Class: CSV::Table

Inherits:
Object
  • Object
show all
Extended by:
Forwardable
Includes:
Enumerable
Defined in:
lib/csv/table.rb

Overview

A CSV::Table is a two-dimensional data structure for representing CSV documents. Tables allow you to work with the data by row or column, manipulate the data, and even convert the results back to CSV, if needed.

All tables returned by CSV will be constructed from this class, if header row processing is activated.

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(array_of_rows, headers: nil) ⇒ Table

Constructs a new CSV::Table from array_of_rows, which are expected to be CSV::Row objects. All rows are assumed to have the same headers.

The optional headers parameter can be set to Array of headers. If headers aren’t set, headers are fetched from CSV::Row objects. Otherwise, headers() method will return headers being set in headers argument.

A CSV::Table object supports the following Array methods through delegation:

  • empty?()

  • length()

  • size()



31
32
33
34
35
36
37
38
39
40
41
42
43
 
# File 'lib/csv/table.rb', line 31

def initialize(array_of_rows, headers: nil)
  @table = array_of_rows
  @headers = headers
  unless @headers
    if @table.empty?
      @headers = []
    else
      @headers = @table.first.headers
    end
  end

  @mode  = :col_or_row
end

Instance Attribute Details

#mode ⇒ Object (readonly)

The current access mode for indexing and iteration.



46
47
48
 
# File 'lib/csv/table.rb', line 46

def mode
  @mode
end

Instance Method Details

#<<(row_or_array) ⇒ Object

:call-seq:

table << row_or_array -> self

If row_or_array is a CSV::Row object, it is appended to the table:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table << CSV::Row.new(table.headers, ['bat', 3])
table[3] # => #<CSV::Row "Name":"bat" "Value":3>

If row_or_array is an Array, it is used to create a new CSV::Row object which is then appended to the table:

table << ['bam', 4]
table[4] # => #<CSV::Row "Name":"bam" "Value":4>



370
371
372
373
374
375
376
377
378
 
# File 'lib/csv/table.rb', line 370

def <<(row_or_array)
  if row_or_array.is_a? Array  # append Array
    @table << Row.new(headers, row_or_array)
  else                         # append Row
    @table << row_or_array
  end

  self # for chaining
end

#==(other) ⇒ Object

Returns true if all each row of self == the corresponding row of other_table, otherwise, false.

The access mode does no affect the result.

Equal tables:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
other_table = CSV.parse(source, headers: true)
table == other_table # => true

Different row count:

other_table.delete(2)
table == other_table # => false

Different last row:

other_table << ['bat', 3]
table == other_table # => false



563
564
565
566
 
# File 'lib/csv/table.rb', line 563

def ==(other)
  return @table == other.table if other.is_a? CSV::Table
  @table == other
end

#[](index_or_header) ⇒ Object

:call-seq:

table[n] -> row
table[range] -> array_of_rows
table[header] -> array_of_fields

Returns data from the table; does not modify the table.

The expression table[n], where n is a non-negative Integer, returns the nth row of the table, if that row exists, and if the access mode is :row or :col_or_row:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.by_row! # => #<CSV::Table mode:row row_count:4>
table[1] # => #<CSV::Row "Name":"bar" "Value":"1">
table.by_col_or_row! # => #<CSV::Table mode:col_or_row row_count:4>
table[1] # => #<CSV::Row "Name":"bar" "Value":"1">

Counts backward from the last row if n is negative:

table[-1] # => #<CSV::Row "Name":"baz" "Value":"2">

Returns nil if n is too large or too small:

table[4] # => nil
table[-4] => nil

Raises an exception if the access mode is :row and n is not an Integer-convertible object.

table.by_row! # => #<CSV::Table mode:row row_count:4>
# Raises TypeError (no implicit conversion of String into Integer):
table['Name']

