dbf 5.2.0 → 5.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 96b3352f6c6d2cc3a667c6f390021c4f99be7c83ff2b9c911ac38c75d4c261aa
-  data.tar.gz: b787cd0e14a34ea66e670f88e793bcbebc1ccbdc260721da08ccee5d3de2624a
+  metadata.gz: d2d1ca393ae41262fd052c74ffe0ec497a963a2c036797686061312a0c492888
+  data.tar.gz: 7e92c62ace7713b883b6b2e925c34f7a8114ebd6345b36a81f97fb47109711e2
 SHA512:
-  metadata.gz: 52ceab2c71a3093ff691be23a0a0590e4309fd88218e683032e12805585dea05dd97bebd0170f003e7678641810d5d9aa6b9731bb0af647a8508d015875312d2
-  data.tar.gz: 41b87283de2deea377a3c27436d621486af57bc0421e5bc84dc1bd1f304abf8d264d72d8dda80350666895ca604efd335b829fd1efc79720721027146fe56b7d
+  metadata.gz: f0f3b1b02ab53f505ab3efc1dff89d2ad1169698ef7d9443b3eec3b758b921ae2df51bf443d4e877ac072bdef87c547607d1d39d913a5c361b4c31e21fee4b07
+  data.tar.gz: 6d826dd35aa6e80b671aeb804dc9991776851bbb9ca0ff5fb9ed6cd7139cbd40b0bad68327fc8f17b03b9ba2f2b77479331b7ad998c53e7ee32e5caee59241e6

data/CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## main branch
+
+- Add support for 8 more code pages (issue #98): Mazovia cp620 and Kamenický cp895 via vendored translation tables (new DBF::Encoder), plus macRoman, cp1255, cp1256, macCyrillic, macCentEuro and macGreek
+- Blank Visual FoxPro "T" (DateTime) columns now return nil instead of a Julian day-0 date
+- Blank "F" (Float) columns now return nil instead of 0.0, matching "N" (Number) behavior
+- Extract CLI logic into DBF::CLI class (bin/dbf is now a thin shim)
+- DBF::Table#to_csv now accepts an IO in addition to a String path
+- Fix broken link in README
+
 ## 5.2.0
 
 - Drop support for Ruby 3.1 and 3.2

data/README.md

@@ -10,7 +10,7 @@
 DBF is a small, fast Ruby library for reading dBase, xBase, Clipper, and FoxPro database files.
 
 * Project page: <https://github.com/infused/dbf>
-* API Documentation: <https://rubydoc.info/github/infused/dbf>
+* API Documentation: <https://rdoc.info/gems/dbf>
 * Report bugs: <https://github.com/infused/dbf/issues>
 * Questions: Email <mailto:keithm@infused.org> and put DBF somewhere in the
   subject line
@@ -21,7 +21,7 @@ NOTE: Beginning with version 5.2 we have dropped support for Ruby 3.2 and earlie
 NOTE: Beginning with version 4.3 we have dropped support for Ruby 3.0 and earlier.
 
 NOTE: Beginning with version 4 we have dropped support for Ruby 2.0, 2.1, 2.2, and 2.3. If you need support for these older Rubies,
-please use 3.0.x (<https://github.com/infused.org/dbf/tree/3_stable>)
+please use 3.0.x (<https://github.com/infused/dbf/tree/3_stable>)
 
