|
08:35
Ven` joined
09:11
Ven` left
09:20
Ven` joined
|
|||
| Ven` | Oh, #328 got merged. I remember we talked about that. Very interesting. | 10:54 | |
| do we have default arguments? | |||
| Ven` takes a look | 10:56 | ||
| Mh, no. I was interested if `func _(infix:<+>, y = 3 + 4) { }` worked :P. | 10:59 | ||
|
11:15
Ven` left
|
|||
| masak | :) | 11:16 | |
| parameter defaults are in an issue as... not exactly a macro, but a language extension of sorts | |||
| I have a 007 issue queued up but not yet written about an extensible signature binder. been having some new insights about that lately. | 11:17 | ||
| with your example, I would definitely expect it to work, as a consequence of parameter defaults causing scopes to be "layered like an onion", like in ES6 | 11:18 | ||
| (my thoughts lately about a signature binder: each user-defined routine gets the signature binder that was lexically active at the routine's definition. when a routine value is invoked, the signature binder enters into a "negotiation" with the arguments list, leading eventually to the routine being entered (or a binding error) | 11:24 | ||
| just like we have a "current active parser", we have a "current active signature binder", that can be extended in a scope and then discarded when the scope is exited) | 11:25 | ||
|
12:09
Ven` joined
|
|||
| masak | Ven`: there's some backlog :) | 12:33 | |
| Ven` | oh, right. I really need to get myself a vps or something, this is getting... | ||
| masak | I feel this summer is going to be special. I'm going to spend quite a bit of effort taking 007 to where I want it to be. | ||
| Ven` | I'm still wondering... I was thinking, maybe I'll try and tackle the 007-regexps-use-perl6-regexps-underneath. | 12:34 | |
| masak | specifically, I know the stavro-parser effort will give us `is parsed`, which will be a watershed moment for 007 | ||
| Ven` | As an alternative to the new parser. | ||
| masak | feel free | ||
| I'm not opposed to that at all, I just want us to be in control, that's the important part | |||
| so if we do it with Perl 6 regexes, they'd have to be constantly prepared to hand back control to 007 | 12:35 | ||
| Ven` | No no I understand. It just seems like a good temporary solution. It's been 3 years and... I'd feel warm&fuzzy if we got is parsed :). | 12:36 | |
| Feels a bit like touching the promised land, if only for a fleeting moment. Well, anyway. | |||
| masak | yes. | ||
| we have the same end goal, just different "good temporary solution" ideas ;) | 12:37 | ||
| which, again, is fine | |||
| let a thousand parsers bloom | |||
| Ven` | I think your solution is the long-term one. | ||
| masak | if your solution ends up being faster, I will happily accept it as the long-term one | ||
| Ven` | Well, you know what we end up talking about often with regards to cheating and getting caught... Our own parser means we can always "cheat further" (or not cheat anymore). We'll definitely get caught at some point with the "regex rewriter"/"regex compiler"/"regex transpiler". | 12:40 | |
| masak | :) | 12:41 | |
| I notice I'm putting off working on the stavro-parser branch. mostly because it's a big undertaking, and I need to fit all of it in my brain at the same time | |||
| on the bright side, I've done more minor issues and cage-cleaning the past few weeks than in the previous entire year ;) | 12:43 | ||
| Ven` | Yeah, I've been "slacking" as well. I tried several times to implement is parsed by myself and hit walls everytime and I've felt a bit disheartened with it for sure. | 12:45 | |
| masak | `is parsed` is hard to implement with what we have now | 12:46 | |
| the stavro-parser branch is meant to expose the parts that `is parsed` needs so that it becomes trivial | 12:47 | ||
|
13:21
Ven` left
13:22
Ven` joined
14:21
Ven` left
14:30
Ven` joined
|
|||
| masak | I was looking into github.com/masak/007/issues/297 yesterday and how to fix that | 14:42 | |
| I realized that I need to lift up the special infixes (= && || //) and all the postfixes to be as close to *actual* built-in macros | 14:43 | ||
| and if I do that first, then the whole Q::Infix::* hierarchy becomes vestigial and I can just remove it | |||
| also, I think I have a way to remove &.hook completely | |||
| Ven` | \o/ | 15:02 | |
| masak | I hope the stavro-parser work also leads us to being able to (finally) close #30, which (as the OP states) HN requested almost 3 years ago | 15:04 | |
| besides, you know, the fact that unquotes are a pretty central part of the 007 story in general | |||
| Ven` | Oh, yeah -- "We happily oblige". Famous last words :P. | 15:05 | |
| masak | ugh, yes | 15:06 | |
| more like "we happily run headfirst into conceptual problems" | |||
| Ven` | .oO( Joke's on you, we were gonna do that anyway! ) |
15:20 | |
|
16:46
Ven` left
|
|||
| masak | yes, totally | 17:07 | |
|
18:21
Ven`` joined
18:46
ilogger2 joined
20:43
Ven`` left,
Ven`` joined
20:56
Ven`` left
|
|||