module Enumerable - Ruby 2.2.4 Core API Reference API Reference

Ruby 2.2 Core API

module Enumerable

The Enumerable mixin provides collection classes with several traversal and searching methods, and with the ability to sort. The class must provide a method each, which yields successive members of the collection. If Enumerable#max, #min, or #sort is used, the objects in the collection must also implement a meaningful <=> operator, as these methods rely on an ordering between members of the collection.

In Files

  • enum.c
  • enumerator.c

Public Instance Methods

all? [{ |obj| block } ] → true or false click to toggle source

Passes each element of the collection to the given block. The method returns true if the block never returns false or nil. If the block is not given, Ruby adds an implicit block of { |obj| obj } which will cause all? to return true when none of the collection members are false or nil.

%w[ant bear cat].all? { |word| word.length >= 3 } #=> true
%w[ant bear cat].all? { |word| word.length >= 4 } #=> false
[nil, true, 99].all?                              #=> false
static VALUE
enum_all(VALUE obj)
{
    NODE *memo = NEW_MEMO(Qtrue, 0, 0);
    rb_block_call(obj, id_each, 0, 0, ENUMFUNC(all), (VALUE)memo);
    return memo->u1.value;
}
any? [{ |obj| block }] → true or false click to toggle source

Passes each element of the collection to the given block. The method returns true if the block ever returns a value other than false or nil. If the block is not given, Ruby adds an implicit block of { |obj| obj } that will cause any? to return true if at least one of the collection members is not false or nil.

%w[ant bear cat].any? { |word| word.length >= 3 } #=> true
%w[ant bear cat].any? { |word| word.length >= 4 } #=> true
[nil, true, 99].any?                              #=> true
static VALUE
enum_any(VALUE obj)
{
    NODE *memo = NEW_MEMO(Qfalse, 0, 0);
    rb_block_call(obj, id_each, 0, 0, ENUMFUNC(any), (VALUE)memo);
    return memo->u1.value;
}
chunk { |elt| ... } → an_enumerator click to toggle source
chunk(initial_state) { |elt, state| ... } → an_enumerator (deprecated)

Enumerates over the items, chunking them together based on the return value of the block.

Consecutive elements which return the same block value are chunked together.

For example, consecutive even numbers and odd numbers can be chunked as follows.

[3, 1, 4, 1, 5, 9, 2, 6, 5, 3, 5].chunk { |n|
  n.even?
}.each { |even, ary|
  p [even, ary]
}
#=> [false, [3, 1]]
#   [true, [4]]
#   [false, [1, 5, 9]]
#   [true, [2, 6]]
#   [false, [5, 3, 5]]

This method is especially useful for sorted series of elements. The following example counts words for each initial letter.

open("/usr/share/dict/words", "r:iso-8859-1") { |f|
  f.chunk { |line| line.ord }.each { |ch, lines| p [ch.chr, lines.length] }
}
#=> ["\n", 1]
#   ["A", 1327]
#   ["B", 1372]
#   ["C", 1507]
#   ["D", 791]
#   ...

The following key values have special meaning:

  • nil and :_separator specifies that the elements should be dropped.

  • :_alone specifies that the element should be chunked by itself.

Any other symbols that begin with an underscore will raise an error:

items.chunk { |item| :_underscore }
#=> RuntimeError: symbols beginning with an underscore are reserved

nil and :_separator can be used to ignore some elements.

For example, the sequence of hyphens in svn log can be eliminated as follows:

sep = "-"*72 + "\n"
IO.popen("svn log README") { |f|
  f.chunk { |line|
    line != sep || nil
  }.each { |_, lines|
    pp lines
  }
}
#=> ["r20018 | knu | 2008-10-29 13:20:42 +0900 (Wed, 29 Oct 2008) | 2 lines\n",
#    "\n",
#    "* README, README.ja: Update the portability section.\n",
#    "\n"]
#   ["r16725 | knu | 2008-05-31 23:34:23 +0900 (Sat, 31 May 2008) | 2 lines\n",
#    "\n",
#    "* README, README.ja: Add a note about default C flags.\n",
#    "\n"]
#   ...

Paragraphs separated by empty lines can be parsed as follows:

File.foreach("README").chunk { |line|
  /\A\s*\z/ !~ line || nil
}.each { |_, lines|
  pp lines
}

:_alone can be used to force items into their own chunk. For example, you can put lines that contain a URL by themselves, and chunk the rest of the lines together, like this:

pattern = /http/
open(filename) { |f|
  f.chunk { |line| line =~ pattern ? :_alone : true }.each { |key, lines|
    pp lines
  }
}
static VALUE
enum_chunk(int argc, VALUE *argv, VALUE enumerable)
{
    VALUE initial_state;
    VALUE enumerator;
    int n;

    if (!rb_block_given_p())
        rb_raise(rb_eArgError, "no block given");
    n = rb_scan_args(argc, argv, "01", &initial_state);
    if (n != 0)
        rb_warn("initial_state given for chunk.  (Use local variables.)");

    enumerator = rb_obj_alloc(rb_cEnumerator);
    rb_ivar_set(enumerator, rb_intern("chunk_enumerable"), enumerable);
    rb_ivar_set(enumerator, rb_intern("chunk_categorize"), rb_block_proc());
    rb_ivar_set(enumerator, rb_intern("chunk_initial_state"), initial_state);
    rb_block_call(enumerator, idInitialize, 0, 0, chunk_i, enumerator);
    return enumerator;
}
collect { |obj| block } → array click to toggle source
map { |obj| block } → array
collect → an_enumerator
map → an_enumerator

Returns a new array with the results of running block once for every element in enum.

If no block is given, an enumerator is returned instead.

(1..4).map { |i| i*i }      #=> [1, 4, 9, 16]
(1..4).collect { "cat"  }   #=> ["cat", "cat", "cat", "cat"]
static VALUE
enum_collect(VALUE obj)
{
    VALUE ary;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, collect_i, ary);

    return ary;
}
flat_map { |obj| block } → array click to toggle source
collect_concat { |obj| block } → array
flat_map → an_enumerator
collect_concat → an_enumerator

Returns a new array with the concatenated results of running block once for every element in enum.

If no block is given, an enumerator is returned instead.

[1, 2, 3, 4].flat_map { |e| [e, -e] } #=> [1, -1, 2, -2, 3, -3, 4, -4]
[[1, 2], [3, 4]].flat_map { |e| e + [100] } #=> [1, 2, 100, 3, 4, 100]
static VALUE
enum_flat_map(VALUE obj)
{
    VALUE ary;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, flat_map_i, ary);

    return ary;
}
count → int click to toggle source
count(item) → int
count { |obj| block } → int

Returns the number of items in enum through enumeration. If an argument is given, the number of items in enum that are equal to item are counted. If a block is given, it counts the number of elements yielding a true value.

ary = [1, 2, 4, 2]
ary.count               #=> 4
ary.count(2)            #=> 2
ary.count{ |x| x%2==0 } #=> 3
static VALUE
enum_count(int argc, VALUE *argv, VALUE obj)
{
    VALUE item = Qnil;
    NODE *memo;
    rb_block_call_func *func;

    if (argc == 0) {
        if (rb_block_given_p()) {
            func = count_iter_i;
        }
        else {
            func = count_all_i;
        }
    }
    else {
        rb_scan_args(argc, argv, "1", &item);
        if (rb_block_given_p()) {
            rb_warn("given block not used");
        }
        func = count_i;
    }

    memo = NEW_MEMO(item, 0, 0);
    rb_block_call(obj, id_each, 0, 0, func, (VALUE)memo);
    return INT2NUM(memo->u3.cnt);
}
cycle(n=nil) { |obj| block } → nil click to toggle source
cycle(n=nil) → an_enumerator

Calls block for each element of enum repeatedly n times or forever if none or nil is given. If a non-positive number is given or the collection is empty, does nothing. Returns nil if the loop has finished without getting interrupted.

