csv 3.2.1 → 3.2.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 6aee33c500a979f9f1a9afdbc394545932700ca3442c1b4326b417e59b654ef4
-  data.tar.gz: 6622e0c4f190f10aa6d3a49b023a6c24540b1133da4d0ef91c07016ba224073f
+  metadata.gz: 3ef88fc9b205f8f64c5e817b22f121d5d03534c155cb8d27dc9d87aa62e6b7e0
+  data.tar.gz: e12b86cee946837a96ae609314a50246e2135fab855dca58e800de1ddc50e524
 SHA512:
-  metadata.gz: 7ba35310fd8dc9ffd4075ca31d786b7458fefefaee4c5a00fc326063fb13d020d1959d9970dcf17c222dd5d2422e3d20b9ddcd503caf5a4aad04e0b93159ff60
-  data.tar.gz: 9d1971baae109ad7396124cfa6bda15bc6db635ff09c6166a750dc22633b552679e5d589b15a57724f7b4aef3ae33c2ccd2fcfcfb1084e71df7f3734347e9926
+  metadata.gz: d663d0917d63315e4fc5802f9538727731f39ea06997adea485f5f137f37639f0791db1f9697476e7d0c4a8048a0339b9e4fb7aecd79f3ec8ee8eb24a4cb676e
+  data.tar.gz: f9d22fe50d227b8f8ca0d075ed0fbf1f39a10bf59a1adb68835fede727f3d1dfd1d2f88754caa1098fe9c24df738c71b29ae5fc3080a25d80df484f76530344e

data/NEWS.md

@@ -1,5 +1,118 @@
 # News
 
+## 3.2.4 - 2022-08-22
+
+### Improvements
+
+  * Cleaned up internal implementations.
+    [[GitHub#249](https://github.com/ruby/csv/pull/249)]
+    [[GitHub#250](https://github.com/ruby/csv/pull/250)]
+    [[GitHub#251](https://github.com/ruby/csv/pull/251)]
+    [Patch by Mau Magnaguagno]
+
+  * Added support for RFC 3339 style time.
+    [[GitHub#248](https://github.com/ruby/csv/pull/248)]
+    [Patch by Thierry Lambert]
+
+  * Added support for transcoding String CSV. Syntax is
+    `from-encoding:to-encoding`.
+    [[GitHub#254](https://github.com/ruby/csv/issues/254)]
+    [Reported by Richard Stueven]
+
+  * Added quoted information to `CSV::FieldInfo`.
+    [[GitHub#254](https://github.com/ruby/csv/pull/253)]
+    [Reported by Hirokazu SUZUKI]
+
+### Fixes
+
+  * Fixed a link in documents.
+    [[GitHub#244](https://github.com/ruby/csv/pull/244)]
+    [Patch by Peter Zhu]
+
+### Thanks
+
+  * Peter Zhu
+
+  * Mau Magnaguagno
+
+  * Thierry Lambert
+
+  * Richard Stueven
+
+  * Hirokazu SUZUKI
+
+## 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

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

@@ -44,7 +44,7 @@ class CSV
       @converters.empty?
     end
 
-    def convert(fields, headers, lineno)
+    def convert(fields, headers, lineno, quoted_fields)
       return fields unless need_convert?
 
       fields.collect.with_index do |field, index|
@@ -63,7 +63,8 @@ class CSV
             else
               header = nil
             end
-            field = converter[field, FieldInfo.new(index, lineno, header)]
+            quoted = quoted_fields[index]
+            field = converter[field, FieldInfo.new(index, lineno, header, quoted)]
           end
           break unless field.is_a?(String)  # short-circuit pipeline for speed
         end

data/lib/csv/input_record_separator.rb

@@ -4,20 +4,7 @@ require "stringio"
 class CSV
   module InputRecordSeparator
     class << self
-      is_input_record_separator_deprecated = false
-      verbose, $VERBOSE = $VERBOSE, true
-      stderr, $stderr = $stderr, StringIO.new
-      input_record_separator = $INPUT_RECORD_SEPARATOR
-      begin
-        $INPUT_RECORD_SEPARATOR = "\r\n"
-        is_input_record_separator_deprecated = (not $stderr.string.empty?)
-      ensure
-        $INPUT_RECORD_SEPARATOR = input_record_separator
-        $stderr = stderr
-        $VERBOSE = verbose
-      end
-
-      if is_input_record_separator_deprecated
+      if RUBY_VERSION >= "3.0.0"
         def value
           "\n"
         end

data/lib/csv/parser.rb

@@ -2,15 +2,10 @@
 
 require "strscan"
 
