Port liquid-c bug compatible whitespace trimming

The parse_context no longer needs to be passed in because it is available
through through an attr_reader on the instance. However, the markup and
tokenizer weren't made available.  This refactor also makes the parameters
given to Document#unknown_tag consistent with Block#unknown_tag.

lib/liquid/block.rb

@@ -28,8 +28,8 @@ module Liquid
       @body.nodelist
     end
 
-    def unknown_tag(tag, _params, _tokens)
+    def unknown_tag(tag_name, _markup, _tokenizer)
-      Block.raise_unknown_tag(tag, block_name, block_delimiter, parse_context)
+      Block.raise_unknown_tag(tag_name, block_name, block_delimiter, parse_context)
     end
 
     # @api private

lib/liquid/block_body.rb

@@ -142,7 +142,11 @@ module Liquid
       if token[2] == WhitespaceControl
         previous_token = @nodelist.last
         if previous_token.is_a?(String)
+          first_byte = previous_token.getbyte(0)
           previous_token.rstrip!
+          if previous_token.empty? && parse_context[:bug_compatible_whitespace_trimming] && first_byte
+            previous_token << first_byte
+          end
         end
       end
       parse_context.trim_whitespace = (token[-3] == WhitespaceControl)

lib/liquid/document.rb

@@ -19,16 +19,15 @@ module Liquid
       @body.nodelist
     end
 
-    def parse(tokens, parse_context)
+    def parse(tokenizer, parse_context)
-      @body.parse(tokens, parse_context) do |end_tag_name, _end_tag_params|
+      while parse_body(tokenizer)
-        unknown_tag(end_tag_name, parse_context) if end_tag_name
       end
     rescue SyntaxError => e
       e.line_number ||= parse_context.line_number
       raise
     end
 
-    def unknown_tag(tag, parse_context)
+    def unknown_tag(tag, _markup, _tokenizer)
       case tag
       when 'else', 'end'
         raise SyntaxError, parse_context.locale.t("errors.syntax.unexpected_outer_tag", tag: tag)
@@ -50,5 +49,16 @@ module Liquid
     def new_body
       Liquid::BlockBody.new
     end
+
+    def parse_body(tokenizer)
+      @body.parse(tokenizer, parse_context) do |unknown_tag_name, unknown_tag_markup|
+        if unknown_tag_name
+          unknown_tag(unknown_tag_name, unknown_tag_markup, tokenizer)
+          true
+        else
+          false
+        end
+      end
+    end
   end
 end

test/integration/trim_mode_test.rb

@@ -528,4 +528,21 @@ class TrimModeTest < Minitest::Test
     END_EXPECTED
     assert_template_result(expected, text)
   end
+
+  def test_bug_compatible_pre_trim
+    template = Liquid::Template.parse("\n {%- raw %}{% endraw %}", bug_compatible_whitespace_trimming: true)
+    assert_equal("\n", template.render)
+
+    template = Liquid::Template.parse("\n {%- if true %}{% endif %}", bug_compatible_whitespace_trimming: true)
+    assert_equal("\n", template.render)
+
+    template = Liquid::Template.parse("{{ 'B' }} \n{%- if true %}C{% endif %}", bug_compatible_whitespace_trimming: true)
+    assert_equal("B C", template.render)
+
+    template = Liquid::Template.parse("B\n {%- raw %}{% endraw %}", bug_compatible_whitespace_trimming: true)
+    assert_equal("B", template.render)
+
+    template = Liquid::Template.parse("B\n {%- if true %}{% endif %}", bug_compatible_whitespace_trimming: true)
+    assert_equal("B", template.render)
+  end
 end # TrimModeTest