csv 3.2.0 → 3.2.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c48c0d15454e002ff10270a9c56cf4311ce635a8a9dfb527f7a7541f29f801b2
-  data.tar.gz: 505d1d0dbb4cff0a544b2e00925cb1101ed71642a584d534f443405fba8bd820
+  metadata.gz: 915b3ed5a51bf4836f08f7bb06efc3b07bdc90e09209a5253092130e2cad2ab6
+  data.tar.gz: 6bce2e39329afcf200691b4b2f422b6a48d45da66368f5d5e136e0c761cd6217
 SHA512:
-  metadata.gz: 1c9ecd18d5b9a4f663c0676694ffc133a4657e2f7a07cafe2f0a5d9ddd2d7846f505bc62c21698fcf1117126efc6978b7aa1b497d2fef8d532a8a4246c58bff2
-  data.tar.gz: e4fe05b49f92c68c011060d1dcd39ead1785d886eabbd3689a12884df9eb30124694417314e39bcf656cd05d6d0dea6a80a701dd5fe6cac42efc33c67be54926
+  metadata.gz: 5c1434c8e91c16de40d19d4d1200f193248e786720b67f2bbecf26a481859fe814b8cbaa02d22027668ff02588541266c8ff5d00b9fc1cfc2163b358b8e9ece9
+  data.tar.gz: 1978e933549049129f0ec99e80a10f2838b3c75a282103aa177d8421fe7589d428308e2786b29a961a4a7a5565ede77e3b1ef44ba8f4bc91b593a5a884ded7aa

data/NEWS.md

@@ -1,5 +1,104 @@
 # News
 
+## 3.2.3 - 2022-04-09
+
+### Improvements
+
+  * Added contents summary to `CSV::Table#inspect`.
+    [GitHub#229][Patch by Eriko Sugiyama]
+    [GitHub#235][Patch by Sampat Badhe]
+
+  * Suppressed `$INPUT_RECORD_SEPARATOR` deprecation warning by
+    `Warning.warn`.
+    [GitHub#233][Reported by Jean byroot Boussier]
+
+  * Improved error message for liberal parsing with quoted values.
+    [GitHub#231][Patch by Nikolay Rys]
+
+  * Fixed typos in documentation.
+    [GitHub#236][Patch by Sampat Badhe]
+
+  * Added `:max_field_size` option and deprecated `:field_size_limit` option.
+    [GitHub#238][Reported by Dan Buettner]
+
+  * Added `:symbol_raw` to built-in header converters.
+    [GitHub#237][Reported by taki]
+    [GitHub#239][Patch by Eriko Sugiyama]
+
+### Fixes
+
+  * Fixed a bug that some texts may be dropped unexpectedly.
+    [Bug #18245][ruby-core:105587][Reported by Hassan Abdul Rehman]
+
+  * Fixed a bug that `:field_size_limit` doesn't work with not complex row.
+    [GitHub#238][Reported by Dan Buettner]
+
+### Thanks
+
+  * Hassan Abdul Rehman
+
+  * Eriko Sugiyama
+
+  * Jean byroot Boussier
+
+  * Nikolay Rys
+
+  * Sampat Badhe
+
+  * Dan Buettner
+
+  * taki
+
+## 3.2.2 - 2021-12-24
+
+### Improvements
+
+  * Added a validation for invalid option combination.
+    [GitHub#225][Patch by adamroyjones]
+
+  * Improved documentation for developers.
+    [GitHub#227][Patch by Eriko Sugiyama]
+
+### Fixes
+
+  * Fixed a bug that all of `ARGF` contents may not be consumed.
+    [GitHub#228][Reported by Rafael Navaza]
+
+### Thanks
+
+  * adamroyjones
+
+  * Eriko Sugiyama
+
+  * Rafael Navaza
+
+## 3.2.1 - 2021-10-23
+
+### Improvements
+
+  * doc: Fixed wrong class name.
+    [GitHub#217][Patch by Vince]
+
+  * Changed to always use `"\n"` for the default row separator on Ruby
+    3.0 or later because `$INPUT_RECORD_SEPARATOR` was deprecated
+    since Ruby 3.0.
+
+  * Added support for Ractor.
+    [GitHub#218][Patch by rm155]
+
+    * Users who want to use the built-in converters in non-main
+      Ractors need to call `Ractor.make_shareable(CSV::Converters)`
+      and/or `Ractor.make_shareable(CSV::HeaderConverters)` before
+      creating non-main Ractors.
+
+### Thanks
+
+  * Vince
+
+  * Joakim Antman
+
+  * rm155
+
 ## 3.2.0 - 2021-06-06
 
 ### Improvements

data/README.md

@@ -35,7 +35,7 @@ end
 
 ## Development
 
-After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake test` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
+After checking out the repo, run `ruby run-test.rb` to check if your changes can pass the test.
 