-require_relative "delete_suffix"
 require_relative "input_record_separator"
-require_relative "match_p"
 require_relative "row"
 require_relative "table"
 
-using CSV::DeleteSuffix if CSV.const_defined?(:DeleteSuffix)
-using CSV::MatchP if CSV.const_defined?(:MatchP)
-
 class CSV
   # Note: Don't use this class directly. This is an internal class.
   class Parser
@@ -27,6 +22,10 @@ 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+,
@@ -78,16 +77,17 @@ 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 = []
@@ -95,11 +95,13 @@ 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
@@ -139,25 +141,28 @@ 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
 
@@ -166,76 +171,135 @@ 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
@@ -243,6 +307,7 @@ class CSV
             end
             true
           else
+            # trace(__method__, :no_chunk)
             @scanner = StringScanner.new("".encode(@encoding))
             @inputs.shift
             @last_scanner = @inputs.empty?
@@ -277,7 +342,11 @@ 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
@@ -345,6 +414,16 @@ 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
 
@@ -361,6 +440,7 @@ class CSV
       prepare_skip_lines
       prepare_strip
       prepare_separators
+      validate_strip_and_col_sep_options
       prepare_quoted
       prepare_unquoted
       prepare_line
@@ -388,7 +468,7 @@ 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]
@@ -531,6 +611,28 @@ class CSV
       @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 +
@@ -656,9 +758,10 @@ class CSV
       case headers
       when Array
         @raw_headers = headers
+        quoted_fields = [false] * @raw_headers.size
         @use_headers = true
       when String
-        @raw_headers = parse_headers(headers)
+        @raw_headers, quoted_fields = parse_headers(headers)
         @use_headers = true
       when nil, false
         @raw_headers = nil
@@ -668,21 +771,28 @@ class CSV
         @use_headers = true
       end
       if @raw_headers
-        @headers = adjust_headers(@raw_headers)
+        @headers = adjust_headers(@raw_headers, quoted_fields)
       else
         @headers = nil
       end
     end
 
     def parse_headers(row)
-      CSV.parse_line(row,
-                     col_sep:    @column_separator,
-                     row_sep:    @row_separator,
-                     quote_char: @quote_character)
+      quoted_fields = []
+      converter = lambda do |field, info|
+        quoted_fields << info.quoted?
+        field
+      end
+      headers = CSV.parse_line(row,
+                               col_sep:    @column_separator,
+                               row_sep:    @row_separator,
+                               quote_char: @quote_character,
+                               converters: [converter])
+      [headers, quoted_fields]
     end
 
-    def adjust_headers(headers)
-      adjusted_headers = @header_fields_converter.convert(headers, nil, @lineno)
+    def adjust_headers(headers, quoted_fields)
+      adjusted_headers = @header_fields_converter.convert(headers, nil, @lineno, quoted_fields)
       adjusted_headers.each {|h| h.freeze if h.is_a? String}
       adjusted_headers
     end
@@ -705,28 +815,28 @@ 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_CHUNK_SIZE =
-        Integer((ENV["CSV_PARSER_SCANNER_TEST_CHUNK_SIZE"] || "1"), 10)
+    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)
@@ -736,16 +846,27 @@ class CSV
         else
           inputs << @input
         end
+        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: SCANNER_TEST_CHUNK_SIZE)
+                          @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
@@ -764,7 +885,7 @@ class CSV
             StringIO.new(sample)
           end
           inputs << @input
-          InputsScanner.new(inputs, @encoding)
+          InputsScanner.new(inputs, @encoding, @row_separator)
         end
       end
     end
@@ -798,6 +919,14 @@ 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)
@@ -807,9 +936,16 @@ class CSV
         if line.empty?
           next if @skip_blanks
           row = []
+          quoted_fields = []
         else
           line = strip_value(line)
           row = line.split(@split_column_separator, -1)
+          quoted_fields = [false] * row.size
+          if @max_field_size
+            row.each do |column|
+              validate_field_size(column)
+            end
+          end
           n_columns = row.size
           i = 0
           while i < n_columns
@@ -818,7 +954,7 @@ class CSV
           end
         end
         @last_line = original_line
-        emit_row(row, &block)
+        emit_row(row, quoted_fields, &block)
       end
     end
 
