Module: SmarterCSV

Defined in:
lib/smarter_csv/parse.rb,
lib/smarter_csv/file_io.rb,
lib/smarter_csv/headers.rb,
lib/smarter_csv/version.rb,
lib/smarter_csv/variables.rb,
lib/smarter_csv/smarter_csv.rb,
lib/smarter_csv/auto_detection.rb,
lib/smarter_csv/options_processing.rb,
ext/smarter_csv/smarter_csv.c

Defined Under Namespace

Classes: DuplicateHeaders, HeaderSizeMismatch, IncorrectOption, KeyMappingError, MissingKeys, NoColSepDetected, SmarterCSVException, ValidationError

Constant Summary collapse

VERSION = 
"1.9.3"
DEFAULT_OPTIONS = 
{
  acceleration: true,
  auto_row_sep_chars: 500,
  chunk_size: nil,
  col_sep: :auto, # was: ',',
  comment_regexp: nil, # was: /\A#/,
  convert_values_to_numeric: true,
  downcase_header: true,
  duplicate_header_suffix: nil,
  file_encoding: 'utf-8',
  force_simple_split: false,
  force_utf8: false,
  headers_in_file: true,
  invalid_byte_sequence: '',
  keep_original_headers: false,
  key_mapping: nil,
  quote_char: '"',
  remove_empty_hashes: true,
  remove_empty_values: true,
  remove_unmapped_keys: false,
  remove_values_matching: nil,
  remove_zero_values: false,
  required_headers: nil,
  required_keys: nil,
  row_sep: :auto, # was: $/,
  silence_missing_keys: false,
  skip_lines: nil,
  strings_as_keys: false,
  strip_chars_from_headers: nil,
  strip_whitespace: true,
  user_provided_headers: nil,
  value_converters: nil,
  verbose: false,
  with_line_numbers: false,
}.freeze

Class Attribute Summary collapse

Class Method Summary collapse

Class Attribute Details

.chunk_count ⇒ Object (readonly)

Returns the value of attribute chunk_count.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def chunk_count
  @chunk_count
end

.csv_line_count ⇒ Object (readonly)

Returns the value of attribute csv_line_count.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def csv_line_count
  @csv_line_count
end

.errors ⇒ Object (readonly)

Returns the value of attribute errors.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def errors
  @errors
end

.file_line_count ⇒ Object (readonly)

Returns the value of attribute file_line_count.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def file_line_count
  @file_line_count
end

.headers ⇒ Object (readonly)

Returns the value of attribute headers.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def headers
  @headers
end

.raw_header ⇒ Object (readonly)

Returns the value of attribute raw_header.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def raw_header
  @raw_header
end

.result ⇒ Object (readonly)

Returns the value of attribute result.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def result
  @result
end

.warnings ⇒ Object (readonly)

Returns the value of attribute warnings.



5
6
7
 
# File 'lib/smarter_csv/variables.rb', line 5

def warnings
  @warnings
end

Class Method Details

.count_quote_chars(line, quote_char) ⇒ Object

  • the ‘scan` method iterates through the string and finds all occurrences of the pattern

  • The reqular expression:

    • (?<!\) : Negative lookbehind to ensure the quote character is not preceded by an unescaped backslash.

    • (?:\\)* : Non-capturing group for an even number of backslashes (escaped backslashes).

      This allows for any number of escaped backslashes before the quote character.

    • #Regexp.escape(quote_char) : Dynamically inserts the quote_char into the regex,

      ensuring it's properly escaped for use in the regex.