 To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
 

data/doc/csv/options/generating/write_headers.rdoc

@@ -19,7 +19,7 @@ Without +write_headers+:
 
 With +write_headers+":
   CSV.open(file_path,'w',
-      :write_headers=> true,
+      :write_headers => true,
       :headers => ['Name','Value']
     ) do |csv|
       csv << ['foo', '0']

data/doc/csv/recipes/generating.rdoc

@@ -148,7 +148,7 @@ This example defines and uses a custom write converter to strip whitespace from
 
 ==== Recipe: Specify Multiple Write Converters
 
-Use option <tt>:write_converters</tt> and multiple custom coverters
+Use option <tt>:write_converters</tt> and multiple custom converters
 to convert field values when generating \CSV.
 
 This example defines and uses two custom write converters to strip and upcase generated fields:

data/doc/csv/recipes/parsing.rdoc

@@ -83,7 +83,7 @@ Use instance method CSV#each with option +headers+ to read a source \String one
   CSV.new(string, headers: true).each do |row|
     p row
   end
-Ouput:
+Output:
   #<CSV::Row "Name":"foo" "Value":"0">
   #<CSV::Row "Name":"bar" "Value":"1">
   #<CSV::Row "Name":"baz" "Value":"2">

data/lib/csv/fields_converter.rb

@@ -16,7 +16,7 @@ class CSV
       @empty_value = options[:empty_value]
       @empty_value_is_empty_string = (@empty_value == "")
       @accept_nil = options[:accept_nil]
-      @builtin_converters = options[:builtin_converters]
+      @builtin_converters_name = options[:builtin_converters_name]
       @need_static_convert = need_static_convert?
     end
 
@@ -24,7 +24,7 @@ class CSV
       if name.nil?  # custom converter
         @converters << converter
       else          # named converter
-        combo = @builtin_converters[name]
+        combo = builtin_converters[name]
         case combo
         when Array  # combo converter
           combo.each do |sub_name|
@@ -80,5 +80,9 @@ class CSV
       @need_static_convert or
         (not @converters.empty?)
     end
+
+    def builtin_converters
+      @builtin_converters ||= ::CSV.const_get(@builtin_converters_name)
+    end
   end
 end

data/lib/csv/input_record_separator.rb

@@ -0,0 +1,18 @@
+require "English"
+require "stringio"
+
+class CSV
+  module InputRecordSeparator
+    class << self
+      if RUBY_VERSION >= "3.0.0"
+        def value
+          "\n"
+        end
+      else
+        def value
+          $INPUT_RECORD_SEPARATOR
+        end
+      end
+    end
+  end
+end

data/lib/csv/parser.rb

@@ -3,6 +3,7 @@
 require "strscan"
 
 require_relative "delete_suffix"
+require_relative "input_record_separator"
 require_relative "match_p"
 require_relative "row"
 require_relative "table"
@@ -26,6 +27,10 @@ class CSV
     class InvalidEncoding < StandardError
     end
 
+    # Raised when unexpected case is happen.
+    class UnexpectedError < StandardError
+    end
+
     #
     # CSV::Scanner receives a CSV output, scans it and return the content.
     # It also controls the life cycle of the object with its methods +keep_start+,
@@ -77,16 +82,17 @@ class CSV
     # +keep_end+, +keep_back+, +keep_drop+.
     #
     # CSV::InputsScanner.scan() tries to match with pattern at the current position.
-    # If there's a match, the scanner advances the “scan pointer” and returns the matched string.
+    # If there's a match, the scanner advances the "scan pointer" and returns the matched string.
     # Otherwise, the scanner returns nil.
     #
-    # CSV::InputsScanner.rest() returns the “rest” of the string (i.e. everything after the scan pointer).
+    # CSV::InputsScanner.rest() returns the "rest" of the string (i.e. everything after the scan pointer).
     # If there is no more data (eos? = true), it returns "".
     #
     class InputsScanner
-      def initialize(inputs, encoding, chunk_size: 8192)
+      def initialize(inputs, encoding, row_separator, chunk_size: 8192)
         @inputs = inputs.dup
         @encoding = encoding
+        @row_separator = row_separator
         @chunk_size = chunk_size
         @last_scanner = @inputs.empty?
         @keeps = []
@@ -94,11 +100,13 @@ class CSV
       end
 
       def each_line(row_separator)
+        return enum_for(__method__, row_separator) unless block_given?
         buffer = nil
         input = @scanner.rest
         position = @scanner.pos
         offset = 0
         n_row_separator_chars = row_separator.size
+        # trace(__method__, :start, line, input)
         while true
           input.each_line(row_separator) do |line|
             @scanner.pos += line.bytesize
@@ -138,25 +146,28 @@ class CSV
       end
 
       def scan(pattern)
+        # trace(__method__, pattern, :start)
         value = @scanner.scan(pattern)
+        # trace(__method__, pattern, :done, :last, value) if @last_scanner
         return value if @last_scanner
 
