smarter_csv 1.6.1 → 1.7.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.rubocop.yml +133 -0
- data/CHANGELOG.md +22 -1
- data/CONTRIBUTORS.md +3 -0
- data/Gemfile +7 -4
- data/README.md +8 -6
- data/Rakefile +15 -13
- data/ext/smarter_csv/extconf.rb +14 -0
- data/ext/smarter_csv/smarter_csv.c +86 -0
- data/lib/extensions/hash.rb +4 -2
- data/lib/smarter_csv/version.rb +3 -1
- data/lib/smarter_csv.rb +519 -10
- data/smarter_csv.gemspec +22 -7
- metadata +54 -176
- data/.gitignore +0 -10
- data/.rspec +0 -2
- data/.travis.yml +0 -27
- data/lib/smarter_csv/smarter_csv.rb +0 -461
- data/spec/fixtures/additional_separator.csv +0 -6
- data/spec/fixtures/basic.csv +0 -8
- data/spec/fixtures/binary.csv +0 -1
- data/spec/fixtures/carriage_returns_n.csv +0 -18
- data/spec/fixtures/carriage_returns_quoted.csv +0 -3
- data/spec/fixtures/carriage_returns_r.csv +0 -1
- data/spec/fixtures/carriage_returns_rn.csv +0 -18
- data/spec/fixtures/chunk_cornercase.csv +0 -10
- data/spec/fixtures/duplicate_headers.csv +0 -3
- data/spec/fixtures/empty.csv +0 -5
- data/spec/fixtures/empty_columns_1.csv +0 -2
- data/spec/fixtures/empty_columns_2.csv +0 -2
- data/spec/fixtures/hard_sample.csv +0 -2
- data/spec/fixtures/ignore_comments.csv +0 -11
- data/spec/fixtures/ignore_comments2.csv +0 -3
- data/spec/fixtures/key_mapping.csv +0 -2
- data/spec/fixtures/line_endings_n.csv +0 -4
- data/spec/fixtures/line_endings_r.csv +0 -1
- data/spec/fixtures/line_endings_rn.csv +0 -4
- data/spec/fixtures/lots_of_columns.csv +0 -2
- data/spec/fixtures/malformed.csv +0 -3
- data/spec/fixtures/malformed_header.csv +0 -3
- data/spec/fixtures/money.csv +0 -3
- data/spec/fixtures/no_header.csv +0 -7
- data/spec/fixtures/numeric.csv +0 -5
- data/spec/fixtures/pets.csv +0 -5
- data/spec/fixtures/problematic.csv +0 -8
- data/spec/fixtures/quote_char.csv +0 -9
- data/spec/fixtures/quoted.csv +0 -5
- data/spec/fixtures/quoted2.csv +0 -4
- data/spec/fixtures/separator_colon.csv +0 -4
- data/spec/fixtures/separator_comma.csv +0 -4
- data/spec/fixtures/separator_pipe.csv +0 -4
- data/spec/fixtures/separator_semi.csv +0 -4
- data/spec/fixtures/separator_tab.csv +0 -4
- data/spec/fixtures/skip_lines.csv +0 -8
- data/spec/fixtures/trading.csv +0 -3
- data/spec/fixtures/user_import.csv +0 -3
- data/spec/fixtures/valid_unicode.csv +0 -5
- data/spec/fixtures/with_dashes.csv +0 -8
- data/spec/fixtures/with_dates.csv +0 -4
- data/spec/smarter_csv/additional_separator_spec.rb +0 -45
- data/spec/smarter_csv/binary_file2_spec.rb +0 -24
- data/spec/smarter_csv/binary_file_spec.rb +0 -22
- data/spec/smarter_csv/blank_spec.rb +0 -55
- data/spec/smarter_csv/carriage_return_spec.rb +0 -190
- data/spec/smarter_csv/chunked_reading_spec.rb +0 -14
- data/spec/smarter_csv/close_file_spec.rb +0 -15
- data/spec/smarter_csv/column_separator_spec.rb +0 -95
- data/spec/smarter_csv/convert_values_to_numeric_spec.rb +0 -48
- data/spec/smarter_csv/duplicate_headers_spec.rb +0 -76
- data/spec/smarter_csv/empty_columns_spec.rb +0 -74
- data/spec/smarter_csv/extenstions_spec.rb +0 -17
- data/spec/smarter_csv/hard_sample_spec.rb +0 -24
- data/spec/smarter_csv/header_transformation_spec.rb +0 -21
- data/spec/smarter_csv/ignore_comments_spec.rb +0 -45
- data/spec/smarter_csv/invalid_headers_spec.rb +0 -38
- data/spec/smarter_csv/keep_headers_spec.rb +0 -24
- data/spec/smarter_csv/key_mapping_spec.rb +0 -56