 NOTE: Beginning with version 3 we have dropped support for Ruby 1.8 and 1.9. If you need support for older Rubies,
 please use 2.0.x (<https://github.com/infused/dbf/tree/2_stable>)
@@ -151,6 +151,7 @@ table = DBF::Table.new('dbf/books.dbf', nil, 'cp866')
 | 01 | cp437 | U.S. MS–DOS |
 | 02 | cp850 | International MS–DOS |
 | 03 | cp1252 | Windows ANSI |
+| 04 | macRoman | Standard Macintosh |
 | 08 | cp865 | Danish OEM |
 | 09 | cp437 | Dutch OEM |
 | 0a | cp850 | Dutch OEM* |
@@ -191,6 +192,8 @@ table = DBF::Table.new('dbf/books.dbf', nil, 'cp866')
 | 65 | cp866 | Russian MS–DOS |
 | 66 | cp865 | Nordic MS–DOS |
 | 67 | cp861 | Icelandic MS–DOS |
+| 68 | cp895 | Kamenický (Czech) MS–DOS |
+| 69 | cp620 | Mazovia (Polish) MS–DOS |
 | 6a | cp737 | Greek MS–DOS (437G) |
 | 6b | cp857 | Turkish MS–DOS |
 | 6c | cp863 | French–Canadian MS–DOS |
@@ -199,9 +202,14 @@ table = DBF::Table.new('dbf/books.dbf', nil, 'cp866')
 | 7a | cp936 | PRC GBK |
 | 7b | cp932 | Japanese Shift-JIS |
 | 7c | cp874 | Thai Windows/MS–DOS |
+| 7d | cp1255 | Hebrew Windows |
+| 7e | cp1256 | Arabic Windows |
 | 86 | cp737 | Greek OEM |
 | 87 | cp852 | Slovenian OEM |
 | 88 | cp857 | Turkish OEM |
+| 96 | macCyrillic | Russian Macintosh |
+| 97 | macCentEuro | Macintosh EE |
+| 98 | macGreek | Greek Macintosh |
 | c8 | cp1250 | Eastern European Windows |
 | c9 | cp1251 | Russian Windows |
 | ca | cp1254 | Turkish Windows |
@@ -328,16 +336,16 @@ at this time.
 | 04      | dBase IV without memo file                          | Y | Y | Y | Y | - | - | - | - | - | - | - | - | - | - | - | - | - |
 | 05      | dBase V without memo file                           | Y | Y | Y | Y | - | - | - | - | - | - | - | - | - | - | - | - | - |
 | 07      | Visual Objects 1.x                                  | Y | Y | Y | Y | - | - | - | - | - | - | - | - | - | - | - | - | - |
-| 30      | Visual FoxPro                                       | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | N | Y | N | N | N | N | - |
-| 31      | Visual FoxPro with AutoIncrement                    | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | N | Y | N | N | N | N | N |
-| 32      | Visual FoxPro with field type Varchar or Varbinary  | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | N | Y | N | N | N | N | N |
+| 30      | Visual FoxPro                                       | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | Y | Y | N | N | N | N | - |
+| 31      | Visual FoxPro with AutoIncrement                    | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | Y | Y | N | N | N | N | N |
+| 32      | Visual FoxPro with field type Varchar or Varbinary  | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | Y | Y | N | N | N | N | N |
 | 7b      | dBase IV with memo file                             | Y | Y | Y | Y | Y | Y | - | - | - | - | - | - | - | - | - | - | - |
 | 83      | dBase III with memo file                            | Y | Y | Y | Y | Y | - | - | - | - | - | - | - | - | - | - | - | - |
 | 87      | Visual Objects 1.x with memo file                   | Y | Y | Y | Y | Y | - | - | - | - | - | - | - | - | - | - | - | - |
 | 8b      | dBase IV with memo file                             | Y | Y | Y | Y | Y | - | - | - | - | - | - | - | - | N | - | - | - |
 | 8e      | dBase IV with SQL table                             | Y | Y | Y | Y | Y | - | - | - | - | - | - | - | - | N | - | - | - |
-| f5      | FoxPro with memo file                               | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | N | Y | N | N | N | N | N |
-| fb      | FoxPro without memo file                            | Y | Y | Y | Y | - | Y | Y | Y | N | Y | N | Y | N | N | N | N | N |
+| f5      | FoxPro with memo file                               | Y | Y | Y | Y | Y | Y | Y | Y | N | Y | Y | Y | N | N | N | N | N |
+| fb      | FoxPro without memo file                            | Y | Y | Y | Y | - | Y | Y | Y | N | Y | Y | Y | N | N | N | N | N |
 
 Data type descriptions
 

data/bin/dbf

@@ -1,60 +1,8 @@
 #!/usr/bin/env ruby
+# frozen_string_literal: true
 
 Signal.trap('PIPE', 'SYSTEM_DEFAULT')
 
 require 'dbf'
-require 'dbf/version'
-require 'optparse'
 