The expression table[range], where range is a Range object, returns rows from the table, beginning at row range.first, if those rows exist, and if the access mode is :row or :col_or_row:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.by_row! # => #<CSV::Table mode:row row_count:4>
rows = table[1..2] # => #<CSV::Row "Name":"bar" "Value":"1">
rows # => [#<CSV::Row "Name":"bar" "Value":"1">, #<CSV::Row "Name":"baz" "Value":"2">]
table.by_col_or_row! # => #<CSV::Table mode:col_or_row row_count:4>
rows = table[1..2] # => #<CSV::Row "Name":"bar" "Value":"1">
rows # => [#<CSV::Row "Name":"bar" "Value":"1">, #<CSV::Row "Name":"baz" "Value":"2">]

If there are too few rows, returns all from range.first to the end:

rows = table[1..50] # => #<CSV::Row "Name":"bar" "Value":"1">
rows # => [#<CSV::Row "Name":"bar" "Value":"1">, #<CSV::Row "Name":"baz" "Value":"2">]

Special case: if range.start == table.size, returns an empty Array:

table[table.size..50] # => []

If range.end is negative, calculates the ending index from the end:

rows = table[0..-1]
rows # => [#<CSV::Row "Name":"foo" "Value":"0">, #<CSV::Row "Name":"bar" "Value":"1">, #<CSV::Row "Name":"baz" "Value":"2">]

If range.start is negative, calculates the starting index from the end:

rows = table[-1..2]
rows # => [#<CSV::Row "Name":"baz" "Value":"2">]

If range.start is larger than table.size, returns nil:

table[4..4] # => nil

The expression table[header], where header is a String, returns column values (Array of Strings) if the column exists and if the access mode is :col or :col_or_row:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.by_col! # => #<CSV::Table mode:col row_count:4>
table['Name'] # => ["foo", "bar", "baz"]
table.by_col_or_row! # => #<CSV::Table mode:col_or_row row_count:4>
col = table['Name']
col # => ["foo", "bar", "baz"]

Modifying the returned column values does not modify the table:

col[0] = 'bat'
col # => ["bat", "bar", "baz"]
table['Name'] # => ["foo", "bar", "baz"]

Returns an Array of nil values if there is no such column:

table['Nosuch'] # => [nil, nil, nil]



232
233
234
235
236
237
238
239
 
# File 'lib/csv/table.rb', line 232

def [](index_or_header)
  if @mode == :row or  # by index
     (@mode == :col_or_row and (index_or_header.is_a?(Integer) or index_or_header.is_a?(Range)))
    @table[index_or_header]
  else                 # by header
    @table.map { |row| row[index_or_header] }
  end
end

#[]=(index_or_header, value) ⇒ Object

In the default mixed mode, this method assigns rows for index access and columns for header access. You can force the index association by first calling by_col!() or by_row!().

Rows may be set to an Array of values (which will inherit the table’s headers()) or a CSV::Row.

Columns may be set to a single value, which is copied to each row of the column, or an Array of values. Arrays of values are assigned to rows top to bottom in row major order. Excess values are ignored and if the Array does not have a value for each row the extra rows will receive a nil.

Assigning to an existing column or row clobbers the data. Assigning to new columns creates them at the right end of the table.



257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
 
# File 'lib/csv/table.rb', line 257

def []=(index_or_header, value)
  if @mode == :row or  # by index
     (@mode == :col_or_row and index_or_header.is_a? Integer)
    if value.is_a? Array
      @table[index_or_header] = Row.new(headers, value)
    else
      @table[index_or_header] = value
    end
  else                 # set column
    unless index_or_header.is_a? Integer
      index = @headers.index(index_or_header) || @headers.size
      @headers[index] = index_or_header
    end
    if value.is_a? Array  # multiple values
      @table.each_with_index do |row, i|
        if row.header_row?
          row[index_or_header] = index_or_header
        else
          row[index_or_header] = value[i]
        end
      end
    else                  # repeated value
      @table.each do |row|
        if row.header_row?
          row[index_or_header] = index_or_header
        else
          row[index_or_header] = value
        end
      end
    end
  end
end

#by_col ⇒ Object

Returns a duplicate table object, in column mode. This is handy for chaining in a single call without changing the table mode, but be aware that this method can consume a fair amount of memory for bigger data sets.