- data/spec/smarter_csv/line_ending_spec.rb +0 -43
- data/spec/smarter_csv/load_basic_spec.rb +0 -20
- data/spec/smarter_csv/malformed_spec.rb +0 -25
- data/spec/smarter_csv/no_header_spec.rb +0 -29
- data/spec/smarter_csv/not_downcase_header_spec.rb +0 -24
- data/spec/smarter_csv/parse/column_separator_spec.rb +0 -61
- data/spec/smarter_csv/parse/old_csv_library_spec.rb +0 -74
- data/spec/smarter_csv/parse/rfc4180_and_more_spec.rb +0 -170
- data/spec/smarter_csv/problematic.rb +0 -34
- data/spec/smarter_csv/quoted_spec.rb +0 -52
- data/spec/smarter_csv/remove_empty_values_spec.rb +0 -13
- data/spec/smarter_csv/remove_keys_from_hashes_spec.rb +0 -25
- data/spec/smarter_csv/remove_not_mapped_keys_spec.rb +0 -35
- data/spec/smarter_csv/remove_values_matching_spec.rb +0 -26
- data/spec/smarter_csv/remove_zero_values_spec.rb +0 -25
- data/spec/smarter_csv/skip_lines_spec.rb +0 -29
- data/spec/smarter_csv/strings_as_keys_spec.rb +0 -24
- data/spec/smarter_csv/strip_chars_from_headers_spec.rb +0 -24
- data/spec/smarter_csv/trading_spec.rb +0 -25
- data/spec/smarter_csv/valid_unicode_spec.rb +0 -94
- data/spec/smarter_csv/value_converters_spec.rb +0 -52
- data/spec/spec/spec_helper.rb +0 -17
- data/spec/spec.opts +0 -2
- data/spec/spec_helper.rb +0 -21
data/lib/smarter_csv.rb
CHANGED
@@ -1,12 +1,521 @@
|
|
1
|
-
|
2
|
-
|
3
|
-
|
4
|
-
|
5
|
-
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require_relative "extensions/hash"
|
4
|
+
require_relative "smarter_csv/version"
|
5
|
+
require_relative "smarter_csv/smarter_csv" unless ENV['CI'] # does not compile/link in CI?
|
6
|
+
|
7
|
+
module SmarterCSV
|
8
|
+
class SmarterCSVException < StandardError; end
|
9
|
+
class HeaderSizeMismatch < SmarterCSVException; end
|
10
|
+
class IncorrectOption < SmarterCSVException; end
|
11
|
+
class DuplicateHeaders < SmarterCSVException; end
|
12
|
+
class MissingHeaders < SmarterCSVException; end
|
13
|
+
class NoColSepDetected < SmarterCSVException; end
|
14
|
+
class KeyMappingError < SmarterCSVException; end
|
15
|
+
class MalformedCSVError < SmarterCSVException; end
|
16
|
+
|
17
|
+
# first parameter: filename or input object which responds to readline method
|
18
|
+
def SmarterCSV.process(input, options = {}, &block)
|
19
|
+
options = default_options.merge(options)
|
20
|
+
options[:invalid_byte_sequence] = '' if options[:invalid_byte_sequence].nil?
|
21
|
+
puts "SmarterCSV OPTIONS: #{options.inspect}" if options[:verbose]
|
22
|
+
|
23
|
+
headerA = []
|
24
|
+
result = []
|
25
|
+
@file_line_count = 0
|
26
|
+
@csv_line_count = 0
|
27
|
+
has_rails = !!defined?(Rails)
|
28
|
+
begin
|
29
|
+
fh = input.respond_to?(:readline) ? input : File.open(input, "r:#{options[:file_encoding]}")
|
30
|
+
|
31
|
+
# auto-detect the row separator
|
32
|
+
options[:row_sep] = guess_line_ending(fh, options) if options[:row_sep]&.to_sym == :auto
|
33
|
+
# attempt to auto-detect column separator
|
34
|
+
options[:col_sep] = guess_column_separator(fh, options) if options[:col_sep]&.to_sym == :auto
|
35
|
+
|
36
|
+
if (options[:force_utf8] || options[:file_encoding] =~ /utf-8/i) && (fh.respond_to?(:external_encoding) && fh.external_encoding != Encoding.find('UTF-8') || fh.respond_to?(:encoding) && fh.encoding != Encoding.find('UTF-8'))
|
37
|
+
puts 'WARNING: you are trying to process UTF-8 input, but did not open the input with "b:utf-8" option. See README file "NOTES about File Encodings".'