-        if value
-          read_chunk if @scanner.eos?
-          return value
-        else
-          nil
-        end
+        read_chunk if value and @scanner.eos?
+        # trace(__method__, pattern, :done, value)
+        value
       end
 
       def scan_all(pattern)
+        # trace(__method__, pattern, :start)
         value = @scanner.scan(pattern)
+        # trace(__method__, pattern, :done, :last, value) if @last_scanner
         return value if @last_scanner
 
         return nil if value.nil?
         while @scanner.eos? and read_chunk and (sub_value = @scanner.scan(pattern))
+          # trace(__method__, pattern, :sub, sub_value)
           value << sub_value
         end
+        # trace(__method__, pattern, :done, value)
         value
       end
 
@@ -165,76 +176,135 @@ class CSV
       end
 
       def keep_start
-        @keeps.push([@scanner.pos, nil])
+        # trace(__method__, :start)
+        adjust_last_keep
+        @keeps.push([@scanner, @scanner.pos, nil])
+        # trace(__method__, :done)
       end
 
       def keep_end
-        start, buffer = @keeps.pop
-        keep = @scanner.string.byteslice(start, @scanner.pos - start)
+        # trace(__method__, :start)
+        scanner, start, buffer = @keeps.pop
+        if scanner == @scanner
+          keep = @scanner.string.byteslice(start, @scanner.pos - start)
+        else
+          keep = @scanner.string.byteslice(0, @scanner.pos)
+        end
         if buffer
           buffer << keep
           keep = buffer
         end
+        # trace(__method__, :done, keep)
         keep
       end
 
       def keep_back
-        start, buffer = @keeps.pop
+        # trace(__method__, :start)
+        scanner, start, buffer = @keeps.pop
         if buffer
+          # trace(__method__, :rescan, start, buffer)
           string = @scanner.string
-          keep = string.byteslice(start, string.bytesize - start)
+          if scanner == @scanner
+            keep = string.byteslice(start, string.bytesize - start)
+          else
+            keep = string
+          end
           if keep and not keep.empty?
             @inputs.unshift(StringIO.new(keep))
             @last_scanner = false
           end
           @scanner = StringScanner.new(buffer)
         else
+          if @scanner != scanner
+            message = "scanners are different but no buffer: "
+            message += "#{@scanner.inspect}(#{@scanner.object_id}): "
+            message += "#{scanner.inspect}(#{scanner.object_id})"
+            raise UnexpectedError, message
+          end
+          # trace(__method__, :repos, start, buffer)
           @scanner.pos = start
         end
         read_chunk if @scanner.eos?
       end
 
       def keep_drop
-        @keeps.pop
+        _, _, buffer = @keeps.pop
+        # trace(__method__, :done, :empty) unless buffer
+        return unless buffer
+
+        last_keep = @keeps.last
+        # trace(__method__, :done, :no_last_keep) unless last_keep
+        return unless last_keep
+
+        if last_keep[2]
+          last_keep[2] << buffer
+        else
+          last_keep[2] = buffer
+        end
+        # trace(__method__, :done)
       end
 
       def rest
         @scanner.rest
       end
 
+      def check(pattern)
+        @scanner.check(pattern)
+      end
+
       private
-      def read_chunk
-        return false if @last_scanner
+      def trace(*args)
+        pp([*args, @scanner, @scanner&.string, @scanner&.pos, @keeps])
+      end
 
-        unless @keeps.empty?
-          keep = @keeps.last
-          keep_start = keep[0]
-          string = @scanner.string
-          keep_data = string.byteslice(keep_start, @scanner.pos - keep_start)
-          if keep_data
-            keep_buffer = keep[1]
-            if keep_buffer
-              keep_buffer << keep_data
-            else
-              keep[1] = keep_data.dup
-            end
+      def adjust_last_keep
+        # trace(__method__, :start)
+
+        keep = @keeps.last
+        # trace(__method__, :done, :empty) if keep.nil?
+        return if keep.nil?
+
+        scanner, start, buffer = keep
+        string = @scanner.string
+        if @scanner != scanner
+          start = 0
+        end
+        if start == 0 and @scanner.eos?
+          keep_data = string
+        else
+          keep_data = string.byteslice(start, @scanner.pos - start)
+        end
+        if keep_data
+          if buffer
+            buffer << keep_data
+          else
+            keep[2] = keep_data.dup
           end
-          keep[0] = 0
         end
 
