If the ship has sailed, why, then, the ship has sailed. I only send this
now because I'd already written it before I saw your reply, Alex.
I do not think `any?` is a "gigantic" mistake, though I do think it is
confusing. I expect to see a few some?/any? bugs from beginners. `some?`
and `any? essentially represent two English synonyms, both defined in the
same namespace, each of which represents a check for existence, under two
competing ontological frameworks.
The bandit stopped me, asking, "Do you have Anything?"
"I do," I said, "please don't hurt me."
"If you have Something, hand it over!"
I handed him nothing.
"You said you had Something!"
"No, you asked me if I had Anything!"
"Just hand it over!"
As he was irate and carried a big stick, I gave up and handed him Nothing.
Ultimately the source of confusions such as this arise from the fact that
Clojure is a language with a concrete token (Something) representing
Nothing. some/any just compounds the confusion. But this problem arises in
certain philosophies of ontology, and there already exists a distinction
that strikes me as being quite natural in common English and a good fit for
Clojure and spec: the distinction between something/nothing and
presence/absence.
I would propose `present?` as the name to replace `any?`.
----
But if the ship has sailed, then onward ho! Clojure will still be a great
tool.
~Timothy Dean
On Friday, October 6, 2017 at 5:55:31 PM UTC-6, Alex Miller wrote:
>
> This ship has sailed.
>
> On Oct 6, 2017, at 6:04 PM, Alan Thompson <cloo...@gmail.com <javascript:>>
> wrote:
>
> Before 1.9.0 is officially released, I would like to propose a revisit to
> the semantic mismatch introduced by the *clojure.core/any? *function.
>
> Many, many people are dissatisfied by the choice of *clojure.core/any?*
> to be defined as *(constantly true)*, which is completely in conflict
> with *clojure.core/not-any?* . After all, any logical person would
> automatically assume that:
>
> (= (not-any? args...) (not (any? args...))
>
>
> for any set of legal arguments. This follows the well-established
> tradition in Clojure of having negated pairs such as* if* vs *if-not*,
> *when *vs *when-not*, *every? *vs* not-every?*, etc.
>
> However, I can see that it is convenient to say something like this:
>
> (s/fdef clojure.core/declare
> :args (s/cat :names (s/* simple-symbol?))
> :ret any?)
>
>
> It seems a simple solution to the problem would be to just define some
> keyword specs in place of the globally visible *any? *function. The
> following example shows that we could define *:clojure.spec/pass-all *and
> *:clojure.spec/pass-none* which would could serve as an exact replacement
> for *any? *(& its negative).
>
> (:require [clojure.spec.alpha :as s] ...)
> (deftest demo
> (s/def ::s/pass-all (constantly true))
> (s/def ::s/pass-none (constantly false))
>
> (is (s/valid? ::s/pass-all 5 ))
> (is (s/valid? ::s/pass-all "joe" ))
> (is (s/valid? ::s/pass-all { :blah 42 :blue 66 :hut! 'hut! }))
> (is (not (s/valid? ::s/pass-none 5 ))))
>
>
> Since 1.9.0 is not out yet, is not too late to avoid a permanent pollution
> of the language with a gigantic mistake such as *any?*. At the very
> least, the function could be moved to *clojure.spec/any?* from
> *clojure.core*. If we insist on adding this blatant contradiction to
> *clojure.core*, we won't even have the excuse of a committee to blame it
> on.
>
> Alan Thompson
>
>
>
> On Sun, Nov 13, 2016 at 4:15 PM, Nathan Smutz <nsm...@gmail.com
> <https://mail.google.com/mail/?view=cm&fs=1&tf=1&to=nsm...@gmail.com>>
> wrote:
>
>> Is there a Tricky Names for Nubies page?
>> We might save some stack-overflow searches.
>>
>> In the spirit of Honest Trailers: if we named functions for what they do:
>> or -> first-truthy
>> some -> first-satisfying
>> some? -> not-nil?
>> any? -> return-true
>>
>> Are there others?
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Clojure" group.
>> To post to this group, send email to clojure@googlegroups.com
>> <https://mail.google.com/mail/?view=cm&fs=1&tf=1&to=clojure@googlegroups.com>
>> Note that posts from new members are moderated - please be patient with
>> your first post.
>> To unsubscribe from this group, send email to
>> clojure+unsubscr...@googlegroups.com
>> <https://mail.google.com/mail/?view=cm&fs=1&tf=1&to=clojure%2bunsubscr...@googlegroups.com>
>> For more options, visit this group at
>> http://groups.google.com/group/clojure?hl=en
>> ---
>> You received this message because you are subscribed to the Google Groups
>> "Clojure" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to clojure+unsubscr...@googlegroups.com
>> <https://mail.google.com/mail/?view=cm&fs=1&tf=1&to=clojure%2bunsubscr...@googlegroups.com>
>> .
>> For more options, visit https://groups.google.com/d/optout.
>>
>
> --
> You received this message because you are subscribed to the Google Groups
> "Clojure Dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to clojure-dev...@googlegroups.com <javascript:>.
> To post to this group, send email to cloju...@googlegroups.com
> <javascript:>.
> Visit this group at https://groups.google.com/group/clojure-dev.
> For more options, visit https://groups.google.com/d/optout.
>
>
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
---
You received this message because you are subscribed to the Google Groups
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to clojure+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