|
38
|
+
end
|
39
|
+
|
40
|
+
if options[:skip_lines].to_i > 0
|
41
|
+
options[:skip_lines].to_i.times do
|
42
|
+
readline_with_counts(fh, options)
|
43
|
+
end
|
44
|
+
end
|
45
|
+
|
46
|
+
headerA, header_size = process_headers(fh, options)
|
47
|
+
|
48
|
+
# in case we use chunking.. we'll need to set it up..
|
49
|
+
if !options[:chunk_size].nil? && options[:chunk_size].to_i > 0
|
50
|
+
use_chunks = true
|
51
|
+
chunk_size = options[:chunk_size].to_i
|
52
|
+
chunk_count = 0
|
53
|
+
chunk = []
|
54
|
+
else
|
55
|
+
use_chunks = false
|
56
|
+
end
|
57
|
+
|
58
|
+
# now on to processing all the rest of the lines in the CSV file:
|
59
|
+
until fh.eof? # we can't use fh.readlines() here, because this would read the whole file into memory at once, and eof => true
|
60
|
+
line = readline_with_counts(fh, options)
|
61
|
+
|
62
|
+
# replace invalid byte sequence in UTF-8 with question mark to avoid errors
|
63
|
+
line = line.force_encoding('utf-8').encode('utf-8', invalid: :replace, undef: :replace, replace: options[:invalid_byte_sequence]) if options[:force_utf8] || options[:file_encoding] !~ /utf-8/i
|
64
|
+
|
65
|
+
print "processing file line %10d, csv line %10d\r" % [@file_line_count, @csv_line_count] if options[:verbose]
|
66
|
+
|
67
|
+
next if options[:comment_regexp] && line =~ options[:comment_regexp] # ignore all comment lines if there are any
|
68
|
+
|
69
|
+
# cater for the quoted csv data containing the row separator carriage return character
|
70
|
+
# in which case the row data will be split across multiple lines (see the sample content in spec/fixtures/carriage_returns_rn.csv)
|
71
|
+
# by detecting the existence of an uneven number of quote characters
|
72
|
+
|
73
|
+
multiline = line.count(options[:quote_char]).odd? # should handle quote_char nil
|
74
|
+
while line.count(options[:quote_char]).odd? # should handle quote_char nil
|
75
|
+
next_line = fh.readline(options[:row_sep])
|
76
|
+
next_line = next_line.force_encoding('utf-8').encode('utf-8', invalid: :replace, undef: :replace, replace: options[:invalid_byte_sequence]) if options[:force_utf8] || options[:file_encoding] !~ /utf-8/i
|
77
|
+
line += next_line
|
78
|
+
@file_line_count += 1
|
79
|
+
end
|
80
|
+
print "\nline contains uneven number of quote chars so including content through file line %d\n" % @file_line_count if options[:verbose] && multiline
|
81
|
+
|
82
|
+
line.chomp!(options[:row_sep])
|
83
|
+
|
84
|
+
dataA, _data_size = parse(line, options, header_size)
|
85
|
+
|
86
|
+
dataA.map!{|x| x.strip} if options[:strip_whitespace]
|
87
|
+
|
88
|
+
# if all values are blank, then ignore this line
|
89
|
+
next if options[:remove_empty_hashes] && (dataA.empty? || blank?(dataA))
|
90
|
+
|
91
|
+
hash = Hash.zip(headerA, dataA) # from Facets of Ruby library
|
92
|
+
|
93
|
+
# make sure we delete any key/value pairs from the hash, which the user wanted to delete:
|
94
|
+
# Note: Ruby < 1.9 doesn't allow empty symbol literals!
|
95
|
+
hash.delete(nil)
|
96
|
+
hash.delete('')
|
97
|
+
eval('hash.delete(:"")') if RUBY_VERSION.to_f > 1.8
|
98
|
+
|
99
|
+
if options[:remove_empty_values] == true
|
100
|
+
hash.delete_if{|_k, v| has_rails ? v.blank? : blank?(v)}
|
101
|
+
end
|
102
|
+
|
103
|
+
hash.delete_if{|_k, v| !v.nil? && v =~ /^(\d+|\d+\.\d+)$/ && v.to_f == 0} if options[:remove_zero_values] # values are typically Strings!
