|||
| masak waves from 6h into the future | 00:47 | ||
| Ven: "colon is over used" -- noted. I want you to know that I'm aware of this :) we'll see where it leads. | 00:52 | ||
| overused* | |||
| basically, the flight allowed me to mull over exactly how much quasis need to late-bind things we've traditionally associated with *parsing*. | 00:53 | ||
| not just macro expansion -- which is blazingly obvious once you start thinking about it | |||
| but things like the exact shape of the AST in an expression involving unquoted operators -- due to them having unknown prec/assoc at parse time | 00:54 | ||
| masak wants to blog about this | |||
| Ven: elixir is very readable: github.com/msaraiva/exsamples/blob...samples.ex -- thanks for sending along that link | 00:58 | ||
| there's something attractive and honest about the quote/unquote keywords :) | |||
| but I suspect TimToady wouldn't like it that `unquote` looked like a sub call | 00:59 | ||
| thought... hm | |||
| if we managed to express unquote as a *macro*... :> | 01:00 | ||
| masak rubs hands evilly | |||
| Ven: quick/rough suggestion to reduce the density of colons: `my count is Int = 0;` `sub add(lhs is Int, rhs is Int) { ... }` `-> s is Str { ... }` `sub add(lhs, rhs) returns Int { ... }` | 01:30 | ||
| and, strangely-consistently, `{{{id is Q::Identifier}}}` | 01:32 | ||
| masak wishes he had a shorter keyword than `returns` for that function | 01:45 | ||
| I guess `of` kind of works... or `give`... or `to`... | 01:48 | ||
|
06:38
Ven joined
|
|||
| Ven uses mostly --> nowadays :) | 06:49 | ||
| masak | I like the colons, so if you're lukewarm to the suggestion, I'll keep them :) | 06:52 | |
| Ven | oh, go ahead. | 07:06 | |
| it's just a "I think I've seen colons somewhere" | 07:07 | ||
| masak | :) | 07:12 | |
| the overloading of colons in both object literals and type annotations bother me a tiny bit | 07:17 | ||
| mostly because one is <general>: <specific> and the other <specific>: <general>, kind of | 07:18 | ||
|
07:28
Ven joined
|
|||
| Ven | right. | 07:36 | |
| masak | which is partly why I suggested `is` instead. | 07:38 | |
| but I'm not necessarily more fond of the `is` proposal than the `:` proposal. | |||
|
07:50
Ven joined
|
|||
| Ven | I'm not either | 07:52 | |
|
09:13
Ven joined
11:34
Ven joined
12:51
Ven_ joined
13:19
Ven joined
13:54
Ven joined
14:13
FROGGS_ joined
14:57
Ven joined
15:03
Ven joined
|
|||
| Ven | masak: I found the paper bracha.org/pluggableTypesPosition.pdf | 16:16 | |
|
17:56
vendethiel joined
|
|||
| vendethiel | masak: so, just to make something "clear" | 18:00 | |
| after reading your blog post | 18:01 | ||
| I don't think languages have experimented with what you're doing | |||
| lisps can "just" use "read" to "splice in" unquotes | 18:02 | ||
| languages like scala have predefined precedence | 18:03 | ||
|
19:00
FROGGS_ joined
19:10
vendethiel joined
19:19
vendethiel joined
19:50
vendethiel joined
21:30
vendethiel joined
|
|||
| vendethiel | I need to finish up my PR, heh | 21:33 | |
|
23:09
Ven_ joined
|
|||