185
186
187
 
# File 'lib/smarter_csv/smarter_csv.rb', line 185

def count_quote_chars(line, quote_char)
  line.scan(/(?<!\\)(?:\\\\)*#{Regexp.escape(quote_char)}/).count
end

.default_options ⇒ Object

NOTE: this is not called when “parse” methods are tested by themselves

ONLY FOR BACKWARDS-COMPATIBILITY



58
59
60
 
# File 'lib/smarter_csv/options_processing.rb', line 58

def default_options
  DEFAULT_OPTIONS
end

.has_acceleration? ⇒ Boolean

Returns:

  • (Boolean) 


189
190
191
 
# File 'lib/smarter_csv/smarter_csv.rb', line 189

def has_acceleration?
  @has_acceleration ||= !!defined?(parse_csv_line_c)
end

.headerA ⇒ Object

:nocov:



20
21
22
23
 
# File 'lib/smarter_csv/variables.rb', line 20

def headerA
  warn "Deprecarion Warning: 'headerA' will be removed in future versions. Use 'headders'"
  @headerA
end

.initialize_variables ⇒ Object 



7
8
9
10
11
12
13
14
15
16
17
 
# File 'lib/smarter_csv/variables.rb', line 7

def initialize_variables
  @csv_line_count = 0
  @chunk_count = 0
  @errors = {}
  @file_line_count = 0
  @headerA = []
  @headers = nil
  @raw_header = nil # header as it appears in the file
  @result = []
  @warnings = {}
end

.parse_csv_line_c(line, col_sep, quote_char, max_size) ⇒ Object

max_size: pass nil if no limit is specified



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
 
# File 'ext/smarter_csv/smarter_csv.c', line 15

static VALUE rb_parse_csv_line(VALUE self, VALUE line, VALUE col_sep, VALUE quote_char, VALUE max_size) {
  if (RB_TYPE_P(line, T_NIL) == 1) {
    return rb_ary_new();
  }

  if (RB_TYPE_P(line, T_STRING) != 1) {
    rb_raise(rb_eTypeError, "ERROR in SmarterCSV.parse_line: line has to be a string or nil");
  }

  rb_encoding *encoding = rb_enc_get(line); /* get the encoding from the input line */
  char *startP = RSTRING_PTR(line); /* may not be null terminated */
  long line_len = RSTRING_LEN(line);
  char *endP = startP + line_len ; /* points behind the string */
  char *p = startP;

  char *col_sepP = RSTRING_PTR(col_sep);
  long col_sep_len = RSTRING_LEN(col_sep);

  char *quoteP = RSTRING_PTR(quote_char);
  long quote_count = 0;

  bool col_sep_found = true;

  VALUE elements = rb_ary_new();
  VALUE field;
  long i;

  char prev_char = '\0'; // Store the previous character for comparison against an escape character
  long backslash_count = 0; // to count consecutive backslash characters

  while (p < endP) {
    /* does the remaining string start with col_sep ? */
    col_sep_found = true;
    for(i=0; (i < col_sep_len) && (p+i < endP) ; i++) {
      col_sep_found = col_sep_found && (*(p+i) == *(col_sepP+i));
    }
    /* if col_sep was found and we have even quotes */
    if (col_sep_found && (quote_count % 2 == 0)) {
      /* if max_size != nil && lements.size >= header_size */
      if ((max_size != Qnil) && RARRAY_LEN(elements) >= NUM2INT(max_size)) {
        break;
      } else {
        /* push that field with original encoding onto the results */
        field = rb_enc_str_new(startP, p - startP, encoding);
        rb_ary_push(elements, field);

        p += col_sep_len;
        startP = p;
      }
    } else {
      if (*p == '\\') {
        backslash_count++;
      } else {
        if (*p == *quoteP && (backslash_count % 2 == 0)) {
          quote_count++;
        }
        backslash_count = 0; // no more consecutive backslash characters
      }
      p++;
    }

    prev_char = *(p - 1); // Update the previous character
  } /* while */

  /* check if the last part of the line needs to be processed */
  if ((max_size == Qnil) || RARRAY_LEN(elements) < NUM2INT(max_size)) {
    /* copy the remaining line as a field with original encoding onto the results */
    field = rb_enc_str_new(startP, endP - startP, encoding);
    rb_ary_push(elements, field);
  }

  return elements;
}

.process(input, given_options = {}, &block) ⇒ Object

first parameter: filename or input object which responds to readline method



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
 
# File 'lib/smarter_csv/smarter_csv.rb', line 14

def SmarterCSV.process(input, given_options = {}, &block) # rubocop:disable Lint/UnusedMethodArgument
  options = process_options(given_options)

  initialize_variables

  has_rails = !!defined?(Rails)
  begin
    fh = input.respond_to?(:readline) ? input : File.open(input, "r:#{options[:file_encoding]}")