|
104
|
+
hash.delete_if{|_k, v| v =~ options[:remove_values_matching]} if options[:remove_values_matching]
|
105
|
+
|
106
|
+
if options[:convert_values_to_numeric]
|
107
|
+
hash.each do |k, v|
|
108
|
+
# deal with the :only / :except options to :convert_values_to_numeric
|
109
|
+
next if only_or_except_limit_execution(options, :convert_values_to_numeric, k)
|
110
|
+
|
111
|
+
# convert if it's a numeric value:
|
112
|
+
case v
|
113
|
+
when /^[+-]?\d+\.\d+$/
|
114
|
+
hash[k] = v.to_f
|
115
|
+
when /^[+-]?\d+$/
|
116
|
+
hash[k] = v.to_i
|
117
|
+
end
|
118
|
+
end
|
119
|
+
end
|
120
|
+
|
121
|
+
if options[:value_converters]
|
122
|
+
hash.each do |k, v|
|
123
|
+
converter = options[:value_converters][k]
|
124
|
+
next unless converter
|
125
|
+
|
126
|
+
hash[k] = converter.convert(v)
|
127
|
+
end
|
128
|
+
end
|
129
|
+
|
130
|
+
next if options[:remove_empty_hashes] && hash.empty?
|
131
|
+
|
132
|
+
if use_chunks
|
133
|
+
chunk << hash # append temp result to chunk
|
134
|
+
|
135
|
+
if chunk.size >= chunk_size || fh.eof? # if chunk if full, or EOF reached
|
136
|
+
# do something with the chunk
|
137
|
+
if block_given?
|
138
|
+
yield chunk # do something with the hashes in the chunk in the block
|
139
|
+
else
|
140
|
+
result << chunk # not sure yet, why anybody would want to do this without a block
|
141
|
+
end
|
142
|
+
chunk_count += 1
|
143
|
+
chunk = [] # initialize for next chunk of data
|
144
|
+
else
|
145
|
+
|
146
|
+
# the last chunk may contain partial data, which also needs to be returned (BUG / ISSUE-18)
|
147
|
+
|
148
|
+
end
|
149
|
+
|
150
|
+
# while a chunk is being filled up we don't need to do anything else here
|
151
|
+
|
152
|
+
else # no chunk handling
|
153
|
+
if block_given?
|
154
|
+
yield [hash] # do something with the hash in the block (better to use chunking here)
|
155
|
+
else
|
156
|
+
result << hash
|
157
|
+
end
|
158
|
+
end
|
159
|
+
end
|
160
|
+
|
161
|
+
# print new line to retain last processing line message
|
162
|
+
print "\n" if options[:verbose]
|
163
|
+
|
164
|
+
# last chunk:
|
165
|
+
if !chunk.nil? && chunk.size > 0
|
166
|
+
# do something with the chunk
|
167
|
+
if block_given?
|
168
|
+
yield chunk # do something with the hashes in the chunk in the block
|
169
|
+
else
|
170
|
+
result << chunk # not sure yet, why anybody would want to do this without a block
|
171
|
+
end
|
172
|
+
chunk_count += 1
|
173
|
+
chunk = [] # initialize for next chunk of data
|
174
|
+
end
|
175
|
+
ensure
|
176
|
+
fh.close if fh.respond_to?(:close)
|
177
|
+
end
|
178
|
+
if block_given?
|
179
|
+
return chunk_count # when we do processing through a block we only care how many chunks we processed
|
180
|
+
else
|
181
|
+
return result # returns either an Array of Hashes, or an Array of Arrays of Hashes (if in chunked mode)
|
182
|
+
end
|
6
183
|
end
|
7
|
-
end
|
8
184
|
|
9
|
-
|
10
|
-
|
11
|
-
|
12
|
-
|
185
|
+
class << self
|
186
|
+
def has_acceleration?