This method returns the duplicate table for chaining. Don’t chain destructive methods (like []=()) this way though, since you are working with a duplicate.



66
67
68
 
# File 'lib/csv/table.rb', line 66

def by_col
  self.class.new(@table.dup).by_col!
end

#by_col! ⇒ Object

Switches the mode of this table to column mode. All calls to indexing and iteration methods will work with columns until the mode is changed again.

This method returns the table and is safe to chain.



76
77
78
79
80
 
# File 'lib/csv/table.rb', line 76

def by_col!
  @mode = :col

  self
end

#by_col_or_row ⇒ Object

Returns a duplicate table object, in mixed mode. This is handy for chaining in a single call without changing the table mode, but be aware that this method can consume a fair amount of memory for bigger data sets.

This method returns the duplicate table for chaining. Don’t chain destructive methods (like []=()) this way though, since you are working with a duplicate.



91
92
93
 
# File 'lib/csv/table.rb', line 91

def by_col_or_row
  self.class.new(@table.dup).by_col_or_row!
end

#by_col_or_row! ⇒ Object

Switches the mode of this table to mixed mode. All calls to indexing and iteration methods will use the default intelligent indexing system until the mode is changed again. In mixed mode an index is assumed to be a row reference while anything else is assumed to be column access by headers.

This method returns the table and is safe to chain.



103
104
105
106
107
 
# File 'lib/csv/table.rb', line 103

def by_col_or_row!
  @mode = :col_or_row

  self
end

#by_row ⇒ Object

Returns a duplicate table object, in row mode. This is handy for chaining in a single call without changing the table mode, but be aware that this method can consume a fair amount of memory for bigger data sets.

This method returns the duplicate table for chaining. Don’t chain destructive methods (like []=()) this way though, since you are working with a duplicate.



118
119
120
 
# File 'lib/csv/table.rb', line 118

def by_row
  self.class.new(@table.dup).by_row!
end

#by_row! ⇒ Object

Switches the mode of this table to row mode. All calls to indexing and iteration methods will work with rows until the mode is changed again.

This method returns the table and is safe to chain.



128
129
130
131
132
 
# File 'lib/csv/table.rb', line 128

def by_row!
  @mode = :row

  self
end

#delete(*indexes_or_headers) ⇒ Object

:call-seq:

table.delete(*indexes) -> deleted_values
table.delete(*headers) -> deleted_values

If the access mode is :row or :col_or_row, and each argument is either an Integer or a Range, returns deleted rows. Otherwise, returns deleted columns data.

In either case, the returned values are in the order specified by the arguments. Arguments may be repeated.

Returns rows as an Array of CSV::Row objects.

One index:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
deleted_values = table.delete(0)
deleted_values # => [#<CSV::Row "Name":"foo" "Value":"0">]

Two indexes:

table = CSV.parse(source, headers: true)
deleted_values = table.delete(2, 0)
deleted_values # => [#<CSV::Row "Name":"baz" "Value":"2">, #<CSV::Row "Name":"foo" "Value":"0">]

Returns columns data as column Arrays.

One header:

table = CSV.parse(source, headers: true)
deleted_values = table.delete('Name')
deleted_values # => ["foo", "bar", "baz"]

Two headers:

table = CSV.parse(source, headers: true)
deleted_values = table.delete('Value', 'Name')
deleted_values # => [["0", "1", "2"], ["foo", "bar", "baz"]]



442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
 
# File 'lib/csv/table.rb', line 442

def delete(*indexes_or_headers)
  if indexes_or_headers.empty?
    raise ArgumentError, "wrong number of arguments (given 0, expected 1+)"
  end
  deleted_values = indexes_or_headers.map do |index_or_header|
    if @mode == :row or  # by index
        (@mode == :col_or_row and index_or_header.is_a? Integer)
      @table.delete_at(index_or_header)
    else                 # by header
      if index_or_header.is_a? Integer
        @headers.delete_at(index_or_header)
      else
        @headers.delete(index_or_header)
      end
      @table.map { |row| row.delete(index_or_header).last }
    end
  end
  if indexes_or_headers.size == 1
    deleted_values[0]
  else
    deleted_values
  end
end

#delete_if(&block) ⇒ Object

Removes rows or columns for which the block returns a truthy value; returns self.

