GSP
Quick Navigator

Search Site

Unix VPS
A - Starter
B - Basic
C - Preferred
D - Commercial
MPS - Dedicated
Previous VPSs
* Sign Up! *

Support
Contact Us
Online Help
Handbooks
Domain Status
Man Pages

FAQ
Virtual Servers
Pricing
Billing
Technical

Network
Facilities
Connectivity
Topology Map

Miscellaneous
Server Agreement
Year 2038
Credits
 

USA Flag

 

 

Man Pages
lfe_clj(3) lfe_clj(3)

clj - LFE Clojure interface library.

This module provides Clojure-inpired functions and macros for use in LFE.

N.B. Instead of making fully-qualified calls to the macros exported from clj, you may (include-lib "lfe/include/clj.lfe") and then call them directly, e.g.

(include-lib "lfe/include/clj.lfe")
(-> 2 (+ 2) (=:= 4)) ; 'true

    

(defn name [arg ...] {{doc-string}} ...)

(defn {{doc-string}} ([argpat ...] ...))

Define and automatically export a function.

(defn- name [arg ...] {{doc-string}} ...)

(defn- {{doc-string}} ([argpat ...] ...))

Equivalent to defun.

(fn (arg ...) ...)

Equivalent to lambda.

Note: The original versions were copied from Tim Dysinger's lfesl repo here:

https://github.com/lfex/lfesl/blob/master/include/thread.lfe

    

They have since been modified to be safely exportable.

(-> ...)

Thread first.

Example usage, demonstrating ordering:


> (set o '(#(a 1) #(b 2) #(c 3)))
(#(a 1) #(b 2) #(c 3))
> (clj:-> o
>         (++ '(#(d 4)))
>         (++ '(#(e 5)))
>         (++ '(#(f 6))))
(#(a 1) #(b 2) #(c 3) #(d 4) #(e 5) #(f 6))

    

Note that the use of -> in this example results in each successive value being appended to the input list.

Another example showing how this works:


> (lists:sublist
>   (lists:reverse
>     (lists:sort
>       (lists:merge
>         (string:tokens
>           (string:to_upper "a b c d e")
>           " ")
>         '("X" "F" "L"))))
>   2 3)
("L" "F" "E")

    

Can be rewritten as this:


> (clj:-> "a b c d e"
>         (string:to_upper)
>         (string:tokens " ")
>         (lists:merge '("X" "F" "L"))
>         (lists:sort)
>         (lists:reverse)
>         (lists:sublist 2 3))
("L" "F" "E")

    

(->> ...)

Thread last.

Example usage, demonstrating ordering:


> (set o '(#(a 1) #(b 2) #(c 3)))
(#(a 1) #(b 2) #(c 3))
> (clj:->> o
>          (++ '(#(d 4)))
>          (++ '(#(e 5)))
>          (++ '(#(f 6))))
(#(f 6) #(e 5) #(d 4) #(a 1) #(b 2) #(c 3))

    

Note that the use of ->> in this example results in each successive value being prepended to the input list.

Another example showing how this:


> (lists:foldl #'+/2 0
>   (clj:take 10
>     (lists:filter
>       (clj:comp #'clj:even?/1 #'clj:round/1)
>       (lists:map
>         (lambda (x)
>           (math:pow x 2))
>         (clj:seq 42)))))
1540.0

    

Can be rewritten as this:


> (clj:->> (clj:seq 42)
>          (lists:map (lambda (x) (math:pow x 2)))
>          (lists:filter (clj:comp #'clj:even?/1 #'clj:round/1))
>          (clj:take 10)
>          (lists:foldl #'+/2 0))
1540.0

    

(as-> expr name . sexps)

Bind name to expr, evaluate the first sexp in the lexical context of that binding, then bind name to that result, repeating for each successive sexp in sexps, returning the result of the last sexp.

(cond-> expr . clauses)

Given an expression and a set of test/sexp pairs, thread x (via ->) through each sexp for which the corresponding test expression is truthy, i.e. neither 'false nor 'undefined. Note that, unlike cond branching, cond-> threading does not short circuit after the first truthy test expression.

(cond->> expr . clauses)

Given an expression and a set of test/sexp pairs, thread x (via ->>) through each sexp for which the corresponding test expression is truthy, i.e. neither 'false nor 'undefined. Note that, unlike cond branching, cond->> threading does not short circuit after the first truthy test expression.

(some-> x . sexps)

When x is not 'undefined, thread it into the first sexp (via ->), and when that result is not 'undefined, through the next, etc.

(some->> x . sexps)

When x is not 'undefined, thread it into the first sexp (via ->>), and when that result is not 'undefined , through the next, etc.

(if-let ((patt test)) then {{else}})

If test evaluates to anything other than 'false or 'undefined, evaluate then with patt bound to the value of test, otherwise else, if supplied, else 'undefined.

(iff-let ((patt test)) . body)

When test evaluates to anything other than 'false or 'undefined, evaluate body with patt bound to the value of test, otherwise return 'undefined.