|
187
|
+
@has_acceleration ||= !!defined?(parse_csv_line_c)
|
188
|
+
end
|
189
|
+
|
190
|
+
def raw_header
|
191
|
+
@raw_header
|
192
|
+
end
|
193
|
+
|
194
|
+
def headers
|
195
|
+
@headers
|
196
|
+
end
|
197
|
+
|
198
|
+
protected
|
199
|
+
|
200
|
+
# NOTE: this is not called when "parse" methods are tested by themselves
|
201
|
+
def default_options
|
202
|
+
{
|
203
|
+
acceleration: true,
|
204
|
+
auto_row_sep_chars: 500,
|
205
|
+
chunk_size: nil,
|
206
|
+
col_sep: ',',
|
207
|
+
comment_regexp: nil, # was: /\A#/,
|
208
|
+
convert_values_to_numeric: true,
|
209
|
+
downcase_header: true,
|
210
|
+
duplicate_header_suffix: nil,
|
211
|
+
file_encoding: 'utf-8',
|
212
|
+
force_simple_split: false,
|
213
|
+
force_utf8: false,
|
214
|
+
headers_in_file: true,
|
215
|
+
invalid_byte_sequence: '',
|
216
|
+
keep_original_headers: false,
|
217
|
+
key_mapping_hash: nil,
|
218
|
+
quote_char: '"',
|
219
|
+
remove_empty_hashes: true,
|
220
|
+
remove_empty_values: true,
|
221
|
+
remove_unmapped_keys: false,
|
222
|
+
remove_values_matching: nil,
|
223
|
+
remove_zero_values: false,
|
224
|
+
required_headers: nil,
|
225
|
+
row_sep: $/,
|
226
|
+
skip_lines: nil,
|
227
|
+
strings_as_keys: false,
|
228
|
+
strip_chars_from_headers: nil,
|
229
|
+
strip_whitespace: true,
|
230
|
+
user_provided_headers: nil,
|
231
|
+
value_converters: nil,
|
232
|
+
verbose: false,
|
233
|
+
}
|
234
|
+
end
|
235
|
+
|
236
|
+
def readline_with_counts(filehandle, options)
|
237
|
+
line = filehandle.readline(options[:row_sep])
|
238
|
+
@file_line_count += 1
|
239
|
+
@csv_line_count += 1
|
240
|
+
line
|
241
|
+
end
|
242
|
+
|
243
|
+
###
|
244
|
+
### Thin wrapper around C-extension
|
245
|
+
###
|
246
|
+
def parse(line, options, header_size = nil)
|
247
|
+
# puts "SmarterCSV.parse OPTIONS: #{options[:acceleration]}" if options[:verbose]
|
248
|
+
|
249
|
+
if options[:acceleration] && has_acceleration?
|
250
|
+
# puts "NOTICE: Accelerated SmarterCSV / #{options[:acceleration]}" if options[:verbose]
|
251
|
+
has_quotes = line =~ /#{options[:quote_char]}/
|
252
|
+
elements = parse_csv_line_c(line, options[:col_sep], options[:quote_char], header_size)
|
253
|
+
elements.map!{|x| cleanup_quotes(x, options[:quote_char])} if has_quotes
|
254
|
+
return [elements, elements.size]
|
255
|
+
|
256
|
+
else
|
257
|
+
# puts "WARNING: SmarterCSV is using un-accelerated parsing of lines. Check options[:acceleration]"
|
258
|
+
return parse_csv_line_ruby(line, options, header_size)
|
259
|
+
end
|
260
|
+
end
|
261
|
+
|
262
|
+
# ------------------------------------------------------------------
|
263
|
+
# Ruby equivalent of the C-extension for parse_line
|
264
|
+
#
|
265
|
+
# parses a single line: either a CSV header and body line
|
266
|
+
# - quoting rules compared to RFC-4180 are somewhat relaxed
|
267
|
+
# - we are not assuming that quotes inside a fields need to be doubled
|
268
|
+
# - we are not assuming that all fields need to be quoted (0 is even)
|
269
|
+
# - works with multi-char col_sep
|
270
|
+
# - if header_size is given, only up to header_size fields are parsed
|
271
|
+
#
|
272
|
+
# We use header_size for parsing the body lines to make sure we always match the number of headers
|
273
|
+
# in case there are trailing col_sep characters in line
|
274
|
+
#
|
275
|
+
# Our convention is that empty fields are returned as empty strings, not as nil.
|
276
|
+
#
|
277
|
+
#
|
278
|
+
# the purpose of the max_size parameter is to handle a corner case where
|
279
|
+
# CSV lines contain more fields than the header.
|
280
|
+
# In which case the remaining fields in the line are ignored
|
281
|
+
#
|
282
|
+
def parse_csv_line_ruby(line, options, header_size = nil)
|
283
|
+
return [] if line.nil?
|
284
|
+
|
285
|
+
line_size = line.size
|
286
|
+
col_sep = options[:col_sep]
|
287
|
+
col_sep_size = col_sep.size
|
288
|
+
quote = options[:quote_char]
|
289
|
+
quote_count = 0
|
290
|
+
elements = []
|
291
|
+
start = 0
|
292
|
+
i = 0
|
293
|
+
|
294
|
+
while i < line_size
|
295
|
+
if line[i...i+col_sep_size] == col_sep && quote_count.even?