+        # trace(__method__, :done)
+      end
+
+      def read_chunk
+        return false if @last_scanner
+
+        adjust_last_keep
+
         input = @inputs.first
         case input
         when StringIO
           string = input.read
           raise InvalidEncoding unless string.valid_encoding?
+          # trace(__method__, :stringio, string)
           @scanner = StringScanner.new(string)
           @inputs.shift
           @last_scanner = @inputs.empty?
           true
         else
-          chunk = input.gets(nil, @chunk_size)
+          chunk = input.gets(@row_separator, @chunk_size)
           if chunk
             raise InvalidEncoding unless chunk.valid_encoding?
+            # trace(__method__, :chunk, chunk)
             @scanner = StringScanner.new(chunk)
             if input.respond_to?(:eof?) and input.eof?
               @inputs.shift
@@ -242,6 +312,7 @@ class CSV
             end
             true
           else
+            # trace(__method__, :no_chunk)
             @scanner = StringScanner.new("".encode(@encoding))
             @inputs.shift
             @last_scanner = @inputs.empty?
@@ -276,7 +347,11 @@ class CSV
     end
 
     def field_size_limit
-      @field_size_limit
+      @max_field_size&.succ
+    end
+
+    def max_field_size
+      @max_field_size
     end
 
     def skip_lines
@@ -344,6 +419,16 @@ class CSV
         end
         message = "Invalid byte sequence in #{@encoding}"
         raise MalformedCSVError.new(message, lineno)
+      rescue UnexpectedError => error
+        if @scanner
+          ignore_broken_line
+          lineno = @lineno
+        else
+          lineno = @lineno + 1
+        end
+        message = "This should not be happen: #{error.message}: "
+        message += "Please report this to https://github.com/ruby/csv/issues"
+        raise MalformedCSVError.new(message, lineno)
       end
     end
 
@@ -360,6 +445,7 @@ class CSV
       prepare_skip_lines
       prepare_strip
       prepare_separators
+      validate_strip_and_col_sep_options
       prepare_quoted
       prepare_unquoted
       prepare_line
@@ -387,7 +473,7 @@ class CSV
         @backslash_quote = false
       end
       @unconverted_fields = @options[:unconverted_fields]
-      @field_size_limit = @options[:field_size_limit]
+      @max_field_size = @options[:max_field_size]
       @skip_blanks = @options[:skip_blanks]
       @fields_converter = @options[:fields_converter]
       @header_fields_converter = @options[:header_fields_converter]
@@ -479,9 +565,9 @@ class CSV
     begin
       StringScanner.new("x").scan("x")
     rescue TypeError
-      @@string_scanner_scan_accept_string = false
+      STRING_SCANNER_SCAN_ACCEPT_STRING = false
     else
-      @@string_scanner_scan_accept_string = true
+      STRING_SCANNER_SCAN_ACCEPT_STRING = true
     end
 
     def prepare_separators
@@ -505,7 +591,7 @@ class CSV
         @first_column_separators = Regexp.new(@escaped_first_column_separator +
                                               "+".encode(@encoding))
       else
-        if @@string_scanner_scan_accept_string
+        if STRING_SCANNER_SCAN_ACCEPT_STRING
           @column_end = @column_separator
         else
           @column_end = Regexp.new(@escaped_column_separator)
@@ -526,10 +612,32 @@ class CSV
 
       @cr = "\r".encode(@encoding)
       @lf = "\n".encode(@encoding)
-      @cr_or_lf = Regexp.new("[\r\n]".encode(@encoding))
+      @line_end = Regexp.new("\r\n|\n|\r".encode(@encoding))
       @not_line_end = Regexp.new("[^\r\n]+".encode(@encoding))
     end
 
+    # This method verifies that there are no (obvious) ambiguities with the
+    # provided +col_sep+ and +strip+ parsing options. For example, if +col_sep+
+    # and +strip+ were both equal to +\t+, then there would be no clear way to
+    # parse the input.
+    def validate_strip_and_col_sep_options
+      return unless @strip
+
+      if @strip.is_a?(String)
+        if @column_separator.start_with?(@strip) || @column_separator.end_with?(@strip)
+          raise ArgumentError,
+                "The provided strip (#{@escaped_strip}) and " \
+                "col_sep (#{@escaped_column_separator}) options are incompatible."
+        end
+      else
+        if Regexp.new("\\A[#{@escaped_strip}]|[#{@escaped_strip}]\\z").match?(@column_separator)
+          raise ArgumentError,
+                "The provided strip (true) and " \
+                "col_sep (#{@escaped_column_separator}) options are incompatible."
+        end
+      end
+    end
+
     def prepare_quoted
       if @quote_character
         @quotes = Regexp.new(@escaped_quote_character +
@@ -605,7 +713,7 @@ class CSV
             # do nothing:  ensure will set default
           end
         end
-        separator = $INPUT_RECORD_SEPARATOR if separator == :auto
+        separator = InputRecordSeparator.value if separator == :auto
       end
       separator.to_s.encode(@encoding)
     end
@@ -704,26 +812,28 @@ class CSV
       sample[0, 128].index(@quote_character)
     end
 
-    SCANNER_TEST = (ENV["CSV_PARSER_SCANNER_TEST"] == "yes")
-    if SCANNER_TEST
-      class UnoptimizedStringIO
-        def initialize(string)
-          @io = StringIO.new(string, "rb:#{string.encoding}")
-        end
+    class UnoptimizedStringIO # :nodoc:
+      def initialize(string)
+        @io = StringIO.new(string, "rb:#{string.encoding}")
+      end
 