-params = ARGV.getopts('h', 's', 'a', 'c', 'r', 'v')
-
-if params['v']
-  puts "dbf version: #{DBF::VERSION}"
-
-elsif params['h']
-  puts "usage: #{File.basename(__FILE__)} [-h|-s|-a|-c|-r] filename"
-  puts '  -h = print this message'
-  puts '  -v = print the DBF gem version'
-  puts '  -s = print summary information'
-  puts '  -a = create an ActiveRecord::Schema'
-  puts '  -r = create a Sequel migration'
-  puts '  -c = export as CSV'
-else
-
-  filename = ARGV.shift
-  abort 'You must supply a filename on the command line' unless filename
-
-  # create an ActiveRecord::Schema
-  if params['a']
-    table = DBF::Table.new filename
-    puts table.schema(:activerecord)
-  end
-
-  # create an Sequel::Migration
-  if params['r']
-    table = DBF::Table.new filename
-    puts table.schema(:sequel)
-  end
-
-  if params['s']
-    table = DBF::Table.new filename
-    puts
-    puts "Database: #{filename}"
-    puts "Type: (#{table.version}) #{table.version_description}"
-    puts "Encoding: #{table.header_encoding}" if table.header_encoding
-    puts "Memo File: #{table.has_memo_file? ? 'true' : 'false'}"
-    puts "Records: #{table.record_count}"
-
-    puts "\nFields:"
-    puts 'Name             Type       Length     Decimal'
-    puts '-' * 78
-    table.columns.each do |f|
-      puts format('%-16s %-10s %-10s %-10s', f.name, f.type, f.length, f.decimal)
-    end
-  end
-
-  if params['c']
-    table = DBF::Table.new filename
-    table.to_csv
-  end
-end
+exit DBF::CLI.run(ARGV)

data/lib/dbf/cli.rb

@@ -0,0 +1,95 @@
+# frozen_string_literal: true
+
+require 'optparse'
+
+module DBF
+  class CLI
+    USAGE = <<~HELP
+      usage: dbf [-h|-s|-a|-c|-r] filename
+        -h = print this message
+        -v = print the DBF gem version
+        -s = print summary information
+        -a = create an ActiveRecord::Schema
+        -r = create a Sequel migration
+        -c = export as CSV
+    HELP
+
+    def self.run(argv, stdout: $stdout, stderr: $stderr)
+      new(argv, stdout: stdout, stderr: stderr).run
+    end
+
+    def initialize(argv, stdout: $stdout, stderr: $stderr)
+      @argv = argv.dup
+      @stdout = stdout
+      @stderr = stderr
+    end
+
+    def run
+      params = OptionParser.new.getopts(@argv, 'h', 's', 'a', 'c', 'r', 'v')
+
+      if params['v']
+        print_version
+      elsif params['h']
+        print_help
+      else
+        filename = @argv.shift
+        return missing_filename unless filename
+
+        action = %w[a r s c].find { |flag| params[flag] }
+        case action
+        when 'a' then print_ar_schema(filename)
+        when 'r' then print_sequel_schema(filename)
+        when 's' then print_summary(filename)
+        when 'c' then print_csv(filename)
+        end
+      end
+      0
+    rescue DBF::FileNotFoundError => e
+      @stderr.puts "DBF::FileNotFoundError: #{e.message}"
+      1
+    end
+
+    private
+
+    def print_version
+      @stdout.puts "dbf version: #{DBF::VERSION}"
+    end
+
+    def print_help
+      @stdout.puts USAGE
+    end
+
+    def missing_filename
+      @stderr.puts 'You must supply a filename on the command line'
+      1
+    end
+
+    def print_ar_schema(filename)
+      @stdout.puts DBF::Table.new(filename).schema(:activerecord)
+    end
+
+    def print_sequel_schema(filename)
+      @stdout.puts DBF::Table.new(filename).schema(:sequel)
+    end
+
+    def print_summary(filename)
+      table = DBF::Table.new(filename)
+      @stdout.puts
+      @stdout.puts "Database: #{filename}"
+      @stdout.puts "Type: (#{table.version}) #{table.version_description}"
+      @stdout.puts "Encoding: #{table.header_encoding}" if table.header_encoding
+      @stdout.puts "Memo File: #{table.has_memo_file? ? 'true' : 'false'}"
+      @stdout.puts "Records: #{table.record_count}"
+      @stdout.puts "\nFields:"
+      @stdout.puts 'Name             Type       Length     Decimal'
+      @stdout.puts '-' * 78
+      table.columns.each do |f|
+        @stdout.puts format('%-16s %-10s %-10s %-10s', f.name, f.type, f.length, f.decimal)
+      end
+    end
+
+    def print_csv(filename)
+      DBF::Table.new(filename).to_csv(@stdout)
+    end
+  end
+end

data/lib/dbf/column.rb

@@ -58,8 +58,8 @@ module DBF
     # @param raw [String]
     # @yield [raw] for memo column resolution
     # @return decoded value
-    def decode(raw, &memo_handler)
-      type_cast_class.decode(raw, &memo_handler)
+    def decode(raw, &)
+      type_cast_class.decode(raw, &)
     end
 
