dbf 5.1.0 → 5.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ea8e27ddd0d6d12797336f59d245bc8b62245e59c8fad0ecbf12c97513ac244f
-  data.tar.gz: d3cc1955ebb54357b5c724ef1b8687a8f69a47f794c445a08840d5dc08fa5861
+  metadata.gz: 96b3352f6c6d2cc3a667c6f390021c4f99be7c83ff2b9c911ac38c75d4c261aa
+  data.tar.gz: b787cd0e14a34ea66e670f88e793bcbebc1ccbdc260721da08ccee5d3de2624a
 SHA512:
-  metadata.gz: 87ec87d1e3a45f4ec1f787f4e56d64aad00779aebe5e391656fe457f278219ffa72de97cc4ae679d2b8a14c7f03035fb059397251a290bc21587513382207c82
-  data.tar.gz: a931c1aa3ab7a5128824cb43bd4ec03217a52d548a071209495dce1647f68f269c78f84c15534ecf35ed9badac227e96a46ab53e0f33a9271d440d832fd26959
+  metadata.gz: 52ceab2c71a3093ff691be23a0a0590e4309fd88218e683032e12805585dea05dd97bebd0170f003e7678641810d5d9aa6b9731bb0af647a8508d015875312d2
+  data.tar.gz: 41b87283de2deea377a3c27436d621486af57bc0421e5bc84dc1bd1f304abf8d264d72d8dda80350666895ca604efd335b829fd1efc79720721027146fe56b7d

data/CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 5.2.0
+
+- Drop support for Ruby 3.1 and 3.2
+
+## 5.1.1
+
+- Frozen string literals
+
 ## 5.1.0
 
 - Drop support for Ruby 3.0.x

data/README.md

@@ -16,6 +16,8 @@ DBF is a small, fast Ruby library for reading dBase, xBase, Clipper, and FoxPro
   subject line
 * Change log: <https://github.com/infused/dbf/blob/main/CHANGELOG.md>
 
+NOTE: Beginning with version 5.2 we have dropped support for Ruby 3.2 and earlier.
+
 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,
@@ -28,7 +30,7 @@ please use 2.0.x (<https://github.com/infused/dbf/tree/2_stable>)
 
 DBF is tested to work with the following versions of Ruby:
 
-* Ruby 3.1.x, 3.2.x, 3.3.x
+* Ruby 3.3.x, 3.4.x, 4.0.x
 
 ## Installation
 
@@ -251,7 +253,7 @@ class Book < Sequel::Model; end
 Sequel.migration do
   up do
     table = DBF::Table.new('db/dbf/books.dbf')
-    eval(table.schema(:sequel, true)) # passing true to limit output to create_table() only
+    eval(table.schema(:sequel, table_only: true)) # limit output to create_table() only
 
     Book.reset_column_information
     table.each do |record|

data/dbf.gemspec

@@ -1,24 +1,21 @@
-lib = File.expand_path('lib', __dir__)
-$LOAD_PATH.unshift lib unless $LOAD_PATH.include?(lib)
-require 'dbf/version'
+require_relative 'lib/dbf/version'
 
 Gem::Specification.new do |s|
   s.name = 'dbf'
   s.version = DBF::VERSION
   s.authors = ['Keith Morrison']
   s.email = 'keithm@infused.org'
-  s.homepage = 'http://github.com/infused/dbf'
+  s.homepage = 'https://github.com/infused/dbf'
   s.summary = 'Read xBase files'
   s.description = 'A small fast library for reading dBase, xBase, Clipper and FoxPro database files.'
   s.license = 'MIT'
   s.bindir = 'bin'
   s.executables = ['dbf']
-  s.rdoc_options = ['--charset=UTF-8']
-  s.extra_rdoc_files = ['README.md', 'CHANGELOG.md', 'LICENSE']
-  s.files = Dir['README.md', 'CHANGELOG.md', 'LICENSE', '{bin,lib,spec}/**/*', 'dbf.gemspec']
+  s.files = Dir['README.md', 'CHANGELOG.md', 'LICENSE', '{bin,lib}/**/*', 'dbf.gemspec']
   s.require_paths = ['lib']
-  s.required_rubygems_version = Gem::Requirement.new('>= 3.2.3')
-  s.required_ruby_version = Gem::Requirement.new('>= 3.1.0')
+  s.required_ruby_version = '>= 3.3.0'
   s.metadata['rubygems_mfa_required'] = 'true'
+  s.metadata['source_code_uri'] = 'https://github.com/infused/dbf'
+  s.metadata['changelog_uri'] = 'https://github.com/infused/dbf/blob/main/CHANGELOG.md'
   s.add_dependency 'csv'
 end