#cycle saves elements in an internal array so changes to enum after the first pass have no effect.

If no block is given, an enumerator is returned instead.

a = ["a", "b", "c"]
a.cycle { |x| puts x }  # print, a, b, c, a, b, c,.. forever.
a.cycle(2) { |x| puts x }  # print, a, b, c, a, b, c.
static VALUE
enum_cycle(int argc, VALUE *argv, VALUE obj)
{
    VALUE ary;
    VALUE nv = Qnil;
    long n, i, len;

    rb_scan_args(argc, argv, "01", &nv);

    RETURN_SIZED_ENUMERATOR(obj, argc, argv, enum_cycle_size);
    if (NIL_P(nv)) {
        n = -1;
    }
    else {
        n = NUM2LONG(nv);
        if (n <= 0) return Qnil;
    }
    ary = rb_ary_new();
    RBASIC_CLEAR_CLASS(ary);
    rb_block_call(obj, id_each, 0, 0, cycle_i, ary);
    len = RARRAY_LEN(ary);
    if (len == 0) return Qnil;
    while (n < 0 || 0 < --n) {
        for (i=0; i<len; i++) {
            rb_yield(RARRAY_AREF(ary, i));
        }
    }
    return Qnil;
}
detect(ifnone = nil) { |obj| block } → obj or nil click to toggle source
find(ifnone = nil) { |obj| block } → obj or nil
detect(ifnone = nil) → an_enumerator
find(ifnone = nil) → an_enumerator

Passes each entry in enum to block. Returns the first for which block is not false. If no object matches, calls ifnone and returns its result when it is specified, or returns nil otherwise.

If no block is given, an enumerator is returned instead.

(1..10).detect   { |i| i % 5 == 0 and i % 7 == 0 }   #=> nil
(1..100).find    { |i| i % 5 == 0 and i % 7 == 0 }   #=> 35
static VALUE
enum_find(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;
    VALUE if_none;

    rb_scan_args(argc, argv, "01", &if_none);
    RETURN_ENUMERATOR(obj, argc, argv);
    memo = NEW_MEMO(Qundef, 0, 0);
    rb_block_call(obj, id_each, 0, 0, find_i, (VALUE)memo);
    if (memo->u3.cnt) {
        return memo->u1.value;
    }
    if (!NIL_P(if_none)) {
        return rb_funcall(if_none, id_call, 0, 0);
    }
    return Qnil;
}
drop(n) → array click to toggle source

Drops first n elements from enum, and returns rest elements in an array.

a = [1, 2, 3, 4, 5, 0]
a.drop(3)             #=> [4, 5, 0]
static VALUE
enum_drop(VALUE obj, VALUE n)
{
    VALUE result;
    NODE *memo;
    long len = NUM2LONG(n);

    if (len < 0) {
        rb_raise(rb_eArgError, "attempt to drop negative size");
    }

    result = rb_ary_new();
    memo = NEW_MEMO(result, 0, len);
    rb_block_call(obj, id_each, 0, 0, drop_i, (VALUE)memo);
    return result;
}
drop_while { |arr| block } → array click to toggle source
drop_while → an_enumerator

Drops elements up to, but not including, the first element for which the block returns nil or false and returns an array containing the remaining elements.

If no block is given, an enumerator is returned instead.

a = [1, 2, 3, 4, 5, 0]
a.drop_while { |i| i < 3 }   #=> [3, 4, 5, 0]
static VALUE
enum_drop_while(VALUE obj)
{
    VALUE result;
    NODE *memo;

    RETURN_ENUMERATOR(obj, 0, 0);
    result = rb_ary_new();
    memo = NEW_MEMO(result, 0, FALSE);
    rb_block_call(obj, id_each, 0, 0, drop_while_i, (VALUE)memo);
    return result;
}
each_cons(n) { ... } → nil click to toggle source
each_cons(n) → an_enumerator

Iterates the given block for each array of consecutive <n> elements. If no block is given, returns an enumerator.

e.g.:

(1..10).each_cons(3) { |a| p a }
# outputs below
[1, 2, 3]
[2, 3, 4]
[3, 4, 5]
[4, 5, 6]
[5, 6, 7]
[6, 7, 8]
[7, 8, 9]
[8, 9, 10]
static VALUE
enum_each_cons(VALUE obj, VALUE n)
{
    long size = NUM2LONG(n);
    NODE *memo;
    int arity;

    if (size <= 0) rb_raise(rb_eArgError, "invalid size");
    RETURN_SIZED_ENUMERATOR(obj, 1, &n, enum_each_cons_size);
    arity = rb_block_arity();
    memo = NEW_MEMO(rb_ary_new2(size), dont_recycle_block_arg(arity), size);
    rb_block_call(obj, id_each, 0, 0, each_cons_i, (VALUE)memo);

    return Qnil;
}
each_entry { |obj| block } → enum click to toggle source
each_entry → an_enumerator

Calls block once for each element in self, passing that element as a parameter, converting multiple values from yield to an array.

If no block is given, an enumerator is returned instead.

class Foo
  include Enumerable
  def each
    yield 1
    yield 1, 2
    yield
  end
end
Foo.new.each_entry{ |o| p o }

produces:

1
[1, 2]
nil
static VALUE
enum_each_entry(int argc, VALUE *argv, VALUE obj)
{
    RETURN_SIZED_ENUMERATOR(obj, argc, argv, enum_size);
    rb_block_call(obj, id_each, argc, argv, each_val_i, 0);
    return obj;
}
each_slice(n) { ... } → nil click to toggle source
each_slice(n) → an_enumerator

Iterates the given block for each slice of <n> elements. If no block is given, returns an enumerator.

(1..10).each_slice(3) { |a| p a }
# outputs below
[1, 2, 3]
[4, 5, 6]
[7, 8, 9]
[10]
static VALUE
enum_each_slice(VALUE obj, VALUE n)
{
    long size = NUM2LONG(n);
    VALUE ary;
    NODE *memo;
    int arity;

    if (size <= 0) rb_raise(rb_eArgError, "invalid slice size");
    RETURN_SIZED_ENUMERATOR(obj, 1, &n, enum_each_slice_size);
    ary = rb_ary_new2(size);
    arity = rb_block_arity();
    memo = NEW_MEMO(ary, dont_recycle_block_arg(arity), size);
    rb_block_call(obj, id_each, 0, 0, each_slice_i, (VALUE)memo);
    ary = memo->u1.value;
    if (RARRAY_LEN(ary) > 0) rb_yield(ary);

    return Qnil;
}
each_with_index(*args) { |obj, i| block } → enum click to toggle source
each_with_index(*args) → an_enumerator

Calls block with two arguments, the item and its index, for each item in enum. Given arguments are passed through to each().

If no block is given, an enumerator is returned instead.

hash = Hash.new
%w(cat dog wombat).each_with_index { |item, index|
  hash[item] = index
}
hash   #=> {"cat"=>0, "dog"=>1, "wombat"=>2}
static VALUE
enum_each_with_index(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;

    RETURN_SIZED_ENUMERATOR(obj, argc, argv, enum_size);

    memo = NEW_MEMO(0, 0, 0);
    rb_block_call(obj, id_each, argc, argv, each_with_index_i, (VALUE)memo);
    return obj;
}
each_with_object(obj) { |(*args), memo_obj| ... } → obj click to toggle source
each_with_object(obj) → an_enumerator

Iterates the given block for each element with an arbitrary object given, and returns the initially given object.