     # Returns a Hash with :name, :type, :length, and :decimal keys

data/lib/dbf/column_type.rb

@@ -5,8 +5,7 @@ module DBF
     class Base
       attr_reader :decimal, :encoding
 
-      # @param decimal [Integer]
-      # @param encoding [String, Encoding]
+      # @param column [DBF::Column]
       def initialize(column)
         @decimal = column.decimal
         @encoding = column.encoding
@@ -20,7 +19,7 @@ module DBF
         false
       end
 
-      def decode(raw, &_memo_handler)
+      def decode(raw, &)
         if skip_blank? && raw.count(' ') == raw.length
           blank_value
         else
@@ -71,8 +70,12 @@ module DBF
     end
 
     class Float < Base
+      def skip_blank? = true
+
       # @param value [String]
       def type_cast(value)
+        return nil if value.empty?
+
         value.to_f
       end
     end
@@ -111,6 +114,8 @@ module DBF
       # @param value [String]
       def type_cast(value)
         days, msecs = value.unpack('l2')
+        return nil if days.nil? || days.zero?
+
         secs = (msecs / 1000).to_i
         ::DateTime.jd(days, (secs / 3600).to_i, (secs / 60).to_i % 60, secs % 60).to_time
       rescue StandardError
@@ -119,8 +124,8 @@ module DBF
     end
 
     class Memo < Base
-      def decode(raw, &memo_handler)
-        memo_content = memo_handler.call(raw)
+      def decode(raw, &)
+        memo_content = yield(raw)
         memo_content ? type_cast(memo_content) : nil
       end
 
@@ -128,7 +133,7 @@ module DBF
       def type_cast(value)
         return value unless encoding && value
 
-        value.dup.force_encoding(encoding).encode(Encoding.default_external, undef: :replace, invalid: :replace)
+        Encoder.encode(value.dup, encoding)
       end
     end
 
@@ -158,8 +163,9 @@ module DBF
       private
 
       def encode(value)
-        value.force_encoding(encoding)
-        @needs_encode ? value.encode(@target_encoding, undef: :replace, invalid: :replace) : value
+        return value.force_encoding(encoding) unless @needs_encode
+
+        Encoder.encode(value, encoding, @target_encoding)
       end
     end
   end

