avoid using StringScanner eos

fix lexer parsing ID 'contains' as comparison

lib/liquid/block_body.rb

@@ -246,10 +246,17 @@ module Liquid
     end
 
     def create_variable(token, parse_context)
-      if token =~ ContentOfVariable
-        markup = Regexp.last_match(1)
+      if token.end_with?("}}")
+        i = 2
+        i = 3 if token[i] == "-"
+        parse_end = token.length - 3
+        parse_end -= 1 if token[parse_end] == "-"
+        markup_end = parse_end - i + 1
+        markup = markup_end <= 0 ? "" : token.slice(i, markup_end)
+
         return Variable.new(markup, parse_context)
       end
+
       BlockBody.raise_missing_variable_terminator(token, parse_context)
     end
 

lib/liquid/lexer.rb

@@ -73,7 +73,6 @@ module Liquid
     COMPARISON_LESS_THAN = [:comparison, "<"].freeze
     COMPARISON_LESS_THAN_OR_EQUAL = [:comparison, "<="].freeze
     COMPARISON_NOT_EQUAL_ALT = [:comparison, "<>"].freeze
-    CONTAINS = /contains(?=\s)/
     DASH = [:dash, "-"].freeze
     DOT = [:dot, "."].freeze
     DOTDOT = [:dotdot, ".."].freeze
@@ -90,7 +89,12 @@ module Liquid
     SINGLE_STRING_LITERAL = /'[^\']*'/
     WHITESPACE_OR_NOTHING = /\s*/
 
-    COMPARISON_JUMP_TABLE = [].tap do |table|
+    SINGLE_COMPARISON_TOKENS = [].tap do |table|
+      table["<".ord] = COMPARISON_LESS_THAN
+      table[">".ord] = COMPARISON_GREATER_THAN
+    end
+
+    TWO_CHARS_COMPARISON_JUMP_TABLE = [].tap do |table|
       table["=".ord] = [].tap do |sub_table|
         sub_table["=".ord] = COMPARISON_EQUAL
         sub_table.freeze
@@ -99,6 +103,9 @@ module Liquid
         sub_table["=".ord] = COMPARISION_NOT_EQUAL
         sub_table.freeze
       end
+    end
+
+    COMPARISON_JUMP_TABLE = [].tap do |table|
       table["<".ord] = [].tap do |sub_table|
         sub_table["=".ord] = COMPARISON_LESS_THAN_OR_EQUAL
         sub_table[">".ord] = COMPARISON_NOT_EQUAL_ALT
@@ -163,6 +170,7 @@ module Liquid
 
         break if @ss.eos?
 
+        start_pos = @ss.pos
         peeked = @ss.peek_byte
 
         if (special = SPECIAL_TABLE[peeked])
@@ -173,7 +181,7 @@ module Liquid
             @output << DOTDOT
           elsif special == DASH
             # Special case for negative numbers
-            if NUMBER_TABLE[@ss.peek_byte]
+            if (peeked_byte = @ss.peek_byte) && NUMBER_TABLE[peeked_byte]
               @ss.pos -= 1
               @output << [:number, @ss.scan(NUMBER_LITERAL)]
             else
@@ -182,26 +190,34 @@ module Liquid
           else
             @output << special
           end
-        elsif (sub_table = COMPARISON_JUMP_TABLE[peeked])
+        elsif (sub_table = TWO_CHARS_COMPARISON_JUMP_TABLE[peeked])
           @ss.scan_byte
-          if (found = sub_table[@ss.peek_byte])
+          if (peeked_byte = @ss.peek_byte) && (found = sub_table[peeked_byte])
             @output << found
             @ss.scan_byte
           else
-            raise SyntaxError, "Unexpected character #{peeked.chr}"
+            raise_syntax_error(start_pos)
+          end
+        elsif (sub_table = COMPARISON_JUMP_TABLE[peeked])
+          @ss.scan_byte
+          if (peeked_byte = @ss.peek_byte) && (found = sub_table[peeked_byte])
+            @output << found
+            @ss.scan_byte
+          else
+            @output << SINGLE_COMPARISON_TOKENS[peeked]
           end
         else
           type, pattern = NEXT_MATCHER_JUMP_TABLE[peeked]
 
           if type && (t = @ss.scan(pattern))
             # Special case for "contains"
-            @output << if type == :id && t == "contains"
+            @output << if type == :id && t == "contains" && @output.last&.first != :dot
               COMPARISON_CONTAINS
             else
               [type, t]
             end
           else
-            raise SyntaxError, "Unexpected character #{peeked.chr}"
+            raise_syntax_error(start_pos)
           end
         end
       end
@@ -209,6 +225,12 @@ module Liquid
 