If no block is given, returns an enumerator.

evens = (1..10).each_with_object([]) { |i, a| a << i*2 }
#=> [2, 4, 6, 8, 10, 12, 14, 16, 18, 20]
static VALUE
enum_each_with_object(VALUE obj, VALUE memo)
{
    RETURN_SIZED_ENUMERATOR(obj, 1, &memo, enum_size);

    rb_block_call(obj, id_each, 0, 0, each_with_object_i, memo);

    return memo;
}
to_a(*args) → array click to toggle source
entries(*args) → array

Returns an array containing the items in enum.

(1..7).to_a                       #=> [1, 2, 3, 4, 5, 6, 7]
{ 'a'=>1, 'b'=>2, 'c'=>3 }.to_a   #=> [["a", 1], ["b", 2], ["c", 3]]

require 'prime'
Prime.entries 10                  #=> [2, 3, 5, 7]
static VALUE
enum_to_a(int argc, VALUE *argv, VALUE obj)
{
    VALUE ary = rb_ary_new();

    rb_block_call(obj, id_each, argc, argv, collect_all, ary);
    OBJ_INFECT(ary, obj);

    return ary;
}
detect(ifnone = nil) { |obj| block } → obj or nil click to toggle source
find(ifnone = nil) { |obj| block } → obj or nil
detect(ifnone = nil) → an_enumerator
find(ifnone = nil) → an_enumerator

Passes each entry in enum to block. Returns the first for which block is not false. If no object matches, calls ifnone and returns its result when it is specified, or returns nil otherwise.

If no block is given, an enumerator is returned instead.

(1..10).detect   { |i| i % 5 == 0 and i % 7 == 0 }   #=> nil
(1..100).find    { |i| i % 5 == 0 and i % 7 == 0 }   #=> 35
static VALUE
enum_find(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;
    VALUE if_none;

    rb_scan_args(argc, argv, "01", &if_none);
    RETURN_ENUMERATOR(obj, argc, argv);
    memo = NEW_MEMO(Qundef, 0, 0);
    rb_block_call(obj, id_each, 0, 0, find_i, (VALUE)memo);
    if (memo->u3.cnt) {
        return memo->u1.value;
    }
    if (!NIL_P(if_none)) {
        return rb_funcall(if_none, id_call, 0, 0);
    }
    return Qnil;
}
find_all { |obj| block } → array click to toggle source
select { |obj| block } → array
find_all → an_enumerator
select → an_enumerator

Returns an array containing all elements of enum for which the given block returns a true value.

If no block is given, an Enumerator is returned instead.

(1..10).find_all { |i|  i % 3 == 0 }   #=> [3, 6, 9]

[1,2,3,4,5].select { |num|  num.even?  }   #=> [2, 4]

See also #reject.

static VALUE
enum_find_all(VALUE obj)
{
    VALUE ary;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, find_all_i, ary);

    return ary;
}
find_index(value) → int or nil click to toggle source
find_index { |obj| block } → int or nil
find_index → an_enumerator

Compares each entry in enum with value or passes to block. Returns the index for the first for which the evaluated value is non-false. If no object matches, returns nil

If neither block nor argument is given, an enumerator is returned instead.

(1..10).find_index  { |i| i % 5 == 0 and i % 7 == 0 }  #=> nil
(1..100).find_index { |i| i % 5 == 0 and i % 7 == 0 }  #=> 34
(1..100).find_index(50)                                #=> 49
static VALUE
enum_find_index(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo; /* [return value, current index, ] */
    VALUE condition_value = Qnil;
    rb_block_call_func *func;

    if (argc == 0) {
        RETURN_ENUMERATOR(obj, 0, 0);
        func = find_index_iter_i;
    }
    else {
        rb_scan_args(argc, argv, "1", &condition_value);
        if (rb_block_given_p()) {
            rb_warn("given block not used");
        }
        func = find_index_i;
    }

    memo = NEW_MEMO(Qnil, condition_value, 0);
    rb_block_call(obj, id_each, 0, 0, func, (VALUE)memo);
    return memo->u1.value;
}
first → obj or nil click to toggle source
first(n) → an_array

Returns the first element, or the first n elements, of the enumerable. If the enumerable is empty, the first form returns nil, and the second form returns an empty array.

%w[foo bar baz].first     #=> "foo"
%w[foo bar baz].first(2)  #=> ["foo", "bar"]
%w[foo bar baz].first(10) #=> ["foo", "bar", "baz"]
[].first                  #=> nil
static VALUE
enum_first(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;
    rb_check_arity(argc, 0, 1);
    if (argc > 0) {
        return enum_take(obj, argv[0]);
    }
    else {
        memo = NEW_MEMO(Qnil, 0, 0);
        rb_block_call(obj, id_each, 0, 0, first_i, (VALUE)memo);
        return memo->u1.value;
    }
}
flat_map { |obj| block } → array click to toggle source
collect_concat { |obj| block } → array
flat_map → an_enumerator
collect_concat → an_enumerator

Returns a new array with the concatenated results of running block once for every element in enum.

If no block is given, an enumerator is returned instead.

[1, 2, 3, 4].flat_map { |e| [e, -e] } #=> [1, -1, 2, -2, 3, -3, 4, -4]
[[1, 2], [3, 4]].flat_map { |e| e + [100] } #=> [1, 2, 100, 3, 4, 100]
static VALUE
enum_flat_map(VALUE obj)
{
    VALUE ary;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, flat_map_i, ary);

    return ary;
}
grep(pattern) → array click to toggle source
grep(pattern) { |obj| block } → array

Returns an array of every element in enum for which Pattern === element. If the optional block is supplied, each matching element is passed to it, and the block's result is stored in the output array.

(1..100).grep 38..44   #=> [38, 39, 40, 41, 42, 43, 44]
c = IO.constants
c.grep(/SEEK/)         #=> [:SEEK_SET, :SEEK_CUR, :SEEK_END]
res = c.grep(/SEEK/) { |v| IO.const_get(v) }
res                    #=> [0, 1, 2]
static VALUE
enum_grep(VALUE obj, VALUE pat)
{
    VALUE ary = rb_ary_new();
    NODE *memo = NEW_MEMO(pat, ary, 0);

    rb_block_call(obj, id_each, 0, 0, rb_block_given_p() ? grep_iter_i : grep_i, (VALUE)memo);

    return ary;
}
group_by { |obj| block } → a_hash click to toggle source
group_by → an_enumerator

Groups the collection by result of the block. Returns a hash where the keys are the evaluated result from the block and the values are arrays of elements in the collection that correspond to the key.

If no block is given an enumerator is returned.

(1..6).group_by { |i| i%3 }   #=> {0=>[3, 6], 1=>[1, 4], 2=>[2, 5]}
static VALUE
enum_group_by(VALUE obj)
{
    VALUE hash;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    hash = rb_hash_new();
    rb_block_call(obj, id_each, 0, 0, group_by_i, hash);
    OBJ_INFECT(hash, obj);

    return hash;
}
include?(obj) → true or false click to toggle source
member?(obj) → true or false

Returns true if any member of enum equals obj. Equality is tested using ==.

IO.constants.include? :SEEK_SET          #=> true
IO.constants.include? :SEEK_NO_FURTHER   #=> false
static VALUE
enum_member(VALUE obj, VALUE val)
{
    NODE *memo = NEW_MEMO(val, Qfalse, 0);

    rb_block_call(obj, id_each, 0, 0, member_i, (VALUE)memo);
    return memo->u2.value;
}
inject(initial, sym) → obj click to toggle source
inject(sym) → obj
inject(initial) { |memo, obj| block } → obj
inject { |memo, obj| block } → obj
reduce(initial, sym) → obj
reduce(sym) → obj
reduce(initial) { |memo, obj| block } → obj
reduce { |memo, obj| block } → obj

Combines all elements of enum by applying a binary operation, specified by a block or a symbol that names a method or operator.

