Update predicates with improvements from Tim Dysinger

[oubiwann]

@dysinger created some nice implementations of various functions and macros in his lfesl project (so far only his thrushing macros have been copied over). In particular, I think I've seen a couple improvements there over what I've done for one or two predicates. That should be reviewed more carefully and then updated with his code.

See https://github.com/dysinger/lfesl/tree/master/include

[oubiwann]

I think this is the one I was thinking of:

(defun dict?
  (x) (-> (is_tuple x)
          (andalso (=:= 'dict (element 1 x)))))

Tim does an extra check there that lutil doesn't.

[arpunk]

Hi @oubiwann,

I was not aware of lfesl repository but now that I look at it it seems most of the lutil code could be backported into lfesl and start working on it to experiment on what "lfe core" experience would look like. If it is the other way around then please let me know, but it looks lfesl had already some work being done in.

What are your thoughts about it?

[oubiwann]

Well, there's more stuff in lutil and it uses the latest forms and syntax in LFE. I don't really want 100% of the Clojurisms in lutil; I've got most of what I liked in lfesl.

Futhermore, lutil is mostly oriented towards offering a set of modules as a library. The includes are fairly recent changes, allowing one to treat them like language extensions. We talked about moving in the direction of a standard library for LFE at the first LFE "language summit" in Stockholm, and @rvirding was very encouraging about experimenting and brining the good stuff that was generally useful into a stdlib for LFE. As such, there's a chance that some of this will make it into LFE at some point.

So for all of these reasons, I don't see a need (or any benefit, really), in forking Tim's repo and re-creating lutil with it. I will definitely be creating more macros as LFE extensions in lutil, though. And that will probably increase greatly once reader macros land ;-)

[oubiwann]

Moved to: lfe-deprecated/clj#3