Removes rows when the access mode is :row or :col_or_row; calls the block with each CSV::Row object:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.by_row! # => #<CSV::Table mode:row row_count:4>
table.size # => 3
table.delete_if {|row| row['Name'].start_with?('b') }
table.size # => 1

Removes columns when the access mode is :col; calls the block with each column as a 2-element array containing the header and an Array of column fields:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.by_col! # => #<CSV::Table mode:col row_count:4>
table.headers.size # => 2
table.delete_if {|column_data| column_data[1].include?('2') }
table.headers.size # => 1

Returns a new Enumerator if no block is given:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.delete_if # => #<Enumerator: #<CSV::Table mode:col_or_row row_count:4>:delete_if>



492
493
494
495
496
497
498
499
500
501
502
503
504
505
 
# File 'lib/csv/table.rb', line 492

def delete_if(&block)
  return enum_for(__method__) { @mode == :row or @mode == :col_or_row ? size : headers.size } unless block_given?

  if @mode == :row or @mode == :col_or_row  # by index
    @table.delete_if(&block)
  else                                      # by header
    deleted = []
    headers.each do |header|
      deleted << delete(header) if yield([header, self[header]])
    end
  end

  self # for chaining
end

#dig(index_or_header, *index_or_headers) ⇒ Object

Extracts the nested value specified by the sequence of index or header objects by calling dig at each step, returning nil if any intermediate step is nil.



602
603
604
605
606
607
608
609
610
611
612
613
614
 
# File 'lib/csv/table.rb', line 602

def dig(index_or_header, *index_or_headers)
  value = self[index_or_header]
  if value.nil?
    nil
  elsif index_or_headers.empty?
    value
  else
    unless value.respond_to?(:dig)
      raise TypeError, "#{value.class} does not have \#dig method"
    end
    value.dig(*index_or_headers)
  end
end

#each(&block) ⇒ Object

Calls the block with each row or column; returns self.

When the access mode is :row or :col_or_row, calls the block with each CSV::Row object:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.by_row! # => #<CSV::Table mode:row row_count:4>
table.each {|row| p row }

Output:

#<CSV::Row "Name":"foo" "Value":"0">
#<CSV::Row "Name":"bar" "Value":"1">
#<CSV::Row "Name":"baz" "Value":"2">

When the access mode is :col, calls the block with each column as a 2-element array containing the header and an Array of column fields:

table.by_col! # => #<CSV::Table mode:col row_count:4>
table.each {|column_data| p column_data }

Output:

["Name", ["foo", "bar", "baz"]]
["Value", ["0", "1", "2"]]

Returns a new Enumerator if no block is given:

table.each # => #<Enumerator: #<CSV::Table mode:col row_count:4>:each>



533
534
535
536
537
538
539
540
541
542
543
 
# File 'lib/csv/table.rb', line 533

def each(&block)
  return enum_for(__method__) { @mode == :col ? headers.size : size } unless block_given?

  if @mode == :col
    headers.each { |header| yield([header, self[header]]) }
  else
    @table.each(&block)
  end

  self # for chaining
end

#headers ⇒ Object

Returns the headers for the first row of this table (assumed to match all other rows). The headers Array passed to CSV::Table.new is returned for empty tables.



139
140
141
142
143
144
145
 
# File 'lib/csv/table.rb', line 139

def headers
  if @table.empty?
    @headers.dup
  else
    @table.first.headers
  end
end

#inspect ⇒ Object

Shows the mode and size of this table in a US-ASCII String.