-        def gets(*args)
-          @io.gets(*args)
-        end
+      def gets(*args)
+        @io.gets(*args)
+      end
 
-        def each_line(*args, &block)
-          @io.each_line(*args, &block)
-        end
+      def each_line(*args, &block)
+        @io.each_line(*args, &block)
+      end
 
-        def eof?
-          @io.eof?
-        end
+      def eof?
+        @io.eof?
       end
+    end
 
+    SCANNER_TEST = (ENV["CSV_PARSER_SCANNER_TEST"] == "yes")
+    if SCANNER_TEST
+      SCANNER_TEST_CHUNK_SIZE_NAME = "CSV_PARSER_SCANNER_TEST_CHUNK_SIZE"
+      SCANNER_TEST_CHUNK_SIZE_VALUE = ENV[SCANNER_TEST_CHUNK_SIZE_NAME]
       def build_scanner
         inputs = @samples.collect do |sample|
           UnoptimizedStringIO.new(sample)
@@ -733,17 +843,27 @@ class CSV
         else
           inputs << @input
         end
-        chunk_size = ENV["CSV_PARSER_SCANNER_TEST_CHUNK_SIZE"] || "1"
+        begin
+          chunk_size_value = ENV[SCANNER_TEST_CHUNK_SIZE_NAME]
+        rescue # Ractor::IsolationError
+          # Ractor on Ruby 3.0 can't read ENV value.
+          chunk_size_value = SCANNER_TEST_CHUNK_SIZE_VALUE
+        end
+        chunk_size = Integer((chunk_size_value || "1"), 10)
         InputsScanner.new(inputs,
                           @encoding,
-                          chunk_size: Integer(chunk_size, 10))
+                          @row_separator,
+                          chunk_size: chunk_size)
       end
     else
       def build_scanner
         string = nil
         if @samples.empty? and @input.is_a?(StringIO)
           string = @input.read
-        elsif @samples.size == 1 and @input.respond_to?(:eof?) and @input.eof?
+        elsif @samples.size == 1 and
+              @input != ARGF and
+              @input.respond_to?(:eof?) and
+              @input.eof?
           string = @samples[0]
         end
         if string
@@ -762,7 +882,7 @@ class CSV
             StringIO.new(sample)
           end
           inputs << @input
-          InputsScanner.new(inputs, @encoding)
+          InputsScanner.new(inputs, @encoding, @row_separator)
         end
       end
     end
@@ -796,6 +916,14 @@ class CSV
       end
     end
 
+    def validate_field_size(field)
+      return unless @max_field_size
+      return if field.size <= @max_field_size
+      ignore_broken_line
+      message = "Field size exceeded: #{field.size} > #{@max_field_size}"
+      raise MalformedCSVError.new(message, @lineno)
+    end
+
     def parse_no_quote(&block)
       @scanner.each_line(@row_separator) do |line|
         next if @skip_lines and skip_line?(line)
@@ -808,6 +936,11 @@ class CSV
         else
           line = strip_value(line)
           row = line.split(@split_column_separator, -1)
+          if @max_field_size
+            row.each do |column|
+              validate_field_size(column)
+            end
+          end
           n_columns = row.size
           i = 0
           while i < n_columns
@@ -863,6 +996,7 @@ class CSV
                 @need_robust_parsing = true
                 return parse_quotable_robust(&block)
               end
+              validate_field_size(row[i])
             end
             i += 1
           end
@@ -886,10 +1020,7 @@ class CSV
         value = parse_column_value
         if value
           @scanner.scan_all(@strip_value) if @strip_value
-          if @field_size_limit and value.size >= @field_size_limit
-            ignore_broken_line
-            raise MalformedCSVError.new("Field size exceeded", @lineno)
-          end
+          validate_field_size(value)
         end
         if parse_column_end
           row << value
@@ -910,11 +1041,17 @@ class CSV
           break
         else
           if @quoted_column_value
+            if liberal_parsing? and (new_line = @scanner.check(@line_end))
+              message =
+                "Illegal end-of-line sequence outside of a quoted field " +
+                "<#{new_line.inspect}>"
+            else
+              message = "Any value after quoted field isn't allowed"
+            end
             ignore_broken_line
-            message = "Any value after quoted field isn't allowed"
             raise MalformedCSVError.new(message, @lineno)
           elsif @unquoted_column_value and
-                (new_line = @scanner.scan(@cr_or_lf))
+                (new_line = @scanner.scan(@line_end))
             ignore_broken_line
             message = "Unquoted fields do not allow new line " +
                       "<#{new_line.inspect}>"
@@ -923,7 +1060,7 @@ class CSV
             ignore_broken_line
             message = "Illegal quoting"
             raise MalformedCSVError.new(message, @lineno)
-          elsif (new_line = @scanner.scan(@cr_or_lf))
+          elsif (new_line = @scanner.scan(@line_end))
             ignore_broken_line
             message = "New line must be <#{@row_separator.inspect}> " +
                       "not <#{new_line.inspect}>"
@@ -1089,7 +1226,7 @@ class CSV
 
