Thank you, that's very helpful.
I'm wondering if you know of a ClojureScript debugging library (something
on the level of xlib for php) that I can use to watch the evaluation of
forms programmatically with breakpoints.
I'm trying to develop a tool that runs a Clojure program by stepping
throu
>
> Are you saying the contrib process is deliberatly made to be difficult for
> the community to contribute to it?
No, not at all, just that it's deliberately designed to be exactly the way
it is, so dedicating a lot of time to trying to change that is likely to be
frustrating and fruitless.
I
> The contrib process is in place because some want it that way - it's very
> deliberately by design and AFAICT unlikely to change.
Are you saying the contrib process is deliberatly made to be difficult for the
community to contribute to it?
If so, maybe if it had more obvious tenets, I find it
On 22/07/17 22:02, Timothy Baldridge wrote:
> Once a transducer is completed it should never be called again.
Thank you, Timothy, this was probably obvious but I don't remember
reading it anywhere so this really helped.
David
--
You received this message because you are subscribed to the Google
Perhaps you want to study the implementation in Medley, those are always
very good quality:
https://github.com/weavejester/medley/blob/254989ed3de83c30ce0101d66c7ce1b6ee257b4d/src/medley/core.cljc#L173
David
--
You received this message because you are subscribed to the Google
Groups "Clojure"
If we think about what we're doing here is a stateful filter, then maybe we
could leverage a few more core Clojure functions:
(defn distinct-by [f coll]
(let [seen (atom #{})]
(filter (fn [itm]
(let [m (f itm)]
(when-not (@seen m)
(swap!
Once a transducer is completed it should never be called again. This is why
transduce takes both a transducer and a reducing function and combines them
internally. The thought here is that it will be harder to shoot yourself in
the foot by reusing a stateful reducing function if you don't have to
c
Hello all,
I found an edge case where stateful transducers in core differ in
whether they reset or clear state in the completion arity
(partition-all) or don’t (take).
Given the two transformed reducing functions
(def conj-partitioning-all-3 ((partition-all 3) conj))
(def conj-taking-3 (
Hi,
Here's a function and a simple use of it. It works, but the code may not be
good Clojure. I'm wondering how it might be better.
(defn distinct-by
"Returns a sequence with no repeats of (f item).
A set is built up internally as the sequence is consumed, so don't use it
on
an infinite
Hi Moe,
Thanks for your suggestions, yes, I've been mostly practicing the first
option suggested. docker-lambda is a good option as well. It does help that
Lambda provides test messages, which is what I've been using to test
locally. Seems like the best way to go about it is to test as much as
refer-clojure doesn't ever remove mappings, it only adds them
The reason a refer-clojure clause in your ns form can prevent bindings is
because your refer-clojure clause (which is likely more qualified than the
default) overrides the args that ns would otherwise provide to
refer-clojure.
On Sat
Apologies, didn't mean to follow that thread!
Phillip Lord writes:
> I am confused about how to exclude clojure.core from a namespace already
> exists. That is, I have not just created the namespace, but it's been
> given to me by a tool.
>
> Consider, for example:
>
>
>
>
> lein repl
> nREPL s
I am confused about how to exclude clojure.core from a namespace already
exists. That is, I have not just created the namespace, but it's been
given to me by a tool.
Consider, for example:
lein repl
nREPL server started on port 41054 on host 127.0.0.1 - nrepl://127.0.0.1:41054
REPL-y 0.3.7, n
>
> I'd much rather see nREPL stay within contrib and the renewed effort, that
> you propose, to go into ironing out kinks in the contrib process
FWIW I don't think this is a realistic option, certainly not for anyone
outside of Clojure core. The contrib process is in place because some want
it t
14 matches
Mail list logo