|
296
|
+
break if !header_size.nil? && elements.size >= header_size
|
297
|
+
|
298
|
+
elements << cleanup_quotes(line[start...i], quote)
|
299
|
+
i += col_sep.size
|
300
|
+
start = i
|
301
|
+
else
|
302
|
+
quote_count += 1 if line[i] == quote
|
303
|
+
i += 1
|
304
|
+
end
|
305
|
+
end
|
306
|
+
elements << cleanup_quotes(line[start..-1], quote) if header_size.nil? || elements.size < header_size
|
307
|
+
[elements, elements.size]
|
308
|
+
end
|
309
|
+
|
310
|
+
def cleanup_quotes(field, quote)
|
311
|
+
return field if field.nil?
|
312
|
+
|
313
|
+
# return if field !~ /#{quote}/ # this check can probably eliminated
|
314
|
+
|
315
|
+
if field.start_with?(quote) && field.end_with?(quote)
|
316
|
+
field.delete_prefix!(quote)
|
317
|
+
field.delete_suffix!(quote)
|
318
|
+
end
|
319
|
+
field.gsub!("#{quote}#{quote}", quote)
|
320
|
+
field
|
321
|
+
end
|
322
|
+
|
323
|
+
# SEE: https://github.com/rails/rails/blob/32015b6f369adc839c4f0955f2d9dce50c0b6123/activesupport/lib/active_support/core_ext/object/blank.rb#L121
|
324
|
+
# and in the future we might also include UTF-8 space characters: https://www.compart.com/en/unicode/category/Zs
|
325
|
+
BLANK_RE = /\A\s*\z/.freeze
|
326
|
+
|
327
|
+
def blank?(value)
|
328
|
+
case value
|
329
|
+
when String
|
330
|
+
value.empty? || BLANK_RE.match?(value)
|
331
|
+
|
332
|
+
when NilClass
|
333
|
+
true
|
334
|
+
|
335
|
+
when Array
|
336
|
+
value.empty? || value.inject(true){|result, x| result &&= elem_blank?(x)}
|
337
|
+
|
338
|
+
when Hash
|
339
|
+
value.empty? || value.values.inject(true){|result, x| result &&= elem_blank?(x)}
|
340
|
+
|
341
|
+
else
|
342
|
+
false
|
343
|
+
end
|
344
|
+
end
|
345
|
+
|
346
|
+
def elem_blank?(value)
|
347
|
+
case value
|
348
|
+
when String
|
349
|
+
value.empty? || BLANK_RE.match?(value)
|
350
|
+
|
351
|
+
when NilClass
|
352
|
+
true
|
353
|
+
|
354
|
+
else
|
355
|
+
false
|
356
|
+
end
|
357
|
+
end
|
358
|
+
|
359
|
+
# acts as a road-block to limit processing when iterating over all k/v pairs of a CSV-hash:
|
360
|
+
def only_or_except_limit_execution(options, option_name, key)
|
361
|
+
if options[option_name].is_a?(Hash)
|
362
|
+
if options[option_name].has_key?(:except)
|
363
|
+
return true if Array(options[option_name][:except]).include?(key)
|
364
|
+
elsif options[option_name].has_key?(:only)
|
365
|
+
return true unless Array(options[option_name][:only]).include?(key)
|
366
|
+
end
|
367
|
+
end
|
368
|
+
return false
|
369
|
+
end
|
370
|
+
|
371
|
+
# raise exception if none is found
|
372
|
+
def guess_column_separator(filehandle, options)
|
373
|
+
del = [',', "\t", ';', ':', '|']
|
374
|
+
n = Hash.new(0)
|
375
|
+
|
376
|
+
5.times do
|
377
|
+
line = filehandle.readline(options[:row_sep])
|
378
|
+
del.each do |d|
|
379
|
+
n[d] += line.scan(d).count
|
380
|
+
end
|
381
|
+
rescue EOFError # short files
|
382
|
+
break
|
383
|
+
end
|
384
|
+
|
385
|
+
filehandle.rewind
|
386
|
+
raise SmarterCSV::NoColSepDetected if n.values.max == 0
|
387
|
+
|
388
|
+
col_sep = n.key(n.values.max)
|
389
|
+
end
|
390
|
+
|
391
|
+
# limitation: this currently reads the whole file in before making a decision
|
392
|
+
def guess_line_ending(filehandle, options)
|
393
|
+
counts = {"\n" => 0, "\r" => 0, "\r\n" => 0}
|
394
|
+
quoted_char = false
|
395
|
+
|
396
|
+
# count how many of the pre-defined line-endings we find
|
397
|
+
# ignoring those contained within quote characters
|
398
|
+
last_char = nil
|
399
|
+
lines = 0
|
400
|
+
filehandle.each_char do |c|
|
401
|
+
quoted_char = !quoted_char if c == options[:quote_char]
|
402
|
+
next if quoted_char
|
403
|
+
|
404
|
+
if last_char == "\r"
|
405
|
+
if c == "\n"
|
406
|
+
counts["\r\n"] += 1
|
407
|
+
else
|
408
|
+
counts["\r"] += 1 # \r are counted after they appeared
|
409
|
+
end
|
410
|
+
elsif c == "\n"
|
411
|
+
counts["\n"] += 1
|
412
|
+
end
|
413
|
+
last_char = c
|
414
|
+
lines += 1
|
415
|
+
break if options[:auto_row_sep_chars] && options[:auto_row_sep_chars] > 0 && lines >= options[:auto_row_sep_chars]
|
416
|
+
end
|
417
|
+
filehandle.rewind
|
418
|
+
|
419
|
+
counts["\r"] += 1 if last_char == "\r"
|
420
|
+
# find the most frequent key/value pair:
|
421
|
+
k, _ = counts.max_by{|_, v| v}
|
422
|
+
return k
|
423
|
+
end
|
424
|
+
|
425
|
+
def process_headers(filehandle, options)