     def ignore_broken_line
       @scanner.scan_all(@not_line_end)
-      @scanner.scan_all(@cr_or_lf)
+      @scanner.scan_all(@line_end)
       @lineno += 1
     end
 

data/lib/csv/table.rb

@@ -999,9 +999,15 @@ class CSV
     # Omits the headers if option +write_headers+ is given as +false+
     # (see {Option +write_headers+}[../CSV.html#class-CSV-label-Option+write_headers]):
     #   table.to_csv(write_headers: false) # => "foo,0\nbar,1\nbaz,2\n"
-    def to_csv(write_headers: true, **options)
+    #
+    # Limit rows if option +limit+ is given like +2+:
+    #   table.to_csv(limit: 2) # => "Name,Value\nfoo,0\nbar,1\n"
+    def to_csv(write_headers: true, limit: nil, **options)
       array = write_headers ? [headers.to_csv(**options)] : []
-      @table.each do |row|
+      limit ||= @table.size
+      limit = @table.size + 1 + limit if limit < 0
+      limit = 0 if limit < 0
+      @table.first(limit).each do |row|
         array.push(row.fields.to_csv(**options)) unless row.header_row?
       end
 
@@ -1038,9 +1044,13 @@ class CSV
     # Example:
     #   source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"
     #   table = CSV.parse(source, headers: true)
-    #   table.inspect # => "#<CSV::Table mode:col_or_row row_count:4>"
+    #   table.inspect # => "#<CSV::Table mode:col_or_row row_count:4>\nName,Value\nfoo,0\nbar,1\nbaz,2\n"
+    #
     def inspect
-      "#<#{self.class} mode:#{@mode} row_count:#{to_a.size}>".encode("US-ASCII")
+      inspected = +"#<#{self.class} mode:#{@mode} row_count:#{to_a.size}>"
+      summary = to_csv(limit: 5)
+      inspected << "\n" << summary if summary.encoding.ascii_compatible?
+      inspected
     end
   end
 end

data/lib/csv/version.rb

@@ -2,5 +2,5 @@
 
 class CSV
   # The version of the installed library.
-  VERSION = "3.2.0"
+  VERSION = "3.2.3"
 end

data/lib/csv/writer.rb

@@ -1,5 +1,6 @@
 # frozen_string_literal: true
 
+require_relative "input_record_separator"
 require_relative "match_p"
 require_relative "row"
 
@@ -133,7 +134,7 @@ class CSV
       @column_separator = @options[:column_separator].to_s.encode(@encoding)
       row_separator = @options[:row_separator]
       if row_separator == :auto
-        @row_separator = $INPUT_RECORD_SEPARATOR.encode(@encoding)
+        @row_separator = InputRecordSeparator.value.encode(@encoding)
       else
         @row_separator = row_separator.to_s.encode(@encoding)
       end

data/lib/csv.rb

@@ -90,11 +90,11 @@
 # with any questions.
 
 require "forwardable"
-require "English"
 require "date"
 require "stringio"
 