data/lib/dbf/encoder.rb

@@ -0,0 +1,116 @@
+# frozen_string_literal: true
+
+module DBF
+  # Transcodes strings from a source encoding to a target encoding.
+  #
+  # Most DBF code pages map to encodings built into Ruby and are converted
+  # with String#encode. A few (Mazovia cp620, Kamenický cp895) have no Ruby
+  # equivalent and are decoded through vendored byte-to-Unicode tables.
+  module Encoder
+    # Byte-to-Unicode tables generated from
+    # https://github.com/SheetJS/js-codepage/blob/master/codepages/620.TBL and
+    # https://github.com/SheetJS/js-codepage/blob/master/codepages/895.TBL
+
+    # Mazovia (Polish) MS-DOS
+    CP620_TO_UNICODE = [
+      0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007,
+      0x0008, 0x0009, 0x000A, 0x000B, 0x000C, 0x000D, 0x000E, 0x000F,
+      0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017,
+      0x0018, 0x0019, 0x001A, 0x001B, 0x001C, 0x001D, 0x001E, 0x001F,
+      0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027,
+      0x0028, 0x0029, 0x002A, 0x002B, 0x002C, 0x002D, 0x002E, 0x002F,
+      0x0030, 0x0031, 0x0032, 0x0033, 0x0034, 0x0035, 0x0036, 0x0037,
+      0x0038, 0x0039, 0x003A, 0x003B, 0x003C, 0x003D, 0x003E, 0x003F,
+      0x0040, 0x0041, 0x0042, 0x0043, 0x0044, 0x0045, 0x0046, 0x0047,
+      0x0048, 0x0049, 0x004A, 0x004B, 0x004C, 0x004D, 0x004E, 0x004F,
+      0x0050, 0x0051, 0x0052, 0x0053, 0x0054, 0x0055, 0x0056, 0x0057,
+      0x0058, 0x0059, 0x005A, 0x005B, 0x005C, 0x005D, 0x005E, 0x005F,
+      0x0060, 0x0061, 0x0062, 0x0063, 0x0064, 0x0065, 0x0066, 0x0067,
+      0x0068, 0x0069, 0x006A, 0x006B, 0x006C, 0x006D, 0x006E, 0x006F,
+      0x0070, 0x0071, 0x0072, 0x0073, 0x0074, 0x0075, 0x0076, 0x0077,
+      0x0078, 0x0079, 0x007A, 0x007B, 0x007C, 0x007D, 0x007E, 0x007F,
+      0x00C7, 0x00FC, 0x00E9, 0x00E2, 0x00E4, 0x00E0, 0x0105, 0x00E7,
+      0x00EA, 0x00EB, 0x00E8, 0x00EF, 0x00EE, 0x0107, 0x00C4, 0x0104,
+      0x0118, 0x0119, 0x0142, 0x00F4, 0x00F6, 0x0106, 0x00FB, 0x00F9,
+      0x015A, 0x00D6, 0x00DC, 0x00A2, 0x0141, 0x00A5, 0x015B, 0x0192,
+      0x0179, 0x017B, 0x00F3, 0x00D3, 0x0144, 0x0143, 0x017A, 0x017C,
+      0x00BF, 0x2310, 0x00AC, 0x00BD, 0x00BC, 0x00A1, 0x00AB, 0x00BB,
+      0x2591, 0x2592, 0x2593, 0x2502, 0x2524, 0x2561, 0x2562, 0x2556,
+      0x2555, 0x2563, 0x2551, 0x2557, 0x255D, 0x255C, 0x255B, 0x2510,
+      0x2514, 0x2534, 0x252C, 0x251C, 0x2500, 0x253C, 0x255E, 0x255F,
+      0x255A, 0x2554, 0x2569, 0x2566, 0x2560, 0x2550, 0x256C, 0x2567,
+      0x2568, 0x2564, 0x2565, 0x2559, 0x2558, 0x2552, 0x2553, 0x256B,
+      0x256A, 0x2518, 0x250C, 0x2588, 0x2584, 0x258C, 0x2590, 0x2580,
+      0x03B1, 0x00DF, 0x0393, 0x03C0, 0x03A3, 0x03C3, 0x00B5, 0x03C4,
+      0x03A6, 0x0398, 0x03A9, 0x03B4, 0x221E, 0x03C6, 0x03B5, 0x2229,
+      0x2261, 0x00B1, 0x2265, 0x2264, 0x2320, 0x2321, 0x00F7, 0x2248,
+      0x00B0, 0x2219, 0x00B7, 0x221A, 0x207F, 0x00B2, 0x25A0, 0x00A0
+    ].freeze
+
+    # Kamenický (Czech) MS-DOS
+    CP895_TO_UNICODE = [
+      0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007,
+      0x0008, 0x0009, 0x000A, 0x000B, 0x000C, 0x000D, 0x000E, 0x000F,
+      0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017,