If you specify a block, then for each element in enum the block is passed an accumulator value (memo) and the element. If you specify a symbol instead, then each element in the collection will be passed to the named method of memo. In either case, the result becomes the new value for memo. At the end of the iteration, the final value of memo is the return value for the method.

If you do not explicitly specify an initial value for memo, then the first element of collection is used as the initial value of memo.

# Sum some numbers
(5..10).reduce(:+)                             #=> 45
# Same using a block and inject
(5..10).inject { |sum, n| sum + n }            #=> 45
# Multiply some numbers
(5..10).reduce(1, :*)                          #=> 151200
# Same using a block
(5..10).inject(1) { |product, n| product * n } #=> 151200
# find the longest word
longest = %w{ cat sheep bear }.inject do |memo, word|
   memo.length > word.length ? memo : word
end
longest                                        #=> "sheep"
static VALUE
enum_inject(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;
    VALUE init, op;
    rb_block_call_func *iter = inject_i;
    ID id;

    switch (rb_scan_args(argc, argv, "02", &init, &op)) {
      case 0:
        init = Qundef;
        break;
      case 1:
        if (rb_block_given_p()) {
            break;
        }
        id = rb_check_id(&init);
        op = id ? ID2SYM(id) : init;
        init = Qundef;
        iter = inject_op_i;
        break;
      case 2:
        if (rb_block_given_p()) {
            rb_warning("given block not used");
        }
        id = rb_check_id(&op);
        if (id) op = ID2SYM(id);
        iter = inject_op_i;
        break;
    }
    memo = NEW_MEMO(init, Qnil, op);
    rb_block_call(obj, id_each, 0, 0, iter, (VALUE)memo);
    if (memo->u1.value == Qundef) return Qnil;
    return memo->u1.value;
}
lazy → lazy_enumerator click to toggle source

Returns a lazy enumerator, whose methods map/collect, flat_map/collect_concat, select/find_all, reject, grep, zip, take, #take_while, drop, and #drop_while enumerate values only on an as-needed basis. However, if a block is given to zip, values are enumerated immediately.

Example

The following program finds pythagorean triples:

def pythagorean_triples
  (1..Float::INFINITY).lazy.flat_map {|z|
    (1..z).flat_map {|x|
      (x..z).select {|y|
        x**2 + y**2 == z**2
      }.map {|y|
        [x, y, z]
      }
    }
  }
end
# show first ten pythagorean triples
p pythagorean_triples.take(10).force # take is lazy, so force is needed
p pythagorean_triples.first(10)      # first is eager
# show pythagorean triples less than 100
p pythagorean_triples.take_while { |*, z| z < 100 }.force
static VALUE
enumerable_lazy(VALUE obj)
{
    VALUE result = lazy_to_enum_i(obj, sym_each, 0, 0, lazyenum_size);
    /* Qfalse indicates that the Enumerator::Lazy has no method name */
    rb_ivar_set(result, id_method, Qfalse);
    return result;
}
collect { |obj| block } → array click to toggle source
map { |obj| block } → array
collect → an_enumerator
map → an_enumerator

Returns a new array with the results of running block once for every element in enum.

If no block is given, an enumerator is returned instead.

(1..4).map { |i| i*i }      #=> [1, 4, 9, 16]
(1..4).collect { "cat"  }   #=> ["cat", "cat", "cat", "cat"]
static VALUE
enum_collect(VALUE obj)
{
    VALUE ary;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, collect_i, ary);

    return ary;
}
max → obj click to toggle source
max { |a, b| block } → obj
max(n) → obj
max(n) {|a,b| block } → obj

Returns the object in enum with the maximum value. The first form assumes all objects implement Comparable; the second uses the block to return a <=> b.

a = %w(albatross dog horse)
a.max                                   #=> "horse"
a.max { |a, b| a.length <=> b.length }  #=> "albatross"

If the n argument is given, maximum n elements are returned as an array.

a = %w[albatross dog horse]
a.max(2)                                  #=> ["horse", "dog"]
a.max(2) {|a, b| a.length <=> b.length }  #=> ["albatross", "horse"]
static VALUE
enum_max(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo = NEW_MEMO(Qundef, 0, 0);
    VALUE result;
    VALUE num;

    rb_scan_args(argc, argv, "01", &num);

    if (!NIL_P(num))
       return nmin_run(obj, num, 0, 1);

    if (rb_block_given_p()) {
        rb_block_call(obj, id_each, 0, 0, max_ii, (VALUE)memo);
    }
    else {
        rb_block_call(obj, id_each, 0, 0, max_i, (VALUE)memo);
    }
    result = memo->u1.value;
    if (result == Qundef) return Qnil;
    return result;
}
max_by {|obj| block } → obj click to toggle source
max_by → an_enumerator
max_by(n) {|obj| block } → obj
max_by(n) → an_enumerator

Returns the object in enum that gives the maximum value from the given block.

If no block is given, an enumerator is returned instead.

a = %w(albatross dog horse)
a.max_by { |x| x.length }   #=> "albatross"

If the n argument is given, minimum n elements are returned as an array.

a = %w[albatross dog horse]
a.max_by(2) {|x| x.length } #=> ["albatross", "horse"]

enum.max_by(n) can be used to implement weighted random sampling. Following example implements and use Enumerable#wsample.

module Enumerable
  # weighted random sampling.
  #
  # Pavlos S. Efraimidis, Paul G. Spirakis
  # Weighted random sampling with a reservoir
  # Information Processing Letters
  # Volume 97, Issue 5 (16 March 2006)
  def wsample(n)
    self.max_by(n) {|v| rand ** (1.0/yield(v)) }
  end
end
e = (-20..20).to_a*10000
a = e.wsample(20000) {|x|
  Math.exp(-(x/5.0)**2) # normal distribution
}
# a is 20000 samples from e.
p a.length #=> 20000
h = a.group_by {|x| x }
-10.upto(10) {|x| puts "*" * (h[x].length/30.0).to_i if h[x] }
#=> *
#   ***
#   ******
#   ***********
#   ******************
#   *****************************
#   *****************************************
#   ****************************************************
#   ***************************************************************
#   ********************************************************************
#   ***********************************************************************
#   ***********************************************************************
#   **************************************************************
#   ****************************************************
#   ***************************************
#   ***************************
#   ******************
#   ***********
#   *******
#   ***
#   *
static VALUE
enum_max_by(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;
    VALUE num;

    rb_scan_args(argc, argv, "01", &num);

    RETURN_SIZED_ENUMERATOR(obj, argc, argv, enum_size);

    if (!NIL_P(num))
        return nmin_run(obj, num, 1, 1);

    memo = NEW_MEMO(Qundef, Qnil, 0);
    rb_block_call(obj, id_each, 0, 0, max_by_i, (VALUE)memo);
    return memo->u2.value;
}
include?(obj) → true or false click to toggle source
member?(obj) → true or false

Returns true if any member of enum equals obj. Equality is tested using ==.