 require_relative "csv/fields_converter"
+require_relative "csv/input_record_separator"
 require_relative "csv/match_p"
 require_relative "csv/parser"
 require_relative "csv/row"
@@ -341,6 +341,7 @@ using CSV::MatchP if CSV.const_defined?(:MatchP)
 #     liberal_parsing:    false,
 #     nil_value:          nil,
 #     empty_value:        "",
+#     strip:              false,
 #     # For generating.
 #     write_headers:      nil,
 #     quote_empty:        true,
@@ -348,7 +349,6 @@ using CSV::MatchP if CSV.const_defined?(:MatchP)
 #     write_converters:   nil,
 #     write_nil_value:    nil,
 #     write_empty_value:  "",
-#     strip:              false,
 #   }
 #
 # ==== Options for Parsing
@@ -357,7 +357,9 @@ using CSV::MatchP if CSV.const_defined?(:MatchP)
 # - +row_sep+: Specifies the row separator; used to delimit rows.
 # - +col_sep+: Specifies the column separator; used to delimit fields.
 # - +quote_char+: Specifies the quote character; used to quote fields.
-# - +field_size_limit+: Specifies the maximum field size allowed.
+# - +field_size_limit+: Specifies the maximum field size + 1 allowed.
+#   Deprecated since 3.2.3. Use +max_field_size+ instead.
+# - +max_field_size+: Specifies the maximum field size allowed.
 # - +converters+: Specifies the field converters to be used.
 # - +unconverted_fields+: Specifies whether unconverted fields are to be available.
 # - +headers+: Specifies whether data contains headers,
@@ -366,8 +368,9 @@ using CSV::MatchP if CSV.const_defined?(:MatchP)
 # - +header_converters+: Specifies the header converters to be used.
 # - +skip_blanks+: Specifies whether blanks lines are to be ignored.
 # - +skip_lines+: Specifies how comments lines are to be recognized.
-# - +strip+: Specifies whether leading and trailing whitespace are
-#   to be stripped from fields..
+# - +strip+: Specifies whether leading and trailing whitespace are to be
+#   stripped from fields. This must be compatible with +col_sep+; if it is not,
+#   then an +ArgumentError+ exception will be raised.
 # - +liberal_parsing+: Specifies whether \CSV should attempt to parse
 #   non-compliant data.
 # - +nil_value+: Specifies the object that is to be substituted for each null (no-text) field.
@@ -513,7 +516,7 @@ using CSV::MatchP if CSV.const_defined?(:MatchP)
 #  [" 1 ", #<struct CSV::FieldInfo index=1, line=2, header=nil>]
 #  [" baz ", #<struct CSV::FieldInfo index=0, line=3, header=nil>]
 #  [" 2 ", #<struct CSV::FieldInfo index=1, line=3, header=nil>]
-# Each CSV::Info object shows:
+# Each CSV::FieldInfo object shows:
 # - The 0-based field index.
 # - The 1-based line index.
 # - The field header, if any.
@@ -547,6 +550,14 @@ using CSV::MatchP if CSV.const_defined?(:MatchP)
 #
 # There is no such storage structure for write headers.
 #
+# In order for the parsing methods to access stored converters in non-main-Ractors, the
+# storage structure must be made shareable first.
+# Therefore, <tt>Ractor.make_shareable(CSV::Converters)</tt> and
+# <tt>Ractor.make_shareable(CSV::HeaderConverters)</tt> must be called before the creation
+# of Ractors that use the converters stored in these structures. (Since making the storage
+# structures shareable involves freezing them, any custom converters that are to be used
+# must be added first.)
+#
 # ===== Converter Lists
 #
 # A _converter_ _list_ is an \Array that may include any assortment of:
@@ -917,8 +928,10 @@ class CSV
     symbol:   lambda { |h|
       h.encode(ConverterEncoding).downcase.gsub(/[^\s\w]+/, "").strip.
                                            gsub(/\s+/, "_").to_sym
-    }
+    },
+    symbol_raw: lambda { |h| h.encode(ConverterEncoding).to_sym }
   }
+
   # Default values for method options.
   DEFAULT_OPTIONS = {
     # For both parsing and generating.
@@ -927,6 +940,7 @@ class CSV
     quote_char:         '"',
     # For parsing.
     field_size_limit:   nil,
+    max_field_size:     nil,
     converters:         nil,
     unconverted_fields: nil,
     headers:            false,
@@ -937,6 +951,7 @@ class CSV
     liberal_parsing:    false,
     nil_value:          nil,
     empty_value:        "",
+    strip:              false,
     # For generating.
     write_headers:      nil,
     quote_empty:        true,
@@ -944,7 +959,6 @@ class CSV
     write_converters:   nil,
     write_nil_value:    nil,
     write_empty_value:  "",
-    strip:              false,
   }.freeze
 
   class << self
@@ -957,6 +971,8 @@ class CSV
     # Creates or retrieves cached \CSV objects.
     # For arguments and options, see CSV.new.
     #
+    # This API is not Ractor-safe.
+    #
     # ---
     #
     # With no block given, returns a \CSV object.
@@ -1187,7 +1203,7 @@ class CSV
     #   See {Options for Parsing}[#class-CSV-label-Options+for+Parsing].
     def filter(input=nil, output=nil, **options)
       # parse options for input, output, or both
-      in_options, out_options = Hash.new, {row_sep: $INPUT_RECORD_SEPARATOR}
+      in_options, out_options = Hash.new, {row_sep: InputRecordSeparator.value}
       options.each do |key, value|
         case key.to_s
         when /\Ain(?:put)?_(.+)\Z/
@@ -1407,8 +1423,8 @@ class CSV
     # Argument +ary+ must be an \Array.
     #
     # Special options:
-    # * Option <tt>:row_sep</tt> defaults to <tt>$INPUT_RECORD_SEPARATOR</tt>
-    #   (<tt>$/</tt>).:
+    # * Option <tt>:row_sep</tt> defaults to <tt>"\n"> on Ruby 3.0 or later
+    #   and <tt>$INPUT_RECORD_SEPARATOR</tt> (<tt>$/</tt>) otherwise.:
     #     $INPUT_RECORD_SEPARATOR # => "\n"
     # * This method accepts an additional option, <tt>:encoding</tt>, which sets the base
     #   Encoding for the output. This method will try to guess your Encoding from
@@ -1430,7 +1446,7 @@ class CSV
     #   CSV.generate_line(:foo)
     #
     def generate_line(row, **options)
-      options = {row_sep: $INPUT_RECORD_SEPARATOR}.merge(options)
+      options = {row_sep: InputRecordSeparator.value}.merge(options)
       str = +""
       if options[:encoding]
         str.force_encoding(options[:encoding])
@@ -1853,6 +1869,7 @@ class CSV
                  row_sep: :auto,
                  quote_char: '"',
                  field_size_limit: nil,
+                 max_field_size: nil,
                  converters: nil,
                  unconverted_fields: nil,
                  headers: false,
@@ -1868,11 +1885,11 @@ class CSV
                  encoding: nil,
                  nil_value: nil,
                  empty_value: "",
+                 strip: false,
                  quote_empty: true,
                  write_converters: nil,
                  write_nil_value: nil,
-                 write_empty_value: "",
-                 strip: false)
+                 write_empty_value: "")
     raise ArgumentError.new("Cannot parse nil as CSV") if data.nil?
 