+      0x0018, 0x0019, 0x001A, 0x001B, 0x001C, 0x001D, 0x001E, 0x001F,
+      0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027,
+      0x0028, 0x0029, 0x002A, 0x002B, 0x002C, 0x002D, 0x002E, 0x002F,
+      0x0030, 0x0031, 0x0032, 0x0033, 0x0034, 0x0035, 0x0036, 0x0037,
+      0x0038, 0x0039, 0x003A, 0x003B, 0x003C, 0x003D, 0x003E, 0x003F,
+      0x0040, 0x0041, 0x0042, 0x0043, 0x0044, 0x0045, 0x0046, 0x0047,
+      0x0048, 0x0049, 0x004A, 0x004B, 0x004C, 0x004D, 0x004E, 0x004F,
+      0x0050, 0x0051, 0x0052, 0x0053, 0x0054, 0x0055, 0x0056, 0x0057,
+      0x0058, 0x0059, 0x005A, 0x005B, 0x005C, 0x005D, 0x005E, 0x005F,
+      0x0060, 0x0061, 0x0062, 0x0063, 0x0064, 0x0065, 0x0066, 0x0067,
+      0x0068, 0x0069, 0x006A, 0x006B, 0x006C, 0x006D, 0x006E, 0x006F,
+      0x0070, 0x0071, 0x0072, 0x0073, 0x0074, 0x0075, 0x0076, 0x0077,
+      0x0078, 0x0079, 0x007A, 0x007B, 0x007C, 0x007D, 0x007E, 0x007F,
+      0x010C, 0x00FC, 0x00E9, 0x010F, 0x00E4, 0x010E, 0x0164, 0x010D,
+      0x011B, 0x011A, 0x0139, 0x00CD, 0x013E, 0x01EA, 0x00C4, 0x00C1,
+      0x00C9, 0x017E, 0x017D, 0x00F4, 0x00F6, 0x00D3, 0x016F, 0x00DA,
+      0x00FD, 0x00D6, 0x00DC, 0x0160, 0x013D, 0x00DD, 0x0158, 0x0165,
+      0x00E1, 0x00ED, 0x00F3, 0x00FA, 0x0148, 0x0147, 0x016E, 0x00D4,
+      0x0161, 0x0159, 0x0155, 0x0154, 0x00BC, 0x00A7, 0x00AB, 0x00BB,
+      0x2591, 0x2592, 0x2593, 0x2502, 0x2524, 0x2561, 0x2562, 0x2556,
+      0x2555, 0x2563, 0x2551, 0x2557, 0x255D, 0x255C, 0x255B, 0x2510,
+      0x2514, 0x2534, 0x252C, 0x251C, 0x2500, 0x253C, 0x255E, 0x255F,
+      0x255A, 0x2554, 0x2569, 0x2566, 0x2560, 0x2550, 0x256C, 0x2567,
+      0x2568, 0x2564, 0x2565, 0x2559, 0x2558, 0x2552, 0x2553, 0x256B,
+      0x256A, 0x2518, 0x250C, 0x2588, 0x2584, 0x258C, 0x2590, 0x2580,
+      0x03B1, 0x00DF, 0x0393, 0x03C0, 0x03A3, 0x03C3, 0x00B5, 0x03C4,
+      0x03A6, 0x0398, 0x03A9, 0x03B4, 0x221E, 0x03C6, 0x03B5, 0x2229,
+      0x2261, 0x00B1, 0x2265, 0x2264, 0x2320, 0x2321, 0x00F7, 0x2248,
+      0x00B0, 0x2219, 0x00B7, 0x221A, 0x207F, 0x00B2, 0x25A0, 0x00A0
+    ].freeze
+
+    CUSTOM_TABLES = {
+      'cp620' => CP620_TO_UNICODE,
+      'cp895' => CP895_TO_UNICODE
+    }.freeze
+
+    # Is the encoding a code page without a Ruby Encoding, decoded through
+    # a vendored table?
+    #
+    # @param encoding [String, Encoding, nil]
+    # @return [Boolean]
+    def self.custom?(encoding)
+      encoding.is_a?(::String) && CUSTOM_TABLES.key?(encoding.downcase)
+    end
+
+    # Transcode a string from source_encoding to target_encoding
+    #
+    # @param string [String]
+    # @param source_encoding [String, Encoding]
+    # @param target_encoding [Encoding]
+    # @return [String]
+    def self.encode(string, source_encoding, target_encoding = Encoding.default_external)
+      table = source_encoding.is_a?(::String) && CUSTOM_TABLES[source_encoding.downcase]
+      if table
+        decoded = string.each_byte.map { |byte| table[byte] }.pack('U*')
+        target_encoding == Encoding::UTF_8 ? decoded : decoded.encode(target_encoding, undef: :replace, invalid: :replace)
+      else
+        string.force_encoding(source_encoding).encode(target_encoding, undef: :replace, invalid: :replace)
+      end
+    end
+  end
+end