data/lib/dbf/column.rb

@@ -1,16 +1,14 @@
+# frozen_string_literal: true
+
 module DBF
   class Column
-    extend Forwardable
-
     class LengthError < StandardError
     end
 
     class NameError < StandardError
     end
 
-    attr_reader :table, :name, :type, :length, :decimal, :encoding
-
-    def_delegator :type_cast_class, :type_cast
+    attr_reader :name, :type, :length, :decimal
 
     # rubocop:disable Style/MutableConstant
     TYPE_CAST_CLASS = {
@@ -24,7 +22,7 @@ module DBF
       M: ColumnType::Memo,
       B: ColumnType::Double,
       G: ColumnType::General,
-      :+ => ColumnType::SignedLong2
+      :+ => ColumnType::AutoIncrement
     }
     # rubocop:enable Style/MutableConstant
     TYPE_CAST_CLASS.default = ColumnType::String
@@ -38,31 +36,37 @@ module DBF
     # @param length [Integer]
     # @param decimal [Integer]
     def initialize(table, name, type, length, decimal)
-      @encoding = table.encoding
-
       @table = table
       @name = clean(name)
       @type = type
       @length = length
       @decimal = decimal
-      @version = table.version
 
       validate_length
       validate_name
     end
 
-    # Returns true if the column is a memo
+    def encoding = @table.encoding
+
+    # @param value [String]
+    def type_cast(value)
+      type_cast_class.type_cast(value)
+    end
+
+    # Decodes a raw column value, handling memo, blank, and type cast cases
     #
-    # @return [Boolean]
-    def memo?
-      @memo ||= type == 'M'
+    # @param raw [String]
+    # @yield [raw] for memo column resolution
+    # @return decoded value
+    def decode(raw, &memo_handler)
+      type_cast_class.decode(raw, &memo_handler)
     end
 
     # Returns a Hash with :name, :type, :length, and :decimal keys
     #
     # @return [Hash]
     def to_hash
-      {name: name, type: type, length: length, decimal: decimal}
+      {name:, type:, length:, decimal:}
     end
 
     # Underscored name
@@ -78,7 +82,7 @@ module DBF
     private
 
     def clean(value) # :nodoc:
-      table.encode_string(value.strip.partition("\x00").first)
+      @table.encode_string(value.strip.split("\x00", 2).first || +'')
     end
 
     def type_cast_class # :nodoc:

data/lib/dbf/column_builder.rb

@@ -0,0 +1,31 @@
+# frozen_string_literal: true
+
+module DBF
+  class ColumnBuilder
+    def initialize(table, data, version_config)
+      @table = table
+      @data = data
+      @version_config = version_config
+    end
+
+    def build
+      safe_seek do
+        @data.seek(@version_config.header_size)
+        [].tap do |columns|
+          columns << Column.new(*@version_config.read_column_args(@table, @data)) until end_of_record?
+        end
+      end
+    end
+
+    private
+
+    def end_of_record?
+      safe_seek { @data.read(1).ord == 13 }
+    end
+
+    def safe_seek
+      original_pos = @data.pos
+      yield.tap { @data.seek(original_pos) }
+    end
+  end
+end

data/lib/dbf/column_type.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module DBF
   module ColumnType
     class Base
@@ -9,6 +11,22 @@ module DBF
         @decimal = column.decimal
         @encoding = column.encoding
       end
+
+      def blank_value
+        nil
+      end
+
+      def skip_blank?
+        false
+      end
+
+      def decode(raw, &_memo_handler)
+        if skip_blank? && raw.count(' ') == raw.length
+          blank_value
+        else
+          type_cast(raw)
+        end
+      end
     end
 
     class Nil < Base
@@ -19,11 +37,13 @@ module DBF
     end
 
     class Number < Base
+      def skip_blank? = true
+
       # @param value [String]
       def type_cast(value)
-        return nil if value.strip.empty?
+        return nil if value.empty?
 
