04:43 Ven` joined 04:48 Ven` left 08:17 AlexDaniel left 08:32 Ven` joined
Ven` it might sound crazy for a moment, but I have something to muse about... signature binders are like EXPORT subroutines 08:46
you take a bunch of "stuff" *handwaves* and return a list of variables to be installed in the scope it's been invoked 08:47
The only difference is that the values are supplied instead of being in the EXPORT's "parent scope".
Which variables pop into existance and what they contain is the responsability of the EXPORT subroutine. 08:48
masak yeah, that connection sounds crazy to me :) 09:03
Ven` oh, fair enough :D.
masak I'm not sure after writing that issue that a signature binder would be "worth it" 09:04
Ven` It's not like I actually mean "we should implement it in these terms".
masak in the sense that (a) we only have a finite, limited set of ways to extend the signature binder, namely those that are already in the issue queue, (b) I have a hard time imagining more, (c) the way the signature binder will have to be set up to be extended already presupposes a lot of knowledge about the way we extend it 09:06
Ven` I've always felt like the signature binder in Perl 6 is not on the right peg (stealing that saying).
masak ok. I don't know which peg it's hanging on, actually 09:07
Ven` Well, then maybe I'm mis-stealing the saying :D. The idea, anyway, is that this is a very general pattern-matching need, and you have to use (or abuse) signatures in Perl 6 to get that behavior.
Though my last experiments of using a signature in a given/when were a few years back by now, I seem to remember it not doing all I had hoped it would. 09:08
masak aha
well, it did occur to me that destructuring in variable declarations and in parameter declarations are very similar, to the extent that it'd make sense to have just one single code path 09:09
Ven` Yup. And it's very much not in Perl 6.
masak right
I'm having the horrible realization that there's years of work here 09:10
.oO( Damn you, Damian, for only having 10 years worth of foresight and not 15! )
masak which, don't get me wrong, would be lots of fun
but the fact that it will take a long time and mostly detract from macros-in-Perl-6...
...makes me want to put it on the backburner a little 09:11
probably more of a concern for Alma than 007
Ven` Alm-wha?
ah, ok
masak github.com/masak/007/issues/277
github.com/masak/alma 09:12
Ven` usually refuses to write Java :)
masak I know, Java has that effect on people 09:13
not me, though 09:14
Ven` oh yeah, for sure 09:15
masak I mean, I'm well aware of Java's (many) weaknesses. but it's also a capable language in many ways, and I don't see that acknowledged as much
Ven` I was of that opinion too, and then schooled has forced me to write hibernate+spring+... java on Spring Tool Suite for months, and now I refuse to even acknowledge that. 09:16
masak yes, understood 09:17
oh, and also, good case statements will need destructuring too, so those would also need to be considered as a client 10:45
10:51 Ven` left 12:40 Ven` joined
Ven` yeah I need to chek how much Perl 6 allow there when you when a :() sig 12:41
12:51 Ven` left 12:53 Ven` joined 18:51 Ven` left 20:54 AlexDaniel joined