data/lib/dbf/encodings.rb

@@ -5,6 +5,7 @@ module DBF
     '01' => 'cp437',       # U.S. MS-DOS
     '02' => 'cp850',       # International MS-DOS
     '03' => 'cp1252',      # Windows ANSI
+    '04' => 'macRoman',    # Standard Macintosh
     '08' => 'cp865',       # Danish OEM
     '09' => 'cp437',       # Dutch OEM
     '0a' => 'cp850',       # Dutch OEM*
@@ -45,6 +46,8 @@ module DBF
     '65' => 'cp866',       # Russian MS-DOS
     '66' => 'cp865',       # Nordic MS-DOS
     '67' => 'cp861',       # Icelandic MS-DOS
+    '68' => 'cp895',       # Kamenicky (Czech) MS-DOS
+    '69' => 'cp620',       # Mazovia (Polish) MS-DOS
     '6a' => 'cp737',       # Greek MS-DOS (437G)
     '6b' => 'cp857',       # Turkish MS-DOS
     '6c' => 'cp863',       # French-Canadian MS-DOS
@@ -53,9 +56,14 @@ module DBF
     '7a' => 'cp936',       # PRC GBK
     '7b' => 'cp932',       # Japanese Shift-JIS
     '7c' => 'cp874',       # Thai Windows/MS-DOS
+    '7d' => 'cp1255',      # Hebrew Windows
+    '7e' => 'cp1256',      # Arabic Windows
     '86' => 'cp737',       # Greek OEM
     '87' => 'cp852',       # Slovenian OEM
     '88' => 'cp857',       # Turkish OEM
+    '96' => 'macCyrillic', # Russian Macintosh
+    '97' => 'macCentEuro', # Macintosh EE
+    '98' => 'macGreek',    # Greek Macintosh
     'c8' => 'cp1250',      # Eastern European Windows
     'c9' => 'cp1251',      # Russian Windows
     'ca' => 'cp1254',      # Turkish Windows

data/lib/dbf/table.rb

@@ -140,9 +140,14 @@ module DBF
     # Dumps all records to a CSV file.  If no filename is given then CSV is
     # output to STDOUT.
     #
-    # @param [optional String] path Defaults to STDOUT
-    def to_csv(path = nil)
-      csv = CSV.new(path ? File.open(path, 'w') : $stdout, force_quotes: true)
+    # @param [optional String, IO] path_or_io String path, IO-like object, or nil for STDOUT
+    def to_csv(path_or_io = nil)
+      io = case path_or_io
+      when nil then $stdout
+      when String then File.open(path_or_io, 'w')
+      else path_or_io
+      end
+      csv = CSV.new(io, force_quotes: true)
       csv << column_names
       each { |record| csv << record.to_a }
     end
@@ -159,7 +164,7 @@ module DBF
     # @param [String] string
     # @return [String]
     def encode_string(string) # :nodoc:
-      string.force_encoding(@encoding).encode(Encoding.default_external, undef: :replace, invalid: :replace)
+      Encoder.encode(string, @encoding)
     end
 
     # Encoding specified in the file header

data/lib/dbf/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module DBF
-  VERSION = '5.2.0'
+  VERSION = '5.3.0'
 end

data/lib/dbf.rb

@@ -14,6 +14,7 @@ require 'dbf/record_context'
 require 'dbf/column_type'
 require 'dbf/column'
 require 'dbf/encodings'
+require 'dbf/encoder'
 require 'dbf/header'
 require 'dbf/version_config'
 require 'dbf/file_handler'
@@ -25,3 +26,4 @@ require 'dbf/memo/dbase3'
 require 'dbf/memo/dbase4'
 require 'dbf/memo/foxpro'
 require 'dbf/database/foxpro'
+require 'dbf/cli'

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: dbf
 version: !ruby/object:Gem::Version
-  version: 5.2.0
+  version: 5.3.0
 platform: ruby
 authors:
 - Keith Morrison
@@ -37,10 +37,12 @@ files:
 - bin/dbf
 - dbf.gemspec
 - lib/dbf.rb
+- lib/dbf/cli.rb
 - lib/dbf/column.rb
 - lib/dbf/column_builder.rb
 - lib/dbf/column_type.rb
 - lib/dbf/database/foxpro.rb
+- lib/dbf/encoder.rb
 - lib/dbf/encodings.rb
 - lib/dbf/file_handler.rb
 - lib/dbf/find.rb
@@ -77,7 +79,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.7.2
+rubygems_version: 4.0.10
 specification_version: 4
 summary: Read xBase files
 test_files: []