617
618
619
 
# File 'lib/csv/table.rb', line 617

def inspect
  "#<#{self.class} mode:#{@mode} row_count:#{to_a.size}>".encode("US-ASCII")
end

#push(*rows) ⇒ Object

:call-seq:

table.push(*rows_or_arrays) -> self

A shortcut for appending multiple rows. Equivalent to:

rows.each {|row| self << row }

Each argument may be either a CSV::Row object or an Array:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
rows = [
  CSV::Row.new(table.headers, ['bat', 3]),
  ['bam', 4]
]
table.push(*rows)
table[3..4] # => [#<CSV::Row "Name":"bat" "Value":3>, #<CSV::Row "Name":"bam" "Value":4>]



396
397
398
399
400
 
# File 'lib/csv/table.rb', line 396

def push(*rows)
  rows.each { |row| self << row }

  self # for chaining
end

#to_a ⇒ Object

Returns the table as an Array of Arrays. Headers will be the first row, then all of the field rows will follow.



572
573
574
575
576
577
578
579
 
# File 'lib/csv/table.rb', line 572

def to_a
  array = [headers]
  @table.each do |row|
    array.push(row.fields) unless row.header_row?
  end

  array
end

#to_csv(write_headers: true, **options) ⇒ Object Also known as: to_s

Returns the table as a complete CSV String. Headers will be listed first, then all of the field rows.

This method assumes you want the Table.headers(), unless you explicitly pass :write_headers => false.



588
589
590
591
592
593
594
595
 
# File 'lib/csv/table.rb', line 588

def to_csv(write_headers: true, **options)
  array = write_headers ? [headers.to_csv(**options)] : []
  @table.each do |row|
    array.push(row.fields.to_csv(**options)) unless row.header_row?
  end

  array.join("")
end

#values_at(*indices_or_headers) ⇒ Object

:call-seq:

table.values_at(*indexes) -> array_of_rows
table.values_at(*headers) -> array_of_columns_data

If the access mode is :row or :col_or_row, and each argument is either an Integer or a Range, returns rows. Otherwise, returns columns data.

In either case, the returned values are in the order specified by the arguments. Arguments may be repeated.

Returns rows as an Array of CSV::Row objects.

No argument:

source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
table = CSV.parse(source, headers: true)
table.values_at # => []

One index:

values = table.values_at(0)
values # => [#<CSV::Row "Name":"foo" "Value":"0">]

Two indexes:

values = table.values_at(2, 0)
values # => [#<CSV::Row "Name":"baz" "Value":"2">, #<CSV::Row "Name":"foo" "Value":"0">]

One Range:

values = table.values_at(1..2)
values # => [#<CSV::Row "Name":"bar" "Value":"1">, #<CSV::Row "Name":"baz" "Value":"2">]

Ranges and indexes:

values = table.values_at(0..1, 1..2, 0, 2)
pp values

Output:

[#<CSV::Row "Name":"foo" "Value":"0">,
 #<CSV::Row "Name":"bar" "Value":"1">,
 #<CSV::Row "Name":"bar" "Value":"1">,
 #<CSV::Row "Name":"baz" "Value":"2">,
 #<CSV::Row "Name":"foo" "Value":"0">,
 #<CSV::Row "Name":"baz" "Value":"2">]

Returns columns data as row Arrays, each consisting of the specified columns data for that row:

values = table.values_at('Name')
values # => [["foo"], ["bar"], ["baz"]]
values = table.values_at('Value', 'Name')
values # => [["0", "foo"], ["1", "bar"], ["2", "baz"]]



342
343
344
345
346
347
348
349
350
351
352
353
354
 
# File 'lib/csv/table.rb', line 342

def values_at(*indices_or_headers)
  if @mode == :row or  # by indices
     ( @mode == :col_or_row and indices_or_headers.all? do |index|
                                  index.is_a?(Integer)         or
                                  ( index.is_a?(Range)         and
                                    index.first.is_a?(Integer) and
                                    index.last.is_a?(Integer) )
                                end )
    @table.values_at(*indices_or_headers)
  else                 # by headers
    @table.map { |row| row.values_at(*indices_or_headers) }
  end
end