Fix a parse bug when split character exists in middle of column value

kou · kou · commit 398b3564c5b0 · 2019-12-25T07:01:48.000+09:00
GitHub: fix #115 Reported by TOMITA Masahiro. Thanks!!!
diff --git a/lib/csv/parser.rb b/lib/csv/parser.rb
@@ -446,6 +446,7 @@ def prepare_strip
       @strip = @options[:strip]
       @escaped_strip = nil
       @strip_value = nil
+      @rstrip_value = nil
       if @strip.is_a?(String)
         case @strip.length
         when 0
@@ -460,13 +461,16 @@ def prepare_strip
         if @quote_character
           @strip_value = Regexp.new(@escaped_strip +
                                     "+".encode(@encoding))
+          @rstrip_value = Regexp.new(@escaped_strip +
+                                     "+\\z".encode(@encoding))
         end
         @need_robust_parsing = true
       elsif @strip
         strip_values = " \t\f\v"
         @escaped_strip = strip_values.encode(@encoding)
         if @quote_character
           @strip_value = Regexp.new("[#{strip_values}]+".encode(@encoding))
+          @rstrip_value = Regexp.new("[#{strip_values}]+\\z".encode(@encoding))
         end
         @need_robust_parsing = true
       end
@@ -561,9 +565,6 @@ def prepare_unquoted
       unless @liberal_parsing
         no_unquoted_values << @escaped_quote_character
       end
-      if @escaped_strip
-        no_unquoted_values << @escaped_strip
-      end
       @unquoted_value = Regexp.new("[^".encode(@encoding) +
                                    no_unquoted_values +
                                    "]+".encode(@encoding))
@@ -939,6 +940,7 @@ def parse_column_value
       if @liberal_parsing
         quoted_value = parse_quoted_column_value
         if quoted_value
+          @scanner.scan_all(@strip_value) if @strip_value
           unquoted_value = parse_unquoted_column_value
           if unquoted_value
             if @double_quote_outside_quote
@@ -986,6 +988,9 @@ def parse_unquoted_column_value
         end
       end
       value.gsub!(@backslash_quote_character, @quote_character) if @backslash_quote
+      if @rstrip_value
+        value.gsub!(@rstrip_value, "")
+      end
       value
     end
 
diff --git a/test/csv/parse/test_strip.rb b/test/csv/parse/test_strip.rb
@@ -21,6 +21,11 @@ def test_right
                  CSV.parse_line(%Q{a  ,b  }, strip: true))
   end
 
+  def test_middle
+    assert_equal(["a b"],
+                 CSV.parse_line(%Q{a b}, strip: true))
+  end
+
   def test_quoted
     assert_equal(["  a  ", "  b  "],
                  CSV.parse_line(%Q{"  a  ","  b  "}, strip: true))