(condp pred expr . clauses)

Given a binary predicate, an expression and a set of clauses of the form:


test-expr result-expr
test-expr >> result-fn

    

where result-fn is a unary function, if (pred test-expr expr) returns anything other than 'undefined or 'false, the clause is a match.

If a binary clause matches, return result-expr. If a ternary clause matches, call result-fn with the result of the predicate and return the result.

If no clause matches and a single default expression is given after the clauses, return it. If no default expression is given and no clause matches, throw a no-matching-clause error.

(if-not test then)

(if-not test then else)

If test evaluates to 'false or 'undefined, evaluate and return then, otherwise else, if supplied, else 'undefined.

(iff test . body)

Like Clojure's when. If test evaluates to anything other than 'false or 'undefined, evaluate body in an implicit progn.

(when-not test . body)

If test evaluates to 'false or 'undefined, evaluate body in an implicit progn. Otherwise return 'undefined.

(not= x)

(not= x y)

(not= x y . more)

Same as (not (== ...)).

Allowed in guards, unless otherwise stated.

(tuple? x)

Return 'true if x is a tuple.

(atom? x)

Return 'true if x is an atom.

(binary? x)

Return 'true if x is a binary.

(bitstring? x)

Return 'true if x is a bitstring.

(boolean? x)

(bool? x)

Return 'true if x is a boolean.

(float? x)

Return 'true if x is a float.

(function? f)

(func? f)

Return 'true if x is a function.

(function? f n)

(func? f n)

Return 'true if f is an n-ary function.

(integer? x)

(int? x)

Return 'true if x is an integer.

(number? x)

Return 'true if x is a number.

(record? x record-tag)

(record? x record-tag size)

Return 'true if x is a tuple and its first element is record-tag. If size is given, check that x is a record-tag record of size size.

N.B. record?/2 may yield unexpected results, due to difference between the Erlang and LFE compilers. As such, whenever possible, prefer record?/3."

(reference? x)

Return 'true if x is a reference.

(map? x)

Return 'true if x is a map. Return 'false on versions of Erlang without maps.

(undefined? x)

(undef? x)

Return 'true if x is the atom 'undefined.

(nil? x)

Return 'true if x is the atom 'nil or the empty list.

(true? x)

Return 'true if x is the atom 'true.

(false? x)

Return 'true if x is the atom 'false.

(falsy? x)

Return 'true if x is one of the atoms 'false and 'undefined.

(odd? x)

Return 'true if x is odd.

(even? x)

Return 'true if x is even.

(zero? x)

Return 'true if x is zero.

(pos? x)

Return 'true if x is greater than zero.

(neg? x)

Return 'true if x is less than zero.

(identical? x)

Return 'true if x is exactly equal to y.

(str x1, x2 ... xn)

Given arbitrary number of arguments, return a string consisting of each of their string representations.

N.B. Because Erlang characters are represented as integers, this will not work for chars, e.g. #\a, which will be presented in the return value as its integer value, i.e. "97".


> (clj:str #\a "bc")
"97bc"
> (clj:str "a" "bc")
"abc"

    

(lazy-seq)

(lazy-seq seq)

Return a (possibly infinite) lazy sequence from a given lazy sequence seq or a finite lazy sequence from given list seq. A lazy sequence is treated as finite if at any iteration it produces the empty list, instead of a cons cell with data as the head and a nullary function for the next iteration as the tail.

(conj coll . xs)

conj[oin] a value onto an existing collection. Prepend to a list, append to a tuple, and merge maps.

(if test then)

(if test then else)

If test evaluates to anything other than 'false or 'undefined, return then, otherwise else, if given, else 'undefined.

(comp f g)

Right to left function composition.

(comp fs x)

Compose a list of functions fs, right to left, and apply the resulting function to x.

(comp f g x)

Equivalent to (funcall (comp f g) x).

(comp fs)

Compose a list of functions fs from right to left.

(comp)

Equivalent to #'identity/1.

The following examples assume #'1+/1 is defined:

> (defun 1+ (x) (+ x 1))
1+

    

> (funcall (clj:comp #'math:sin/1 #'math:asin/1) 0.5)
0.49999999999999994
> (funcall (clj:comp (list #'1+/1 #'math:sin/1 #'math:asin/1) 0.5))
1.5

    

Or used in another function call:


> (lists:filter (clj:comp #'not/1 #'zero?/1)
    '(0 1 0 2 0 3 0 4))
(1 2 3 4)

    

The usage above is best when comp will be called by higher-order functions like lists:foldl/3 or lists:filter/2, etc. However, one may also call comp in the following manner, best suited for direct usage:


> (clj:comp #'math:sin/1 #'math:asin/1 0.5)
0.49999999999999994
> (clj:comp (list #'1+/1 #'math:sin/1 #'math:asin/1) 0.5)
1.5

    

(partial f args)

(partial f arg-1)

Partially apply f to a given argument arg-1 or list of args.