    # auto-detect the row separator
    options[:row_sep] = guess_line_ending(fh, options) if options[:row_sep]&.to_sym == :auto
    # attempt to auto-detect column separator
    options[:col_sep] = guess_column_separator(fh, options) if options[:col_sep]&.to_sym == :auto

    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'))
      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".'
    end

    skip_lines(fh, options)

    @headers, header_size = process_headers(fh, options)
    @headerA = @headers # @headerA is deprecated, use @headers

    # in case we use chunking.. we'll need to set it up..
    if options[:chunk_size].to_i > 0
      use_chunks = true
      chunk_size = options[:chunk_size].to_i
      @chunk_count = 0
      chunk = []
    else
      use_chunks = false
    end

    # now on to processing all the rest of the lines in the CSV file:
    until fh.eof? # we can't use fh.readlines() here, because this would read the whole file into memory at once, and eof => true
      line = readline_with_counts(fh, options)

      # replace invalid byte sequence in UTF-8 with question mark to avoid errors
      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

      print "processing file line %10d, csv line %10d\r" % [@file_line_count, @csv_line_count] if options[:verbose]

      next if options[:comment_regexp] && line =~ options[:comment_regexp] # ignore all comment lines if there are any

      # cater for the quoted csv data containing the row separator carriage return character
      # in which case the row data will be split across multiple lines (see the sample content in spec/fixtures/carriage_returns_rn.csv)
      # by detecting the existence of an uneven number of quote characters

      multiline = count_quote_chars(line, options[:quote_char]).odd? # should handle quote_char nil
      while count_quote_chars(line, options[:quote_char]).odd? # should handle quote_char nil
        next_line = fh.readline(options[:row_sep])
        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
        line += next_line
        @file_line_count += 1
      end
      print "\nline contains uneven number of quote chars so including content through file line %d\n" % @file_line_count if options[:verbose] && multiline

      line.chomp!(options[:row_sep])

      dataA, _data_size = parse(line, options, header_size)

      dataA.map!{|x| x.strip} if options[:strip_whitespace]

      # if all values are blank, then ignore this line
      next if options[:remove_empty_hashes] && (dataA.empty? || blank?(dataA))

      hash = @headers.zip(dataA).to_h

      # make sure we delete any key/value pairs from the hash, which the user wanted to delete:
      hash.delete(nil)
      hash.delete('')
      hash.delete(:"")

      if options[:remove_empty_values] == true
        hash.delete_if{|_k, v| has_rails ? v.blank? : blank?(v)}
      end

      hash.delete_if{|_k, v| !v.nil? && v =~ /^(0+|0+\.0+)$/} if options[:remove_zero_values] # values are Strings
      hash.delete_if{|_k, v| v =~ options[:remove_values_matching]} if options[:remove_values_matching]

      if options[:convert_values_to_numeric]
        hash.each do |k, v|
          # deal with the :only / :except options to :convert_values_to_numeric
          next if limit_execution_for_only_or_except(options, :convert_values_to_numeric, k)

          # convert if it's a numeric value:
          case v
          when /^[+-]?\d+\.\d+$/
            hash[k] = v.to_f
          when /^[+-]?\d+$/
            hash[k] = v.to_i
          end
        end
      end

      if options[:value_converters]
        hash.each do |k, v|
          converter = options[:value_converters][k]
          next unless converter

          hash[k] = converter.convert(v)
        end
      end

      next if options[:remove_empty_hashes] && hash.empty?

      hash[:csv_line_number] = @csv_line_count if options[:with_line_numbers]

      if use_chunks
        chunk << hash # append temp result to chunk

        if chunk.size >= chunk_size || fh.eof? # if chunk if full, or EOF reached
          # do something with the chunk
          if block_given?
            yield chunk # do something with the hashes in the chunk in the block
          else
            @result << chunk # not sure yet, why anybody would want to do this without a block
          end
          @chunk_count += 1
          chunk = [] # initialize for next chunk of data
        else

