|
This diff has been collapsed as it changes many lines, (1984 lines changed)
Show them
Hide them
|
|
|
@@
-0,0
+1,1984
|
|
|
|
|
|
1
|
#!/usr/local/bin/ruby -w
|
|
|
|
|
|
2
|
|
|
|
|
|
|
3
|
# = faster_csv.rb -- Faster CSV Reading and Writing
|
|
|
|
|
|
4
|
#
|
|
|
|
|
|
5
|
# Created by James Edward Gray II on 2005-10-31.
|
|
|
|
|
|
6
|
# Copyright 2005 Gray Productions. All rights reserved.
|
|
|
|
|
|
7
|
#
|
|
|
|
|
|
8
|
# See FasterCSV for documentation.
|
|
|
|
|
|
9
|
|
|
|
|
|
|
10
|
if RUBY_VERSION >= "1.9"
|
|
|
|
|
|
11
|
abort <<-VERSION_WARNING.gsub(/^\s+/, "")
|
|
|
|
|
|
12
|
Please switch to Ruby 1.9's standard CSV library. It's FasterCSV plus
|
|
|
|
|
|
13
|
support for Ruby 1.9's m17n encoding engine.
|
|
|
|
|
|
14
|
VERSION_WARNING
|
|
|
|
|
|
15
|
end
|
|
|
|
|
|
16
|
|
|
|
|
|
|
17
|
require "forwardable"
|
|
|
|
|
|
18
|
require "English"
|
|
|
|
|
|
19
|
require "enumerator"
|
|
|
|
|
|
20
|
require "date"
|
|
|
|
|
|
21
|
require "stringio"
|
|
|
|
|
|
22
|
|
|
|
|
|
|
23
|
#
|
|
|
|
|
|
24
|
# This class provides a complete interface to CSV files and data. It offers
|
|
|
|
|
|
25
|
# tools to enable you to read and write to and from Strings or IO objects, as
|
|
|
|
|
|
26
|
# needed.
|
|
|
|
|
|
27
|
#
|
|
|
|
|
|
28
|
# == Reading
|
|
|
|
|
|
29
|
#
|
|
|
|
|
|
30
|
# === From a File
|
|
|
|
|
|
31
|
#
|
|
|
|
|
|
32
|
# ==== A Line at a Time
|
|
|
|
|
|
33
|
#
|
|
|
|
|
|
34
|
# FasterCSV.foreach("path/to/file.csv") do |row|
|
|
|
|
|
|
35
|
# # use row here...
|
|
|
|
|
|
36
|
# end
|
|
|
|
|
|
37
|
#
|
|
|
|
|
|
38
|
# ==== All at Once
|
|
|
|
|
|
39
|
#
|
|
|
|
|
|
40
|
# arr_of_arrs = FasterCSV.read("path/to/file.csv")
|
|
|
|
|
|
41
|
#
|
|
|
|
|
|
42
|
# === From a String
|
|
|
|
|
|
43
|
#
|
|
|
|
|
|
44
|
# ==== A Line at a Time
|
|
|
|
|
|
45
|
#
|
|
|
|
|
|
46
|
# FasterCSV.parse("CSV,data,String") do |row|
|
|
|
|
|
|
47
|
# # use row here...
|
|
|
|
|
|
48
|
# end
|
|
|
|
|
|
49
|
#
|
|
|
|
|
|
50
|
# ==== All at Once
|
|
|
|
|
|
51
|
#
|
|
|
|
|
|
52
|
# arr_of_arrs = FasterCSV.parse("CSV,data,String")
|
|
|
|
|
|
53
|
#
|
|
|
|
|
|
54
|
# == Writing
|
|
|
|
|
|
55
|
#
|
|
|
|
|
|
56
|
# === To a File
|
|
|
|
|
|
57
|
#
|
|
|
|
|
|
58
|
# FasterCSV.open("path/to/file.csv", "w") do |csv|
|
|
|
|
|
|
59
|
# csv << ["row", "of", "CSV", "data"]
|
|
|
|
|
|
60
|
# csv << ["another", "row"]
|
|
|
|
|
|
61
|
# # ...
|
|
|
|
|
|
62
|
# end
|
|
|
|
|
|
63
|
#
|
|
|
|
|
|
64
|
# === To a String
|
|
|
|
|
|
65
|
#
|
|
|
|
|
|
66
|
# csv_string = FasterCSV.generate do |csv|
|
|
|
|
|
|
67
|
# csv << ["row", "of", "CSV", "data"]
|
|
|
|
|
|
68
|
# csv << ["another", "row"]
|
|
|
|
|
|
69
|
# # ...
|
|
|
|
|
|
70
|
# end
|
|
|
|
|
|
71
|
#
|
|
|
|
|
|
72
|
# == Convert a Single Line
|
|
|
|
|
|
73
|
#
|
|
|
|
|
|
74
|
# csv_string = ["CSV", "data"].to_csv # to CSV
|
|
|
|
|
|
75
|
# csv_array = "CSV,String".parse_csv # from CSV
|
|
|
|
|
|
76
|
#
|
|
|
|
|
|
77
|
# == Shortcut Interface
|
|
|
|
|
|
78
|
#
|
|
|
|
|
|
79
|
# FCSV { |csv_out| csv_out << %w{my data here} } # to $stdout
|
|
|
|
|
|
80
|
# FCSV(csv = "") { |csv_str| csv_str << %w{my data here} } # to a String
|
|
|
|
|
|
81
|
# FCSV($stderr) { |csv_err| csv_err << %w{my data here} } # to $stderr
|
|
|
|
|
|
82
|
#
|
|
|
|
|
|
83
|
class FasterCSV
|
|
|
|
|
|
84
|
# The version of the installed library.
|
|
|
|
|
|
85
|
VERSION = "1.5.0".freeze
|
|
|
|
|
|
86
|
|
|
|
|
|
|
87
|
#
|
|
|
|
|
|
88
|
# A FasterCSV::Row is part Array and part Hash. It retains an order for the
|
|
|
|
|
|
89
|
# fields and allows duplicates just as an Array would, but also allows you to
|
|
|
|
|
|
90
|
# access fields by name just as you could if they were in a Hash.
|
|
|
|
|
|
91
|
#
|
|
|
|
|
|
92
|
# All rows returned by FasterCSV will be constructed from this class, if
|
|
|
|
|
|
93
|
# header row processing is activated.
|
|
|
|
|
|
94
|
#
|
|
|
|
|
|
95
|
class Row
|
|
|
|
|
|
96
|
#
|
|
|
|
|
|
97
|
# Construct a new FasterCSV::Row from +headers+ and +fields+, which are
|
|
|
|
|
|
98
|
# expected to be Arrays. If one Array is shorter than the other, it will be
|
|
|
|
|
|
99
|
# padded with +nil+ objects.
|
|
|
|
|
|
100
|
#
|
|
|
|
|
|
101
|
# The optional +header_row+ parameter can be set to +true+ to indicate, via
|
|
|
|
|
|
102
|
# FasterCSV::Row.header_row?() and FasterCSV::Row.field_row?(), that this is
|
|
|
|
|
|
103
|
# a header row. Otherwise, the row is assumes to be a field row.
|
|
|
|
|
|
104
|
#
|
|
|
|
|
|
105
|
# A FasterCSV::Row object supports the following Array methods through
|
|
|
|
|
|
106
|
# delegation:
|
|
|
|
|
|
107
|
#
|
|
|
|
|
|
108
|
# * empty?()
|
|
|
|
|
|
109
|
# * length()
|
|
|
|
|
|
110
|
# * size()
|
|
|
|
|
|
111
|
#
|
|
|
|
|
|
112
|
def initialize(headers, fields, header_row = false)
|
|
|
|
|
|
113
|
@header_row = header_row
|
|
|
|
|
|
114
|
|
|
|
|
|
|
115
|
# handle extra headers or fields
|
|
|
|
|
|
116
|
@row = if headers.size > fields.size
|
|
|
|
|
|
117
|
headers.zip(fields)
|
|
|
|
|
|
118
|
else
|
|
|
|
|
|
119
|
fields.zip(headers).map { |pair| pair.reverse }
|
|
|
|
|
|
120
|
end
|
|
|
|
|
|
121
|
end
|
|
|
|
|
|
122
|
|
|
|
|
|
|
123
|
# Internal data format used to compare equality.
|
|
|
|
|
|
124
|
attr_reader :row
|
|
|
|
|
|
125
|
protected :row
|
|
|
|
|
|
126
|
|
|
|
|
|
|
127
|
### Array Delegation ###
|
|
|
|
|
|
128
|
|
|
|
|
|
|
129
|
extend Forwardable
|
|
|
|
|
|
130
|
def_delegators :@row, :empty?, :length, :size
|
|
|
|
|
|
131
|
|
|
|
|
|
|
132
|
# Returns +true+ if this is a header row.
|
|
|
|
|
|
133
|
def header_row?
|
|
|
|
|
|
134
|
@header_row
|
|
|
|
|
|
135
|
end
|
|
|
|
|
|
136
|
|
|
|
|
|
|
137
|
# Returns +true+ if this is a field row.
|
|
|
|
|
|
138
|
def field_row?
|
|
|
|
|
|
139
|
not header_row?
|
|
|
|
|
|
140
|
end
|
|
|
|
|
|
141
|
|
|
|
|
|
|
142
|
# Returns the headers of this row.
|
|
|
|
|
|
143
|
def headers
|
|
|
|
|
|
144
|
@row.map { |pair| pair.first }
|
|
|
|
|
|
145
|
end
|
|
|
|
|
|
146
|
|
|
|
|
|
|
147
|
#
|
|
|
|
|
|
148
|
# :call-seq:
|
|
|
|
|
|
149
|
# field( header )
|
|
|
|
|
|
150
|
# field( header, offset )
|
|
|
|
|
|
151
|
# field( index )
|
|
|
|
|
|
152
|
#
|
|
|
|
|
|
153
|
# This method will fetch the field value by +header+ or +index+. If a field
|
|
|
|
|
|
154
|
# is not found, +nil+ is returned.
|
|
|
|
|
|
155
|
#
|
|
|
|
|
|
156
|
# When provided, +offset+ ensures that a header match occurrs on or later
|
|
|
|
|
|
157
|
# than the +offset+ index. You can use this to find duplicate headers,
|
|
|
|
|
|
158
|
# without resorting to hard-coding exact indices.
|
|
|
|
|
|
159
|
#
|
|
|
|
|
|
160
|
def field(header_or_index, minimum_index = 0)
|
|
|
|
|
|
161
|
# locate the pair
|
|
|
|
|
|
162
|
finder = header_or_index.is_a?(Integer) ? :[] : :assoc
|
|
|
|
|
|
163
|
pair = @row[minimum_index..-1].send(finder, header_or_index)
|
|
|
|
|
|
164
|
|
|
|
|
|
|
165
|
# return the field if we have a pair
|
|
|
|
|
|
166
|
pair.nil? ? nil : pair.last
|
|
|
|
|
|
167
|
end
|
|
|
|
|
|
168
|
alias_method :[], :field
|
|
|
|
|
|
169
|
|
|
|
|
|
|
170
|
#
|
|
|
|
|
|
171
|
# :call-seq:
|
|
|
|
|
|
172
|
# []=( header, value )
|
|
|
|
|
|
173
|
# []=( header, offset, value )
|
|
|
|
|
|
174
|
# []=( index, value )
|
|
|
|
|
|
175
|
#
|
|
|
|
|
|
176
|
# Looks up the field by the semantics described in FasterCSV::Row.field()
|
|
|
|
|
|
177
|
# and assigns the +value+.
|
|
|
|
|
|
178
|
#
|
|
|
|
|
|
179
|
# Assigning past the end of the row with an index will set all pairs between
|
|
|
|
|
|
180
|
# to <tt>[nil, nil]</tt>. Assigning to an unused header appends the new
|
|
|
|
|
|
181
|
# pair.
|
|
|
|
|
|
182
|
#
|
|
|
|
|
|
183
|
def []=(*args)
|
|
|
|
|
|
184
|
value = args.pop
|
|
|
|
|
|
185
|
|
|
|
|
|
|
186
|
if args.first.is_a? Integer
|
|
|
|
|
|
187
|
if @row[args.first].nil? # extending past the end with index
|
|
|
|
|
|
188
|
@row[args.first] = [nil, value]
|
|
|
|
|
|
189
|
@row.map! { |pair| pair.nil? ? [nil, nil] : pair }
|
|
|
|
|
|
190
|
else # normal index assignment
|
|
|
|
|
|
191
|
@row[args.first][1] = value
|
|
|
|
|
|
192
|
end
|
|
|
|
|
|
193
|
else
|
|
|
|
|
|
194
|
index = index(*args)
|
|
|
|
|
|
195
|
if index.nil? # appending a field
|
|
|
|
|
|
196
|
self << [args.first, value]
|
|
|
|
|
|
197
|
else # normal header assignment
|
|
|
|
|
|
198
|
@row[index][1] = value
|
|
|
|
|
|
199
|
end
|
|
|
|
|
|
200
|
end
|
|
|
|
|
|
201
|
end
|
|
|
|
|
|
202
|
|
|
|
|
|
|
203
|
#
|
|
|
|
|
|
204
|
# :call-seq:
|
|
|
|
|
|
205
|
# <<( field )
|
|
|
|
|
|
206
|
# <<( header_and_field_array )
|
|
|
|
|
|
207
|
# <<( header_and_field_hash )
|
|
|
|
|
|
208
|
#
|
|
|
|
|
|
209
|
# If a two-element Array is provided, it is assumed to be a header and field
|
|
|
|
|
|
210
|
# and the pair is appended. A Hash works the same way with the key being
|
|
|
|
|
|
211
|
# the header and the value being the field. Anything else is assumed to be
|
|
|
|
|
|
212
|
# a lone field which is appended with a +nil+ header.
|
|
|
|
|
|
213
|
#
|
|
|
|
|
|
214
|
# This method returns the row for chaining.
|
|
|
|
|
|
215
|
#
|
|
|
|
|
|
216
|
def <<(arg)
|
|
|
|
|
|
217
|
if arg.is_a?(Array) and arg.size == 2 # appending a header and name
|
|
|
|
|
|
218
|
@row << arg
|
|
|
|
|
|
219
|
elsif arg.is_a?(Hash) # append header and name pairs
|
|
|
|
|
|
220
|
arg.each { |pair| @row << pair }
|
|
|
|
|
|
221
|
else # append field value
|
|
|
|
|
|
222
|
@row << [nil, arg]
|
|
|
|
|
|
223
|
end
|
|
|
|
|
|
224
|
|
|
|
|
|
|
225
|
self # for chaining
|
|
|
|
|
|
226
|
end
|
|
|
|
|
|
227
|
|
|
|
|
|
|
228
|
#
|
|
|
|
|
|
229
|
# A shortcut for appending multiple fields. Equivalent to:
|
|
|
|
|
|
230
|
#
|
|
|
|
|
|
231
|
# args.each { |arg| faster_csv_row << arg }
|
|
|
|
|
|
232
|
#
|
|
|
|
|
|
233
|
# This method returns the row for chaining.
|
|
|
|
|
|
234
|
#
|
|
|
|
|
|
235
|
def push(*args)
|
|
|
|
|
|
236
|
args.each { |arg| self << arg }
|
|
|
|
|
|
237
|
|
|
|
|
|
|
238
|
self # for chaining
|
|
|
|
|
|
239
|
end
|
|
|
|
|
|
240
|
|
|
|
|
|
|
241
|
#
|
|
|
|
|
|
242
|
# :call-seq:
|
|
|
|
|
|
243
|
# delete( header )
|
|
|
|
|
|
244
|
# delete( header, offset )
|
|
|
|
|
|
245
|
# delete( index )
|
|
|
|
|
|
246
|
#
|
|
|
|
|
|
247
|
# Used to remove a pair from the row by +header+ or +index+. The pair is
|
|
|
|
|
|
248
|
# located as described in FasterCSV::Row.field(). The deleted pair is
|
|
|
|
|
|
249
|
# returned, or +nil+ if a pair could not be found.
|
|
|
|
|
|
250
|
#
|
|
|
|
|
|
251
|
def delete(header_or_index, minimum_index = 0)
|
|
|
|
|
|
252
|
if header_or_index.is_a? Integer # by index
|
|
|
|
|
|
253
|
@row.delete_at(header_or_index)
|
|
|
|
|
|
254
|
else # by header
|
|
|
|
|
|
255
|
@row.delete_at(index(header_or_index, minimum_index))