> (set f (clj:partial #'+/2 1))
#Fun<clj.3.121115395>
> (funcall f 2)
3
> (set f (clj:partial #'+/3 1))
#Fun<clj.3.121115395>
> (funcall f '(2 3))
6
> (set f (clj:partial #'+/3 '(2 3)))
#Fun<clj.3.121115395>
> (funcall f 4)
9
> (set f (clj:partial #'+/4 '(2 3)))
#Fun<clj.3.121115395>
> (funcall f '(4 5))
14

    

Note that to partially apply a function that expects a list, you must wrap said list into a (singleton) list.


> (set double (clj:partial #'*/2 2))
#Fun<clj.5.16146786>
> (set f (clj:partial #'lists:map/2 double))
#Fun<clj.5.16146786>
> (funcall f '((1 2 3)))
(2 4 6)

    

N.B. These functions may not be used in guards.

(string? data)

Return 'true if data is a flat list of printable characters.

(unicode? data)

Return 'true if data is a flat list of printable Unicode characters.

(list? data)

Return 'true if data is a list and not a string.

(set? data)

Return 'true if data is appears to be a (possibly ordered) set.

(dict? data)

Return 'true if data is a dictionary.

(proplist? lst)

Return 'true if lst is a list where proplist-kv?/1 returns 'true for all elements in lst.

(proplist-kv? data)

Return 'true if a data is a key/value tuple or an atom.

(queue? x)

Return 'true if x is a queue.

(empty? x)

Return 'true if x is the empty list, tuple, map, dictionary, queue, or general balanced tree.

(every? pred lst)

(all? pred lst)

Return 'true if (pred x) returns 'true for every x in lst.

(any? pred lst)

Return 'true if (pred x) returns 'true for any x in lst.

(not-any? pred lst)

Return 'false if (pred x) returns 'true for any x in lst.

(element? elem data)

Return 'true if elem is an element of data, where data is a list, set or ordset.

(seq end)

Equivalent to (seq 1 end).

(seq start end)

Equivalent to (seq start end 1).

(seq start end step)

Return a sequence of integers, starting with start, containing the successive results of adding step to the previous element, until end has been reached or password. In the latter case, end is not an element of the sequence.

(next func)

Equivalent to (next func 1 1).

(next func start)

Equivalent to (next func start 1).

(next func start step)

Return a nullary function that returns a cons cell with start as the head and a nullary function, (next func (funcall func start step) step) as the tail. The result can be treated as a (possibly infinite) lazy list, which only computes subseqeuent values as needed.

(lazy-seq seq)

Return a lazy sequence (possibly infinite) from given lazy sequence seq or finite lazy sequence from given list seq. Lazy sequence is treated as finite if at any iteration it produces empty list instead of data as its head and nullary function for next iteration as its tail.

(cycle lst)

Return a lazy infinite sequence with all elements from a given list lst or another lazy sequence cycled.

See next/3 for details on the structure.

(range)

Equivalent to (range 1 1).

(range start)

Equivalent to (range start 1).

(range start step)

Return a lazy list of integers, starting with start and increasing by step. Equivalent to (next #'+/2 start step). See also: next/3.

(drop n lst)

(drop 'all lst)

Return a list of all but the first n elements in lst. If n is the atom all, return the empty list.

(take n lst)

(take 'all lst)

Given a (possibly lazy) list lst, return a list of the first n elements of lst, or all elements if there are fewer than n. If n is the atom all and lst is a "normal" list, return lst.

(split-at n lst)

Return a tuple of `#(,(take n lst) ,(drop n lst)).

(partition n lst)

Equivalent to (partition n n lst).

(partition n step lst)

Equivalent to (partition n step () lst).

(partition n step pad lst)

Return a list of lists of n items each, at offsets step apart. Use the elements of pad as necessary to complete the last partition up to n elements. In case there are not enough padding elements, return a parition with less than n items.

(partition-all n lst)

Equivalent to (partition-all n n lst).

(partition-all n step lst)

Return a list of lists like partition/3, possibly including partitions with fewer than n elements at the end.

(interleave list-1 list-2)

Return a list of the first element of each list, then the second, etc.

(get-in data keys)

Equivalent to (get-in data keys 'undefined).

(get-in data keys not-found)

Return the value in a nested associative structure, where keys is a list of keys or list indices. Return the atom not-found if the key is not present or index is out of bounds, or the not-found value.

(reduce func (cons head tail))

Equivalent to (reduce func head tail).

(reduce func acc lst)

Equivalent to (lists:foldl func acc lst).

(repeat x)

Return a lazy infinite sequence of xs.

See next/3 for details on the structure.

(repeat n f)

Given a nullary function f, return a list of n applications of f.

(repeat n x)

Given a term x, return a list of n copies of x.

(identity x)

Identity function.

(constantly x)

Return a unary function that returns x. N.B. This is like Haskell's const rather than Clojure's constantly.

(inc x)

Increment x by 1.

(dec x)

Decrement x by 1.

Tim Dysinger, Duncan McGreggor, Eric Bailey.
2015-2016

Search for    or go to Top of page |  Section 3 |  Main Index

Powered by GSP Visit the GSP FreeBSD Man Page Interface.
Output converted with ManDoc.