@@ -840,31 +976,37 @@ class CSV
             next
           end
           row = []
+          quoted_fields = []
         elsif line.include?(@cr) or line.include?(@lf)
           @scanner.keep_back
           @need_robust_parsing = true
           return parse_quotable_robust(&block)
         else
           row = line.split(@split_column_separator, -1)
+          quoted_fields = []
           n_columns = row.size
           i = 0
           while i < n_columns
             column = row[i]
             if column.empty?
+              quoted_fields << false
               row[i] = nil
             else
               n_quotes = column.count(@quote_character)
               if n_quotes.zero?
+                quoted_fields << false
                 # no quote
               elsif n_quotes == 2 and
                    column.start_with?(@quote_character) and
                    column.end_with?(@quote_character)
+                quoted_fields << true
                 row[i] = column[1..-2]
               else
                 @scanner.keep_back
                 @need_robust_parsing = true
                 return parse_quotable_robust(&block)
               end
+              validate_field_size(row[i])
             end
             i += 1
           end
@@ -872,13 +1014,14 @@ class CSV
         @scanner.keep_drop
         @scanner.keep_start
         @last_line = original_line
-        emit_row(row, &block)
+        emit_row(row, quoted_fields, &block)
       end
       @scanner.keep_drop
     end
 
     def parse_quotable_robust(&block)
       row = []
+      quoted_fields = []
       skip_needless_lines
       start_row
       while true
@@ -888,32 +1031,39 @@ 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
+          quoted_fields << @quoted_column_value
         elsif parse_row_end
           if row.empty? and value.nil?
-            emit_row([], &block) unless @skip_blanks
+            emit_row([], [], &block) unless @skip_blanks
           else
             row << value
-            emit_row(row, &block)
+            quoted_fields << @quoted_column_value
+            emit_row(row, quoted_fields, &block)
             row = []
+            quoted_fields = []
           end
           skip_needless_lines
           start_row
         elsif @scanner.eos?
           break if row.empty? and value.nil?
           row << value
-          emit_row(row, &block)
+          quoted_fields << @quoted_column_value
+          emit_row(row, quoted_fields, &block)
           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(@line_end))
@@ -1006,7 +1156,7 @@ class CSV
       if (n_quotes % 2).zero?
         quotes[0, (n_quotes - 2) / 2]
       else
-        value = quotes[0, (n_quotes - 1) / 2]
+        value = quotes[0, n_quotes / 2]
         while true
           quoted_value = @scanner.scan_all(@quoted_value)
           value << quoted_value if quoted_value
@@ -1030,11 +1180,9 @@ class CSV
           n_quotes = quotes.size
           if n_quotes == 1
             break
-          elsif (n_quotes % 2) == 1
-            value << quotes[0, (n_quotes - 1) / 2]
-            break
           else
             value << quotes[0, n_quotes / 2]
+            break if (n_quotes % 2) == 1
           end
         end
         value
@@ -1070,18 +1218,15 @@ class CSV
 
     def strip_value(value)
       return value unless @strip
-      return nil if value.nil?
+      return value if value.nil?
 