|
|
|
|
|
|
256
|
end
|
|
|
|
|
|
257
|
end
|
|
|
|
|
|
258
|
|
|
|
|
|
|
259
|
#
|
|
|
|
|
|
260
|
# The provided +block+ is passed a header and field for each pair in the row
|
|
|
|
|
|
261
|
# and expected to return +true+ or +false+, depending on whether the pair
|
|
|
|
|
|
262
|
# should be deleted.
|
|
|
|
|
|
263
|
#
|
|
|
|
|
|
264
|
# This method returns the row for chaining.
|
|
|
|
|
|
265
|
#
|
|
|
|
|
|
266
|
def delete_if(&block)
|
|
|
|
|
|
267
|
@row.delete_if(&block)
|
|
|
|
|
|
268
|
|
|
|
|
|
|
269
|
self # for chaining
|
|
|
|
|
|
270
|
end
|
|
|
|
|
|
271
|
|
|
|
|
|
|
272
|
#
|
|
|
|
|
|
273
|
# This method accepts any number of arguments which can be headers, indices,
|
|
|
|
|
|
274
|
# Ranges of either, or two-element Arrays containing a header and offset.
|
|
|
|
|
|
275
|
# Each argument will be replaced with a field lookup as described in
|
|
|
|
|
|
276
|
# FasterCSV::Row.field().
|
|
|
|
|
|
277
|
#
|
|
|
|
|
|
278
|
# If called with no arguments, all fields are returned.
|
|
|
|
|
|
279
|
#
|
|
|
|
|
|
280
|
def fields(*headers_and_or_indices)
|
|
|
|
|
|
281
|
if headers_and_or_indices.empty? # return all fields--no arguments
|
|
|
|
|
|
282
|
@row.map { |pair| pair.last }
|
|
|
|
|
|
283
|
else # or work like values_at()
|
|
|
|
|
|
284
|
headers_and_or_indices.inject(Array.new) do |all, h_or_i|
|
|
|
|
|
|
285
|
all + if h_or_i.is_a? Range
|
|
|
|
|
|
286
|
index_begin = h_or_i.begin.is_a?(Integer) ? h_or_i.begin :
|
|
|
|
|
|
287
|
index(h_or_i.begin)
|
|
|
|
|
|
288
|
index_end = h_or_i.end.is_a?(Integer) ? h_or_i.end :
|
|
|
|
|
|
289
|
index(h_or_i.end)
|
|
|
|
|
|
290
|
new_range = h_or_i.exclude_end? ? (index_begin...index_end) :
|
|
|
|
|
|
291
|
(index_begin..index_end)
|
|
|
|
|
|
292
|
fields.values_at(new_range)
|
|
|
|
|
|
293
|
else
|
|
|
|
|
|
294
|
[field(*Array(h_or_i))]
|
|
|
|
|
|
295
|
end
|
|
|
|
|
|
296
|
end
|
|
|
|
|
|
297
|
end
|
|
|
|
|
|
298
|
end
|
|
|
|
|
|
299
|
alias_method :values_at, :fields
|
|
|
|
|
|
300
|
|
|
|
|
|
|
301
|
#
|
|
|
|
|
|
302
|
# :call-seq:
|
|
|
|
|
|
303
|
# index( header )
|
|
|
|
|
|
304
|
# index( header, offset )
|
|
|
|
|
|
305
|
#
|
|
|
|
|
|
306
|
# This method will return the index of a field with the provided +header+.
|
|
|
|
|
|
307
|
# The +offset+ can be used to locate duplicate header names, as described in
|
|
|
|
|
|
308
|
# FasterCSV::Row.field().
|
|
|
|
|
|
309
|
#
|
|
|
|
|
|
310
|
def index(header, minimum_index = 0)
|
|
|
|
|
|
311
|
# find the pair
|
|
|
|
|
|
312
|
index = headers[minimum_index..-1].index(header)
|
|
|
|
|
|
313
|
# return the index at the right offset, if we found one
|
|
|
|
|
|
314
|
index.nil? ? nil : index + minimum_index
|
|
|
|
|
|
315
|
end
|
|
|
|
|
|
316
|
|
|
|
|
|
|
317
|
# Returns +true+ if +name+ is a header for this row, and +false+ otherwise.
|
|
|
|
|
|
318
|
def header?(name)
|
|
|
|
|
|
319
|
headers.include? name
|
|
|
|
|
|
320
|
end
|
|
|
|
|
|
321
|
alias_method :include?, :header?
|
|
|
|
|
|
322
|
|
|
|
|
|
|
323
|
#
|
|
|
|
|
|
324
|
# Returns +true+ if +data+ matches a field in this row, and +false+
|
|
|
|
|
|
325
|
# otherwise.
|
|
|
|
|
|
326
|
#
|
|
|
|
|
|
327
|
def field?(data)
|
|
|
|
|
|
328
|
fields.include? data
|
|
|
|
|
|
329
|
end
|
|
|
|
|
|
330
|
|
|
|
|
|
|
331
|
include Enumerable
|
|
|
|
|
|
332
|
|
|
|
|
|
|
333
|
#
|
|
|
|
|
|
334
|
# Yields each pair of the row as header and field tuples (much like
|
|
|
|
|
|
335
|
# iterating over a Hash).
|
|
|
|
|
|
336
|
#
|
|
|
|
|
|
337
|
# Support for Enumerable.
|
|
|
|
|
|
338
|
#
|
|
|
|
|
|
339
|
# This method returns the row for chaining.
|
|
|
|
|
|
340
|
#
|
|
|
|
|
|
341
|
def each(&block)
|
|
|
|
|
|
342
|
@row.each(&block)
|
|
|
|
|
|
343
|
|
|
|
|
|
|
344
|
self # for chaining
|
|
|
|
|
|
345
|
end
|
|
|
|
|
|
346
|
|
|
|
|
|
|
347
|
#
|
|
|
|
|
|
348
|
# Returns +true+ if this row contains the same headers and fields in the
|
|
|
|
|
|
349
|
# same order as +other+.
|
|
|
|
|
|
350
|
#
|
|
|
|
|
|
351
|
def ==(other)
|
|
|
|
|
|
352
|
@row == other.row
|
|
|
|
|
|
353
|
end
|
|
|
|
|
|
354
|
|
|
|
|
|
|
355
|
#
|
|
|
|
|
|
356
|
# Collapses the row into a simple Hash. Be warning that this discards field
|
|
|
|
|
|
357
|
# order and clobbers duplicate fields.
|
|
|
|
|
|
358
|
#
|
|
|
|
|
|
359
|
def to_hash
|
|
|
|
|
|
360
|
# flatten just one level of the internal Array
|
|
|
|
|
|
361
|
Hash[*@row.inject(Array.new) { |ary, pair| ary.push(*pair) }]
|
|
|
|
|
|
362
|
end
|
|
|
|
|
|
363
|
|
|
|
|
|
|
364
|
#
|
|
|
|
|
|
365
|
# Returns the row as a CSV String. Headers are not used. Equivalent to:
|
|
|
|
|
|
366
|
#
|
|
|
|
|
|
367
|
# faster_csv_row.fields.to_csv( options )
|
|
|
|
|
|
368
|
#
|
|
|
|
|
|
369
|
def to_csv(options = Hash.new)
|
|
|
|
|
|
370
|
fields.to_csv(options)
|
|
|
|
|
|
371
|
end
|
|
|
|
|
|
372
|
alias_method :to_s, :to_csv
|
|
|
|
|
|
373
|
|
|
|
|
|
|
374
|
# A summary of fields, by header.
|
|
|
|
|
|
375
|
def inspect
|
|
|
|
|
|
376
|
str = "#<#{self.class}"
|
|
|
|
|
|
377
|
each do |header, field|
|
|
|
|
|
|
378
|
str << " #{header.is_a?(Symbol) ? header.to_s : header.inspect}:" <<
|
|
|
|
|
|
379
|
field.inspect
|
|
|
|
|
|
380
|
end
|
|
|
|
|
|
381
|
str << ">"
|
|
|
|
|
|
382
|
end
|
|
|
|
|
|
383
|
end
|
|
|
|
|
|
384
|
|
|
|
|
|
|
385
|
#
|
|
|
|
|
|
386
|
# A FasterCSV::Table is a two-dimensional data structure for representing CSV
|
|
|
|
|
|
387
|
# documents. Tables allow you to work with the data by row or column,
|
|
|
|
|
|
388
|
# manipulate the data, and even convert the results back to CSV, if needed.
|
|
|
|
|
|
389
|
#
|
|
|
|
|
|
390
|
# All tables returned by FasterCSV will be constructed from this class, if
|
|
|
|
|
|
391
|
# header row processing is activated.
|
|
|
|
|
|
392
|
#
|
|
|
|
|
|
393
|
class Table
|
|
|
|
|
|
394
|
#
|
|
|
|
|
|
395
|
# Construct a new FasterCSV::Table from +array_of_rows+, which are expected
|
|
|
|
|
|
396
|
# to be FasterCSV::Row objects. All rows are assumed to have the same
|
|
|
|
|
|
397
|
# headers.
|
|
|
|
|
|
398
|
#
|
|
|
|
|
|
399
|
# A FasterCSV::Table object supports the following Array methods through
|
|
|
|
|
|
400
|
# delegation:
|
|
|
|
|
|
401
|
#
|
|
|
|
|
|
402
|
# * empty?()
|
|
|
|
|
|
403
|
# * length()
|
|
|
|
|
|
404
|
# * size()
|
|
|
|
|
|
405
|
#
|
|
|
|
|
|
406
|
def initialize(array_of_rows)
|
|
|
|
|
|
407
|
@table = array_of_rows
|
|
|
|
|
|
408
|
@mode = :col_or_row
|
|
|
|
|
|
409
|
end
|
|
|
|
|
|
410
|
|
|
|
|
|
|
411
|
# The current access mode for indexing and iteration.
|
|
|
|
|
|
412
|
attr_reader :mode
|
|
|
|
|
|
413
|
|
|
|
|
|
|
414
|
# Internal data format used to compare equality.
|
|
|
|
|
|
415
|
attr_reader :table
|
|
|
|
|
|
416
|
protected :table
|
|
|
|
|
|
417
|
|
|
|
|
|
|
418
|
### Array Delegation ###
|
|
|
|
|
|
419
|
|
|
|
|
|
|
420
|
extend Forwardable
|
|
|
|
|
|
421
|
def_delegators :@table, :empty?, :length, :size
|
|
|
|
|
|
422
|
|
|
|
|
|
|
423
|
#
|
|
|
|
|
|
424
|
# Returns a duplicate table object, in column mode. This is handy for
|
|
|
|
|
|
425
|
# chaining in a single call without changing the table mode, but be aware
|
|
|
|
|
|
426
|
# that this method can consume a fair amount of memory for bigger data sets.
|
|
|
|
|
|
427
|
#
|
|
|
|
|
|
428
|
# This method returns the duplicate table for chaining. Don't chain
|
|
|
|
|
|
429
|
# destructive methods (like []=()) this way though, since you are working
|
|
|
|
|
|
430
|
# with a duplicate.
|
|
|
|
|
|
431
|
#
|
|
|
|
|
|
432
|
def by_col
|
|
|
|
|
|
433
|
self.class.new(@table.dup).by_col!
|
|
|
|
|
|
434
|
end
|
|
|
|
|
|
435
|
|
|
|
|
|
|
436
|
#
|
|
|
|
|
|
437
|
# Switches the mode of this table to column mode. All calls to indexing and
|
|
|
|
|
|
438
|
# iteration methods will work with columns until the mode is changed again.
|
|
|
|
|
|
439
|
#
|
|
|
|
|
|
440
|
# This method returns the table and is safe to chain.
|
|
|
|
|
|
441
|
#
|
|
|
|
|
|
442
|
def by_col!
|
|
|
|
|
|
443
|
@mode = :col
|
|
|
|
|
|
444
|
|
|
|
|
|
|
445
|
self
|
|
|
|
|
|
446
|
end
|
|
|
|
|
|
447
|
|
|
|
|
|
|
448
|
#
|
|
|
|
|
|
449
|
# Returns a duplicate table object, in mixed mode. This is handy for
|
|
|
|
|
|
450
|
# chaining in a single call without changing the table mode, but be aware
|
|
|
|
|
|
451
|
# that this method can consume a fair amount of memory for bigger data sets.
|
|
|
|
|
|
452
|
#
|
|
|
|
|
|
453
|
# This method returns the duplicate table for chaining. Don't chain
|
|
|
|
|
|
454
|
# destructive methods (like []=()) this way though, since you are working
|
|
|
|
|
|
455
|
# with a duplicate.
|
|
|
|
|
|
456
|
#
|
|
|
|
|
|
457
|
def by_col_or_row
|
|
|
|
|
|
458
|
self.class.new(@table.dup).by_col_or_row!
|
|
|
|
|
|
459
|
end
|
|
|
|
|
|
460
|
|
|
|
|
|
|
461
|
#
|
|
|
|
|
|
462
|
# Switches the mode of this table to mixed mode. All calls to indexing and
|
|
|
|
|
|
463
|
# iteration methods will use the default intelligent indexing system until
|
|
|
|
|
|
464
|
# the mode is changed again. In mixed mode an index is assumed to be a row
|
|
|
|
|
|
465
|
# reference while anything else is assumed to be column access by headers.
|
|
|
|
|
|
466
|
#
|
|
|
|
|
|
467
|
# This method returns the table and is safe to chain.
|
|
|
|
|
|
468
|
#
|
|
|
|
|
|
469
|
def by_col_or_row!
|
|
|
|
|
|
470
|
@mode = :col_or_row
|
|
|
|
|
|
471
|
|
|
|
|
|
|
472
|
self
|
|
|
|
|
|
473
|
end
|
|
|
|
|
|
474
|
|
|
|
|
|
|
475
|
#
|
|
|
|
|
|
476
|
# Returns a duplicate table object, in row mode. This is handy for chaining
|
|
|
|
|
|
477
|
# in a single call without changing the table mode, but be aware that this
|
|
|
|
|
|
478
|
# method can consume a fair amount of memory for bigger data sets.
|
|
|
|
|
|
479
|
#
|
|
|
|
|
|
480
|
# This method returns the duplicate table for chaining. Don't chain
|
|
|
|
|
|
481
|
# destructive methods (like []=()) this way though, since you are working
|
|
|
|
|
|
482
|
# with a duplicate.
|
|
|
|
|
|
483
|
#
|
|
|
|
|
|
484
|
def by_row
|
|
|
|
|
|
485
|
self.class.new(@table.dup).by_row!
|
|
|
|
|
|
486
|
end
|
|
|
|
|
|
487
|
|
|
|
|
|
|
488
|
#
|
|
|
|
|
|
489
|
# Switches the mode of this table to row mode. All calls to indexing and
|
|
|
|
|
|
490
|
# iteration methods will work with rows until the mode is changed again.
|
|
|
|
|
|
491
|
#
|
|
|
|
|
|
492
|
# This method returns the table and is safe to chain.
|
|
|
|
|
|
493
|
#
|
|
|
|
|
|
494
|
def by_row!
|
|
|
|
|
|
495
|
@mode = :row
|
|
|
|
|
|
496
|
|
|
|
|
|
|
497
|
self
|
|
|
|
|
|
498
|
end
|
|
|
|
|
|
499
|
|
|
|
|
|
|
500
|
#
|
|
|
|
|
|
501
|
# Returns the headers for the first row of this table (assumed to match all
|
|
|
|
|
|
502
|
# other rows). An empty Array is returned for empty tables.
|
|
|
|
|
|
503
|
#
|
|
|
|
|
|
504
|
def headers
|
|
|
|
|
|
505
|
if @table.empty?