       @output << EOS
     end
+
+    def raise_syntax_error(start_pos)
+      @ss.pos = start_pos
+      # the character could be a UTF-8 character, use getch to get all the bytes
+      raise SyntaxError, "Unexpected character #{@ss.getch}"
+    end
   end
 
   Lexer = StringScanner.instance_methods.include?(:scan_byte) ? Lexer2 : Lexer1

lib/liquid/standardfilters.rb

@@ -877,7 +877,7 @@ module Liquid
     #   - [`nil`](/docs/api/liquid/basics#nil)
     # @liquid_syntax variable | default: variable
     # @liquid_return [untyped]
-    # @liquid_optional_param allow_false [boolean] Whether to use false values instead of the default.
+    # @liquid_optional_param allow_false: [boolean] Whether to use false values instead of the default.
     def default(input, default_value = '', options = {})
       options = {} unless options.is_a?(Hash)
       false_check = options['allow_false'] ? input.nil? : !Liquid::Utils.to_liquid_value(input)

test/integration/parsing_quirks_test.rb

@@ -131,4 +131,16 @@ class ParsingQuirksTest < Minitest::Test
   def test_contains_in_id
     assert_template_result(' YES ', '{% if containsallshipments == true %} YES {% endif %}', { 'containsallshipments' => true })
   end
+
+  def test_incomplete_expression
+    with_error_mode(:lax) do
+      assert_template_result("false", "{% liquid assign foo = false -\n%}{{ foo }}")
+      assert_template_result("false", "{% liquid assign foo = false >\n%}{{ foo }}")
+      assert_template_result("false", "{% liquid assign foo = false <\n%}{{ foo }}")
+      assert_template_result("false", "{% liquid assign foo = false =\n%}{{ foo }}")
+      assert_template_result("false", "{% liquid assign foo = false !\n%}{{ foo }}")
+      assert_template_result("false", "{% liquid assign foo = false 1\n%}{{ foo }}")
+      assert_template_result("false", "{% liquid assign foo = false a\n%}{{ foo }}")
+    end
+  end
 end # ParsingQuirksTest

test/unit/block_unit_test.rb

@@ -32,6 +32,12 @@ class BlockUnitTest < Minitest::Test
     assert_equal(String, template.root.nodelist[2].class)
   end
 
+  def test_variable_with_multibyte_character
+    template = Liquid::Template.parse("{{ '❤️' }}")
+    assert_equal(1, template.root.nodelist.size)
+    assert_equal(Variable, template.root.nodelist[0].class)
+  end
+
   def test_variable_many_embedded_fragments
     template = Liquid::Template.parse("  {{funk}} {{so}} {{brother}} ")
     assert_equal(7, template.root.nodelist.size)

test/unit/lexer_unit_test.rb

@@ -6,58 +6,134 @@ class LexerUnitTest < Minitest::Test
   include Liquid
 
   def test_strings
-    tokens = Lexer.new(%( 'this is a test""' "wat 'lol'")).tokenize
-    assert_equal([[:string, %('this is a test""')], [:string, %("wat 'lol'")], [:end_of_string]], tokens)
+    assert_equal(
+      [[:string, %('this is a test""')], [:string, %("wat 'lol'")], [:end_of_string]],
+      tokenize(%( 'this is a test""' "wat 'lol'")),
+    )
   end
 
   def test_integer
-    tokens = Lexer.new('hi 50').tokenize
-    assert_equal([[:id, 'hi'], [:number, '50'], [:end_of_string]], tokens)
+    assert_equal(
+      [[:id, 'hi'], [:number, '50'], [:end_of_string]],
+      tokenize('hi 50'),
+    )
   end
 
   def test_float
-    tokens = Lexer.new('hi 5.0').tokenize
-    assert_equal([[:id, 'hi'], [:number, '5.0'], [:end_of_string]], tokens)
+    assert_equal(
+      [[:id, 'hi'], [:number, '5.0'], [:end_of_string]],
+      tokenize('hi 5.0'),
+    )
   end
 
