summaryrefslogtreecommitdiff
path: root/test/csv/test_csv_parsing.rb
blob: e65bbad92ec019bb53b49aecdd8626a432b4d548 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
#!/usr/bin/env ruby -w
# encoding: UTF-8
# frozen_string_literal: false

# tc_csv_parsing.rb
#
# Created by James Edward Gray II on 2005-10-31.

require "timeout"

require_relative "base"

#
# Following tests are my interpretation of the
# {CSV RCF}[http://www.ietf.org/rfc/rfc4180.txt].  I only deviate from that
# document in one place (intentionally) and that is to make the default row
# separator <tt>$/</tt>.
#
class TestCSV::Parsing < TestCSV
  extend DifferentOFS

  BIG_DATA = "123456789\n" * 1024

  def test_mastering_regex_example
    ex = %Q{Ten Thousand,10000, 2710 ,,"10,000","It's ""10 Grand"", baby",10K}
    assert_equal( [ "Ten Thousand", "10000", " 2710 ", nil, "10,000",
                    "It's \"10 Grand\", baby", "10K" ],
                  CSV.parse_line(ex) )
  end

  # Old Ruby 1.8 CSV library tests.
  def test_std_lib_csv
    [ ["\t", ["\t"]],
      ["foo,\"\"\"\"\"\",baz", ["foo", "\"\"", "baz"]],
      ["foo,\"\"\"bar\"\"\",baz", ["foo", "\"bar\"", "baz"]],
      ["\"\"\"\n\",\"\"\"\n\"", ["\"\n", "\"\n"]],
      ["foo,\"\r\n\",baz", ["foo", "\r\n", "baz"]],
      ["\"\"", [""]],
      ["foo,\"\"\"\",baz", ["foo", "\"", "baz"]],
      ["foo,\"\r.\n\",baz", ["foo", "\r.\n", "baz"]],
      ["foo,\"\r\",baz", ["foo", "\r", "baz"]],
      ["foo,\"\",baz", ["foo", "", "baz"]],
      ["\",\"", [","]],
      ["foo", ["foo"]],
      [",,", [nil, nil, nil]],
      [",", [nil, nil]],
      ["foo,\"\n\",baz", ["foo", "\n", "baz"]],
      ["foo,,baz", ["foo", nil, "baz"]],
      ["\"\"\"\r\",\"\"\"\r\"", ["\"\r", "\"\r"]],
      ["\",\",\",\"", [",", ","]],
      ["foo,bar,", ["foo", "bar", nil]],
      [",foo,bar", [nil, "foo", "bar"]],
      ["foo,bar", ["foo", "bar"]],
      [";", [";"]],
      ["\t,\t", ["\t", "\t"]],
      ["foo,\"\r\n\r\",baz", ["foo", "\r\n\r", "baz"]],
      ["foo,\"\r\n\n\",baz", ["foo", "\r\n\n", "baz"]],
      ["foo,\"foo,bar\",baz", ["foo", "foo,bar", "baz"]],
      [";,;", [";", ";"]] ].each do |csv_test|
      assert_equal(csv_test.last, CSV.parse_line(csv_test.first))
    end

    [ ["foo,\"\"\"\"\"\",baz", ["foo", "\"\"", "baz"]],
      ["foo,\"\"\"bar\"\"\",baz", ["foo", "\"bar\"", "baz"]],
      ["foo,\"\r\n\",baz", ["foo", "\r\n", "baz"]],
      ["\"\"", [""]],
      ["foo,\"\"\"\",baz", ["foo", "\"", "baz"]],
      ["foo,\"\r.\n\",baz", ["foo", "\r.\n", "baz"]],
      ["foo,\"\r\",baz", ["foo", "\r", "baz"]],
      ["foo,\"\",baz", ["foo", "", "baz"]],
      ["foo", ["foo"]],
      [",,", [nil, nil, nil]],
      [",", [nil, nil]],
      ["foo,\"\n\",baz", ["foo", "\n", "baz"]],
      ["foo,,baz", ["foo", nil, "baz"]],
      ["foo,bar", ["foo", "bar"]],
      ["foo,\"\r\n\n\",baz", ["foo", "\r\n\n", "baz"]],
      ["foo,\"foo,bar\",baz", ["foo", "foo,bar", "baz"]] ].each do |csv_test|
      assert_equal(csv_test.last, CSV.parse_line(csv_test.first))
    end
  end

  # From:  http://ruby-talk.org/cgi-bin/scat.rb/ruby/ruby-core/6496
  def test_aras_edge_cases
    [ [%Q{a,b},               ["a", "b"]],
      [%Q{a,"""b"""},         ["a", "\"b\""]],
      [%Q{a,"""b"},           ["a", "\"b"]],
      [%Q{a,"b"""},           ["a", "b\""]],
      [%Q{a,"\nb"""},         ["a", "\nb\""]],
      [%Q{a,"""\nb"},         ["a", "\"\nb"]],
      [%Q{a,"""\nb\n"""},     ["a", "\"\nb\n\""]],
      [%Q{a,"""\nb\n""",\nc}, ["a", "\"\nb\n\"", nil]],
      [%Q{a,,,},              ["a", nil, nil, nil]],
      [%Q{,},                 [nil, nil]],
      [%Q{"",""},             ["", ""]],
      [%Q{""""},              ["\""]],
      [%Q{"""",""},           ["\"",""]],
      [%Q{,""},               [nil,""]],
      [%Q{,"\r"},             [nil,"\r"]],
      [%Q{"\r\n,"},           ["\r\n,"]],
      [%Q{"\r\n,",},          ["\r\n,", nil]] ].each do |edge_case|
        assert_equal(edge_case.last, CSV.parse_line(edge_case.first))
      end
  end