|
|
|
|
|
|
506
|
Array.new
|
|
|
|
|
|
507
|
else
|
|
|
|
|
|
508
|
@table.first.headers
|
|
|
|
|
|
509
|
end
|
|
|
|
|
|
510
|
end
|
|
|
|
|
|
511
|
|
|
|
|
|
|
512
|
#
|
|
|
|
|
|
513
|
# In the default mixed mode, this method returns rows for index access and
|
|
|
|
|
|
514
|
# columns for header access. You can force the index association by first
|
|
|
|
|
|
515
|
# calling by_col!() or by_row!().
|
|
|
|
|
|
516
|
#
|
|
|
|
|
|
517
|
# Columns are returned as an Array of values. Altering that Array has no
|
|
|
|
|
|
518
|
# effect on the table.
|
|
|
|
|
|
519
|
#
|
|
|
|
|
|
520
|
def [](index_or_header)
|
|
|
|
|
|
521
|
if @mode == :row or # by index
|
|
|
|
|
|
522
|
(@mode == :col_or_row and index_or_header.is_a? Integer)
|
|
|
|
|
|
523
|
@table[index_or_header]
|
|
|
|
|
|
524
|
else # by header
|
|
|
|
|
|
525
|
@table.map { |row| row[index_or_header] }
|
|
|
|
|
|
526
|
end
|
|
|
|
|
|
527
|
end
|
|
|
|
|
|
528
|
|
|
|
|
|
|
529
|
#
|
|
|
|
|
|
530
|
# In the default mixed mode, this method assigns rows for index access and
|
|
|
|
|
|
531
|
# columns for header access. You can force the index association by first
|
|
|
|
|
|
532
|
# calling by_col!() or by_row!().
|
|
|
|
|
|
533
|
#
|
|
|
|
|
|
534
|
# Rows may be set to an Array of values (which will inherit the table's
|
|
|
|
|
|
535
|
# headers()) or a FasterCSV::Row.
|
|
|
|
|
|
536
|
#
|
|
|
|
|
|
537
|
# Columns may be set to a single value, which is copied to each row of the
|
|
|
|
|
|
538
|
# column, or an Array of values. Arrays of values are assigned to rows top
|
|
|
|
|
|
539
|
# to bottom in row major order. Excess values are ignored and if the Array
|
|
|
|
|
|
540
|
# does not have a value for each row the extra rows will receive a +nil+.
|
|
|
|
|
|
541
|
#
|
|
|
|
|
|
542
|
# Assigning to an existing column or row clobbers the data. Assigning to
|
|
|
|
|
|
543
|
# new columns creates them at the right end of the table.
|
|
|
|
|
|
544
|
#
|
|
|
|
|
|
545
|
def []=(index_or_header, value)
|
|
|
|
|
|
546
|
if @mode == :row or # by index
|
|
|
|
|
|
547
|
(@mode == :col_or_row and index_or_header.is_a? Integer)
|
|
|
|
|
|
548
|
if value.is_a? Array
|
|
|
|
|
|
549
|
@table[index_or_header] = Row.new(headers, value)
|
|
|
|
|
|
550
|
else
|
|
|
|
|
|
551
|
@table[index_or_header] = value
|
|
|
|
|
|
552
|
end
|
|
|
|
|
|
553
|
else # set column
|
|
|
|
|
|
554
|
if value.is_a? Array # multiple values
|
|
|
|
|
|
555
|
@table.each_with_index do |row, i|
|
|
|
|
|
|
556
|
if row.header_row?
|
|
|
|
|
|
557
|
row[index_or_header] = index_or_header
|
|
|
|
|
|
558
|
else
|
|
|
|
|
|
559
|
row[index_or_header] = value[i]
|
|
|
|
|
|
560
|
end
|
|
|
|
|
|
561
|
end
|
|
|
|
|
|
562
|
else # repeated value
|
|
|
|
|
|
563
|
@table.each do |row|
|
|
|
|
|
|
564
|
if row.header_row?
|
|
|
|
|
|
565
|
row[index_or_header] = index_or_header
|
|
|
|
|
|
566
|
else
|
|
|
|
|
|
567
|
row[index_or_header] = value
|
|
|
|
|
|
568
|
end
|
|
|
|
|
|
569
|
end
|
|
|
|
|
|
570
|
end
|
|
|
|
|
|
571
|
end
|
|
|
|
|
|
572
|
end
|
|
|
|
|
|
573
|
|
|
|
|
|
|
574
|
#
|
|
|
|
|
|
575
|
# The mixed mode default is to treat a list of indices as row access,
|
|
|
|
|
|
576
|
# returning the rows indicated. Anything else is considered columnar
|
|
|
|
|
|
577
|
# access. For columnar access, the return set has an Array for each row
|
|
|
|
|
|
578
|
# with the values indicated by the headers in each Array. You can force
|
|
|
|
|
|
579
|
# column or row mode using by_col!() or by_row!().
|
|
|
|
|
|
580
|
#
|
|
|
|
|
|
581
|
# You cannot mix column and row access.
|
|
|
|
|
|
582
|
#
|
|
|
|
|
|
583
|
def values_at(*indices_or_headers)
|
|
|
|
|
|
584
|
if @mode == :row or # by indices
|
|
|
|
|
|
585
|
( @mode == :col_or_row and indices_or_headers.all? do |index|
|
|
|
|
|
|
586
|
index.is_a?(Integer) or
|
|
|
|
|
|
587
|
( index.is_a?(Range) and
|
|
|
|
|
|
588
|
index.first.is_a?(Integer) and
|
|
|
|
|
|
589
|
index.last.is_a?(Integer) )
|
|
|
|
|
|
590
|
end )
|
|
|
|
|
|
591
|
@table.values_at(*indices_or_headers)
|
|
|
|
|
|
592
|
else # by headers
|
|
|
|
|
|
593
|
@table.map { |row| row.values_at(*indices_or_headers) }
|
|
|
|
|
|
594
|
end
|
|
|
|
|
|
595
|
end
|
|
|
|
|
|
596
|
|
|
|
|
|
|
597
|
#
|
|
|
|
|
|
598
|
# Adds a new row to the bottom end of this table. You can provide an Array,
|
|
|
|
|
|
599
|
# which will be converted to a FasterCSV::Row (inheriting the table's
|
|
|
|
|
|
600
|
# headers()), or a FasterCSV::Row.
|
|
|
|
|
|
601
|
#
|
|
|
|
|
|
602
|
# This method returns the table for chaining.
|
|
|
|
|
|
603
|
#
|
|
|
|
|
|
604
|
def <<(row_or_array)
|
|
|
|
|
|
605
|
if row_or_array.is_a? Array # append Array
|
|
|
|
|
|
606
|
@table << Row.new(headers, row_or_array)
|
|
|
|
|
|
607
|
else # append Row
|
|
|
|
|
|
608
|
@table << row_or_array
|
|
|
|
|
|
609
|
end
|
|
|
|
|
|
610
|
|
|
|
|
|
|
611
|
self # for chaining
|
|
|
|
|
|
612
|
end
|
|
|
|
|
|
613
|
|
|
|
|
|
|
614
|
#
|
|
|
|
|
|
615
|
# A shortcut for appending multiple rows. Equivalent to:
|
|
|
|
|
|
616
|
#
|
|
|
|
|
|
617
|
# rows.each { |row| self << row }
|
|
|
|
|
|
618
|
#
|
|
|
|
|
|
619
|
# This method returns the table for chaining.
|
|
|
|
|
|
620
|
#
|
|
|
|
|
|
621
|
def push(*rows)
|
|
|
|
|
|
622
|
rows.each { |row| self << row }
|
|
|
|
|
|
623
|
|
|
|
|
|
|
624
|
self # for chaining
|
|
|
|
|
|
625
|
end
|
|
|
|
|
|
626
|
|
|
|
|
|
|
627
|
#
|
|
|
|
|
|
628
|
# Removes and returns the indicated column or row. In the default mixed
|
|
|
|
|
|
629
|
# mode indices refer to rows and everything else is assumed to be a column
|
|
|
|
|
|
630
|
# header. Use by_col!() or by_row!() to force the lookup.
|
|
|
|
|
|
631
|
#
|
|
|
|
|
|
632
|
def delete(index_or_header)
|
|
|
|
|
|
633
|
if @mode == :row or # by index
|
|
|
|
|
|
634
|
(@mode == :col_or_row and index_or_header.is_a? Integer)
|
|
|
|
|
|
635
|
@table.delete_at(index_or_header)
|
|
|
|
|
|
636
|
else # by header
|
|
|
|
|
|
637
|
@table.map { |row| row.delete(index_or_header).last }
|
|
|
|
|
|
638
|
end
|
|
|
|
|
|
639
|
end
|
|
|
|
|
|
640
|
|
|
|
|
|
|
641
|
#
|
|
|
|
|
|
642
|
# Removes any column or row for which the block returns +true+. In the
|
|
|
|
|
|
643
|
# default mixed mode or row mode, iteration is the standard row major
|
|
|
|
|
|
644
|
# walking of rows. In column mode, interation will +yield+ two element
|
|
|
|
|
|
645
|
# tuples containing the column name and an Array of values for that column.
|
|
|
|
|
|
646
|
#
|
|
|
|
|
|
647
|
# This method returns the table for chaining.
|
|
|
|
|
|
648
|
#
|
|
|
|
|
|
649
|
def delete_if(&block)
|
|
|
|
|
|
650
|
if @mode == :row or @mode == :col_or_row # by index
|
|
|
|
|
|
651
|
@table.delete_if(&block)
|
|
|
|
|
|
652
|
else # by header
|
|
|
|
|
|
653
|
to_delete = Array.new
|
|
|
|
|
|
654
|
headers.each_with_index do |header, i|
|
|
|
|
|
|
655
|
to_delete << header if block[[header, self[header]]]
|
|
|
|
|
|
656
|
end
|
|
|
|
|
|
657
|
to_delete.map { |header| delete(header) }
|
|
|
|
|
|
658
|
end
|
|
|
|
|
|
659
|
|
|
|
|
|
|
660
|
self # for chaining
|
|
|
|
|
|
661
|
end
|
|
|
|
|
|
662
|
|
|
|
|
|
|
663
|
include Enumerable
|
|
|
|
|
|
664
|
|
|
|
|
|
|
665
|
#
|
|
|
|
|
|
666
|
# In the default mixed mode or row mode, iteration is the standard row major
|
|
|
|
|
|
667
|
# walking of rows. In column mode, interation will +yield+ two element
|
|
|
|
|
|
668
|
# tuples containing the column name and an Array of values for that column.
|
|
|
|
|
|
669
|
#
|
|
|
|
|
|
670
|
# This method returns the table for chaining.
|
|
|
|
|
|
671
|
#
|
|
|
|
|
|
672
|
def each(&block)
|
|
|
|
|
|
673
|
if @mode == :col
|
|
|
|
|
|
674
|
headers.each { |header| block[[header, self[header]]] }
|
|
|
|
|
|
675
|
else
|
|
|
|
|
|
676
|
@table.each(&block)
|
|
|
|
|
|
677
|
end
|
|
|
|
|
|
678
|
|
|
|
|
|
|
679
|
self # for chaining
|
|
|
|
|
|
680
|
end
|
|
|
|
|
|
681
|
|
|
|
|
|
|
682
|
# Returns +true+ if all rows of this table ==() +other+'s rows.
|
|
|
|
|
|
683
|
def ==(other)
|
|
|
|
|
|
684
|
@table == other.table
|
|
|
|
|
|
685
|
end
|
|
|
|
|
|
686
|
|
|
|
|
|
|
687
|
#
|
|
|
|
|
|
688
|
# Returns the table as an Array of Arrays. Headers will be the first row,
|
|
|
|
|
|
689
|
# then all of the field rows will follow.
|
|
|
|
|
|
690
|
#
|
|
|
|
|
|
691
|
def to_a
|
|
|
|
|
|
692
|
@table.inject([headers]) do |array, row|
|
|
|
|
|
|
693
|
if row.header_row?
|
|
|
|
|
|
694
|
array
|
|
|
|
|
|
695
|
else
|
|
|
|
|
|
696
|
array + [row.fields]
|
|
|
|
|
|
697
|
end
|
|
|
|
|
|
698
|
end
|
|
|
|
|
|
699
|
end
|
|
|
|
|
|
700
|
|
|
|
|
|
|
701
|
#
|
|
|
|
|
|
702
|
# Returns the table as a complete CSV String. Headers will be listed first,
|
|
|
|
|
|
703
|
# then all of the field rows.
|
|
|
|
|
|
704
|
#
|
|
|
|
|
|
705
|
def to_csv(options = Hash.new)
|
|
|
|
|
|
706
|
@table.inject([headers.to_csv(options)]) do |rows, row|
|
|
|
|
|
|
707
|
if row.header_row?
|
|
|
|
|
|
708
|
rows
|
|
|
|
|
|
709
|
else
|
|
|
|
|
|
710
|
rows + [row.fields.to_csv(options)]
|
|
|
|
|
|
711
|
end
|
|
|
|
|
|
712
|
end.join
|
|
|
|
|
|
713
|
end
|
|
|
|
|
|
714
|
alias_method :to_s, :to_csv
|
|
|
|
|
|
715
|
|
|
|
|
|
|
716
|
def inspect
|
|
|
|
|
|
717
|
"#<#{self.class} mode:#{@mode} row_count:#{to_a.size}>"
|
|
|
|
|
|
718
|
end
|
|
|
|
|
|
719
|
end
|
|
|
|
|
|
720
|
|
|
|
|
|
|
721
|
# The error thrown when the parser encounters illegal CSV formatting.
|
|
|
|
|
|
722
|
class MalformedCSVError < RuntimeError; end
|
|
|
|
|
|
723
|
|
|
|
|
|
|
724
|
#
|
|
|
|
|
|
725
|
# A FieldInfo Struct contains details about a field's position in the data
|
|
|
|
|
|
726
|
# source it was read from. FasterCSV will pass this Struct to some blocks
|
|
|
|
|
|
727
|
# that make decisions based on field structure. See
|
|
|
|
|
|
728
|
# FasterCSV.convert_fields() for an example.
|
|
|
|
|
|
729
|
#
|
|
|
|
|
|
730
|
# <b><tt>index</tt></b>:: The zero-based index of the field in its row.
|
|
|
|
|
|
731
|
# <b><tt>line</tt></b>:: The line of the data source this row is from.
|
|
|
|
|
|
732
|
# <b><tt>header</tt></b>:: The header for the column, when available.
|
|
|
|
|
|
733
|
#
|
|
|
|
|
|
734
|
FieldInfo = Struct.new(:index, :line, :header)