IO.constants.include? :SEEK_SET          #=> true
IO.constants.include? :SEEK_NO_FURTHER   #=> false
static VALUE
enum_member(VALUE obj, VALUE val)
{
    NODE *memo = NEW_MEMO(val, Qfalse, 0);

    rb_block_call(obj, id_each, 0, 0, member_i, (VALUE)memo);
    return memo->u2.value;
}
min → obj click to toggle source
min {| a,b | block } → obj
min(n) → array
min(n) {| a,b | block } → array

Returns the object in enum with the minimum value. The first form assumes all objects implement Comparable; the second uses the block to return a <=> b.

a = %w(albatross dog horse)
a.min                                   #=> "albatross"
a.min { |a, b| a.length <=> b.length }  #=> "dog"

If the n argument is given, minimum n elements are returned as an array.

a = %w[albatross dog horse]
a.min(2)                                  #=> ["albatross", "dog"]
a.min(2) {|a, b| a.length <=> b.length }  #=> ["dog", "horse"]
static VALUE
enum_min(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo = NEW_MEMO(Qundef, 0, 0);
    VALUE result;
    VALUE num;

    rb_scan_args(argc, argv, "01", &num);

    if (!NIL_P(num))
       return nmin_run(obj, num, 0, 0);

    if (rb_block_given_p()) {
        rb_block_call(obj, id_each, 0, 0, min_ii, (VALUE)memo);
    }
    else {
        rb_block_call(obj, id_each, 0, 0, min_i, (VALUE)memo);
    }
    result = memo->u1.value;
    if (result == Qundef) return Qnil;
    return result;
}
min_by {|obj| block } → obj click to toggle source
min_by → an_enumerator
min_by(n) {|obj| block } → array
min_by(n) → an_enumerator

Returns the object in enum that gives the minimum value from the given block.

If no block is given, an enumerator is returned instead.

a = %w(albatross dog horse)
a.min_by { |x| x.length }   #=> "dog"

If the n argument is given, minimum n elements are returned as an array.

a = %w[albatross dog horse]
p a.min_by(2) {|x| x.length } #=> ["dog", "horse"]
static VALUE
enum_min_by(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;
    VALUE num;

    rb_scan_args(argc, argv, "01", &num);

    RETURN_SIZED_ENUMERATOR(obj, argc, argv, enum_size);

    if (!NIL_P(num))
        return nmin_run(obj, num, 1, 0);

    memo = NEW_MEMO(Qundef, Qnil, 0);
    rb_block_call(obj, id_each, 0, 0, min_by_i, (VALUE)memo);
    return memo->u2.value;
}
minmax → [min, max] click to toggle source
minmax { |a, b| block } → [min, max]

Returns two elements array which contains the minimum and the maximum value in the enumerable. The first form assumes all objects implement Comparable; the second uses the block to return a <=> b.

a = %w(albatross dog horse)
a.minmax                                  #=> ["albatross", "horse"]
a.minmax { |a, b| a.length <=> b.length } #=> ["dog", "albatross"]
static VALUE
enum_minmax(VALUE obj)
{
    NODE *memo = NEW_MEMO(Qundef, Qundef, Qundef);
    struct minmax_t *m = (struct minmax_t *)&memo->u1.value;
    VALUE ary = rb_ary_new3(2, Qnil, Qnil);

    m->min = Qundef;
    m->last = Qundef;
    if (rb_block_given_p()) {
        rb_block_call(obj, id_each, 0, 0, minmax_ii, (VALUE)memo);
        if (m->last != Qundef)
            minmax_ii_update(m->last, m->last, m);
    }
    else {
        rb_block_call(obj, id_each, 0, 0, minmax_i, (VALUE)memo);
        if (m->last != Qundef)
            minmax_i_update(m->last, m->last, m);
    }
    if (m->min != Qundef) {
        rb_ary_store(ary, 0, m->min);
        rb_ary_store(ary, 1, m->max);
    }
    return ary;
}
minmax_by { |obj| block } → [min, max] click to toggle source
minmax_by → an_enumerator

Returns a two element array containing the objects in enum that correspond to the minimum and maximum values respectively from the given block.

If no block is given, an enumerator is returned instead.

a = %w(albatross dog horse)
a.minmax_by { |x| x.length }   #=> ["dog", "albatross"]
static VALUE
enum_minmax_by(VALUE obj)
{
    VALUE memo;
    struct minmax_by_t *m = NEW_MEMO_FOR(struct minmax_by_t, memo);

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    m->min_bv = Qundef;
    m->max_bv = Qundef;
    m->min = Qnil;
    m->max = Qnil;
    m->last_bv = Qundef;
    m->last = Qundef;
    rb_block_call(obj, id_each, 0, 0, minmax_by_i, memo);
    if (m->last_bv != Qundef)
        minmax_by_i_update(m->last_bv, m->last_bv, m->last, m->last, m);
    m = MEMO_FOR(struct minmax_by_t, memo);
    return rb_assoc_new(m->min, m->max);
}
none? [{ |obj| block }] → true or false click to toggle source

Passes each element of the collection to the given block. The method returns true if the block never returns true for all elements. If the block is not given, none? will return true only if none of the collection members is true.

%w{ant bear cat}.none? { |word| word.length == 5 } #=> true
%w{ant bear cat}.none? { |word| word.length >= 4 } #=> false
[].none?                                           #=> true
[nil].none?                                        #=> true
[nil, false].none?                                 #=> true
static VALUE
enum_none(VALUE obj)
{
    NODE *memo = NEW_MEMO(Qtrue, 0, 0);
    rb_block_call(obj, id_each, 0, 0, ENUMFUNC(none), (VALUE)memo);
    return memo->u1.value;
}
one? [{ |obj| block }] → true or false click to toggle source

Passes each element of the collection to the given block. The method returns true if the block returns true exactly once. If the block is not given, one? will return true only if exactly one of the collection members is true.

%w{ant bear cat}.one? { |word| word.length == 4 }  #=> true
%w{ant bear cat}.one? { |word| word.length > 4 }   #=> false
%w{ant bear cat}.one? { |word| word.length < 4 }   #=> false
[ nil, true, 99 ].one?                             #=> false
[ nil, true, false ].one?                          #=> true
static VALUE
enum_one(VALUE obj)
{
    NODE *memo = NEW_MEMO(Qundef, 0, 0);
    VALUE result;

    rb_block_call(obj, id_each, 0, 0, ENUMFUNC(one), (VALUE)memo);
    result = memo->u1.value;
    if (result == Qundef) return Qfalse;
    return result;
}
partition { |obj| block } → [ true_array, false_array ] click to toggle source
partition → an_enumerator

Returns two arrays, the first containing the elements of enum for which the block evaluates to true, the second containing the rest.

If no block is given, an enumerator is returned instead.

(1..6).partition { |v| v.even? }  #=> [[2, 4, 6], [1, 3, 5]]
static VALUE
enum_partition(VALUE obj)
{
    NODE *memo;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    memo = NEW_MEMO(rb_ary_new(), rb_ary_new(), 0);
    rb_block_call(obj, id_each, 0, 0, partition_i, (VALUE)memo);

    return rb_assoc_new(memo->u1.value, memo->u2.value);
}
inject(initial, sym) → obj click to toggle source
inject(sym) → obj
inject(initial) { |memo, obj| block } → obj
inject { |memo, obj| block } → obj
reduce(initial, sym) → obj
reduce(sym) → obj
reduce(initial) { |memo, obj| block } → obj
reduce { |memo, obj| block } → obj

Combines all elements of enum by applying a binary operation, specified by a block or a symbol that names a method or operator.

If you specify a block, then for each element in enum the block is passed an accumulator value (memo) and the element. If you specify a symbol instead, then each element in the collection will be passed to the named method of memo. In either case, the result becomes the new value for memo. At the end of the iteration, the final value of memo is the return value for the method.

If you do not explicitly specify an initial value for memo, then the first element of collection is used as the initial value of memo.