|
426
|
+
@raw_header = nil
|
427
|
+
@headers = nil
|
428
|
+
if options[:headers_in_file] # extract the header line
|
429
|
+
# process the header line in the CSV file..
|
430
|
+
# the first line of a CSV file contains the header .. it might be commented out, so we need to read it anyhow
|
431
|
+
header = readline_with_counts(filehandle, options)
|
432
|
+
@raw_header = header
|
433
|
+
|
434
|
+
header = header.force_encoding('utf-8').encode('utf-8', invalid: :replace, undef: :replace, replace: options[:invalid_byte_sequence]) if options[:force_utf8] || options[:file_encoding] !~ /utf-8/i
|
435
|
+
header = header.sub(options[:comment_regexp], '') if options[:comment_regexp]
|
436
|
+
header = header.chomp(options[:row_sep])
|
437
|
+
|
438
|
+
header = header.gsub(options[:strip_chars_from_headers], '') if options[:strip_chars_from_headers]
|
439
|
+
|
440
|
+
file_headerA, file_header_size = parse(header, options)
|
441
|
+
|
442
|
+
file_headerA.map!{|x| x.gsub(%r/#{options[:quote_char]}/, '')}
|
443
|
+
file_headerA.map!{|x| x.strip} if options[:strip_whitespace]
|
444
|
+
unless options[:keep_original_headers]
|
445
|
+
file_headerA.map!{|x| x.gsub(/\s+|-+/, '_')}
|
446
|
+
file_headerA.map!{|x| x.downcase} if options[:downcase_header]
|
447
|
+
end
|
448
|
+
else
|
449
|
+
raise SmarterCSV::IncorrectOption, "ERROR: If :headers_in_file is set to false, you have to provide :user_provided_headers" unless options[:user_provided_headers]
|
450
|
+
end
|
451
|
+
if options[:user_provided_headers] && options[:user_provided_headers].class == Array && !options[:user_provided_headers].empty?
|
452
|
+
# use user-provided headers
|
453
|
+
headerA = options[:user_provided_headers]
|
454
|
+
if defined?(file_header_size) && !file_header_size.nil?
|
455
|
+
if headerA.size != file_header_size
|
456
|
+
raise SmarterCSV::HeaderSizeMismatch, "ERROR: :user_provided_headers defines #{headerA.size} headers != CSV-file has #{file_header_size} headers"
|
457
|
+
else
|
458
|
+
# we could print out the mapping of file_headerA to headerA here
|
459
|
+
end
|
460
|
+
end
|
461
|
+
else
|
462
|
+
headerA = file_headerA
|
463
|
+
end
|
464
|
+
|
465
|
+
# detect duplicate headers and disambiguate
|
466
|
+
headerA = process_duplicate_headers(headerA, options) if options[:duplicate_header_suffix]
|
467
|
+
header_size = headerA.size # used for splitting lines
|
468
|
+
|
469
|
+
headerA.map!{|x| x.to_sym } unless options[:strings_as_keys] || options[:keep_original_headers]
|
470
|
+
|
471
|
+
unless options[:user_provided_headers] # wouldn't make sense to re-map user provided headers
|
472
|
+
key_mappingH = options[:key_mapping]
|
473
|
+
|
474
|
+
# do some key mapping on the keys in the file header
|
475
|
+
# if you want to completely delete a key, then map it to nil or to ''
|
476
|
+
if !key_mappingH.nil? && key_mappingH.class == Hash && key_mappingH.keys.size > 0
|
477
|
+
# we can't map keys that are not there
|
478
|
+
missing_keys = key_mappingH.keys - headerA
|
479
|
+
puts "WARNING: missing header(s): #{missing_keys.join(",")}" unless missing_keys.empty?