          # the last chunk may contain partial data, which also needs to be returned (BUG / ISSUE-18)

        end

        # while a chunk is being filled up we don't need to do anything else here

      else # no chunk handling
        if block_given?
          yield [hash] # do something with the hash in the block (better to use chunking here)
        else
          @result << hash
        end
      end
    end

    # print new line to retain last processing line message
    print "\n" if options[:verbose]

    # last chunk:
    if !chunk.nil? && chunk.size > 0
      # do something with the chunk
      if block_given?
        yield chunk # do something with the hashes in the chunk in the block
      else
        @result << chunk # not sure yet, why anybody would want to do this without a block
      end
      @chunk_count += 1
      # chunk = [] # initialize for next chunk of data
    end
  ensure
    fh.close if fh.respond_to?(:close)
  end

  if block_given?
    @chunk_count # when we do processing through a block we only care how many chunks we processed
  else
    @result # returns either an Array of Hashes, or an Array of Arrays of Hashes (if in chunked mode)
  end
end

.process_headers(filehandle, options) ⇒ Object 



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
 
# File 'lib/smarter_csv/headers.rb', line 5

def process_headers(filehandle, options)
  @raw_header = nil # header as it appears in the file
  @headers = nil # the processed headers
  header_array = []
  file_header_size = nil

  # if headers_in_file, get the headers -> We get the number of columns, even when user provided headers
  if options[:headers_in_file] # extract the header line
    # process the header line in the CSV file..
    # the first line of a CSV file contains the header .. it might be commented out, so we need to read it anyhow
    header_line = @raw_header = readline_with_counts(filehandle, options)
    header_line = preprocess_header_line(header_line, options)
    file_header_array, file_header_size = parse_and_modify_headers(header_line, options)
  else
    unless options[:user_provided_headers]
      raise SmarterCSV::IncorrectOption, "ERROR: If :headers_in_file is set to false, you have to provide :user_provided_headers"
    end
  end

  if options[:user_provided_headers]
    unless options[:user_provided_headers].is_a?(Array) && !options[:user_provided_headers].empty?
      raise(SmarterCSV::IncorrectOption, "ERROR: incorrect format for user_provided_headers! Expecting array with headers.")
    end

    # use user-provided headers
    user_header_array = options[:user_provided_headers]
    # user_provided_headers: their count should match the headers_in_file if any
    if defined?(file_header_size) && !file_header_size.nil?
      if user_header_array.size != file_header_size
        raise SmarterCSV::HeaderSizeMismatch, "ERROR: :user_provided_headers defines #{user_header_array.size} headers !=  CSV-file has #{file_header_size} headers"
      else
        # we could print out the mapping of file_header_array to header_array here
      end
    end
    header_array = user_header_array
  else
    header_array = file_header_array
  end

  # detect duplicate headers and disambiguate
  header_array = disambiguate_headers(header_array, options) if options[:duplicate_header_suffix]

  # symbolize headers
  header_array.map!{|x| x.to_sym } unless options[:strings_as_keys] || options[:keep_original_headers]

  # wouldn't make sense to re-map user provided headers
  header_array = remap_headers(header_array, options) if options[:key_mapping] && !options[:user_provided_headers]

  validate_and_deprecate_headers(header_array, options)

  [header_array, header_array.size]
end

.process_options(given_options = {}) ⇒ Object

NOTE: this is not called when “parse” methods are tested by themselves



42
43
44
45
46
47
48
49
50
51
52
53
 
# File 'lib/smarter_csv/options_processing.rb', line 42

def process_options(given_options = {})
  puts "User provided options:\n#{pp(given_options)}\n" if given_options[:verbose]

  # fix invalid input
  given_options[:invalid_byte_sequence] = '' if given_options[:invalid_byte_sequence].nil?

  @options = DEFAULT_OPTIONS.dup.merge!(given_options)
  puts "Computed options:\n#{pp(@options)}\n" if given_options[:verbose]

  validate_options!(@options)
  @options
end