Increase speed of Table#to_csv

benchmark/table.yaml

@@ -0,0 +1,27 @@
+loop_count: 100
+contexts:
+  - gems:
+      csv: 3.3.0
+  - name: "master"
+    prelude: |
+      $LOAD_PATH.unshift(File.expand_path("lib"))
+      require "csv"
+prelude: |-
+  n_columns = Integer(ENV.fetch("N_COLUMNS", "5"), 10)
+  n_rows = Integer(ENV.fetch("N_ROWS", "100"), 10)
+  fields = ["AAAAA"] * n_columns
+  headers = n_columns.times.collect do |i|
+    "header#{i}"
+  end
+  row = CSV::Row.new(headers, fields)
+  rows = [row] * n_rows
+  table = CSV::Table.new(rows)
+  rows = [row] * n_rows * 10
+  large_table = CSV::Table.new(rows)
+benchmark:
+  "to_csv: no encoding": |-
+    table.to_csv
+  "to_csv: encoding": |-
+    table.to_csv(encoding: 'UTF-8')
+  "to_csv: encoding - 10 x rows": |-
+    large_table.to_csv(encoding: 'UTF-8')

lib/csv.rb

@@ -1506,17 +1506,7 @@ def generate_line(row, **options)
       if options[:encoding]
         str.force_encoding(options[:encoding])
       else
-        fallback_encoding = nil
-        output_encoding = nil
-        row.each do |field|
-          next unless field.is_a?(String)
-          fallback_encoding ||= field.encoding
-          next if field.ascii_only?
-          output_encoding = field.encoding
-          break
-        end
-        output_encoding ||= fallback_encoding
-        if output_encoding
+        if output_encoding = row_encoding(row)
           str.force_encoding(output_encoding)
         end
       end
@@ -1960,6 +1950,20 @@ def table(path, **options)
     private_constant :ON_WINDOWS
 
     private
+
+    def row_encoding(row)
+      fallback_encoding = nil
+      output_encoding = nil
+      row.each do |field|
+        next unless field.is_a?(String)
+        fallback_encoding ||= field.encoding
+        next if field.ascii_only?
+        output_encoding = field.encoding
+        break
+      end
+      output_encoding || fallback_encoding
+    end
+
     def may_enable_bom_detection_automatically(filename_or_io,
                                                mode,
                                                options,

lib/csv/table.rb

@@ -1006,10 +1006,15 @@ def to_csv(write_headers: true, limit: nil, **options)
       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
 
+      if options[:encoding]
+        rows = @table.first(limit).select { |row| !row.header_row? }
+        array.push(CSV.generate_lines(rows, **options))
+      else
+        @table.first(limit).each do |row|
+          array.push(row.fields.to_csv(**options)) unless row.header_row?
+        end
+      end
       array.join("")
     end
     alias_method :to_s, :to_csv

test/csv/test_table.rb

@@ -373,6 +373,15 @@ def test_to_csv_limit_negative_over
     CSV
   end
 
+  def test_to_csv_encoding
+    rows  = [ CSV::Row.new(%w{A}, ["\x00\xac".force_encoding("ASCII-8BIT")]),
+              CSV::Row.new(%w{A}, ["\x00\xac"]) ]
+    table = CSV::Table.new(rows)
+
+    assert_equal(Encoding::UTF_8, table.to_csv(encoding: 'UTF-8').encoding)
+    assert_raises(Encoding::CompatibilityError) {table.to_csv}
+  end
+
   def test_append
     # verify that we can chain the call
     assert_equal(@table, @table << [10, 11, 12])