|
|
|
|
|
|
735
|
|
|
|
|
|
|
736
|
# A Regexp used to find and convert some common Date formats.
|
|
|
|
|
|
737
|
DateMatcher = / \A(?: (\w+,?\s+)?\w+\s+\d{1,2},?\s+\d{2,4} |
|
|
|
|
|
|
738
|
\d{4}-\d{2}-\d{2} )\z /x
|
|
|
|
|
|
739
|
# A Regexp used to find and convert some common DateTime formats.
|
|
|
|
|
|
740
|
DateTimeMatcher =
|
|
|
|
|
|
741
|
/ \A(?: (\w+,?\s+)?\w+\s+\d{1,2}\s+\d{1,2}:\d{1,2}:\d{1,2},?\s+\d{2,4} |
|
|
|
|
|
|
742
|
\d{4}-\d{2}-\d{2}\s\d{2}:\d{2}:\d{2} )\z /x
|
|
|
|
|
|
743
|
#
|
|
|
|
|
|
744
|
# This Hash holds the built-in converters of FasterCSV that can be accessed by
|
|
|
|
|
|
745
|
# name. You can select Converters with FasterCSV.convert() or through the
|
|
|
|
|
|
746
|
# +options+ Hash passed to FasterCSV::new().
|
|
|
|
|
|
747
|
#
|
|
|
|
|
|
748
|
# <b><tt>:integer</tt></b>:: Converts any field Integer() accepts.
|
|
|
|
|
|
749
|
# <b><tt>:float</tt></b>:: Converts any field Float() accepts.
|
|
|
|
|
|
750
|
# <b><tt>:numeric</tt></b>:: A combination of <tt>:integer</tt>
|
|
|
|
|
|
751
|
# and <tt>:float</tt>.
|
|
|
|
|
|
752
|
# <b><tt>:date</tt></b>:: Converts any field Date::parse() accepts.
|
|
|
|
|
|
753
|
# <b><tt>:date_time</tt></b>:: Converts any field DateTime::parse() accepts.
|
|
|
|
|
|
754
|
# <b><tt>:all</tt></b>:: All built-in converters. A combination of
|
|
|
|
|
|
755
|
# <tt>:date_time</tt> and <tt>:numeric</tt>.
|
|
|
|
|
|
756
|
#
|
|
|
|
|
|
757
|
# This Hash is intetionally left unfrozen and users should feel free to add
|
|
|
|
|
|
758
|
# values to it that can be accessed by all FasterCSV objects.
|
|
|
|
|
|
759
|
#
|
|
|
|
|
|
760
|
# To add a combo field, the value should be an Array of names. Combo fields
|
|
|
|
|
|
761
|
# can be nested with other combo fields.
|
|
|
|
|
|
762
|
#
|
|
|
|
|
|
763
|
Converters = { :integer => lambda { |f| Integer(f) rescue f },
|
|
|
|
|
|
764
|
:float => lambda { |f| Float(f) rescue f },
|
|
|
|
|
|
765
|
:numeric => [:integer, :float],
|
|
|
|
|
|
766
|
:date => lambda { |f|
|
|
|
|
|
|
767
|
f =~ DateMatcher ? (Date.parse(f) rescue f) : f
|
|
|
|
|
|
768
|
},
|
|
|
|
|
|
769
|
:date_time => lambda { |f|
|
|
|
|
|
|
770
|
f =~ DateTimeMatcher ? (DateTime.parse(f) rescue f) : f
|
|
|
|
|
|
771
|
},
|
|
|
|
|
|
772
|
:all => [:date_time, :numeric] }
|
|
|
|
|
|
773
|
|
|
|
|
|
|
774
|
#
|
|
|
|
|
|
775
|
# This Hash holds the built-in header converters of FasterCSV that can be
|
|
|
|
|
|
776
|
# accessed by name. You can select HeaderConverters with
|
|
|
|
|
|
777
|
# FasterCSV.header_convert() or through the +options+ Hash passed to
|
|
|
|
|
|
778
|
# FasterCSV::new().
|
|
|
|
|
|
779
|
#
|
|
|
|
|
|
780
|
# <b><tt>:downcase</tt></b>:: Calls downcase() on the header String.
|
|
|
|
|
|
781
|
# <b><tt>:symbol</tt></b>:: The header String is downcased, spaces are
|
|
|
|
|
|
782
|
# replaced with underscores, non-word characters
|
|
|
|
|
|
783
|
# are dropped, and finally to_sym() is called.
|
|
|
|
|
|
784
|
#
|
|
|
|
|
|
785
|
# This Hash is intetionally left unfrozen and users should feel free to add
|
|
|
|
|
|
786
|
# values to it that can be accessed by all FasterCSV objects.
|
|
|
|
|
|
787
|
#
|
|
|
|
|
|
788
|
# To add a combo field, the value should be an Array of names. Combo fields
|
|
|
|
|
|
789
|
# can be nested with other combo fields.
|
|
|
|
|
|
790
|
#
|
|
|
|
|
|
791
|
HeaderConverters = {
|
|
|
|
|
|
792
|
:downcase => lambda { |h| h.downcase },
|
|
|
|
|
|
793
|
:symbol => lambda { |h|
|
|
|
|
|
|
794
|
h.downcase.tr(" ", "_").delete("^a-z0-9_").to_sym
|
|
|
|
|
|
795
|
}
|
|
|
|
|
|
796
|
}
|
|
|
|
|
|
797
|
|
|
|
|
|
|
798
|
#
|
|
|
|
|
|
799
|
# The options used when no overrides are given by calling code. They are:
|
|
|
|
|
|
800
|
#
|
|
|
|
|
|
801
|
# <b><tt>:col_sep</tt></b>:: <tt>","</tt>
|
|
|
|
|
|
802
|
# <b><tt>:row_sep</tt></b>:: <tt>:auto</tt>
|
|
|
|
|
|
803
|
# <b><tt>:quote_char</tt></b>:: <tt>'"'</tt>
|
|
|
|
|
|
804
|
# <b><tt>:converters</tt></b>:: +nil+
|
|
|
|
|
|
805
|
# <b><tt>:unconverted_fields</tt></b>:: +nil+
|
|
|
|
|
|
806
|
# <b><tt>:headers</tt></b>:: +false+
|
|
|
|
|
|
807
|
# <b><tt>:return_headers</tt></b>:: +false+
|
|
|
|
|
|
808
|
# <b><tt>:header_converters</tt></b>:: +nil+
|
|
|
|
|
|
809
|
# <b><tt>:skip_blanks</tt></b>:: +false+
|
|
|
|
|
|
810
|
# <b><tt>:force_quotes</tt></b>:: +false+
|
|
|
|
|
|
811
|
#
|
|
|
|
|
|
812
|
DEFAULT_OPTIONS = { :col_sep => ",",
|
|
|
|
|
|
813
|
:row_sep => :auto,
|
|
|
|
|
|
814
|
:quote_char => '"',
|
|
|
|
|
|
815
|
:converters => nil,
|
|
|
|
|
|
816
|
:unconverted_fields => nil,
|
|
|
|
|
|
817
|
:headers => false,
|
|
|
|
|
|
818
|
:return_headers => false,
|
|
|
|
|
|
819
|
:header_converters => nil,
|
|
|
|
|
|
820
|
:skip_blanks => false,
|
|
|
|
|
|
821
|
:force_quotes => false }.freeze
|
|
|
|
|
|
822
|
|
|
|
|
|
|
823
|
#
|
|
|
|
|
|
824
|
# This method will build a drop-in replacement for many of the standard CSV
|
|
|
|
|
|
825
|
# methods. It allows you to write code like:
|
|
|
|
|
|
826
|
#
|
|
|
|
|
|
827
|
# begin
|
|
|
|
|
|
828
|
# require "faster_csv"
|
|
|
|
|
|
829
|
# FasterCSV.build_csv_interface
|
|
|
|
|
|
830
|
# rescue LoadError
|
|
|
|
|
|
831
|
# require "csv"
|
|
|
|
|
|
832
|
# end
|
|
|
|
|
|
833
|
# # ... use CSV here ...
|
|
|
|
|
|
834
|
#
|
|
|
|
|
|
835
|
# This is not a complete interface with completely identical behavior.
|
|
|
|
|
|
836
|
# However, it is intended to be close enough that you won't notice the
|
|
|
|
|
|
837
|
# difference in most cases. CSV methods supported are:
|
|
|
|
|
|
838
|
#
|
|
|
|
|
|
839
|
# * foreach()
|
|
|
|
|
|
840
|
# * generate_line()
|
|
|
|
|
|
841
|
# * open()
|
|
|
|
|
|
842
|
# * parse()
|
|
|
|
|
|
843
|
# * parse_line()
|
|
|
|
|
|
844
|
# * readlines()
|
|
|
|
|
|
845
|
#
|
|
|
|
|
|
846
|
# Be warned that this interface is slower than vanilla FasterCSV due to the
|
|
|
|
|
|
847
|
# extra layer of method calls. Depending on usage, this can slow it down to
|
|
|
|
|
|
848
|
# near CSV speeds.
|
|
|
|
|
|
849
|
#
|
|
|
|
|
|
850
|
def self.build_csv_interface
|
|
|
|
|
|
851
|
Object.const_set(:CSV, Class.new).class_eval do
|
|
|
|
|
|
852
|
def self.foreach(path, rs = :auto, &block) # :nodoc:
|
|
|
|
|
|
853
|
FasterCSV.foreach(path, :row_sep => rs, &block)
|
|
|
|
|
|
854
|
end
|
|
|
|
|
|
855
|
|
|
|
|
|
|
856
|
def self.generate_line(row, fs = ",", rs = "") # :nodoc:
|
|
|
|
|
|
857
|
FasterCSV.generate_line(row, :col_sep => fs, :row_sep => rs)
|
|
|
|
|
|
858
|
end
|
|
|
|
|
|
859
|
|
|
|
|
|
|
860
|
def self.open(path, mode, fs = ",", rs = :auto, &block) # :nodoc:
|
|
|
|
|
|
861
|
if block and mode.include? "r"
|
|
|
|
|
|
862
|
FasterCSV.open(path, mode, :col_sep => fs, :row_sep => rs) do |csv|
|
|
|
|
|
|
863
|
csv.each(&block)
|
|
|
|
|
|
864
|
end
|
|
|
|
|
|
865
|
else
|
|
|
|
|
|
866
|
FasterCSV.open(path, mode, :col_sep => fs, :row_sep => rs, &block)
|
|
|
|
|
|
867
|
end
|
|
|
|
|
|
868
|
end
|
|
|
|
|
|
869
|
|
|
|
|
|
|
870
|
def self.parse(str_or_readable, fs = ",", rs = :auto, &block) # :nodoc:
|
|
|
|
|
|
871
|
FasterCSV.parse(str_or_readable, :col_sep => fs, :row_sep => rs, &block)
|
|
|
|
|
|
872
|
end
|
|
|
|
|
|
873
|
|
|
|
|
|
|
874
|
def self.parse_line(src, fs = ",", rs = :auto) # :nodoc:
|
|
|
|
|
|
875
|
FasterCSV.parse_line(src, :col_sep => fs, :row_sep => rs)
|
|
|
|
|
|
876
|
end
|
|
|
|
|
|
877
|
|
|
|
|
|
|
878
|
def self.readlines(path, rs = :auto) # :nodoc:
|
|
|
|
|
|
879
|
FasterCSV.readlines(path, :row_sep => rs)
|
|
|
|
|
|
880
|
end
|
|
|
|
|
|
881
|
end
|
|
|
|
|
|
882
|
end
|
|
|
|
|
|
883
|
|
|
|
|
|
|
884
|
#
|
|
|
|
|
|
885
|
# This method allows you to serialize an Array of Ruby objects to a String or
|
|
|
|
|
|
886
|
# File of CSV data. This is not as powerful as Marshal or YAML, but perhaps
|
|
|
|
|
|
887
|
# useful for spreadsheet and database interaction.
|
|
|
|
|
|
888
|
#
|
|
|
|
|
|
889
|
# Out of the box, this method is intended to work with simple data objects or
|
|
|
|
|
|
890
|
# Structs. It will serialize a list of instance variables and/or
|
|
|
|
|
|
891
|
# Struct.members().
|
|
|
|
|
|
892
|
#
|
|
|
|
|
|
893
|
# If you need need more complicated serialization, you can control the process
|
|
|
|
|
|
894
|
# by adding methods to the class to be serialized.
|
|
|
|
|
|
895
|
#
|
|
|
|
|
|
896
|
# A class method csv_meta() is responsible for returning the first row of the
|
|
|
|
|
|
897
|
# document (as an Array). This row is considered to be a Hash of the form
|
|
|
|
|
|
898
|
# key_1,value_1,key_2,value_2,... FasterCSV::load() expects to find a class
|
|
|
|
|
|
899
|
# key with a value of the stringified class name and FasterCSV::dump() will
|
|
|
|
|
|
900
|
# create this, if you do not define this method. This method is only called
|
|
|
|
|
|
901
|
# on the first object of the Array.
|
|
|
|
|
|
902
|
#
|
|
|
|
|
|
903
|
# The next method you can provide is an instance method called csv_headers().
|
|
|
|
|
|
904
|
# This method is expected to return the second line of the document (again as
|
|
|
|
|
|
905
|
# an Array), which is to be used to give each column a header. By default,
|
|
|
|
|
|
906
|
# FasterCSV::load() will set an instance variable if the field header starts
|
|
|
|
|
|
907
|
# with an @ character or call send() passing the header as the method name and
|
|
|
|
|
|
908
|
# the field value as an argument. This method is only called on the first
|
|
|
|
|
|
909
|
# object of the Array.
|
|
|
|
|
|
910
|
#
|
|
|
|
|
|
911
|
# Finally, you can provide an instance method called csv_dump(), which will
|
|
|
|
|
|
912
|
# be passed the headers. This should return an Array of fields that can be
|
|
|
|
|
|
913
|
# serialized for this object. This method is called once for every object in
|
|
|
|
|
|
914
|
# the Array.
|
|
|
|
|
|
915
|
#
|
|
|
|
|
|
916
|
# The +io+ parameter can be used to serialize to a File, and +options+ can be
|
|
|
|
|
|
917
|
# anything FasterCSV::new() accepts.
|
|
|
|
|
|
918
|
#
|
|
|
|
|
|
919
|
def self.dump(ary_of_objs, io = "", options = Hash.new)
|
|
|
|
|
|
920
|
obj_template = ary_of_objs.first
|
|
|
|
|
|
921
|
|
|
|
|
|
|
922
|
csv = FasterCSV.new(io, options)
|
|
|
|
|
|
923
|
|
|
|
|
|
|
924
|
# write meta information
|
|
|
|
|
|
925
|
begin
|
|
|
|
|
|
926
|
csv << obj_template.class.csv_meta
|
|
|
|
|
|
927
|
rescue NoMethodError
|
|
|
|
|
|
928
|
csv << [:class, obj_template.class]
|
|
|
|
|
|
929
|
end
|
|
|
|
|
|
930
|
|
|
|
|
|
|
931
|
# write headers
|
|
|
|
|
|
932
|
begin
|
|
|
|
|
|
933
|
headers = obj_template.csv_headers
|
|
|
|
|
|
934
|
rescue NoMethodError
|
|
|
|
|
|
935
|
headers = obj_template.instance_variables.sort
|
|
|
|
|
|
936
|
if obj_template.class.ancestors.find { |cls| cls.to_s =~ /\AStruct\b/ }
|
|
|
|
|
|
937
|
headers += obj_template.members.map { |mem| "#{mem}=" }.sort
|
|
|
|
|
|
938
|
end
|
|
|
|
|
|
939
|
end
|
|
|
|
|
|
940
|
csv << headers
|
|
|
|
|
|
941
|
|
|
|
|
|
|
942
|
# serialize each object
|
|
|
|
|
|
943
|
ary_of_objs.each do |obj|
|
|
|
|
|
|
944
|
begin
|
|
|
|
|
|
945
|
csv << obj.csv_dump(headers)
|
|
|
|
|
|
946
|
rescue NoMethodError
|
|
|
|
|
|
947
|
csv << headers.map do |var|
|
|
|
|
|
|
948
|
if var[0] == ?@
|
|
|
|
|
|
949
|
obj.instance_variable_get(var)
|
|
|
|
|
|
950
|
else
|
|
|
|
|
|
951
|
obj[var[0..-2]]
|
|
|
|
|
|
952
|
end
|
|
|
|
|
|
953
|
end
|
|
|
|
|
|
954
|
end
|
|
|
|
|
|
955
|
end
|
|
|
|
|
|
956
|
|
|
|
|
|
|
957
|
if io.is_a? String
|
|
|
|
|
|
958
|
csv.string
|
|
|
|
|
|
959
|
else
|
|
|
|
|
|
960
|
csv.close
|
|
|
|
|
|
961
|
end
|
|
|
|
|
|
962
|
end
|
|
|
|
|
|
963
|
|
|
|
|
|
|
964
|
#
|
|
|
|
|
|
965
|
# :call-seq:
|
|
|
|
|
|
966
|
# filter( options = Hash.new ) { |row| ... }
|
|
|
|
|
|
967
|
# filter( input, options = Hash.new ) { |row| ... }
|
|
|
|
|
|
968
|
# filter( input, output, options = Hash.new ) { |row| ... }
|
|
|
|
|
|
969
|
#
|
|
|
|
|
|
970
|
# This method is a convenience for building Unix-like filters for CSV data.
|
|
|
|
|
|
971
|
# Each row is yielded to the provided block which can alter it as needed.
|
|
|
|
|
|
972
|
# After the block returns, the row is appended to +output+ altered or not.
|
|
|
|
|
|
973
|
#
|
|
|
|
|
|
974
|
# The +input+ and +output+ arguments can be anything FasterCSV::new() accepts
|
|
|
|
|
|
975
|
# (generally String or IO objects). If not given, they default to
|
|
|
|
|
|
976
|
# <tt>ARGF</tt> and <tt>$stdout</tt>.
|
|
|
|
|
|
977
|
#
|
|
|
|
|
|
978
|
# The +options+ parameter is also filtered down to FasterCSV::new() after some
|
|
|
|
|
|
979
|
# clever key parsing. Any key beginning with <tt>:in_</tt> or
|
|
|
|
|
|
980
|
# <tt>:input_</tt> will have that leading identifier stripped and will only
|
|
|
|
|
|
981
|
# be used in the +options+ Hash for the +input+ object. Keys starting with
|
|
|
|
|
|
982
|
# <tt>:out_</tt> or <tt>:output_</tt> affect only +output+. All other keys
|
|
|
|
|
|
983
|
# are assigned to both objects.
|
|
|
|
|
|
984
|
#
|
|
|
|
|
|
985
|
# The <tt>:output_row_sep</tt> +option+ defaults to
|
|
|
|
|
|
986
|
# <tt>$INPUT_RECORD_SEPARATOR</tt> (<tt>$/</tt>).
|
|
|
|
|
|
987
|
#
|
|
|
|
|
|
988
|
def self.filter(*args)
|
|
|
|
|
|
989
|
# parse options for input, output, or both
|
|
|
|
|
|
990
|
in_options, out_options = Hash.new, {:row_sep => $INPUT_RECORD_SEPARATOR}
|
|
|
|
|
|
991
|
if args.last.is_a? Hash
|
|
|
|
|
|
992
|
args.pop.each do |key, value|
|
|
|
|
|
|
993
|
case key.to_s
|
|
|
|
|
|
994
|
when /\Ain(?:put)?_(.+)\Z/
|
|
|
|
|
|
995
|
in_options[$1.to_sym] = value
|
|
|
|
|
|
996
|
when /\Aout(?:put)?_(.+)\Z/
|
|
|
|
|
|
997
|
out_options[$1.to_sym] = value
|
|
|
|
|
|
998
|
else
|
|
|
|
|
|
999
|
in_options[key] = value
|
|
|
|
|
|
1000
|
out_options[key] = value
|
|
|
|
|
|
1001
|
end
|
|
|
|
|
|
1002
|
end
|
|
|
|
|
|
1003
|
end
|
|
|
|
|
|
1004
|
# build input and output wrappers
|
|
|
|
|
|
1005
|
input = FasterCSV.new(args.shift || ARGF, in_options)
|
|
|
|
|
|
1006
|
output = FasterCSV.new(args.shift || $stdout, out_options)