-        @decimal.zero? ? value.to_i : value.to_f
+        decimal.zero? ? value.to_i : value.to_f
       end
     end
 
@@ -41,12 +61,12 @@ module DBF
       end
     end
 
-    class SignedLong2 < Base
+    class AutoIncrement < Base
       # @param value [String]
       def type_cast(value)
-        s = value.unpack1('B*')
-        sign_multiplier = s[0] == '0' ? -1 : 1
-        s[1, 31].to_i(2) * sign_multiplier
+        bits = value.unpack1('B*')
+        sign_multiplier = bits[0] == '0' ? -1 : 1
+        bits[1, 31].to_i(2) * sign_multiplier
       end
     end
 
@@ -65,13 +85,20 @@ module DBF
     end
 
     class Boolean < Base
+      def skip_blank? = true
+      def blank_value = false
+
       # @param value [String]
       def type_cast(value)
-        value.strip.match?(/^(y|t)$/i)
+        byte = value.getbyte(0)
+        byte == 89 || byte == 121 || byte == 84 || byte == 116 # Y y T t
       end
     end
 
     class Date < Base
+      def skip_blank? = true
+      def blank_value = false
+
       # @param value [String]
       def type_cast(value)
         value.match?(/\d{8}/) && ::Date.strptime(value, '%Y%m%d')
@@ -92,28 +119,47 @@ module DBF
     end
 
     class Memo < Base
+      def decode(raw, &memo_handler)
+        memo_content = memo_handler.call(raw)
+        memo_content ? type_cast(memo_content) : nil
+      end
+
       # @param value [String]
       def type_cast(value)
-        if encoding && !value.nil?
-          value.force_encoding(@encoding).encode(Encoding.default_external, undef: :replace, invalid: :replace)
-        else
-          value
-        end
+        return value unless encoding && value
+
+        value.dup.force_encoding(encoding).encode(Encoding.default_external, undef: :replace, invalid: :replace)
       end
     end
 
     class General < Base
       # @param value [String]
       def type_cast(value)
-        value
+        value&.dup&.force_encoding(Encoding::ASCII_8BIT)
       end
     end
 
     class String < Base
+      def initialize(column)
+        super
+        @target_encoding = Encoding.default_external
+        @needs_encode = encoding && encoding != @target_encoding
+      end
+
+      def skip_blank? = true
+      def blank_value = ''
+
       # @param value [String]
       def type_cast(value)
-        value = value.strip
-        @encoding ? value.force_encoding(@encoding).encode(Encoding.default_external, undef: :replace, invalid: :replace) : value
+        value.strip!
+        encoding ? encode(value) : value
+      end
+
+      private
+
+      def encode(value)
+        value.force_encoding(encoding)
+        @needs_encode ? value.encode(@target_encoding, undef: :replace, invalid: :replace) : value
       end
     end
   end

data/lib/dbf/database/foxpro.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module DBF
   # DBF::Database::Foxpro is the primary interface to a Visual Foxpro database
   # container (.dbc file). When using this database container, long fieldnames
@@ -24,7 +26,7 @@ module DBF
         @db = DBF::Table.new(@path)
         @tables = extract_dbc_data
       rescue Errno::ENOENT
-        raise DBF::FileNotFoundError, "file not found: #{data}"
+        raise DBF::FileNotFoundError, "file not found: #{path}"
       end
 
       def table_names
@@ -36,9 +38,7 @@ module DBF
       # @param name [String]
       # @return [DBF::Table]
       def table(name)
-        Table.new table_path(name) do |table|
-          table.long_names = @tables[name]
-        end
+        Table.new(table_path(name), long_names: @tables[name])
       end
 