|
480
|
+
|
481
|
+
headerA.map!{|x| key_mappingH.has_key?(x) ? (key_mappingH[x].nil? ? nil : key_mappingH[x]) : (options[:remove_unmapped_keys] ? nil : x)}
|
482
|
+
end
|
483
|
+
end
|
484
|
+
|
485
|
+
# header_validations
|
486
|
+
duplicate_headers = []
|
487
|
+
headerA.compact.each do |k|
|
488
|
+
duplicate_headers << k if headerA.select{|x| x == k}.size > 1
|
489
|
+
end
|
490
|
+
|
491
|
+
unless options[:user_provided_headers] || duplicate_headers.empty?
|
492
|
+
raise SmarterCSV::DuplicateHeaders, "ERROR: duplicate headers: #{duplicate_headers.join(',')}"
|
493
|
+
end
|
494
|
+
|
495
|
+
if options[:required_headers] && options[:required_headers].is_a?(Array)
|
496
|
+
missing_headers = []
|
497
|
+
options[:required_headers].each do |k|
|
498
|
+
missing_headers << k unless headerA.include?(k)
|
499
|
+
end
|
500
|
+
raise SmarterCSV::MissingHeaders, "ERROR: missing headers: #{missing_headers.join(',')}" unless missing_headers.empty?
|
501
|
+
end
|
502
|
+
|
503
|
+
@headers = headerA
|
504
|
+
[headerA, header_size]
|
505
|
+
end
|
506
|
+
|
507
|
+
def process_duplicate_headers(headers, options)
|
508
|
+
counts = Hash.new(0)
|
509
|
+
result = []
|
510
|
+
headers.each do |key|
|
511
|
+
counts[key] += 1
|
512
|
+
if counts[key] == 1
|
513
|
+
result << key
|
514
|
+
else
|
515
|
+
result << [key, options[:duplicate_header_suffix], counts[key]].join
|
516
|
+
end
|
517
|
+
end
|
518
|
+
result
|
519
|
+
end
|
520
|
+
end
|
521
|
+
end
|
data/smarter_csv.gemspec
CHANGED
@@ -12,14 +12,29 @@ Gem::Specification.new do |spec|
|
|
12
12
|
spec.homepage = "https://github.com/tilo/smarter_csv"
|
13
13
|
spec.license = 'MIT'
|
14
14
|
|
15
|
-
spec.
|
15
|
+
spec.metadata["homepage_uri"] = spec.homepage
|
16
|
+
spec.metadata["source_code_uri"] = spec.homepage
|
17
|
+
spec.metadata["changelog_uri"] = "https://github.com/tilo/smarter_csv/blob/main/CHANGELOG.md"
|
18
|
+
|
19
|
+
# Specify which files should be added to the gem when it is released.
|
20
|
+
# The `git ls-files -z` loads the files in the RubyGem that have been added into git.
|
21
|
+
spec.files = Dir.chdir(__dir__) do
|
22
|
+
`git ls-files -z`.split("\x0").reject do |f|
|
23
|
+
(f == __FILE__) ||
|
24
|
+
f.match(%r{\A(?:(?:bin|test|spec|features)/|\.(?:git|travis|circleci)|appveyor)}) || f.match(/\.h\z/)
|
25
|
+
end
|
26
|
+
end
|
27
|
+
spec.test_files = spec.files.grep(%r{^(test|spec|features)/})
|
28
|
+
|
16
29
|
spec.executables = spec.files.grep(%r{^bin/}).map{ |f| File.basename(f) }
|
17
|
-
spec.
|
18
|
-
spec.
|
30
|
+
spec.require_paths = ["lib"] # add ext here?
|
31
|
+
spec.extensions = ["ext/smarter_csv/extconf.rb"]
|
32
|
+
|
33
|
+
|
34
|
+
spec.add_development_dependency "awesome_print"
|
35
|
+
spec.add_development_dependency "codecov"
|
36
|
+
spec.add_development_dependency "pry"
|
19
37
|
spec.add_development_dependency "rspec"
|
38
|
+
spec.add_development_dependency "rubocop"
|
20
39
|
spec.add_development_dependency "simplecov"
|
21
|
-
spec.add_development_dependency "awesome_print"
|
22
|
-
# spec.add_development_dependency "guard-rspec"
|
23
|
-
|
24
|
-
spec.metadata["homepage_uri"] = spec.homepage
|
25
40
|
end
|