|
|
|
|
|
|
1007
|
|
|
|
|
|
|
1008
|
# read, yield, write
|
|
|
|
|
|
1009
|
input.each do |row|
|
|
|
|
|
|
1010
|
yield row
|
|
|
|
|
|
1011
|
output << row
|
|
|
|
|
|
1012
|
end
|
|
|
|
|
|
1013
|
end
|
|
|
|
|
|
1014
|
|
|
|
|
|
|
1015
|
#
|
|
|
|
|
|
1016
|
# This method is intended as the primary interface for reading CSV files. You
|
|
|
|
|
|
1017
|
# pass a +path+ and any +options+ you wish to set for the read. Each row of
|
|
|
|
|
|
1018
|
# file will be passed to the provided +block+ in turn.
|
|
|
|
|
|
1019
|
#
|
|
|
|
|
|
1020
|
# The +options+ parameter can be anything FasterCSV::new() understands.
|
|
|
|
|
|
1021
|
#
|
|
|
|
|
|
1022
|
def self.foreach(path, options = Hash.new, &block)
|
|
|
|
|
|
1023
|
open(path, "rb", options) do |csv|
|
|
|
|
|
|
1024
|
csv.each(&block)
|
|
|
|
|
|
1025
|
end
|
|
|
|
|
|
1026
|
end
|
|
|
|
|
|
1027
|
|
|
|
|
|
|
1028
|
#
|
|
|
|
|
|
1029
|
# :call-seq:
|
|
|
|
|
|
1030
|
# generate( str, options = Hash.new ) { |faster_csv| ... }
|
|
|
|
|
|
1031
|
# generate( options = Hash.new ) { |faster_csv| ... }
|
|
|
|
|
|
1032
|
#
|
|
|
|
|
|
1033
|
# This method wraps a String you provide, or an empty default String, in a
|
|
|
|
|
|
1034
|
# FasterCSV object which is passed to the provided block. You can use the
|
|
|
|
|
|
1035
|
# block to append CSV rows to the String and when the block exits, the
|
|
|
|
|
|
1036
|
# final String will be returned.
|
|
|
|
|
|
1037
|
#
|
|
|
|
|
|
1038
|
# Note that a passed String *is* modfied by this method. Call dup() before
|
|
|
|
|
|
1039
|
# passing if you need a new String.
|
|
|
|
|
|
1040
|
#
|
|
|
|
|
|
1041
|
# The +options+ parameter can be anthing FasterCSV::new() understands.
|
|
|
|
|
|
1042
|
#
|
|
|
|
|
|
1043
|
def self.generate(*args)
|
|
|
|
|
|
1044
|
# add a default empty String, if none was given
|
|
|
|
|
|
1045
|
if args.first.is_a? String
|
|
|
|
|
|
1046
|
io = StringIO.new(args.shift)
|
|
|
|
|
|
1047
|
io.seek(0, IO::SEEK_END)
|
|
|
|
|
|
1048
|
args.unshift(io)
|
|
|
|
|
|
1049
|
else
|
|
|
|
|
|
1050
|
args.unshift("")
|
|
|
|
|
|
1051
|
end
|
|
|
|
|
|
1052
|
faster_csv = new(*args) # wrap
|
|
|
|
|
|
1053
|
yield faster_csv # yield for appending
|
|
|
|
|
|
1054
|
faster_csv.string # return final String
|
|
|
|
|
|
1055
|
end
|
|
|
|
|
|
1056
|
|
|
|
|
|
|
1057
|
#
|
|
|
|
|
|
1058
|
# This method is a shortcut for converting a single row (Array) into a CSV
|
|
|
|
|
|
1059
|
# String.
|
|
|
|
|
|
1060
|
#
|
|
|
|
|
|
1061
|
# The +options+ parameter can be anthing FasterCSV::new() understands.
|
|
|
|
|
|
1062
|
#
|
|
|
|
|
|
1063
|
# The <tt>:row_sep</tt> +option+ defaults to <tt>$INPUT_RECORD_SEPARATOR</tt>
|
|
|
|
|
|
1064
|
# (<tt>$/</tt>) when calling this method.
|
|
|
|
|
|
1065
|
#
|
|
|
|
|
|
1066
|
def self.generate_line(row, options = Hash.new)
|
|
|
|
|
|
1067
|
options = {:row_sep => $INPUT_RECORD_SEPARATOR}.merge(options)
|
|
|
|
|
|
1068
|
(new("", options) << row).string
|
|
|
|
|
|
1069
|
end
|
|
|
|
|
|
1070
|
|
|
|
|
|
|
1071
|
#
|
|
|
|
|
|
1072
|
# This method will return a FasterCSV instance, just like FasterCSV::new(),
|
|
|
|
|
|
1073
|
# but the instance will be cached and returned for all future calls to this
|
|
|
|
|
|
1074
|
# method for the same +data+ object (tested by Object#object_id()) with the
|
|
|
|
|
|
1075
|
# same +options+.
|
|
|
|
|
|
1076
|
#
|
|
|
|
|
|
1077
|
# If a block is given, the instance is passed to the block and the return
|
|
|
|
|
|
1078
|
# value becomes the return value of the block.
|
|
|
|
|
|
1079
|
#
|
|
|
|
|
|
1080
|
def self.instance(data = $stdout, options = Hash.new)
|
|
|
|
|
|
1081
|
# create a _signature_ for this method call, data object and options
|
|
|
|
|
|
1082
|
sig = [data.object_id] +
|
|
|
|
|
|
1083
|
options.values_at(*DEFAULT_OPTIONS.keys.sort_by { |sym| sym.to_s })
|
|
|
|
|
|
1084
|
|
|
|
|
|
|
1085
|
# fetch or create the instance for this signature
|
|
|
|
|
|
1086
|
@@instances ||= Hash.new
|
|
|
|
|
|
1087
|
instance = (@@instances[sig] ||= new(data, options))
|
|
|
|
|
|
1088
|
|
|
|
|
|
|
1089
|
if block_given?
|
|
|
|
|
|
1090
|
yield instance # run block, if given, returning result
|
|
|
|
|
|
1091
|
else
|
|
|
|
|
|
1092
|
instance # or return the instance
|
|
|
|
|
|
1093
|
end
|
|
|
|
|
|
1094
|
end
|
|
|
|
|
|
1095
|
|
|
|
|
|
|
1096
|
#
|
|
|
|
|
|
1097
|
# This method is the reading counterpart to FasterCSV::dump(). See that
|
|
|
|
|
|
1098
|
# method for a detailed description of the process.
|
|
|
|
|
|
1099
|
#
|
|
|
|
|
|
1100
|
# You can customize loading by adding a class method called csv_load() which
|
|
|
|
|
|
1101
|
# will be passed a Hash of meta information, an Array of headers, and an Array
|
|
|
|
|
|
1102
|
# of fields for the object the method is expected to return.
|
|
|
|
|
|
1103
|
#
|
|
|
|
|
|
1104
|
# Remember that all fields will be Strings after this load. If you need
|
|
|
|
|
|
1105
|
# something else, use +options+ to setup converters or provide a custom
|
|
|
|
|
|
1106
|
# csv_load() implementation.
|
|
|
|
|
|
1107
|
#
|
|
|
|
|
|
1108
|
def self.load(io_or_str, options = Hash.new)
|
|
|
|
|
|
1109
|
csv = FasterCSV.new(io_or_str, options)
|
|
|
|
|
|
1110
|
|
|
|
|
|
|
1111
|
# load meta information
|
|
|
|
|
|
1112
|
meta = Hash[*csv.shift]
|
|
|
|
|
|
1113
|
cls = meta["class"].split("::").inject(Object) do |c, const|
|
|
|
|
|
|
1114
|
c.const_get(const)
|
|
|
|
|
|
1115
|
end
|
|
|
|
|
|
1116
|
|
|
|
|
|
|
1117
|
# load headers
|
|
|
|
|
|
1118
|
headers = csv.shift
|
|
|
|
|
|
1119
|
|
|
|
|
|
|
1120
|
# unserialize each object stored in the file
|
|
|
|
|
|
1121
|
results = csv.inject(Array.new) do |all, row|
|
|
|
|
|
|
1122
|
begin
|
|
|
|
|
|
1123
|
obj = cls.csv_load(meta, headers, row)
|
|
|
|
|
|
1124
|
rescue NoMethodError
|
|
|
|
|
|
1125
|
obj = cls.allocate
|
|
|
|
|
|
1126
|
headers.zip(row) do |name, value|
|
|
|
|
|
|
1127
|
if name[0] == ?@
|
|
|
|
|
|
1128
|
obj.instance_variable_set(name, value)
|
|
|
|
|
|
1129
|
else
|
|
|
|
|
|
1130
|
obj.send(name, value)
|
|
|
|
|
|
1131
|
end
|
|
|
|
|
|
1132
|
end
|
|
|
|
|
|
1133
|
end
|
|
|
|
|
|
1134
|
all << obj
|
|
|
|
|
|
1135
|
end
|
|
|
|
|
|
1136
|
|
|
|
|
|
|
1137
|
csv.close unless io_or_str.is_a? String
|
|
|
|
|
|
1138
|
|
|
|
|
|
|
1139
|
results
|
|
|
|
|
|
1140
|
end
|
|
|
|
|
|
1141
|
|
|
|
|
|
|
1142
|
#
|
|
|
|
|
|
1143
|
# :call-seq:
|
|
|
|
|
|
1144
|
# open( filename, mode="rb", options = Hash.new ) { |faster_csv| ... }
|
|
|
|
|
|
1145
|
# open( filename, mode="rb", options = Hash.new )
|
|
|
|
|
|
1146
|
#
|
|
|
|
|
|
1147
|
# This method opens an IO object, and wraps that with FasterCSV. This is
|
|
|
|
|
|
1148
|
# intended as the primary interface for writing a CSV file.
|
|
|
|
|
|
1149
|
#
|
|
|
|
|
|
1150
|
# You may pass any +args+ Ruby's open() understands followed by an optional
|
|
|
|
|
|
1151
|
# Hash containing any +options+ FasterCSV::new() understands.
|
|
|
|
|
|
1152
|
#
|
|
|
|
|
|
1153
|
# This method works like Ruby's open() call, in that it will pass a FasterCSV
|
|
|
|
|
|
1154
|
# object to a provided block and close it when the block termminates, or it
|
|
|
|
|
|
1155
|
# will return the FasterCSV object when no block is provided. (*Note*: This
|
|
|
|
|
|
1156
|
# is different from the standard CSV library which passes rows to the block.
|
|
|
|
|
|
1157
|
# Use FasterCSV::foreach() for that behavior.)
|
|
|
|
|
|
1158
|
#
|
|
|
|
|
|
1159
|
# An opened FasterCSV object will delegate to many IO methods, for
|
|
|
|
|
|
1160
|
# convenience. You may call:
|
|
|
|
|
|
1161
|
#
|
|
|
|
|
|
1162
|
# * binmode()
|
|
|
|
|
|
1163
|
# * close()
|
|
|
|
|
|
1164
|
# * close_read()
|
|
|
|
|
|
1165
|
# * close_write()
|
|
|
|
|
|
1166
|
# * closed?()
|
|
|
|
|
|
1167
|
# * eof()
|
|
|
|
|
|
1168
|
# * eof?()
|
|
|
|
|
|
1169
|
# * fcntl()
|
|
|
|
|
|
1170
|
# * fileno()
|
|
|
|
|
|
1171
|
# * flush()
|
|
|
|
|
|
1172
|
# * fsync()
|
|
|
|
|
|
1173
|
# * ioctl()
|
|
|
|
|
|
1174
|
# * isatty()
|
|
|
|
|
|
1175
|
# * pid()
|
|
|
|
|
|
1176
|
# * pos()
|
|
|
|
|
|
1177
|
# * reopen()
|
|
|
|
|
|
1178
|
# * seek()
|
|
|
|
|
|
1179
|
# * stat()
|
|
|
|
|
|
1180
|
# * sync()
|
|
|
|
|
|
1181
|
# * sync=()
|
|
|
|
|
|
1182
|
# * tell()
|
|
|
|
|
|
1183
|
# * to_i()
|
|
|
|
|
|
1184
|
# * to_io()
|
|
|
|
|
|
1185
|
# * tty?()
|
|
|
|
|
|
1186
|
#
|
|
|
|
|
|
1187
|
def self.open(*args)
|
|
|
|
|
|
1188
|
# find the +options+ Hash
|
|
|
|
|
|
1189
|
options = if args.last.is_a? Hash then args.pop else Hash.new end
|
|
|
|
|
|
1190
|
# default to a binary open mode
|
|
|
|
|
|
1191
|
args << "rb" if args.size == 1
|
|
|
|
|
|
1192
|
# wrap a File opened with the remaining +args+
|
|
|
|
|
|
1193
|
csv = new(File.open(*args), options)
|
|
|
|
|
|
1194
|
|
|
|
|
|
|
1195
|
# handle blocks like Ruby's open(), not like the CSV library
|
|
|
|
|
|
1196
|
if block_given?
|
|
|
|
|
|
1197
|
begin
|
|
|
|
|
|
1198
|
yield csv
|
|
|
|
|
|
1199
|
ensure
|
|
|
|
|
|
1200
|
csv.close
|
|
|
|
|
|
1201
|
end
|
|
|
|
|
|
1202
|
else
|
|
|
|
|
|
1203
|
csv
|
|
|
|
|
|
1204
|
end
|
|
|
|
|
|
1205
|
end
|
|
|
|
|
|
1206
|
|
|
|
|
|
|
1207
|
#
|
|
|
|
|
|
1208
|
# :call-seq:
|
|
|
|
|
|
1209
|
# parse( str, options = Hash.new ) { |row| ... }
|
|
|
|
|
|
1210
|
# parse( str, options = Hash.new )
|
|
|
|
|
|
1211
|
#
|
|
|
|
|
|
1212
|
# This method can be used to easily parse CSV out of a String. You may either
|
|
|
|
|
|
1213
|
# provide a +block+ which will be called with each row of the String in turn,
|
|
|
|
|
|
1214
|
# or just use the returned Array of Arrays (when no +block+ is given).
|
|
|
|
|
|
1215
|
#
|
|
|
|
|
|
1216
|
# You pass your +str+ to read from, and an optional +options+ Hash containing
|
|
|
|
|
|
1217
|
# anything FasterCSV::new() understands.
|
|
|
|
|
|
1218
|
#
|
|
|
|
|
|
1219
|
def self.parse(*args, &block)
|
|
|
|
|
|
1220
|
csv = new(*args)
|
|
|
|
|
|
1221
|
if block.nil? # slurp contents, if no block is given
|
|
|
|
|
|
1222
|
begin
|
|
|
|
|
|
1223
|
csv.read
|
|
|
|
|
|
1224
|
ensure
|
|
|
|
|
|
1225
|
csv.close
|
|
|
|
|
|
1226
|
end
|
|
|
|
|
|
1227
|
else # or pass each row to a provided block
|
|
|
|
|
|
1228
|
csv.each(&block)
|
|
|
|
|
|
1229
|
end
|
|
|
|
|
|
1230
|
end
|
|
|
|
|
|
1231
|
|
|
|
|
|
|
1232
|
#
|
|
|
|
|
|
1233
|
# This method is a shortcut for converting a single line of a CSV String into
|
|
|
|
|
|
1234
|
# a into an Array. Note that if +line+ contains multiple rows, anything
|
|
|
|
|
|
1235
|
# beyond the first row is ignored.
|
|
|
|
|
|
1236
|
#
|
|
|
|
|
|
1237
|
# The +options+ parameter can be anthing FasterCSV::new() understands.
|
|
|
|
|
|
1238
|
#
|
|
|
|
|
|
1239
|
def self.parse_line(line, options = Hash.new)
|
|
|
|
|
|
1240
|
new(line, options).shift
|
|
|
|
|
|
1241
|
end
|
|
|
|
|
|
1242
|
|
|
|
|
|
|
1243
|
#
|
|
|
|
|
|
1244
|
# Use to slurp a CSV file into an Array of Arrays. Pass the +path+ to the
|
|
|
|
|
|
1245
|
# file and any +options+ FasterCSV::new() understands.
|
|
|
|
|
|
1246
|
#
|
|
|
|
|
|
1247
|
def self.read(path, options = Hash.new)
|
|
|
|
|
|
1248
|
open(path, "rb", options) { |csv| csv.read }
|
|
|
|
|
|
1249
|
end
|
|
|
|
|
|
1250
|
|
|
|
|
|
|
1251
|
# Alias for FasterCSV::read().
|
|
|
|
|
|
1252
|
def self.readlines(*args)
|
|
|
|
|
|
1253
|
read(*args)
|
|
|
|
|
|
1254
|
end
|
|
|
|
|
|
1255
|
|
|
|
|
|
|
1256
|
#
|
|
|
|
|
|
1257
|
# A shortcut for:
|
|
|
|
|
|
1258
|
#
|
|
|
|
|
|
1259
|
# FasterCSV.read( path, { :headers => true,
|
|
|
|
|
|
1260
|
# :converters => :numeric,
|
|
|
|
|
|
1261
|
# :header_converters => :symbol }.merge(options) )
|
|
|
|
|
|
1262
|
#
|
|
|
|
|
|
1263
|
def self.table(path, options = Hash.new)
|
|
|
|
|
|
1264
|
read( path, { :headers => true,
|
|
|
|
|
|
1265
|
:converters => :numeric,
|
|
|
|
|
|
1266
|
:header_converters => :symbol }.merge(options) )