       # Searches the database directory for the table's dbf file
@@ -56,7 +56,8 @@ module DBF
       end
 
       def method_missing(method, *args) # :nodoc:
-        table_names.index(method.to_s) ? table(method.to_s) : super
+        name = method.to_s
+        table_names.index(name) ? table(name) : super
       end
 
       def respond_to_missing?(method, *)
@@ -70,45 +71,33 @@ module DBF
       # are in the same order as in the linked tables but only the long name
       # is provided.
       def extract_dbc_data # :nodoc:
-        data = {}
-        @db.each do |record|
+        build_table_data.values.to_h { |entry| entry.values_at(:name, :fields) }
+      end
+
+      def build_table_data # :nodoc:
+        @db.each_with_object({}) do |record, hash|
           next unless record
 
+          name = record.objectname
           case record.objecttype
-          when 'Table'
-            # This is a related table
-            process_table record, data
-          when 'Field'
-            # This is a related field. The parentid points to the table object.
-            # Create using the parentid if the parentid is still unknown.
-            process_field record, data
+          when 'Table' then hash[record.objectid] = table_field_hash(name)
+          when 'Field' then (hash[record.parentid] ||= table_field_hash('UNKNOWN'))[:fields] << name
           end
         end
-
-        data.values.to_h { |v| [v[:name], v[:fields]] }
-      end
-
-      def process_table(record, data)
-        id = record.objectid
-        name = record.objectname
-        data[id] = table_field_hash(name)
-      end
-
-      def process_field(record, data)
-        id = record.parentid
-        name = 'UNKNOWN'
-        field = record.objectname
-        data[id] ||= table_field_hash(name)
-        data[id][:fields] << field
       end
 
       def table_field_hash(name)
-        {name: name, fields: []}
+        {name:, fields: []}
       end
     end
 
     class Table < DBF::Table
-      attr_accessor :long_names
+      attr_reader :long_names
+
+      def initialize(path, long_names:)
+        @long_names = long_names
+        super(path)
+      end
 
       def build_columns # :nodoc:
         columns = super

data/lib/dbf/encodings.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module DBF
   ENCODINGS = {
     '01' => 'cp437',       # U.S. MS-DOS

data/lib/dbf/file_handler.rb

@@ -0,0 +1,36 @@
+# frozen_string_literal: true
+
+module DBF
+  module FileHandler
+    module_function
+
+    def open_data(data)
+      case data
+      when StringIO
+        data
+      when String
+        File.open(data, 'rb')
+      else
+        raise ArgumentError, 'data must be a file path or StringIO object'
+      end
+    rescue Errno::ENOENT
+      raise DBF::FileNotFoundError, "file not found: #{data}"
+    end
+
+    def open_memo(data, memo, memo_class, version)
+      if memo
+        meth = memo.is_a?(StringIO) ? :new : :open
+        memo_class.send(meth, memo, version)
+      elsif !data.is_a?(StringIO)
+        path = Dir.glob(memo_search_path(data)).first
+        path && memo_class.open(path, version)
+      end
+    end
+
+    def memo_search_path(io)
+      dirname = File.dirname(io)
+      basename = File.basename(io, '.*')
+      "#{dirname}/#{basename}*.{fpt,FPT,dbt,DBT}"
+    end
+  end
+end

data/lib/dbf/find.rb

@@ -0,0 +1,54 @@
+# frozen_string_literal: true
+
+module DBF
+  # The Find module provides methods for searching and retrieving
+  # records using a simple ActiveRecord-like syntax.
+  #
+  # Examples:
+  #   table = DBF::Table.new 'mydata.dbf'
+  #
+  #   # Find record number 5
+  #   table.find(5)
+  #
+  #   # Find all records for Keith Morrison
+  #   table.find :all, first_name: "Keith", last_name: "Morrison"
+  #
+  #   # Find first record
+  #   table.find :first, first_name: "Keith"
+  #
+  # The <b>command</b> may be a record index, :all, or :first.
+  # <b>options</b> is optional and, if specified, should be a hash where the
+  # keys correspond to column names in the database.  The values will be
+  # matched exactly with the value in the database.  If you specify more
+  # than one key, all values must match in order for the record to be
+  # returned.  The equivalent SQL would be "WHERE key1 = 'value1'
+  # AND key2 = 'value2'".
+  module Find
+    # @param command [Integer, Symbol] command
+    # @param options [optional, Hash] options Hash of search parameters
+    # @yield [optional, DBF::Record, NilClass]
+    def find(command, options = {}, &)
+      case command
+      when Integer then record(command)
+      when Array then command.map { |index| record(index) }
+      when :all then find_all_records(options, &)
+      when :first then find_first_record(options)
+      end
+    end
+
+    private
+
+    def find_all_records(options)
+      select do |record|
+        next unless record&.match?(options)
+
+        yield record if block_given?
+        record
+      end
+    end
+
+    def find_first_record(options)
+      detect { |record| record&.match?(options) }
+    end
+  end
+end