       case @strip
       when String
-        size = value.size
-        while value.start_with?(@strip)
-          size -= 1
-          value = value[1, size]
+        while value.delete_prefix!(@strip)
+          # do nothing
         end
-        while value.end_with?(@strip)
-          size -= 1
-          value = value[0, size]
+        while value.delete_suffix!(@strip)
+          # do nothing
         end
       else
         value.strip!
@@ -1104,22 +1249,22 @@ class CSV
       @scanner.keep_start
     end
 
-    def emit_row(row, &block)
+    def emit_row(row, quoted_fields, &block)
       @lineno += 1
 
       raw_row = row
       if @use_headers
         if @headers.nil?
-          @headers = adjust_headers(row)
+          @headers = adjust_headers(row, quoted_fields)
           return unless @return_headers
           row = Row.new(@headers, row, true)
         else
           row = Row.new(@headers,
-                        @fields_converter.convert(raw_row, @headers, @lineno))
+                        @fields_converter.convert(raw_row, @headers, @lineno, quoted_fields))
         end
       else
         # convert fields, if needed...
-        row = @fields_converter.convert(raw_row, nil, @lineno)
+        row = @fields_converter.convert(raw_row, nil, @lineno, quoted_fields)
       end
 
       # inject unconverted fields and accessor, if requested...

data/lib/csv/row.rb

@@ -703,7 +703,7 @@ class CSV
     # by +index_or_header+ and +specifiers+.
     #
     # The nested objects may be instances of various classes.
-    # See {Dig Methods}[https://docs.ruby-lang.org/en/master/doc/dig_methods_rdoc.html].
+    # See {Dig Methods}[https://docs.ruby-lang.org/en/master/dig_methods_rdoc.html].
     #
     # Examples:
     #   source = "Name,Value\nfoo,0\nbar,1\nbaz,2\n"

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 = "3.2.1"
+  VERSION = "3.2.4"
 end

data/lib/csv/writer.rb

@@ -1,11 +1,8 @@
 # frozen_string_literal: true
 
 require_relative "input_record_separator"
-require_relative "match_p"
 require_relative "row"
 
-using CSV::MatchP if CSV.const_defined?(:MatchP)
-
 class CSV
   # Note: Don't use this class directly. This is an internal class.
   class Writer
@@ -42,7 +39,10 @@ class CSV
       @headers ||= row if @use_headers
       @lineno += 1
 
-      row = @fields_converter.convert(row, nil, lineno) if @fields_converter
+      if @fields_converter
+        quoted_fields = [false] * row.size
+        row = @fields_converter.convert(row, nil, lineno, quoted_fields)
+      end
 
       i = -1
       converted_row = row.collect do |field|
@@ -97,7 +97,7 @@ class CSV
       return unless @headers
 
       converter = @options[:header_fields_converter]
-      @headers = converter.convert(@headers, nil, 0)
+      @headers = converter.convert(@headers, nil, 0, [])
       @headers.each do |header|
         header.freeze if header.is_a?(String)
       end

data/lib/csv.rb

@@ -95,14 +95,11 @@ 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"
 require_relative "csv/table"
 require_relative "csv/writer"
 
-using CSV::MatchP if CSV.const_defined?(:MatchP)
-
 # == \CSV
 #
 # === In a Hurry?
@@ -341,6 +338,7 @@ 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 +346,6 @@ 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 +354,9 @@ 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 +365,9 @@ 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.
@@ -863,8 +863,9 @@ class CSV
   # <b><tt>index</tt></b>::  The zero-based index of the field in its row.
   # <b><tt>line</tt></b>::   The line of the data source this row is from.
   # <b><tt>header</tt></b>:: The header for the column, when available.
+  # <b><tt>quoted?</tt></b>:: True or false, whether the original value is quoted or not.
   #
-  FieldInfo = Struct.new(:index, :line, :header)
+  FieldInfo = Struct.new(:index, :line, :header, :quoted?)
 