     if data.is_a?(String)
@@ -1895,11 +1912,14 @@ class CSV
     @initial_header_converters = header_converters
     @initial_write_converters = write_converters
 
+    if max_field_size.nil? and field_size_limit
+      max_field_size = field_size_limit - 1
+    end
     @parser_options = {
       column_separator: col_sep,
       row_separator: row_sep,
       quote_character: quote_char,
-      field_size_limit: field_size_limit,
+      max_field_size: max_field_size,
       unconverted_fields: unconverted_fields,
       headers: headers,
       return_headers: return_headers,
@@ -1967,10 +1987,24 @@ class CSV
   # Returns the limit for field size; used for parsing;
   # see {Option +field_size_limit+}[#class-CSV-label-Option+field_size_limit]:
   #   CSV.new('').field_size_limit # => nil
+  #
+  # Deprecated since 3.2.3. Use +max_field_size+ instead.
   def field_size_limit
     parser.field_size_limit
   end
 
+  # :call-seq:
+  #   csv.max_field_size -> integer or nil
+  #
+  # Returns the limit for field size; used for parsing;
+  # see {Option +max_field_size+}[#class-CSV-label-Option+max_field_size]:
+  #   CSV.new('').max_field_size # => nil
+  #
+  # Since 3.2.3.
+  def max_field_size
+    parser.max_field_size
+  end
+
   # :call-seq:
   #   csv.skip_lines -> regexp or nil
   #
@@ -1992,6 +2026,10 @@ class CSV
   #   csv.converters # => [:integer]
   #   csv.convert(proc {|x| x.to_s })
   #   csv.converters
+  #
+  # Notes that you need to call
+  # +Ractor.make_shareable(CSV::Converters)+ on the main Ractor to use
+  # this method.
   def converters
     parser_fields_converter.map do |converter|
       name = Converters.rassoc(converter)
@@ -2054,6 +2092,10 @@ class CSV
   # Returns an \Array containing header converters; used for parsing;
   # see {Header Converters}[#class-CSV-label-Header+Converters]:
   #   CSV.new('').header_converters # => []
+  #
+  # Notes that you need to call
+  # +Ractor.make_shareable(CSV::HeaderConverters)+ on the main Ractor
+  # to use this method.
   def header_converters
     header_fields_converter.map do |converter|
       name = HeaderConverters.rassoc(converter)
@@ -2694,7 +2736,7 @@ class CSV
 
   def build_parser_fields_converter
     specific_options = {
-      builtin_converters: Converters,
+      builtin_converters_name: :Converters,
     }
     options = @base_fields_converter_options.merge(specific_options)
     build_fields_converter(@initial_converters, options)
@@ -2706,7 +2748,7 @@ class CSV
 
   def build_header_fields_converter
     specific_options = {
-      builtin_converters: HeaderConverters,
+      builtin_converters_name: :HeaderConverters,
       accept_nil: true,
     }
     options = @base_fields_converter_options.merge(specific_options)
@@ -2774,6 +2816,8 @@ end
 #   io = StringIO.new
 #   CSV(io, col_sep: ";") { |csv| csv << ["a", "b", "c"] }
 #
+# This API is not Ractor-safe.
+#
 def CSV(*args, **options, &block)
   CSV.instance(*args, **options, &block)
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: csv
 version: !ruby/object:Gem::Version
-  version: 3.2.0
+  version: 3.2.3
 platform: ruby
 authors:
 - James Edward Gray II
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-06-05 00:00:00.000000000 Z
+date: 2022-04-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -59,14 +59,14 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 3.4.3
+        version: 3.4.8
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 3.4.3
+        version: 3.4.8
 description: The CSV library provides a complete interface to CSV files and data.
   It offers tools to enable you to read and write to and from Strings or IO objects,
   as needed.
@@ -118,6 +118,7 @@ files:
 - lib/csv/core_ext/string.rb
 - lib/csv/delete_suffix.rb
 - lib/csv/fields_converter.rb
+- lib/csv/input_record_separator.rb
 - lib/csv/match_p.rb
 - lib/csv/parser.rb
 - lib/csv/row.rb
@@ -146,7 +147,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.3.0.dev
+rubygems_version: 3.4.0.dev
 signing_key: 
 specification_version: 4
 summary: CSV Reading and Writing