   def test_comparison
-    tokens = Lexer.new('== <> contains ').tokenize
-    assert_equal([[:comparison, '=='], [:comparison, '<>'], [:comparison, 'contains'], [:end_of_string]], tokens)
+    assert_equal(
+      [[:comparison, '=='], [:comparison, '<>'], [:comparison, 'contains'], [:end_of_string]],
+      tokenize('== <> contains '),
+    )
+  end
+
+  def test_comparison_without_whitespace
+    assert_equal(
+      [[:number, '1'], [:comparison, '>'], [:number, '0'], [:end_of_string]],
+      tokenize('1>0'),
+    )
+  end
+
+  def test_comparison_with_negative_number
+    assert_equal(
+      [[:number, '1'], [:comparison, '>'], [:number, '-1'], [:end_of_string]],
+      tokenize('1>-1'),
+    )
+  end
+
+  def test_raise_for_invalid_comparison
+    assert_raises(SyntaxError) do
+      tokenize('1>!1')
+    end
+
+    assert_raises(SyntaxError) do
+      tokenize('1=<1')
+    end
+
+    assert_raises(SyntaxError) do
+      tokenize('1!!1')
+    end
   end
 
   def test_specials
-    tokens = Lexer.new('| .:').tokenize
-    assert_equal([[:pipe, '|'], [:dot, '.'], [:colon, ':'], [:end_of_string]], tokens)
-    tokens = Lexer.new('[,]').tokenize
-    assert_equal([[:open_square, '['], [:comma, ','], [:close_square, ']'], [:end_of_string]], tokens)
+    assert_equal(
+      [[:pipe, '|'], [:dot, '.'], [:colon, ':'], [:end_of_string]],
+      tokenize('| .:'),
+    )
+
+    assert_equal(
+      [[:open_square, '['], [:comma, ','], [:close_square, ']'], [:end_of_string]],
+      tokenize('[,]'),
+    )
   end
 
   def test_fancy_identifiers
-    tokens = Lexer.new('hi five?').tokenize
-    assert_equal([[:id, 'hi'], [:id, 'five?'], [:end_of_string]], tokens)
+    assert_equal([[:id, 'hi'], [:id, 'five?'], [:end_of_string]], tokenize('hi five?'))
 
-    tokens = Lexer.new('2foo').tokenize
-    assert_equal([[:number, '2'], [:id, 'foo'], [:end_of_string]], tokens)
+    assert_equal([[:number, '2'], [:id, 'foo'], [:end_of_string]], tokenize('2foo'))
   end
 
   def test_whitespace
-    tokens = Lexer.new("five|\n\t ==").tokenize
-    assert_equal([[:id, 'five'], [:pipe, '|'], [:comparison, '=='], [:end_of_string]], tokens)
+    assert_equal(
+      [[:id, 'five'], [:pipe, '|'], [:comparison, '=='], [:end_of_string]],
+      tokenize("five|\n\t =="),
+    )
   end
 
   def test_unexpected_character
     assert_raises(SyntaxError) do
-      Lexer.new("%").tokenize
+      tokenize("%")
     end
   end
 
   def test_negative_numbers
-    tokens = Lexer.new("foo | default: -1").tokenize
-    assert_equal([[:id, 'foo'], [:pipe, '|'], [:id, 'default'], [:colon, ":"], [:number, '-1'], [:end_of_string]], tokens)
+    assert_equal(
+      [[:id, 'foo'], [:pipe, '|'], [:id, 'default'], [:colon, ":"], [:number, '-1'], [:end_of_string]],
+      tokenize("foo | default: -1"),
+    )
   end
 
   def test_greater_than_two_digits
-    tokens = Lexer.new("foo > 12").tokenize
-    assert_equal([[:id, 'foo'], [:comparison, '>'], [:number, '12'], [:end_of_string]], tokens)
+    assert_equal(
+      [[:id, 'foo'], [:comparison, '>'], [:number, '12'], [:end_of_string]],
+      tokenize("foo > 12"),
+    )
+  end
+
+  def test_error_with_utf8_character
+    error = assert_raises(SyntaxError) do
+      tokenize("1 < 1Ø")
+    end
+
+    assert_equal(
+      'Liquid syntax error: Unexpected character Ø',
+      error.message,
+    )
+  end
+
+  def test_contains_as_attribute_name
+    assert_equal(
+      [[:id, "a"], [:dot, "."], [:id, "contains"], [:dot, "."], [:id, "b"], [:end_of_string]],
+      tokenize("a.contains.b"),
+    )
+  end
+
+  def test_tokenize_incomplete_expression
+    assert_equal([[:id, "false"], [:dash, "-"], [:end_of_string]], tokenize("false -"))
+    assert_equal([[:id, "false"], [:comparison, "<"], [:end_of_string]], tokenize("false <"))
+    assert_equal([[:id, "false"], [:comparison, ">"], [:end_of_string]], tokenize("false >"))
+    assert_equal([[:id, "false"], [:number, "1"], [:end_of_string]], tokenize("false 1"))
+  end
+
+  private
+
+  def tokenize(input)
+    Lexer.new(input).tokenize
   end
 end