   # A Regexp used to find and convert some common Date formats.
   DateMatcher     = / \A(?: (\w+,?\s+)?\w+\s+\d{1,2},?\s+\d{2,4} |
@@ -872,10 +873,9 @@ class CSV
   # A Regexp used to find and convert some common DateTime formats.
   DateTimeMatcher =
     / \A(?: (\w+,?\s+)?\w+\s+\d{1,2}\s+\d{1,2}:\d{1,2}:\d{1,2},?\s+\d{2,4} |
-            \d{4}-\d{2}-\d{2}\s\d{2}:\d{2}:\d{2} |
-            # ISO-8601
+            # ISO-8601 and RFC-3339 (space instead of T) recognized by DateTime.parse
             \d{4}-\d{2}-\d{2}
-              (?:T\d{2}:\d{2}(?::\d{2}(?:\.\d+)?(?:[+-]\d{2}(?::\d{2})|Z)?)?)?
+              (?:[T\s]\d{2}:\d{2}(?::\d{2}(?:\.\d+)?(?:[+-]\d{2}(?::\d{2})|Z)?)?)?
         )\z /x
 
   # The encoding used by all converters.
@@ -925,7 +925,8 @@ 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.
@@ -936,6 +937,7 @@ class CSV
     quote_char:         '"',
     # For parsing.
     field_size_limit:   nil,
+    max_field_size:     nil,
     converters:         nil,
     unconverted_fields: nil,
     headers:            false,
@@ -946,6 +948,7 @@ class CSV
     liberal_parsing:    false,
     nil_value:          nil,
     empty_value:        "",
+    strip:              false,
     # For generating.
     write_headers:      nil,
     quote_empty:        true,
@@ -953,7 +956,6 @@ class CSV
     write_converters:   nil,
     write_nil_value:    nil,
     write_empty_value:  "",
-    strip:              false,
   }.freeze
 
   class << self
@@ -1864,6 +1866,7 @@ class CSV
                  row_sep: :auto,
                  quote_char: '"',
                  field_size_limit: nil,
+                 max_field_size: nil,
                  converters: nil,
                  unconverted_fields: nil,
                  headers: false,
@@ -1879,16 +1882,27 @@ 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)
+      if encoding
+        if encoding.is_a?(String)
+          data_external_encoding, data_internal_encoding = encoding.split(":", 2)
+          if data_internal_encoding
+            data = data.encode(data_internal_encoding, data_external_encoding)
+          else
+            data = data.dup.force_encoding(data_external_encoding)
+          end
+        else
+          data = data.dup.force_encoding(encoding)
+        end
+      end
       @io = StringIO.new(data)
-      @io.set_encoding(encoding || data.encoding)
     else
       @io = data
     end
@@ -1906,11 +1920,14 @@ 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,
@@ -1978,10 +1995,24 @@ 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
   #

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: csv
 version: !ruby/object:Gem::Version
-  version: 3.2.1
+  version: 3.2.4
 platform: ruby
 authors:
 - James Edward Gray II
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-10-22 00:00:00.000000000 Z
+date: 2022-08-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -116,10 +116,8 @@ files:
 - lib/csv.rb
 - lib/csv/core_ext/array.rb
 - 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
 - lib/csv/table.rb
@@ -147,7 +145,7 @@ 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

data/lib/csv/delete_suffix.rb

@@ -1,18 +0,0 @@
-# frozen_string_literal: true
-
-# This provides String#delete_suffix? for Ruby 2.4.
-unless String.method_defined?(:delete_suffix)
-  class CSV
-    module DeleteSuffix
-      refine String do
-        def delete_suffix(suffix)
-          if end_with?(suffix)
-            self[0...-suffix.size]
-          else
-            self
-          end
-        end
-      end
-    end
-  end
-end

data/lib/csv/match_p.rb

@@ -1,20 +0,0 @@
-# frozen_string_literal: true
-
-# This provides String#match? and Regexp#match? for Ruby 2.3.
-unless String.method_defined?(:match?)
-  class CSV
-    module MatchP
-      refine String do
-        def match?(pattern)
-          self =~ pattern
-        end
-      end
-
-      refine Regexp do
-        def match?(string)
-          self =~ string
-        end
-      end
-    end
-  end
-end