Returns elements from self; does not modify self.
When a single Integer argument index is given, returns the element at offset index:
a = [:foo, 'bar', 2] a[0] # => :foo a[2] # => 2 a # => [:foo, "bar", 2]
If index is negative, counts relative to the end of self:
a = [:foo, 'bar', 2] a[-1] # => 2 a[-2] # => "bar"
If index is out of range, returns nil.
When two Integer arguments start and length are given, returns a new Array of size length containing successive elements beginning at offset start:
a = [:foo, 'bar', 2] a[0, 2] # => [:foo, "bar"] a[1, 2] # => ["bar", 2]
If start + length is greater than self.length, returns all elements from offset start to the end:
a = [:foo, 'bar', 2] a[0, 4] # => [:foo, "bar", 2] a[1, 3] # => ["bar", 2] a[2, 2] # => [2]
If start == self.size and length >= 0, returns a new empty Array.
If length is negative, returns nil.
When a single Range argument range is given, treats range.min as start above and range.size as length above:
a = [:foo, 'bar', 2] a[0..1] # => [:foo, "bar"] a[1..2] # => ["bar", 2]
Special case: If range.start == a.size, returns a new empty Array.
If range.end is negative, calculates the end index from the end:
a = [:foo, 'bar', 2] a[0..-1] # => [:foo, "bar", 2] a[0..-2] # => [:foo, "bar"] a[0..-3] # => [:foo]
If range.start is negative, calculates the start index from the end:
a = [:foo, 'bar', 2] a[-1..2] # => [2] a[-2..2] # => ["bar", 2] a[-3..2] # => [:foo, "bar", 2]
If range.start is larger than the array size, returns nil.
a = [:foo, 'bar', 2] a[4..1] # => nil a[4..0] # => nil a[4..-1] # => nil
When a single Enumerator::ArithmeticSequence argument aseq is given, returns an Array of elements corresponding to the indexes produced by the sequence.
a = ['--', 'data1', '--', 'data2', '--', 'data3'] a[(1..).step(2)] # => ["data1", "data2", "data3"]
Unlike slicing with range, if the start or the end of the arithmetic sequence is larger than array size, throws RangeError.
a = ['--', 'data1', '--', 'data2', '--', 'data3'] a[(1..11).step(2)] # RangeError (((1..11).step(2)) out of range) a[(7..).step(2)] # RangeError (((7..).step(2)) out of range)
If given a single argument, and its type is not one of the listed, tries to convert it to Integer, and raises if it is impossible:
a = [:foo, 'bar', 2] # Raises TypeError (no implicit conversion of Symbol into Integer): a[:foo]
Array#slice is an alias for Array#[].
Returns the substring of self specified by the arguments.
When the single Integer argument index is given, returns the 1-character substring found in self at offset index:
'bar'[2] # => "r"
Counts backward from the end of self if index is negative:
'foo'[-3] # => "f"
Returns nil if index is out of range:
'foo'[3] # => nil 'foo'[-4] # => nil
When the two Integer arguments start and length are given, returns the substring of the given length found in self at offset start:
'foo'[0, 2] # => "fo" 'foo'[0, 0] # => ""
Counts backward from the end of self if start is negative:
'foo'[-2, 2] # => "oo"
Special case: returns a new empty String if start is equal to the length of self:
'foo'[3, 2] # => ""
Returns nil if start is out of range:
'foo'[4, 2] # => nil 'foo'[-4, 2] # => nil
Returns the trailing substring of self if length is large:
'foo'[1, 50] # => "oo"
Returns nil if length is negative:
'foo'[0, -1] # => nil
When the single Range argument range is given, derives start and length values from the given range, and returns values as above:
'foo'[0..1] is equivalent to 'foo'[0, 2].
'foo'[0...1] is equivalent to 'foo'[0, 1].
When the Regexp argument regexp is given, and the capture argument is 0, returns the first matching substring found in self, or nil if none found:
'foo'[/o/] # => "o" 'foo'[/x/] # => nil s = 'hello there' s[/[aeiou](.)\1/] # => "ell" s[/[aeiou](.)\1/, 0] # => "ell"
If argument capture is given and not 0, it should be either an Integer capture group index or a String or Symbol capture group name; the method call returns only the specified capture (see Regexp Capturing):
s = 'hello there' s[/[aeiou](.)\1/, 1] # => "l" s[/(?<vowel>[aeiou])(?<non_vowel>[^aeiou])/, "non_vowel"] # => "l" s[/(?<vowel>[aeiou])(?<non_vowel>[^aeiou])/, :vowel] # => "e"
If an invalid capture group index is given, nil is returned. If an invalid capture group name is given, IndexError is raised.
When the single String argument substring is given, returns the substring from self if found, otherwise nil:
'foo'['oo'] # => "oo" 'foo'['xx'] # => nil
String#slice is an alias for String#[].
Returns sym.to_s[].
Parses src and return a string which was matched to pattern. pattern should be described as Regexp.
require 'ripper' p Ripper.slice('def m(a) nil end', 'ident') #=> "m" p Ripper.slice('def m(a) nil end', '[ident lparen rparen]+') #=> "m(a)" p Ripper.slice("<<EOS\nstring\nEOS", 'heredoc_beg nl $(tstring_content*) heredoc_end', 1) #=> "string\n"
Returns a new Hash object containing the entries for the given keys:
h = {foo: 0, bar: 1, baz: 2} h.slice(:baz, :foo) # => {:baz=>2, :foo=>0}
Any given keys that are not found are ignored.
Returns a Hash of the given ENV names and their corresponding values:
ENV.replace('foo' => '0', 'bar' => '1', 'baz' => '2', 'bat' => '3') ENV.slice('foo', 'baz') # => {"foo"=>"0", "baz"=>"2"} ENV.slice('baz', 'foo') # => {"baz"=>"2", "foo"=>"0"}
Raises an exception if any of the names is invalid (see Invalid Names and Values):
ENV.slice('foo', 'bar', :bat) # Raises TypeError (no implicit conversion of Symbol into String)
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]
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 to_a, 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 (see chunk_while for a better way):
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| emp = true f.slice_before { |line| prevemp = emp emp = line == "\n" prevemp && line.start_with?("From ") }.each { |mail| mail.pop if mail.last == "\n" pp mail } }
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"]
Creates an enumerator for each chunked elements. The beginnings of chunks are defined by the block.
This method splits 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 follows:
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: (Enumerable#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 Enumerable#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"]]
Enumerable#chunk_while does the same, except splitting when the block returns false instead of true.
Like Enumerable#slice_before, but chains operation to be lazy-evaluated.
Like Enumerable#slice_after, but chains operation to be lazy-evaluated.
Like Enumerable#slice_when, but chains operation to be lazy-evaluated.
define UnicodeNormalize module here so that we don’t have to look it up
Potentially raised when a specification is validated.
Enumerator::ArithmeticSequence is a subclass of Enumerator, that is a representation of sequences of numbers with common difference. Instances of this class can be generated by the Range#step and Numeric#step methods.
The class can be used for slicing Array (see Array#slice) or custom collections.