# Sum some numbers
(5..10).reduce(:+)                             #=> 45
# Same using a block and inject
(5..10).inject { |sum, n| sum + n }            #=> 45
# Multiply some numbers
(5..10).reduce(1, :*)                          #=> 151200
# Same using a block
(5..10).inject(1) { |product, n| product * n } #=> 151200
# find the longest word
longest = %w{ cat sheep bear }.inject do |memo, word|
   memo.length > word.length ? memo : word
end
longest                                        #=> "sheep"
static VALUE
enum_inject(int argc, VALUE *argv, VALUE obj)
{
    NODE *memo;
    VALUE init, op;
    rb_block_call_func *iter = inject_i;
    ID id;

    switch (rb_scan_args(argc, argv, "02", &init, &op)) {
      case 0:
        init = Qundef;
        break;
      case 1:
        if (rb_block_given_p()) {
            break;
        }
        id = rb_check_id(&init);
        op = id ? ID2SYM(id) : init;
        init = Qundef;
        iter = inject_op_i;
        break;
      case 2:
        if (rb_block_given_p()) {
            rb_warning("given block not used");
        }
        id = rb_check_id(&op);
        if (id) op = ID2SYM(id);
        iter = inject_op_i;
        break;
    }
    memo = NEW_MEMO(init, Qnil, op);
    rb_block_call(obj, id_each, 0, 0, iter, (VALUE)memo);
    if (memo->u1.value == Qundef) return Qnil;
    return memo->u1.value;
}
reject { |obj| block } → array click to toggle source
reject → an_enumerator

Returns an array for all elements of enum for which the given block returns false.

If no block is given, an Enumerator is returned instead.

(1..10).reject { |i|  i % 3 == 0 }   #=> [1, 2, 4, 5, 7, 8, 10]

[1, 2, 3, 4, 5].reject { |num| num.even? } #=> [1, 3, 5]

See also #find_all.

static VALUE
enum_reject(VALUE obj)
{
    VALUE ary;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, reject_i, ary);

    return ary;
}
reverse_each(*args) { |item| block } → enum click to toggle source
reverse_each(*args) → an_enumerator

Builds a temporary array and traverses that array in reverse order.

If no block is given, an enumerator is returned instead.

  (1..3).reverse_each { |v| p v }

produces:

  3
  2
  1
static VALUE
enum_reverse_each(int argc, VALUE *argv, VALUE obj)
{
    VALUE ary;
    long i;

    RETURN_SIZED_ENUMERATOR(obj, argc, argv, enum_size);

    ary = enum_to_a(argc, argv, obj);

    for (i = RARRAY_LEN(ary); --i >= 0; ) {
        rb_yield(RARRAY_AREF(ary, i));
    }

    return obj;
}
find_all { |obj| block } → array click to toggle source
select { |obj| block } → array
find_all → an_enumerator
select → an_enumerator

Returns an array containing all elements of enum for which the given block returns a true value.

If no block is given, an Enumerator is returned instead.

(1..10).find_all { |i|  i % 3 == 0 }   #=> [3, 6, 9]

[1,2,3,4,5].select { |num|  num.even?  }   #=> [2, 4]

See also #reject.

static VALUE
enum_find_all(VALUE obj)
{
    VALUE ary;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, find_all_i, ary);

    return ary;
}
slice_after(pattern) → an_enumerator click to toggle source
slice_after { |elt| bool } → an_enumerator

Creates an enumerator for each chunked elements. The ends of chunks are defined by pattern and the block.

If pattern === elt returns true or the block returns true for the element, the element is end of a chunk.

The === and block is called from the first element to the last element of enum.

The result enumerator yields the chunked elements as an array. So each method can be called as follows:

enum.slice_after(pattern).each { |ary| ... }
enum.slice_after { |elt| bool }.each { |ary| ... }

Other methods of the Enumerator class and Enumerable module, such as map, etc., are also usable.

For example, continuation lines (lines end with backslash) can be concatenated as follows:

lines = ["foo\n", "bar\\\n", "baz\n", "\n", "qux\n"]
e = lines.slice_after(/(?<!\)\n\z/)
p e.to_a
#=> [["foo\n"], ["bar\\\n", "baz\n"], ["\n"], ["qux\n"]]
p e.map {|ll| ll[0...-1].map {|l| l.sub(/\\n\z/, "") }.join + ll.last }
#=>["foo\n", "barbaz\n", "\n", "qux\n"]
static VALUE
enum_slice_after(int argc, VALUE *argv, VALUE enumerable)
{
    VALUE enumerator;
    VALUE pat = Qnil, pred = Qnil;

    if (rb_block_given_p()) {
        if (0 < argc)
            rb_raise(rb_eArgError, "both pattan and block are given");
        pred = rb_block_proc();
    }
    else {
        rb_scan_args(argc, argv, "1", &pat);
    }

    enumerator = rb_obj_alloc(rb_cEnumerator);
    rb_ivar_set(enumerator, rb_intern("sliceafter_enum"), enumerable);
    rb_ivar_set(enumerator, rb_intern("sliceafter_pat"), pat);
    rb_ivar_set(enumerator, rb_intern("sliceafter_pred"), pred);

    rb_block_call(enumerator, idInitialize, 0, 0, sliceafter_i, enumerator);
    return enumerator;
}
slice_before(pattern) → an_enumerator click to toggle source
slice_before { |elt| bool } → an_enumerator
slice_before(initial_state) { |elt, state| bool } → an_enumerator (deprecated)

Creates an enumerator for each chunked elements. The beginnings of chunks are defined by pattern and the block.

If pattern === elt returns true or the block returns true for the element, the element is beginning of a chunk.

The === and block is called from the first element to the last element of enum. The result for the first element is ignored.

The result enumerator yields the chunked elements as an array. So each method can be called as follows:

enum.slice_before(pattern).each { |ary| ... }
enum.slice_before { |elt| bool }.each { |ary| ... }

Other methods of the Enumerator class and Enumerable module, such as map, etc., are also usable.

For example, iteration over ChangeLog entries can be implemented as follows:

# iterate over ChangeLog entries.
open("ChangeLog") { |f|
  f.slice_before(/\A\S/).each { |e| pp e }
}

# same as above.  block is used instead of pattern argument.
open("ChangeLog") { |f|
  f.slice_before { |line| /\A\S/ === line }.each { |e| pp e }
}

“svn proplist -R” produces multiline output for each file. They can be chunked as follows:

IO.popen([{"LC_ALL"=>"C"}, "svn", "proplist", "-R"]) { |f|
  f.lines.slice_before(/\AProp/).each { |lines| p lines }
}
#=> ["Properties on '.':\n", "  svn:ignore\n", "  svk:merge\n"]
#   ["Properties on 'goruby.c':\n", "  svn:eol-style\n"]
#   ["Properties on 'complex.c':\n", "  svn:mime-type\n", "  svn:eol-style\n"]
#   ["Properties on 'regparse.c':\n", "  svn:eol-style\n"]
#   ...

If the block needs to maintain state over multiple elements, local variables can be used. For example, three or more consecutive increasing numbers can be squashed as follows:

a = [0, 2, 3, 4, 6, 7, 9]
prev = a[0]
p a.slice_before { |e|
  prev, prev2 = e, prev
  prev2 + 1 != e
}.map { |es|
  es.length <= 2 ? es.join(",") : "#{es.first}-#{es.last}"
}.join(",")
#=> "0,2-4,6,7,9"

However local variables should be used carefully if the result enumerator is enumerated twice or more. The local variables should be initialized for each enumeration. Enumerator.new can be used to do it.