data/lib/dbf/header.rb

@@ -1,21 +1,20 @@
+# frozen_string_literal: true
+
 module DBF
   class Header
     attr_reader :version, :record_count, :header_length, :record_length, :encoding_key, :encoding
 
     def initialize(data)
-      @data = data
-      unpack_header
-    end
-
-    def unpack_header
-      @version = @data.unpack1('H2')
+      @version = data.unpack1('H2')
+      @encoding_key = nil
+      @encoding = nil
 
       case @version
       when '02'
-        @record_count, @record_length = @data.unpack('x v x3 v')
+        @record_count, @record_length = data.unpack('x v x3 v')
         @header_length = 521
       else
-        @record_count, @header_length, @record_length, @encoding_key = @data.unpack('x x3 V v2 x17 H2')
+        @record_count, @header_length, @record_length, @encoding_key = data.unpack('x x3 V v2 x17 H2')
         @encoding = DBF::ENCODINGS[@encoding_key]
       end
     end

data/lib/dbf/memo/base.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module DBF
   module Memo
     class Base
@@ -29,6 +31,8 @@ module DBF
 
       private
 
+      attr_reader :data
+
       def offset(start_block) # :nodoc:
         start_block * block_size
       end

data/lib/dbf/memo/dbase3.rb

@@ -1,11 +1,13 @@
+# frozen_string_literal: true
+
 module DBF
   module Memo
     class Dbase3 < Base
       def build_memo(start_block) # :nodoc:
-        @data.seek offset(start_block)
-        memo_string = ''
+        data.seek offset(start_block)
+        memo_string = +''
         loop do
-          block = @data.read(BLOCK_SIZE).gsub(/(\000|\032)/, '')
+          block = data.read(BLOCK_SIZE).gsub(/(\000|\032)/, '')
           memo_string << block
           break if block.size < BLOCK_SIZE
         end

data/lib/dbf/memo/dbase4.rb

@@ -1,9 +1,11 @@
+# frozen_string_literal: true
+
 module DBF
   module Memo
     class Dbase4 < Base
       def build_memo(start_block) # :nodoc:
-        @data.seek offset(start_block)
-        @data.read(@data.read(BLOCK_HEADER_SIZE).unpack1('x4L'))
+        data.seek offset(start_block)
+        data.read(data.read(BLOCK_HEADER_SIZE).unpack1('x4L'))
       end
     end
   end

data/lib/dbf/memo/foxpro.rb

@@ -1,26 +1,35 @@
+# frozen_string_literal: true
+
 module DBF
   module Memo
     class Foxpro < Base
       FPT_HEADER_SIZE = 512
 
+      def initialize(data, version)
+        @data = data
+        super
+      end
+
       def build_memo(start_block) # :nodoc:
         @data.seek offset(start_block)
-
         memo_type, memo_size, memo_string = @data.read(block_size).unpack('NNa*')
         return nil unless memo_type == 1 && memo_size > 0
 
-        if memo_size > block_content_size
-          memo_string << @data.read(content_size(memo_size))
-        else
-          memo_string = memo_string[0, memo_size]
-        end
-        memo_string
+        read_memo_content(memo_string, memo_size)
       rescue StandardError
         nil
       end
 
       private
 
+      def read_memo_content(memo_string, memo_size) # :nodoc:
+        if memo_size > block_content_size
+          memo_string << @data.read(content_size(memo_size))
+        else
+          memo_string[0, memo_size]
+        end
+      end
+
       def block_size # :nodoc:
         @block_size ||= begin
           @data.rewind