2015-02-14 20:23 GMT+01:00 Steve Miner :
> Clojure doesn't give you direct access to the name of the function you're
> defining. However, you could use a macro to get that. Here’s one way.
> This macro binds the strange symbol %0 to the symbol naming the current
> function.
>
> ;; %0 is bound t
2015-02-14 20:21 GMT+01:00 Jony Hudson :
> Unless I'm mistaken, in the output you show:
>
> Exception ERROR: round [:high|:low|:normal] user/round
> (repl-startup.clj:30)
>
>
> "user/round" is the name of the function, as desired.
>
You are right: I was not looking correctly. :-(
In a way it
You can probably omit try and throw to achieve the same effect.
(let [ste# (aget (.getStackTrace (Exception.)))]
..)
Shantanu
On Sunday, 15 February 2015 00:58:28 UTC+5:30, Shantanu Kumar wrote:
>
> See if you can put this to any use (implies no warranty) - applicable to
> the JVM only:
>
> (
See if you can put this to any use (implies no warranty) - applicable to
the JVM only:
(defmacro whereami
[]
`(try (throw (Exception.))
(catch Exception e#
;; (.printStackTrace e#) ; uncomment this line to inspect stack
trace
(let [ste# (aget (.getStackTrace e#) 0
Clojure doesn't give you direct access to the name of the function you're
defining. However, you could use a macro to get that. Here’s one way. This
macro binds the strange symbol %0 to the symbol naming the current function.
;; %0 is bound to the function's symbolic name within the function
Unless I'm mistaken, in the output you show:
Exception ERROR: round [:high|:low|:normal] user/round
(repl-startup.clj:30)
"user/round" is the name of the function, as desired.
Jony
On Saturday, 14 February 2015 19:09:53 UTC, Cecil Westerhof wrote:
>
> 2015-02-14 20:03 GMT+01:00 Jony Huds
Well, one of the many reasons, that clojure is faster than bash is, that
during compilation, code is divorced from the original source symbols.
That means you can only access information about the original source that
you, or some library code you use, put in there for you.
Jony stated correctly,
2015-02-14 20:03 GMT+01:00 Jony Hudson :
> Ah, I see. I don't know how to do that. But, the function name should be
> in the stack trace associated with the exception. Is there a particular
> reason you also want to put it in the message?
>
Well if I enter in the REPL:
(round :dummy 12.4)
I
Ah, I see. I don't know how to do that. But, the function name should be in
the stack trace associated with the exception. Is there a particular reason
you also want to put it in the message?
Jony
On Saturday, 14 February 2015 18:45:12 UTC, Cecil Westerhof wrote:
>
>
>
> 2015-02-14 18:58 GMT+
2015-02-14 18:58 GMT+01:00 Jony Hudson :
> There might be a neater way, but
>
> (name (:name (meta (var reduce
>
> => "reduce"
>
That is not what I meant. I have the following function:
(defn round
([x] (round :normal x))
([mode x]
(let [fn (case mode
There might be a neater way, but
(name (:name (meta (var reduce
=> "reduce"
Jony
On Saturday, 14 February 2015 16:11:48 UTC, Cecil Westerhof wrote:
>
> In Bash I use the following construct:
> printf "${FUNCNAME} needs an expression\n"
>
> In this way I do not have to change the print
In Bash I use the following construct:
printf "${FUNCNAME} needs an expression\n"
In this way I do not have to change the print statement when the name of
the function changes. Is something like this also possible in clojure?
--
Cecil Westerhof
--
You received this message because you are
12 matches
Mail list logo
