From 970ce272830c424820d1dcae6804f6c5f567cd4c Mon Sep 17 00:00:00 2001
From: Yorick Peterse <yorickpeterse@gmail.com>
Date: Fri, 28 Feb 2014 23:16:01 +0100
Subject: [PATCH] Cleanup of buffering text/strings.

This removes the need to use ||= and such, which should speed things up a bit
and keeps the code cleaner.
---
 lib/oga/lexer.rl | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

diff --git a/lib/oga/lexer.rl b/lib/oga/lexer.rl
index 4ab167b..4a42841 100644
--- a/lib/oga/lexer.rl
+++ b/lib/oga/lexer.rl
@@ -29,6 +29,9 @@ module Oga
       @tokens = []
       @stack  = []
       @top    = 0
+
+      @string_buffer = ''
+      @text_buffer   = ''
     end
 
     def lex(data)
@@ -75,11 +78,17 @@ module Oga
       @tokens << token
     end
 
+    def emit_text_buffer
+      add_token(:T_TEXT, @text_buffer)
+
+      @text_buffer = ''
+    end
+
     def emit_string_buffer
       add_token(:T_STRING, @string_buffer)
       advance_column
 
-      @string_buffer = nil
+      @string_buffer = ''
     end
 
     %%{
@@ -109,8 +118,11 @@ module Oga
       dquote = '"';
       squote = "'";
 
+      action buffer_text {
+        @text_buffer << text
+      }
+
       action buffer_string {
-        @string_buffer ||= ''
         @string_buffer << text
       }
 
@@ -184,8 +196,7 @@ module Oga
 
       cdata := |*
         cdata_end => {
-          add_token(:T_TEXT, @cdata_buffer)
-          @cdata_buffer = nil
+          emit_text_buffer
 
           t(:T_CDATA_END)
 
@@ -194,7 +205,7 @@ module Oga
 
         # Consume everything else character by character and store it in a
         # separate buffer.
-        any => { @cdata_buffer << text };
+        any => buffer_text;
       *|;
 
       main := |*
@@ -211,8 +222,6 @@ module Oga
         cdata_start => {
           t(:T_CDATA_START)
 
-          @cdata_buffer = ''
-
           fgoto cdata;
         };