|
|
|
|
|
|
1267
|
end
|
|
|
|
|
|
1268
|
|
|
|
|
|
|
1269
|
#
|
|
|
|
|
|
1270
|
# This constructor will wrap either a String or IO object passed in +data+ for
|
|
|
|
|
|
1271
|
# reading and/or writing. In addition to the FasterCSV instance methods,
|
|
|
|
|
|
1272
|
# several IO methods are delegated. (See FasterCSV::open() for a complete
|
|
|
|
|
|
1273
|
# list.) If you pass a String for +data+, you can later retrieve it (after
|
|
|
|
|
|
1274
|
# writing to it, for example) with FasterCSV.string().
|
|
|
|
|
|
1275
|
#
|
|
|
|
|
|
1276
|
# Note that a wrapped String will be positioned at at the beginning (for
|
|
|
|
|
|
1277
|
# reading). If you want it at the end (for writing), use
|
|
|
|
|
|
1278
|
# FasterCSV::generate(). If you want any other positioning, pass a preset
|
|
|
|
|
|
1279
|
# StringIO object instead.
|
|
|
|
|
|
1280
|
#
|
|
|
|
|
|
1281
|
# You may set any reading and/or writing preferences in the +options+ Hash.
|
|
|
|
|
|
1282
|
# Available options are:
|
|
|
|
|
|
1283
|
#
|
|
|
|
|
|
1284
|
# <b><tt>:col_sep</tt></b>:: The String placed between each field.
|
|
|
|
|
|
1285
|
# <b><tt>:row_sep</tt></b>:: The String appended to the end of each
|
|
|
|
|
|
1286
|
# row. This can be set to the special
|
|
|
|
|
|
1287
|
# <tt>:auto</tt> setting, which requests
|
|
|
|
|
|
1288
|
# that FasterCSV automatically discover
|
|
|
|
|
|
1289
|
# this from the data. Auto-discovery
|
|
|
|
|
|
1290
|
# reads ahead in the data looking for
|
|
|
|
|
|
1291
|
# the next <tt>"\r\n"</tt>,
|
|
|
|
|
|
1292
|
# <tt>"\n"</tt>, or <tt>"\r"</tt>
|
|
|
|
|
|
1293
|
# sequence. A sequence will be selected
|
|
|
|
|
|
1294
|
# even if it occurs in a quoted field,
|
|
|
|
|
|
1295
|
# assuming that you would have the same
|
|
|
|
|
|
1296
|
# line endings there. If none of those
|
|
|
|
|
|
1297
|
# sequences is found, +data+ is
|
|
|
|
|
|
1298
|
# <tt>ARGF</tt>, <tt>STDIN</tt>,
|
|
|
|
|
|
1299
|
# <tt>STDOUT</tt>, or <tt>STDERR</tt>,
|
|
|
|
|
|
1300
|
# or the stream is only available for
|
|
|
|
|
|
1301
|
# output, the default
|
|
|
|
|
|
1302
|
# <tt>$INPUT_RECORD_SEPARATOR</tt>
|
|
|
|
|
|
1303
|
# (<tt>$/</tt>) is used. Obviously,
|
|
|
|
|
|
1304
|
# discovery takes a little time. Set
|
|
|
|
|
|
1305
|
# manually if speed is important. Also
|
|
|
|
|
|
1306
|
# note that IO objects should be opened
|
|
|
|
|
|
1307
|
# in binary mode on Windows if this
|
|
|
|
|
|
1308
|
# feature will be used as the
|
|
|
|
|
|
1309
|
# line-ending translation can cause
|
|
|
|
|
|
1310
|
# problems with resetting the document
|
|
|
|
|
|
1311
|
# position to where it was before the
|
|
|
|
|
|
1312
|
# read ahead.
|
|
|
|
|
|
1313
|
# <b><tt>:quote_char</tt></b>:: The character used to quote fields.
|
|
|
|
|
|
1314
|
# This has to be a single character
|
|
|
|
|
|
1315
|
# String. This is useful for
|
|
|
|
|
|
1316
|
# application that incorrectly use
|
|
|
|
|
|
1317
|
# <tt>'</tt> as the quote character
|
|
|
|
|
|
1318
|
# instead of the correct <tt>"</tt>.
|
|
|
|
|
|
1319
|
# FasterCSV will always consider a
|
|
|
|
|
|
1320
|
# double sequence this character to be
|
|
|
|
|
|
1321
|
# an escaped quote.
|
|
|
|
|
|
1322
|
# <b><tt>:encoding</tt></b>:: The encoding to use when parsing the
|
|
|
|
|
|
1323
|
# file. Defaults to your <tt>$KDOCE</tt>
|
|
|
|
|
|
1324
|
# setting. Valid values: <tt>`n’</tt> or
|
|
|
|
|
|
1325
|
# <tt>`N’</tt> for none, <tt>`e’</tt> or
|
|
|
|
|
|
1326
|
# <tt>`E’</tt> for EUC, <tt>`s’</tt> or
|
|
|
|
|
|
1327
|
# <tt>`S’</tt> for SJIS, and
|
|
|
|
|
|
1328
|
# <tt>`u’</tt> or <tt>`U’</tt> for UTF-8
|
|
|
|
|
|
1329
|
# (see Regexp.new()).
|
|
|
|
|
|
1330
|
# <b><tt>:field_size_limit</tt></b>:: This is a maximum size FasterCSV will
|
|
|
|
|
|
1331
|
# read ahead looking for the closing
|
|
|
|
|
|
1332
|
# quote for a field. (In truth, it
|
|
|
|
|
|
1333
|
# reads to the first line ending beyond
|
|
|
|
|
|
1334
|
# this size.) If a quote cannot be
|
|
|
|
|
|
1335
|
# found within the limit FasterCSV will
|
|
|
|
|
|
1336
|
# raise a MalformedCSVError, assuming
|
|
|
|
|
|
1337
|
# the data is faulty. You can use this
|
|
|
|
|
|
1338
|
# limit to prevent what are effectively
|
|
|
|
|
|
1339
|
# DoS attacks on the parser. However,
|
|
|
|
|
|
1340
|
# this limit can cause a legitimate
|
|
|
|
|
|
1341
|
# parse to fail and thus is set to
|
|
|
|
|
|
1342
|
# +nil+, or off, by default.
|
|
|
|
|
|
1343
|
# <b><tt>:converters</tt></b>:: An Array of names from the Converters
|
|
|
|
|
|
1344
|
# Hash and/or lambdas that handle custom
|
|
|
|
|
|
1345
|
# conversion. A single converter
|
|
|
|
|
|
1346
|
# doesn't have to be in an Array.
|
|
|
|
|
|
1347
|
# <b><tt>:unconverted_fields</tt></b>:: If set to +true+, an
|
|
|
|
|
|
1348
|
# unconverted_fields() method will be
|
|
|
|
|
|
1349
|
# added to all returned rows (Array or
|
|
|
|
|
|
1350
|
# FasterCSV::Row) that will return the
|
|
|
|
|
|
1351
|
# fields as they were before convertion.
|
|
|
|
|
|
1352
|
# Note that <tt>:headers</tt> supplied
|
|
|
|
|
|
1353
|
# by Array or String were not fields of
|
|
|
|
|
|
1354
|
# the document and thus will have an
|
|
|
|
|
|
1355
|
# empty Array attached.
|
|
|
|
|
|
1356
|
# <b><tt>:headers</tt></b>:: If set to <tt>:first_row</tt> or
|
|
|
|
|
|
1357
|
# +true+, the initial row of the CSV
|
|
|
|
|
|
1358
|
# file will be treated as a row of
|
|
|
|
|
|
1359
|
# headers. If set to an Array, the
|
|
|
|
|
|
1360
|
# contents will be used as the headers.
|
|
|
|
|
|
1361
|
# If set to a String, the String is run
|
|
|
|
|
|
1362
|
# through a call of
|
|
|
|
|
|
1363
|
# FasterCSV::parse_line() with the same
|
|
|
|
|
|
1364
|
# <tt>:col_sep</tt>, <tt>:row_sep</tt>,
|
|
|
|
|
|
1365
|
# and <tt>:quote_char</tt> as this
|
|
|
|
|
|
1366
|
# instance to produce an Array of
|
|
|
|
|
|
1367
|
# headers. This setting causes
|
|
|
|
|
|
1368
|
# FasterCSV.shift() to return rows as
|
|
|
|
|
|
1369
|
# FasterCSV::Row objects instead of
|
|
|
|
|
|
1370
|
# Arrays and FasterCSV.read() to return
|
|
|
|
|
|
1371
|
# FasterCSV::Table objects instead of
|
|
|
|
|
|
1372
|
# an Array of Arrays.
|
|
|
|
|
|
1373
|
# <b><tt>:return_headers</tt></b>:: When +false+, header rows are silently
|
|
|
|
|
|
1374
|
# swallowed. If set to +true+, header
|
|
|
|
|
|
1375
|
# rows are returned in a FasterCSV::Row
|
|
|
|
|
|
1376
|
# object with identical headers and
|
|
|
|
|
|
1377
|
# fields (save that the fields do not go
|
|
|
|
|
|
1378
|
# through the converters).
|
|
|
|
|
|
1379
|
# <b><tt>:write_headers</tt></b>:: When +true+ and <tt>:headers</tt> is
|
|
|
|
|
|
1380
|
# set, a header row will be added to the
|
|
|
|
|
|
1381
|
# output.
|
|
|
|
|
|
1382
|
# <b><tt>:header_converters</tt></b>:: Identical in functionality to
|
|
|
|
|
|
1383
|
# <tt>:converters</tt> save that the
|
|
|
|
|
|
1384
|
# conversions are only made to header
|
|
|
|
|
|
1385
|
# rows.
|
|
|
|
|
|
1386
|
# <b><tt>:skip_blanks</tt></b>:: When set to a +true+ value, FasterCSV
|
|
|
|
|
|
1387
|
# will skip over any rows with no
|
|
|
|
|
|
1388
|
# content.
|
|
|
|
|
|
1389
|
# <b><tt>:force_quotes</tt></b>:: When set to a +true+ value, FasterCSV
|
|
|
|
|
|
1390
|
# will quote all CSV fields it creates.
|
|
|
|
|
|
1391
|
#
|
|
|
|
|
|
1392
|
# See FasterCSV::DEFAULT_OPTIONS for the default settings.
|
|
|
|
|
|
1393
|
#
|
|
|
|
|
|
1394
|
# Options cannot be overriden in the instance methods for performance reasons,
|
|
|
|
|
|
1395
|
# so be sure to set what you want here.
|
|
|
|
|
|
1396
|
#
|
|
|
|
|
|
1397
|
def initialize(data, options = Hash.new)
|
|
|
|
|
|
1398
|
# build the options for this read/write
|
|
|
|
|
|
1399
|
options = DEFAULT_OPTIONS.merge(options)
|
|
|
|
|
|
1400
|
|
|
|
|
|
|
1401
|
# create the IO object we will read from
|
|
|
|
|
|
1402
|
@io = if data.is_a? String then StringIO.new(data) else data end
|
|
|
|
|
|
1403
|
|
|
|
|
|
|
1404
|
init_separators(options)
|
|
|
|
|
|
1405
|
init_parsers(options)
|
|
|
|
|
|
1406
|
init_converters(options)
|
|
|
|
|
|
1407
|
init_headers(options)
|
|
|
|
|
|
1408
|
|
|
|
|
|
|
1409
|
unless options.empty?
|
|
|
|
|
|
1410
|
raise ArgumentError, "Unknown options: #{options.keys.join(', ')}."
|
|
|
|
|
|
1411
|
end
|
|
|
|
|
|
1412
|
|
|
|
|
|
|
1413
|
# track our own lineno since IO gets confused about line-ends is CSV fields
|
|
|
|
|
|
1414
|
@lineno = 0
|
|
|
|
|
|
1415
|
end
|
|
|
|
|
|
1416
|
|
|
|
|
|
|
1417
|
#
|
|
|
|
|
|
1418
|
# The line number of the last row read from this file. Fields with nested
|
|
|
|
|
|
1419
|
# line-end characters will not affect this count.
|
|
|
|
|
|
1420
|
#
|
|
|
|
|
|
1421
|
attr_reader :lineno
|
|
|
|
|
|
1422
|
|
|
|
|
|
|
1423
|
### IO and StringIO Delegation ###
|
|
|
|
|
|
1424
|
|
|
|
|
|
|
1425
|
extend Forwardable
|
|
|
|
|
|
1426
|
def_delegators :@io, :binmode, :close, :close_read, :close_write, :closed?,
|
|
|
|
|
|
1427
|
:eof, :eof?, :fcntl, :fileno, :flush, :fsync, :ioctl,
|
|
|
|
|
|
1428
|
:isatty, :pid, :pos, :reopen, :seek, :stat, :string,
|
|
|
|
|
|
1429
|
:sync, :sync=, :tell, :to_i, :to_io, :tty?
|
|
|
|
|
|
1430
|
|
|
|
|
|
|
1431
|
# Rewinds the underlying IO object and resets FasterCSV's lineno() counter.
|
|
|
|
|
|
1432
|
def rewind
|
|
|
|
|
|
1433
|
@headers = nil
|
|
|
|
|
|
1434
|
@lineno = 0
|
|
|
|
|
|
1435
|
|
|
|
|
|
|
1436
|
@io.rewind
|
|
|
|
|
|
1437
|
end
|
|
|
|
|
|
1438
|
|
|
|
|
|
|
1439
|
### End Delegation ###
|
|
|
|
|
|
1440
|
|
|
|
|
|
|
1441
|
#
|
|
|
|
|
|
1442
|
# The primary write method for wrapped Strings and IOs, +row+ (an Array or
|
|
|
|
|
|
1443
|
# FasterCSV::Row) is converted to CSV and appended to the data source. When a
|
|
|
|
|
|
1444
|
# FasterCSV::Row is passed, only the row's fields() are appended to the
|
|
|
|
|
|
1445
|
# output.
|
|
|
|
|
|
1446
|
#
|
|
|
|
|
|
1447
|
# The data source must be open for writing.
|
|
|
|
|
|
1448
|
#
|
|
|
|
|
|
1449
|
def <<(row)
|
|
|
|
|
|
1450
|
# make sure headers have been assigned
|
|
|
|
|
|
1451
|
if header_row? and [Array, String].include? @use_headers.class
|
|
|
|
|
|
1452
|
parse_headers # won't read data for Array or String
|
|
|
|
|
|
1453
|
self << @headers if @write_headers
|
|
|
|
|
|
1454
|
end
|
|
|
|
|
|
1455
|
|
|
|
|
|
|
1456
|
# Handle FasterCSV::Row objects and Hashes
|
|
|
|
|
|
1457
|
row = case row
|
|
|
|
|
|
1458
|
when self.class::Row then row.fields
|
|
|
|
|
|
1459
|
when Hash then @headers.map { |header| row[header] }
|
|
|
|
|
|
1460
|
else row
|
|
|
|
|
|
1461
|
end
|
|
|
|
|
|
1462
|
|
|
|
|
|
|
1463
|
@headers = row if header_row?