# Word wrapping.  This assumes all characters have same width.
def wordwrap(words, maxwidth)
  Enumerator.new {|y|
    # cols is initialized in Enumerator.new.
    cols = 0
    words.slice_before { |w|
      cols += 1 if cols != 0
      cols += w.length
      if maxwidth < cols
        cols = w.length
        true
      else
        false
      end
    }.each {|ws| y.yield ws }
  }
end
text = (1..20).to_a.join(" ")
enum = wordwrap(text.split(/\s+/), 10)
puts "-"*10
enum.each { |ws| puts ws.join(" ") } # first enumeration.
puts "-"*10
enum.each { |ws| puts ws.join(" ") } # second enumeration generates same result as the first.
puts "-"*10
#=> ----------
#   1 2 3 4 5
#   6 7 8 9 10
#   11 12 13
#   14 15 16
#   17 18 19
#   20
#   ----------
#   1 2 3 4 5
#   6 7 8 9 10
#   11 12 13
#   14 15 16
#   17 18 19
#   20
#   ----------

mbox contains series of mails which start with Unix From line. So each mail can be extracted by slice before Unix From line.

# parse mbox
open("mbox") { |f|
  f.slice_before { |line|
    line.start_with? "From "
  }.each { |mail|
    unix_from = mail.shift
    i = mail.index("\n")
    header = mail[0...i]
    body = mail[(i+1)..-1]
    body.pop if body.last == "\n"
    fields = header.slice_before { |line| !" \t".include?(line[0]) }.to_a
    p unix_from
    pp fields
    pp body
  }
}

# split mails in mbox (slice before Unix From line after an empty line)
open("mbox") { |f|
  f.slice_before(emp: true) { |line, h|
    prevemp = h[:emp]
    h[:emp] = line == "\n"
    prevemp && line.start_with?("From ")
  }.each { |mail|
    mail.pop if mail.last == "\n"
    pp mail
  }
}
static VALUE
enum_slice_before(int argc, VALUE *argv, VALUE enumerable)
{
    VALUE enumerator;

    if (rb_block_given_p()) {
        VALUE initial_state;
        int n;
        n = rb_scan_args(argc, argv, "01", &initial_state);
        if (n != 0)
            rb_warn("initial_state given for slice_before.  (Use local variables.)");
        enumerator = rb_obj_alloc(rb_cEnumerator);
        rb_ivar_set(enumerator, rb_intern("slicebefore_sep_pred"), rb_block_proc());
        rb_ivar_set(enumerator, rb_intern("slicebefore_initial_state"), initial_state);
    }
    else {
        VALUE sep_pat;
        rb_scan_args(argc, argv, "1", &sep_pat);
        enumerator = rb_obj_alloc(rb_cEnumerator);
        rb_ivar_set(enumerator, rb_intern("slicebefore_sep_pat"), sep_pat);
    }
    rb_ivar_set(enumerator, rb_intern("slicebefore_enumerable"), enumerable);
    rb_block_call(enumerator, idInitialize, 0, 0, slicebefore_i, enumerator);
    return enumerator;
}
slice_when {|elt_before, elt_after| bool } → an_enumerator click to toggle source

Creates an enumerator for each chunked elements. The beginnings of chunks are defined by the block.

This method split each chunk using adjacent elements, elt_before and elt_after, in the receiver enumerator. This method split chunks between elt_before and elt_after where the block returns true.

The block is called the length of the receiver enumerator minus one.

The result enumerator yields the chunked elements as an array. So each method can be called as follows:

enum.slice_when { |elt_before, elt_after| bool }.each { |ary| ... }

Other methods of the Enumerator class and Enumerable module, such as to_a, map, etc., are also usable.

For example, one-by-one increasing subsequence can be chunked as follows:

a = [1,2,4,9,10,11,12,15,16,19,20,21]
b = a.slice_when {|i, j| i+1 != j }
p b.to_a #=> [[1, 2], [4], [9, 10, 11, 12], [15, 16], [19, 20, 21]]
c = b.map {|a| a.length < 3 ? a : "#{a.first}-#{a.last}" }
p c #=> [[1, 2], [4], "9-12", [15, 16], "19-21"]
d = c.join(",")
p d #=> "1,2,4,9-12,15,16,19-21"

Near elements (threshold: 6) in sorted array can be chunked as follwos:

a = [3, 11, 14, 25, 28, 29, 29, 41, 55, 57]
p a.slice_when {|i, j| 6 < j - i }.to_a
#=> [[3], [11, 14], [25, 28, 29, 29], [41], [55, 57]]

Increasing (non-decreasing) subsequence can be chunked as follows:

a = [0, 9, 2, 2, 3, 2, 7, 5, 9, 5]
p a.slice_when {|i, j| i > j }.to_a
#=> [[0, 9], [2, 2, 3], [2, 7], [5, 9], [5]]

Adjacent evens and odds can be chunked as follows: (#chunk is another way to do it.)

a = [7, 5, 9, 2, 0, 7, 9, 4, 2, 0]
p a.slice_when {|i, j| i.even? != j.even? }.to_a
#=> [[7, 5, 9], [2, 0], [7, 9], [4, 2, 0]]

Paragraphs (non-empty lines with trailing empty lines) can be chunked as follows: (See #chunk to ignore empty lines.)

lines = ["foo\n", "bar\n", "\n", "baz\n", "qux\n"]
p lines.slice_when {|l1, l2| /\A\s*\z/ =~ l1 && /\S/ =~ l2 }.to_a
#=> [["foo\n", "bar\n", "\n"], ["baz\n", "qux\n"]]
static VALUE
enum_slice_when(VALUE enumerable)
{
    VALUE enumerator;
    VALUE pred;

    pred = rb_block_proc();

    enumerator = rb_obj_alloc(rb_cEnumerator);
    rb_ivar_set(enumerator, rb_intern("slicewhen_enum"), enumerable);
    rb_ivar_set(enumerator, rb_intern("slicewhen_pred"), pred);

    rb_block_call(enumerator, idInitialize, 0, 0, slicewhen_i, enumerator);
    return enumerator;
}
sort → array click to toggle source
sort { |a, b| block } → array

Returns an array containing the items in enum sorted, either according to their own <=> method, or by using the results of the supplied block. The block should return -1, 0, or +1 depending on the comparison between a and b. As of Ruby 1.8, the method Enumerable#sort_by implements a built-in Schwartzian Transform, useful when key computation or comparison is expensive.

%w(rhea kea flea).sort          #=> ["flea", "kea", "rhea"]
(1..10).sort { |a, b| b <=> a }  #=> [10, 9, 8, 7, 6, 5, 4, 3, 2, 1]
static VALUE
enum_sort(VALUE obj)
{
    return rb_ary_sort(enum_to_a(0, 0, obj));
}
sort_by { |obj| block } → array click to toggle source
sort_by → an_enumerator

Sorts enum using a set of keys generated by mapping the values in enum through the given block.

If no block is given, an enumerator is returned instead.