  def test_james_edge_cases
    # A read at eof? should return nil.
    assert_equal(nil, CSV.parse_line(""))
    #
    # With Ruby 1.8 CSV it's impossible to tell an empty line from a line
    # containing a single +nil+ field.  The old CSV library returns
    # <tt>[nil]</tt> in these cases, but <tt>Array.new</tt> makes more sense to
    # me.
    #
    assert_equal(Array.new, CSV.parse_line("\n1,2,3\n"))
  end

  def test_rob_edge_cases
    [ [%Q{"a\nb"},                         ["a\nb"]],
      [%Q{"\n\n\n"},                       ["\n\n\n"]],
      [%Q{a,"b\n\nc"},                     ['a', "b\n\nc"]],
      [%Q{,"\r\n"},                        [nil,"\r\n"]],
      [%Q{,"\r\n."},                       [nil,"\r\n."]],
      [%Q{"a\na","one newline"},           ["a\na", 'one newline']],
      [%Q{"a\n\na","two newlines"},        ["a\n\na", 'two newlines']],
      [%Q{"a\r\na","one CRLF"},            ["a\r\na", 'one CRLF']],
      [%Q{"a\r\n\r\na","two CRLFs"},       ["a\r\n\r\na", 'two CRLFs']],
      [%Q{with blank,"start\n\nfinish"\n}, ['with blank', "start\n\nfinish"]],
    ].each do |edge_case|
      assert_equal(edge_case.last, CSV.parse_line(edge_case.first))
    end
  end

  def test_non_regex_edge_cases
    # An early version of the non-regex parser fails this test
    [ [ "foo,\"foo,bar,baz,foo\",\"foo\"",
        ["foo", "foo,bar,baz,foo", "foo"] ] ].each do |edge_case|
      assert_equal(edge_case.last, CSV.parse_line(edge_case.first))
    end

    assert_raise(CSV::MalformedCSVError) do
      CSV.parse_line("1,\"23\"4\"5\", 6")
    end
  end

  def test_malformed_csv
    assert_raise(CSV::MalformedCSVError) do
      CSV.parse_line("1,2\r,3", row_sep: "\n")
    end

    bad_data = <<-CSV
line,1,abc
line,2,"def\nghi"

line,4,some\rjunk
line,5,jkl
    CSV
    lines = bad_data.lines.to_a
    assert_equal(6, lines.size)
    assert_match(/\Aline,4/, lines.find { |l| l =~ /some\rjunk/ })

    csv = CSV.new(bad_data)
    begin
      loop do
        assert_not_nil(csv.shift)
        assert_send([csv.lineno, :<, 4])
      end
    rescue CSV::MalformedCSVError
      assert_equal( "Unquoted fields do not allow \\r or \\n in line 4.",
                    $!.message )
    end

    assert_raise(CSV::MalformedCSVError) { CSV.parse_line('1,2,"3...') }

    bad_data = <<-CSV
line,1,abc
line,2,"def\nghi"

line,4,8'10"
line,5,jkl
    CSV
    lines = bad_data.lines.to_a
    assert_equal(6, lines.size)
    assert_match(/\Aline,4/, lines.find { |l| l =~ /8'10"/ })

    csv = CSV.new(bad_data)
    begin
      loop do
        assert_not_nil(csv.shift)
        assert_send([csv.lineno, :<, 4])
      end
    rescue CSV::MalformedCSVError
      assert_equal("Illegal quoting in line 4.", $!.message)
    end
  end

  def test_the_parse_fails_fast_when_it_can_for_unquoted_fields
    assert_parse_errors_out('valid,fields,bad start"' + BIG_DATA)
  end

  def test_the_parse_fails_fast_when_it_can_for_unescaped_quotes
    assert_parse_errors_out('valid,fields,"bad start"unescaped' + BIG_DATA)
  end

  def test_field_size_limit_controls_lookahead
    assert_parse_errors_out( 'valid,fields,"' + BIG_DATA + '"',
                             field_size_limit: 2048 )
  end

  def test_field_size_limit_in_extended_column_not_exceeding
    data = <<~DATA
      "a","b"
      "
      2
      ",""
    DATA
    assert_nothing_raised(CSV::MalformedCSVError) do
      CSV.parse(data, field_size_limit: 4)
    end
  end

  def test_field_size_limit_in_extended_column_exceeding
    data = <<~DATA
      "a","b"
      "
      2345
      ",""
    DATA
    assert_parse_errors_out(data, field_size_limit: 5)
  end

  def test_col_sep_comma
    assert_equal([["a", "b", nil, "d"]],
                 CSV.parse("a,b,,d", col_sep: ","))
  end

  def test_col_sep_space
    assert_equal([["a", "b", nil, "d"]],
                 CSV.parse("a b  d", col_sep: " "))
  end

  private

  def assert_parse_errors_out(*args)
    assert_raise(CSV::MalformedCSVError) do
      Timeout.timeout(0.2) do
        CSV.parse(*args)
        fail("Parse didn't error out")
      end
    end
  end
end