|
|
|
|
|
|
1464
|
@lineno += 1
|
|
|
|
|
|
1465
|
|
|
|
|
|
|
1466
|
@io << row.map(&@quote).join(@col_sep) + @row_sep # quote and separate
|
|
|
|
|
|
1467
|
|
|
|
|
|
|
1468
|
self # for chaining
|
|
|
|
|
|
1469
|
end
|
|
|
|
|
|
1470
|
alias_method :add_row, :<<
|
|
|
|
|
|
1471
|
alias_method :puts, :<<
|
|
|
|
|
|
1472
|
|
|
|
|
|
|
1473
|
#
|
|
|
|
|
|
1474
|
# :call-seq:
|
|
|
|
|
|
1475
|
# convert( name )
|
|
|
|
|
|
1476
|
# convert { |field| ... }
|
|
|
|
|
|
1477
|
# convert { |field, field_info| ... }
|
|
|
|
|
|
1478
|
#
|
|
|
|
|
|
1479
|
# You can use this method to install a FasterCSV::Converters built-in, or
|
|
|
|
|
|
1480
|
# provide a block that handles a custom conversion.
|
|
|
|
|
|
1481
|
#
|
|
|
|
|
|
1482
|
# If you provide a block that takes one argument, it will be passed the field
|
|
|
|
|
|
1483
|
# and is expected to return the converted value or the field itself. If your
|
|
|
|
|
|
1484
|
# block takes two arguments, it will also be passed a FieldInfo Struct,
|
|
|
|
|
|
1485
|
# containing details about the field. Again, the block should return a
|
|
|
|
|
|
1486
|
# converted field or the field itself.
|
|
|
|
|
|
1487
|
#
|
|
|
|
|
|
1488
|
def convert(name = nil, &converter)
|
|
|
|
|
|
1489
|
add_converter(:converters, self.class::Converters, name, &converter)
|
|
|
|
|
|
1490
|
end
|
|
|
|
|
|
1491
|
|
|
|
|
|
|
1492
|
#
|
|
|
|
|
|
1493
|
# :call-seq:
|
|
|
|
|
|
1494
|
# header_convert( name )
|
|
|
|
|
|
1495
|
# header_convert { |field| ... }
|
|
|
|
|
|
1496
|
# header_convert { |field, field_info| ... }
|
|
|
|
|
|
1497
|
#
|
|
|
|
|
|
1498
|
# Identical to FasterCSV.convert(), but for header rows.
|
|
|
|
|
|
1499
|
#
|
|
|
|
|
|
1500
|
# Note that this method must be called before header rows are read to have any
|
|
|
|
|
|
1501
|
# effect.
|
|
|
|
|
|
1502
|
#
|
|
|
|
|
|
1503
|
def header_convert(name = nil, &converter)
|
|
|
|
|
|
1504
|
add_converter( :header_converters,
|
|
|
|
|
|
1505
|
self.class::HeaderConverters,
|
|
|
|
|
|
1506
|
name,
|
|
|
|
|
|
1507
|
&converter )
|
|
|
|
|
|
1508
|
end
|
|
|
|
|
|
1509
|
|
|
|
|
|
|
1510
|
include Enumerable
|
|
|
|
|
|
1511
|
|
|
|
|
|
|
1512
|
#
|
|
|
|
|
|
1513
|
# Yields each row of the data source in turn.
|
|
|
|
|
|
1514
|
#
|
|
|
|
|
|
1515
|
# Support for Enumerable.
|
|
|
|
|
|
1516
|
#
|
|
|
|
|
|
1517
|
# The data source must be open for reading.
|
|
|
|
|
|
1518
|
#
|
|
|
|
|
|
1519
|
def each
|
|
|
|
|
|
1520
|
while row = shift
|
|
|
|
|
|
1521
|
yield row
|
|
|
|
|
|
1522
|
end
|
|
|
|
|
|
1523
|
end
|
|
|
|
|
|
1524
|
|
|
|
|
|
|
1525
|
#
|
|
|
|
|
|
1526
|
# Slurps the remaining rows and returns an Array of Arrays.
|
|
|
|
|
|
1527
|
#
|
|
|
|
|
|
1528
|
# The data source must be open for reading.
|
|
|
|
|
|
1529
|
#
|
|
|
|
|
|
1530
|
def read
|
|
|
|
|
|
1531
|
rows = to_a
|
|
|
|
|
|
1532
|
if @use_headers
|
|
|
|
|
|
1533
|
Table.new(rows)
|
|
|
|
|
|
1534
|
else
|
|
|
|
|
|
1535
|
rows
|
|
|
|
|
|
1536
|
end
|
|
|
|
|
|
1537
|
end
|
|
|
|
|
|
1538
|
alias_method :readlines, :read
|
|
|
|
|
|
1539
|
|
|
|
|
|
|
1540
|
# Returns +true+ if the next row read will be a header row.
|
|
|
|
|
|
1541
|
def header_row?
|
|
|
|
|
|
1542
|
@use_headers and @headers.nil?
|
|
|
|
|
|
1543
|
end
|
|
|
|
|
|
1544
|
|
|
|
|
|
|
1545
|
#
|
|
|
|
|
|
1546
|
# The primary read method for wrapped Strings and IOs, a single row is pulled
|
|
|
|
|
|
1547
|
# from the data source, parsed and returned as an Array of fields (if header
|
|
|
|
|
|
1548
|
# rows are not used) or a FasterCSV::Row (when header rows are used).
|
|
|
|
|
|
1549
|
#
|
|
|
|
|
|
1550
|
# The data source must be open for reading.
|
|
|
|
|
|
1551
|
#
|
|
|
|
|
|
1552
|
def shift
|
|
|
|
|
|
1553
|
#########################################################################
|
|
|
|
|
|
1554
|
### This method is purposefully kept a bit long as simple conditional ###
|
|
|
|
|
|
1555
|
### checks are faster than numerous (expensive) method calls. ###
|
|
|
|
|
|
1556
|
#########################################################################
|
|
|
|
|
|
1557
|
|
|
|
|
|
|
1558
|
# handle headers not based on document content
|
|
|
|
|
|
1559
|
if header_row? and @return_headers and
|
|
|
|
|
|
1560
|
[Array, String].include? @use_headers.class
|
|
|
|
|
|
1561
|
if @unconverted_fields
|
|
|
|
|
|
1562
|
return add_unconverted_fields(parse_headers, Array.new)
|
|
|
|
|
|
1563
|
else
|
|
|
|
|
|
1564
|
return parse_headers
|
|
|
|
|
|
1565
|
end
|
|
|
|
|
|
1566
|
end
|
|
|
|
|
|
1567
|
|
|
|
|
|
|
1568
|
# begin with a blank line, so we can always add to it
|
|
|
|
|
|
1569
|
line = String.new
|
|
|
|
|
|
1570
|
|
|
|
|
|
|
1571
|
#
|
|
|
|
|
|
1572
|
# it can take multiple calls to <tt>@io.gets()</tt> to get a full line,
|
|
|
|
|
|
1573
|
# because of \r and/or \n characters embedded in quoted fields
|
|
|
|
|
|
1574
|
#
|
|
|
|
|
|
1575
|
loop do
|
|
|
|
|
|
1576
|
# add another read to the line
|
|
|
|
|
|
1577
|
begin
|
|
|
|
|
|
1578
|
line += @io.gets(@row_sep)
|
|
|
|
|
|
1579
|
rescue
|
|
|
|
|
|
1580
|
return nil
|
|
|
|
|
|
1581
|
end
|
|
|
|
|
|
1582
|
# copy the line so we can chop it up in parsing
|
|
|
|
|
|
1583
|
parse = line.dup
|
|
|
|
|
|
1584
|
parse.sub!(@parsers[:line_end], "")
|
|
|
|
|
|
1585
|
|
|
|
|
|
|
1586
|
#
|
|
|
|
|
|
1587
|
# I believe a blank line should be an <tt>Array.new</tt>, not
|
|
|
|
|
|
1588
|
# CSV's <tt>[nil]</tt>
|
|
|
|
|
|
1589
|
#
|
|
|
|
|
|
1590
|
if parse.empty?
|
|
|
|
|
|
1591
|
@lineno += 1
|
|
|
|
|
|
1592
|
if @skip_blanks
|
|
|
|
|
|
1593
|
line = ""
|
|
|
|
|
|
1594
|
next
|
|
|
|
|
|
1595
|
elsif @unconverted_fields
|
|
|
|
|
|
1596
|
return add_unconverted_fields(Array.new, Array.new)
|
|
|
|
|
|
1597
|
elsif @use_headers
|
|
|
|
|
|
1598
|
return FasterCSV::Row.new(Array.new, Array.new)
|
|
|
|
|
|
1599
|
else
|
|
|
|
|
|
1600
|
return Array.new
|
|
|
|
|
|
1601
|
end
|
|
|
|
|
|
1602
|
end
|
|
|
|
|
|
1603
|
|
|
|
|
|
|
1604
|
# parse the fields with a mix of String#split and regular expressions
|
|
|
|
|
|
1605
|
csv = Array.new
|
|
|
|
|
|
1606
|
current_field = String.new
|
|
|
|
|
|
1607
|
field_quotes = 0
|
|
|
|
|
|
1608
|
parse.split(@col_sep, -1).each do |match|
|
|
|
|
|
|
1609
|
if current_field.empty? && match.count(@quote_and_newlines).zero?
|
|
|
|
|
|
1610
|
csv << (match.empty? ? nil : match)
|
|
|
|
|
|
1611
|
elsif(current_field.empty? ? match[0] : current_field[0]) == @quote_char[0]
|
|
|
|
|
|
1612
|
current_field << match
|
|
|
|
|
|
1613
|
field_quotes += match.count(@quote_char)
|
|
|
|
|
|
1614
|
if field_quotes % 2 == 0
|
|
|
|
|
|
1615
|
in_quotes = current_field[@parsers[:quoted_field], 1]
|
|
|
|
|
|
1616
|
raise MalformedCSVError unless in_quotes
|
|
|
|
|
|
1617
|
current_field = in_quotes
|
|
|
|
|
|
1618
|
current_field.gsub!(@quote_char * 2, @quote_char) # unescape contents
|
|
|
|
|
|
1619
|
csv << current_field
|
|
|
|
|
|
1620
|
current_field = String.new
|
|
|
|
|
|
1621
|
field_quotes = 0
|
|
|
|
|
|
1622
|
else # we found a quoted field that spans multiple lines
|
|
|
|
|
|
1623
|
current_field << @col_sep
|
|
|
|
|
|
1624
|
end
|
|
|
|
|
|
1625
|
elsif match.count("\r\n").zero?
|
|
|
|
|
|
1626
|
raise MalformedCSVError, "Illegal quoting on line #{lineno + 1}."
|
|
|
|
|
|
1627
|
else
|
|
|
|
|
|
1628
|
raise MalformedCSVError, "Unquoted fields do not allow " +
|
|
|
|
|
|
1629
|
"\\r or \\n (line #{lineno + 1})."
|
|
|
|
|
|
1630
|
end
|
|
|
|
|
|
1631
|
end
|
|
|
|
|
|
1632
|
|
|
|
|
|
|
1633
|
# if parse is empty?(), we found all the fields on the line...
|
|
|
|
|
|
1634
|
if field_quotes % 2 == 0
|
|
|
|
|
|
1635
|
@lineno += 1
|
|
|
|
|
|
1636
|
|
|
|
|
|
|
1637
|
# save fields unconverted fields, if needed...
|
|
|
|
|
|
1638
|
unconverted = csv.dup if @unconverted_fields
|
|
|
|
|
|
1639
|
|
|
|
|
|
|
1640
|
# convert fields, if needed...
|
|
|
|
|
|
1641
|
csv = convert_fields(csv) unless @use_headers or @converters.empty?
|
|
|
|
|
|
1642
|
# parse out header rows and handle FasterCSV::Row conversions...
|
|
|
|
|
|
1643
|
csv = parse_headers(csv) if @use_headers
|
|
|
|
|
|
1644
|
|
|
|
|
|
|
1645
|
# inject unconverted fields and accessor, if requested...
|
|
|
|
|
|
1646
|
if @unconverted_fields and not csv.respond_to? :unconverted_fields
|
|
|
|
|
|
1647
|
add_unconverted_fields(csv, unconverted)
|
|
|
|
|
|
1648
|
end
|
|
|
|
|
|
1649
|
|
|
|
|
|
|
1650
|
# return the results
|
|
|
|
|
|
1651
|
break csv
|
|
|
|
|
|
1652
|
end
|
|
|
|
|
|
1653
|
# if we're not empty?() but at eof?(), a quoted field wasn't closed...
|
|
|
|
|
|
1654
|
if @io.eof?
|
|
|
|
|
|
1655
|
raise MalformedCSVError, "Unclosed quoted field on line #{lineno + 1}."
|
|
|
|
|
|
1656
|
elsif @field_size_limit and current_field.size >= @field_size_limit
|
|
|
|
|
|
1657
|
raise MalformedCSVError, "Field size exceeded on line #{lineno + 1}."
|
|
|
|
|
|
1658
|
end
|
|
|
|
|
|
1659
|
# otherwise, we need to loop and pull some more data to complete the row
|
|
|
|
|
|
1660
|
end
|
|
|
|
|
|
1661
|
end
|
|
|
|
|
|
1662
|
alias_method :gets, :shift
|
|
|
|
|
|
1663
|
alias_method :readline, :shift
|
|
|
|
|
|
1664
|
|
|
|
|
|
|
1665
|
# Returns a simplified description of the key FasterCSV attributes.
|
|
|
|
|
|
1666
|
def inspect
|
|
|
|
|
|
1667
|
str = "<##{self.class} io_type:"
|
|
|
|
|
|
1668
|
# show type of wrapped IO
|
|
|
|
|
|
1669
|
if @io == $stdout then str << "$stdout"
|
|
|
|
|
|
1670
|
elsif @io == $stdin then str << "$stdin"
|
|
|
|
|
|
1671
|
elsif @io == $stderr then str << "$stderr"
|
|
|
|
|
|
1672
|
else str << @io.class.to_s
|
|
|
|
|
|
1673
|
end
|
|
|
|
|
|
1674
|
# show IO.path(), if available
|
|
|
|
|
|
1675
|
if @io.respond_to?(:path) and (p = @io.path)
|
|
|
|
|
|
1676
|
str << " io_path:#{p.inspect}"
|
|
|
|
|
|
1677
|
end
|
|
|
|
|
|
1678
|
# show other attributes
|
|
|
|
|
|
1679
|
%w[ lineno col_sep row_sep
|
|
|
|
|
|
1680
|
quote_char skip_blanks encoding ].each do |attr_name|
|
|
|
|
|
|
1681
|
if a = instance_variable_get("@#{attr_name}")
|
|
|
|
|
|
1682
|
str << " #{attr_name}:#{a.inspect}"
|
|
|
|
|
|
1683
|
end
|
|
|
|
|
|
1684
|
end
|
|
|
|
|
|
1685
|
if @use_headers
|
|
|
|
|
|
1686
|
str << " headers:#{(@headers || true).inspect}"
|
|
|
|
|
|
1687
|
end
|
|
|
|
|
|
1688
|
str << ">"
|
|
|
|
|
|
1689
|
end
|
|
|
|
|
|
1690
|
|
|
|
|
|
|
1691
|
private
|
|
|
|
|
|
1692
|
|
|
|
|
|
|
1693
|
#
|
|
|
|
|
|
1694
|
# Stores the indicated separators for later use.
|
|
|
|
|
|
1695
|
#
|
|
|
|
|
|
1696
|
# If auto-discovery was requested for <tt>@row_sep</tt>, this method will read
|
|
|
|
|
|
1697
|
# ahead in the <tt>@io</tt> and try to find one. +ARGF+, +STDIN+, +STDOUT+,
|
|
|
|
|
|
1698
|
# +STDERR+ and any stream open for output only with a default
|
|
|
|
|
|
1699
|
# <tt>@row_sep</tt> of <tt>$INPUT_RECORD_SEPARATOR</tt> (<tt>$/</tt>).
|
|
|
|
|
|
1700
|
#
|
|
|
|
|
|
1701
|
# This method also establishes the quoting rules used for CSV output.
|
|
|
|
|
|
1702
|
#
|
|
|
|
|
|
1703
|
def init_separators(options)
|
|
|
|
|
|
1704
|
# store the selected separators
|
|
|
|
|
|
1705
|
@col_sep = options.delete(:col_sep)
|
|
|
|
|
|
1706
|
@row_sep = options.delete(:row_sep)
|
|
|
|
|
|
1707
|
@quote_char = options.delete(:quote_char)
|
|
|
|
|
|
1708
|
@quote_and_newlines = "#{@quote_char}\r\n"
|
|
|
|
|
|
1709
|
|
|
|
|
|
|
1710
|
if @quote_char.length != 1
|
|
|
|
|
|
1711
|
raise ArgumentError, ":quote_char has to be a single character String"
|
|
|
|
|
|
1712
|
end
|
|
|
|
|
|
1713
|
|
|
|
|
|
|
1714
|
# automatically discover row separator when requested
|
|
|
|
|
|
1715
|
if @row_sep == :auto
|
|
|
|
|
|
1716
|
if [ARGF, STDIN, STDOUT, STDERR].include?(@io) or
|
|
|
|
|
|
1717
|
(defined?(Zlib) and @io.class == Zlib::GzipWriter)
|
|
|
|
|
|
1718
|
@row_sep = $INPUT_RECORD_SEPARATOR
|
|
|
|
|
|
1719
|
else
|
|
|
|
|
|
1720
|
begin
|
|
|
|
|
|
1721
|
saved_pos = @io.pos # remember where we were
|
|
|
|
|
|
1722
|
while @row_sep == :auto
|
|
|
|
|
|
1723
|
#
|
|
|
|
|
|
1724
|
# if we run out of data, it's probably a single line
|
|
|
|
|
|
1725
|
# (use a sensible default)
|
|
|
|
|
|
1726
|
#
|
|
|
|
|
|
1727
|
if @io.eof?
|
|
|
|
|
|
1728
|
@row_sep = $INPUT_RECORD_SEPARATOR
|
|
|
|
|
|
1729
|
break
|
|
|
|
|
|
1730
|
end
|
|
|
|
|
|
1731
|
|
|
|
|
|
|
1732
|
# read ahead a bit
|
|
|
|
|
|
1733
|
sample = @io.read(1024)
|
|
|
|
|
|
1734
|
sample += @io.read(1) if sample[-1..-1] == "\r" and not @io.eof?