%w{apple pear fig}.sort_by { |word| word.length}
              #=> ["fig", "pear", "apple"]

The current implementation of sort_by generates an array of tuples containing the original collection element and the mapped value. This makes sort_by fairly expensive when the keysets are simple.

require 'benchmark'

a = (1..100000).map { rand(100000) }

Benchmark.bm(10) do |b|
  b.report("Sort")    { a.sort }
  b.report("Sort by") { a.sort_by { |a| a } }
end

produces:

user     system      total        real
Sort        0.180000   0.000000   0.180000 (  0.175469)
Sort by     1.980000   0.040000   2.020000 (  2.013586)

However, consider the case where comparing the keys is a non-trivial operation. The following code sorts some files on modification time using the basic sort method.

files = Dir["*"]
sorted = files.sort { |a, b| File.new(a).mtime <=> File.new(b).mtime }
sorted   #=> ["mon", "tues", "wed", "thurs"]

This sort is inefficient: it generates two new File objects during every comparison. A slightly better technique is to use the Kernel#test method to generate the modification times directly.

files = Dir["*"]
sorted = files.sort { |a, b|
  test(M, a) <=> test(M, b)
}
sorted   #=> ["mon", "tues", "wed", "thurs"]

This still generates many unnecessary Time objects. A more efficient technique is to cache the sort keys (modification times in this case) before the sort. Perl users often call this approach a Schwartzian Transform, after Randal Schwartz. We construct a temporary array, where each element is an array containing our sort key along with the filename. We sort this array, and then extract the filename from the result.

sorted = Dir["*"].collect { |f|
   [test(M, f), f]
}.sort.collect { |f| f[1] }
sorted   #=> ["mon", "tues", "wed", "thurs"]

This is exactly what sort_by does internally.

sorted = Dir["*"].sort_by { |f| test(M, f) }
sorted   #=> ["mon", "tues", "wed", "thurs"]
static VALUE
enum_sort_by(VALUE obj)
{
    VALUE ary, buf;
    NODE *memo;
    long i;
    struct sort_by_data *data;

    RETURN_SIZED_ENUMERATOR(obj, 0, 0, enum_size);

    if (RB_TYPE_P(obj, T_ARRAY) && RARRAY_LEN(obj) <= LONG_MAX/2) {
        ary = rb_ary_new2(RARRAY_LEN(obj)*2);
    }
    else {
        ary = rb_ary_new();
    }
    RBASIC_CLEAR_CLASS(ary);
    buf = rb_ary_tmp_new(SORT_BY_BUFSIZE*2);
    rb_ary_store(buf, SORT_BY_BUFSIZE*2-1, Qnil);
    memo = NEW_MEMO(0, 0, 0);
    OBJ_INFECT(memo, obj);
    data = (struct sort_by_data *)&memo->u1;
    data->ary = ary;
    data->buf = buf;
    data->n = 0;
    rb_block_call(obj, id_each, 0, 0, sort_by_i, (VALUE)memo);
    ary = data->ary;
    buf = data->buf;
    if (data->n) {
        rb_ary_resize(buf, data->n*2);
        rb_ary_concat(ary, buf);
    }
    if (RARRAY_LEN(ary) > 2) {
        RARRAY_PTR_USE(ary, ptr,
                      ruby_qsort(ptr, RARRAY_LEN(ary)/2, 2*sizeof(VALUE),
                                 sort_by_cmp, (void *)ary));
    }
    if (RBASIC(ary)->klass) {
        rb_raise(rb_eRuntimeError, "sort_by reentered");
    }
    for (i=1; i<RARRAY_LEN(ary); i+=2) {
        RARRAY_ASET(ary, i/2, RARRAY_AREF(ary, i));
    }
    rb_ary_resize(ary, RARRAY_LEN(ary)/2);
    RBASIC_SET_CLASS_RAW(ary, rb_cArray);
    OBJ_INFECT(ary, memo);

    return ary;
}
take(n) → array click to toggle source

Returns first n elements from enum.

a = [1, 2, 3, 4, 5, 0]
a.take(3)             #=> [1, 2, 3]
a.take(30)            #=> [1, 2, 3, 4, 5, 0]
static VALUE
enum_take(VALUE obj, VALUE n)
{
    NODE *memo;
    VALUE result;
    long len = NUM2LONG(n);

    if (len < 0) {
        rb_raise(rb_eArgError, "attempt to take negative size");
    }

    if (len == 0) return rb_ary_new2(0);
    result = rb_ary_new2(len);
    memo = NEW_MEMO(result, 0, len);
    rb_block_call(obj, id_each, 0, 0, take_i, (VALUE)memo);
    return result;
}
take_while { |arr| block } → array click to toggle source
take_while → an_enumerator

Passes elements to the block until the block returns nil or false, then stops iterating and returns an array of all prior elements.

If no block is given, an enumerator is returned instead.

a = [1, 2, 3, 4, 5, 0]
a.take_while { |i| i < 3 }   #=> [1, 2]
static VALUE
enum_take_while(VALUE obj)
{
    VALUE ary;

    RETURN_ENUMERATOR(obj, 0, 0);
    ary = rb_ary_new();
    rb_block_call(obj, id_each, 0, 0, take_while_i, ary);
    return ary;
}
to_a(*args) → array click to toggle source
entries(*args) → array

Returns an array containing the items in enum.

(1..7).to_a                       #=> [1, 2, 3, 4, 5, 6, 7]
{ 'a'=>1, 'b'=>2, 'c'=>3 }.to_a   #=> [["a", 1], ["b", 2], ["c", 3]]

require 'prime'
Prime.entries 10                  #=> [2, 3, 5, 7]
static VALUE
enum_to_a(int argc, VALUE *argv, VALUE obj)
{
    VALUE ary = rb_ary_new();

    rb_block_call(obj, id_each, argc, argv, collect_all, ary);
    OBJ_INFECT(ary, obj);

    return ary;
}
to_h(*args) → hash click to toggle source

Returns the result of interpreting enum as a list of [key, value] pairs.

%[hello world].each_with_index.to_h
  # => {:hello => 0, :world => 1}
static VALUE
enum_to_h(int argc, VALUE *argv, VALUE obj)
{
    VALUE hash = rb_hash_new();
    rb_block_call(obj, id_each, argc, argv, enum_to_h_i, hash);
    OBJ_INFECT(hash, obj);
    return hash;
}
zip(arg, ...) → an_array_of_array click to toggle source
zip(arg, ...) { |arr| block } → nil

Takes one element from enum and merges corresponding elements from each args. This generates a sequence of n-element arrays, where n is one more than the count of arguments. The length of the resulting sequence will be enum#size. If the size of any argument is less than enum#size, nil values are supplied. If a block is given, it is invoked for each output array, otherwise an array of arrays is returned.

a = [ 4, 5, 6 ]
b = [ 7, 8, 9 ]

a.zip(b)                 #=> [[4, 7], [5, 8], [6, 9]]
[1, 2, 3].zip(a, b)      #=> [[1, 4, 7], [2, 5, 8], [3, 6, 9]]
[1, 2].zip(a, b)         #=> [[1, 4, 7], [2, 5, 8]]
a.zip([1, 2], [8])       #=> [[4, 1, 8], [5, 2, nil], [6, nil, nil]]
static VALUE
enum_zip(int argc, VALUE *argv, VALUE obj)
{
    int i;
    ID conv;
    NODE *memo;
    VALUE result = Qnil;
    VALUE args = rb_ary_new4(argc, argv);
    int allary = TRUE;

    argv = RARRAY_PTR(args);
    for (i=0; i<argc; i++) {
        VALUE ary = rb_check_array_type(argv[i]);
        if (NIL_P(ary)) {
            allary = FALSE;
            break;
        }
        argv[i] = ary;
    }
    if (!allary) {
        CONST_ID(conv, "to_enum");
        for (i=0; i<argc; i++) {
            if (!rb_respond_to(argv[i], id_each)) {
                rb_raise(rb_eTypeError, "wrong argument type %s (must respond to :each)",
                    rb_obj_classname(argv[i]));
            }
            argv[i] = rb_funcall(argv[i], conv, 1, ID2SYM(id_each));
        }
    }
    if (!rb_block_given_p()) {
        result = rb_ary_new();
    }
    /* use NODE_DOT2 as memo(v, v, -) */
    memo = rb_node_newnode(NODE_DOT2, result, args, 0);
    rb_block_call(obj, id_each, 0, 0, allary ? zip_ary : zip_i, (VALUE)memo);

    return result;
}