|
|
|
|
|
|
1735
|
|
|
|
|
|
|
1736
|
# try to find a standard separator
|
|
|
|
|
|
1737
|
if sample =~ /\r\n?|\n/
|
|
|
|
|
|
1738
|
@row_sep = $&
|
|
|
|
|
|
1739
|
break
|
|
|
|
|
|
1740
|
end
|
|
|
|
|
|
1741
|
end
|
|
|
|
|
|
1742
|
# tricky seek() clone to work around GzipReader's lack of seek()
|
|
|
|
|
|
1743
|
@io.rewind
|
|
|
|
|
|
1744
|
# reset back to the remembered position
|
|
|
|
|
|
1745
|
while saved_pos > 1024 # avoid loading a lot of data into memory
|
|
|
|
|
|
1746
|
@io.read(1024)
|
|
|
|
|
|
1747
|
saved_pos -= 1024
|
|
|
|
|
|
1748
|
end
|
|
|
|
|
|
1749
|
@io.read(saved_pos) if saved_pos.nonzero?
|
|
|
|
|
|
1750
|
rescue IOError # stream not opened for reading
|
|
|
|
|
|
1751
|
@row_sep = $INPUT_RECORD_SEPARATOR
|
|
|
|
|
|
1752
|
end
|
|
|
|
|
|
1753
|
end
|
|
|
|
|
|
1754
|
end
|
|
|
|
|
|
1755
|
|
|
|
|
|
|
1756
|
# establish quoting rules
|
|
|
|
|
|
1757
|
do_quote = lambda do |field|
|
|
|
|
|
|
1758
|
@quote_char +
|
|
|
|
|
|
1759
|
String(field).gsub(@quote_char, @quote_char * 2) +
|
|
|
|
|
|
1760
|
@quote_char
|
|
|
|
|
|
1761
|
end
|
|
|
|
|
|
1762
|
@quote = if options.delete(:force_quotes)
|
|
|
|
|
|
1763
|
do_quote
|
|
|
|
|
|
1764
|
else
|
|
|
|
|
|
1765
|
lambda do |field|
|
|
|
|
|
|
1766
|
if field.nil? # represent +nil+ fields as empty unquoted fields
|
|
|
|
|
|
1767
|
""
|
|
|
|
|
|
1768
|
else
|
|
|
|
|
|
1769
|
field = String(field) # Stringify fields
|
|
|
|
|
|
1770
|
# represent empty fields as empty quoted fields
|
|
|
|
|
|
1771
|
if field.empty? or
|
|
|
|
|
|
1772
|
field.count("\r\n#{@col_sep}#{@quote_char}").nonzero?
|
|
|
|
|
|
1773
|
do_quote.call(field)
|
|
|
|
|
|
1774
|
else
|
|
|
|
|
|
1775
|
field # unquoted field
|
|
|
|
|
|
1776
|
end
|
|
|
|
|
|
1777
|
end
|
|
|
|
|
|
1778
|
end
|
|
|
|
|
|
1779
|
end
|
|
|
|
|
|
1780
|
end
|
|
|
|
|
|
1781
|
|
|
|
|
|
|
1782
|
# Pre-compiles parsers and stores them by name for access during reads.
|
|
|
|
|
|
1783
|
def init_parsers(options)
|
|
|
|
|
|
1784
|
# store the parser behaviors
|
|
|
|
|
|
1785
|
@skip_blanks = options.delete(:skip_blanks)
|
|
|
|
|
|
1786
|
@encoding = options.delete(:encoding) # nil will use $KCODE
|
|
|
|
|
|
1787
|
@field_size_limit = options.delete(:field_size_limit)
|
|
|
|
|
|
1788
|
|
|
|
|
|
|
1789
|
# prebuild Regexps for faster parsing
|
|
|
|
|
|
1790
|
esc_col_sep = Regexp.escape(@col_sep)
|
|
|
|
|
|
1791
|
esc_row_sep = Regexp.escape(@row_sep)
|
|
|
|
|
|
1792
|
esc_quote = Regexp.escape(@quote_char)
|
|
|
|
|
|
1793
|
@parsers = {
|
|
|
|
|
|
1794
|
:any_field => Regexp.new( "[^#{esc_col_sep}]+",
|
|
|
|
|
|
1795
|
Regexp::MULTILINE,
|
|
|
|
|
|
1796
|
@encoding ),
|
|
|
|
|
|
1797
|
:quoted_field => Regexp.new( "^#{esc_quote}(.*)#{esc_quote}$",
|
|
|
|
|
|
1798
|
Regexp::MULTILINE,
|
|
|
|
|
|
1799
|
@encoding ),
|
|
|
|
|
|
1800
|
# safer than chomp!()
|
|
|
|
|
|
1801
|
:line_end => Regexp.new("#{esc_row_sep}\\z", nil, @encoding)
|
|
|
|
|
|
1802
|
}
|
|
|
|
|
|
1803
|
end
|
|
|
|
|
|
1804
|
|
|
|
|
|
|
1805
|
#
|
|
|
|
|
|
1806
|
# Loads any converters requested during construction.
|
|
|
|
|
|
1807
|
#
|
|
|
|
|
|
1808
|
# If +field_name+ is set <tt>:converters</tt> (the default) field converters
|
|
|
|
|
|
1809
|
# are set. When +field_name+ is <tt>:header_converters</tt> header converters
|
|
|
|
|
|
1810
|
# are added instead.
|
|
|
|
|
|
1811
|
#
|
|
|
|
|
|
1812
|
# The <tt>:unconverted_fields</tt> option is also actived for
|
|
|
|
|
|
1813
|
# <tt>:converters</tt> calls, if requested.
|
|
|
|
|
|
1814
|
#
|
|
|
|
|
|
1815
|
def init_converters(options, field_name = :converters)
|
|
|
|
|
|
1816
|
if field_name == :converters
|
|
|
|
|
|
1817
|
@unconverted_fields = options.delete(:unconverted_fields)
|
|
|
|
|
|
1818
|
end
|
|
|
|
|
|
1819
|
|
|
|
|
|
|
1820
|
instance_variable_set("@#{field_name}", Array.new)
|
|
|
|
|
|
1821
|
|
|
|
|
|
|
1822
|
# find the correct method to add the coverters
|
|
|
|
|
|
1823
|
convert = method(field_name.to_s.sub(/ers\Z/, ""))
|
|
|
|
|
|
1824
|
|
|
|
|
|
|
1825
|
# load converters
|
|
|
|
|
|
1826
|
unless options[field_name].nil?
|
|
|
|
|
|
1827
|
# allow a single converter not wrapped in an Array
|
|
|
|
|
|
1828
|
unless options[field_name].is_a? Array
|
|
|
|
|
|
1829
|
options[field_name] = [options[field_name]]
|
|
|
|
|
|
1830
|
end
|
|
|
|
|
|
1831
|
# load each converter...
|
|
|
|
|
|
1832
|
options[field_name].each do |converter|
|
|
|
|
|
|
1833
|
if converter.is_a? Proc # custom code block
|
|
|
|
|
|
1834
|
convert.call(&converter)
|
|
|
|
|
|
1835
|
else # by name
|
|
|
|
|
|
1836
|
convert.call(converter)
|
|
|
|
|
|
1837
|
end
|
|
|
|
|
|
1838
|
end
|
|
|
|
|
|
1839
|
end
|
|
|
|
|
|
1840
|
|
|
|
|
|
|
1841
|
options.delete(field_name)
|
|
|
|
|
|
1842
|
end
|
|
|
|
|
|
1843
|
|
|
|
|
|
|
1844
|
# Stores header row settings and loads header converters, if needed.
|
|
|
|
|
|
1845
|
def init_headers(options)
|
|
|
|
|
|
1846
|
@use_headers = options.delete(:headers)
|
|
|
|
|
|
1847
|
@return_headers = options.delete(:return_headers)
|
|
|
|
|
|
1848
|
@write_headers = options.delete(:write_headers)
|
|
|
|
|
|
1849
|
|
|
|
|
|
|
1850
|
# headers must be delayed until shift(), in case they need a row of content
|
|
|
|
|
|
1851
|
@headers = nil
|
|
|
|
|
|
1852
|
|
|
|
|
|
|
1853
|
init_converters(options, :header_converters)
|
|
|
|
|
|
1854
|
end
|
|
|
|
|
|
1855
|
|
|
|
|
|
|
1856
|
#
|
|
|
|
|
|
1857
|
# The actual work method for adding converters, used by both
|
|
|
|
|
|
1858
|
# FasterCSV.convert() and FasterCSV.header_convert().
|
|
|
|
|
|
1859
|
#
|
|
|
|
|
|
1860
|
# This method requires the +var_name+ of the instance variable to place the
|
|
|
|
|
|
1861
|
# converters in, the +const+ Hash to lookup named converters in, and the
|
|
|
|
|
|
1862
|
# normal parameters of the FasterCSV.convert() and FasterCSV.header_convert()
|
|
|
|
|
|
1863
|
# methods.
|
|
|
|
|
|
1864
|
#
|
|
|
|
|
|
1865
|
def add_converter(var_name, const, name = nil, &converter)
|
|
|
|
|
|
1866
|
if name.nil? # custom converter
|
|
|
|
|
|
1867
|
instance_variable_get("@#{var_name}") << converter
|
|
|
|
|
|
1868
|
else # named converter
|
|
|
|
|
|
1869
|
combo = const[name]
|
|
|
|
|
|
1870
|
case combo
|
|
|
|
|
|
1871
|
when Array # combo converter
|
|
|
|
|
|
1872
|
combo.each do |converter_name|
|
|
|
|
|
|
1873
|
add_converter(var_name, const, converter_name)
|
|
|
|
|
|
1874
|
end
|
|
|
|
|
|
1875
|
else # individual named converter
|
|
|
|
|
|
1876
|
instance_variable_get("@#{var_name}") << combo
|
|
|
|
|
|
1877
|
end
|
|
|
|
|
|
1878
|
end
|
|
|
|
|
|
1879
|
end
|
|
|
|
|
|
1880
|
|
|
|
|
|
|
1881
|
#
|
|
|
|
|
|
1882
|
# Processes +fields+ with <tt>@converters</tt>, or <tt>@header_converters</tt>
|
|
|
|
|
|
1883
|
# if +headers+ is passed as +true+, returning the converted field set. Any
|
|
|
|
|
|
1884
|
# converter that changes the field into something other than a String halts
|
|
|
|
|
|
1885
|
# the pipeline of conversion for that field. This is primarily an efficiency
|
|
|
|
|
|
1886
|
# shortcut.
|
|
|
|
|
|
1887
|
#
|
|
|
|
|
|
1888
|
def convert_fields(fields, headers = false)
|
|
|
|
|
|
1889
|
# see if we are converting headers or fields
|
|
|
|
|
|
1890
|
converters = headers ? @header_converters : @converters
|
|
|
|
|
|
1891
|
|
|
|
|
|
|
1892
|
fields.enum_for(:each_with_index).map do |field, index| # map_with_index
|
|
|
|
|
|
1893
|
converters.each do |converter|
|
|
|
|
|
|
1894
|
field = if converter.arity == 1 # straight field converter
|
|
|
|
|
|
1895
|
converter[field]
|
|
|
|
|
|
1896
|
else # FieldInfo converter
|
|
|
|
|
|
1897
|
header = @use_headers && !headers ? @headers[index] : nil
|
|
|
|
|
|
1898
|
converter[field, FieldInfo.new(index, lineno, header)]
|
|
|
|
|
|
1899
|
end
|
|
|
|
|
|
1900
|
break unless field.is_a? String # short-curcuit pipeline for speed
|
|
|
|
|
|
1901
|
end
|
|
|
|
|
|
1902
|
field # return final state of each field, converted or original
|
|
|
|
|
|
1903
|
end
|
|
|
|
|
|
1904
|
end
|
|
|
|
|
|
1905
|
|
|
|
|
|
|
1906
|
#
|
|
|
|
|
|
1907
|
# This methods is used to turn a finished +row+ into a FasterCSV::Row. Header
|
|
|
|
|
|
1908
|
# rows are also dealt with here, either by returning a FasterCSV::Row with
|
|
|
|
|
|
1909
|
# identical headers and fields (save that the fields do not go through the
|
|
|
|
|
|
1910
|
# converters) or by reading past them to return a field row. Headers are also
|
|
|
|
|
|
1911
|
# saved in <tt>@headers</tt> for use in future rows.
|
|
|
|
|
|
1912
|
#
|
|
|
|
|
|
1913
|
# When +nil+, +row+ is assumed to be a header row not based on an actual row
|
|
|
|
|
|
1914
|
# of the stream.
|
|
|
|
|
|
1915
|
#
|
|
|
|
|
|
1916
|
def parse_headers(row = nil)
|
|
|
|
|
|
1917
|
if @headers.nil? # header row
|
|
|
|
|
|
1918
|
@headers = case @use_headers # save headers
|
|
|
|
|
|
1919
|
# Array of headers
|
|
|
|
|
|
1920
|
when Array then @use_headers
|
|
|
|
|
|
1921
|
# CSV header String
|
|
|
|
|
|
1922
|
when String
|
|
|
|
|
|
1923
|
self.class.parse_line( @use_headers,
|
|
|
|
|
|
1924
|
:col_sep => @col_sep,
|
|
|
|
|
|
1925
|
:row_sep => @row_sep,
|
|
|
|
|
|
1926
|
:quote_char => @quote_char )
|
|
|
|
|
|
1927
|
# first row is headers
|
|
|
|
|
|
1928
|
else row
|
|
|
|
|
|
1929
|
end
|
|
|
|
|
|
1930
|
|
|
|
|
|
|
1931
|
# prepare converted and unconverted copies
|
|
|
|
|
|
1932
|
row = @headers if row.nil?
|
|
|
|
|
|
1933
|
@headers = convert_fields(@headers, true)
|
|
|
|
|
|
1934
|
|
|
|
|
|
|
1935
|
if @return_headers # return headers
|
|
|
|
|
|
1936
|
return FasterCSV::Row.new(@headers, row, true)
|
|
|
|
|
|
1937
|
elsif not [Array, String].include? @use_headers.class # skip to field row
|
|
|
|
|
|
1938
|
return shift
|
|
|
|
|
|
1939
|
end
|
|
|
|
|
|
1940
|
end
|
|
|
|
|
|
1941
|
|
|
|
|
|
|
1942
|
FasterCSV::Row.new(@headers, convert_fields(row)) # field row
|
|
|
|
|
|
1943
|
end
|
|
|
|
|
|
1944
|
|
|
|
|
|
|
1945
|
#
|
|
|
|
|
|
1946
|
# Thiw methods injects an instance variable <tt>unconverted_fields</tt> into
|
|
|
|
|
|
1947
|
# +row+ and an accessor method for it called unconverted_fields(). The
|
|
|
|
|
|
1948
|
# variable is set to the contents of +fields+.
|
|
|
|
|
|
1949
|
#
|
|
|
|
|
|
1950
|
def add_unconverted_fields(row, fields)
|
|
|
|
|
|
1951
|
class << row
|
|
|
|
|
|
1952
|
attr_reader :unconverted_fields
|
|
|
|
|
|
1953
|
end
|
|
|
|
|
|
1954
|
row.instance_eval { @unconverted_fields = fields }
|
|
|
|
|
|
1955
|
row
|
|
|
|
|
|
1956
|
end
|
|
|
|
|
|
1957
|
end
|
|
|
|
|
|
1958
|
|
|
|
|
|
|
1959
|
# Another name for FasterCSV.
|
|
|
|
|
|
1960
|
FCSV = FasterCSV
|
|
|
|
|
|
1961
|
|
|
|
|
|
|
1962
|
# Another name for FasterCSV::instance().
|
|
|
|
|
|
1963
|
def FasterCSV(*args, &block)
|
|
|
|
|
|
1964
|
FasterCSV.instance(*args, &block)
|
|
|
|
|
|
1965
|
end
|
|
|
|
|
|
1966
|
|
|
|
|
|
|
1967
|
# Another name for FCSV::instance().
|
|
|
|
|
|
1968
|
def FCSV(*args, &block)
|
|
|
|
|
|
1969
|
FCSV.instance(*args, &block)
|
|
|
|
|
|
1970
|
end
|
|
|
|
|
|
1971
|
|
|
|
|
|
|
1972
|
class Array
|
|
|
|
|
|
1973
|
# Equivalent to <tt>FasterCSV::generate_line(self, options)</tt>.
|
|
|
|
|
|
1974
|
def to_csv(options = Hash.new)
|
|
|
|
|
|
1975
|
FasterCSV.generate_line(self, options)
|
|
|
|
|
|
1976
|
end
|
|
|
|
|
|
1977
|
end
|
|
|
|
|
|
1978
|
|
|
|
|
|
|
1979
|
class String
|
|
|
|
|
|
1980
|
# Equivalent to <tt>FasterCSV::parse_line(self, options)</tt>.
|
|
|
|
|
|
1981
|
def parse_csv(options = Hash.new)
|
|
|
|
|
|
1982
|
FasterCSV.parse_line(self, options)
|
|
|
|
|
|
1983
|
end
|
|
|
|
|
|
1984
|
end
|
