From d7a4adf48149e0d6d1ae0cc79c1a6d73aa02dd3f Mon Sep 17 00:00:00 2001
From: James Bursa <james@netsurf-browser.org>
Date: Sat, 1 May 2004 17:48:38 +0000
Subject: [project @ 2004-05-01 17:48:38 by bursa] CSS parsing improvements:
 new tokeniser using re2c, improve memory-exhaustion behaviour, plug leaks,
 reduce memory usage, clean up code, add source documention.

svn path=/import/netsurf/; revision=806
---
 css/scanner.l | 145 +++++++++++++++++++++++++++++++++-------------------------
 1 file changed, 82 insertions(+), 63 deletions(-)

(limited to 'css/scanner.l')

diff --git a/css/scanner.l b/css/scanner.l
index 1f8cc1d92..36347b3a8 100644
--- a/css/scanner.l
+++ b/css/scanner.l
@@ -2,78 +2,97 @@
  * This file is part of NetSurf, http://netsurf.sourceforge.net/
  * Licensed under the GNU General Public License,
  *                http://www.opensource.org/licenses/gpl-license
- * Copyright 2003 James Bursa <bursa@users.sourceforge.net>
+ * Copyright 2004 James Bursa <bursa@users.sourceforge.net>
  */
 
-%{
-#include "parser.h"
-%}
+/** \file
+ * CSS tokeniser using re2c.
+ *
+ * see CSS2 Specification, chapter 4
+ * http://www.w3.org/TR/REC-CSS2/syndata.html,
+ * and errata
+ * http://www.w3.org/Style/css2-updates/REC-CSS2-19980512-errata
+ */
+
+#include <stdbool.h>
+#define CSS_INTERNALS
+#include "netsurf/css/css.h"
+#include "netsurf/css/parser.h"
+
+#define YYCTYPE unsigned char
+#define YYCURSOR (*buffer)
+#define YYLIMIT end
+#define YYMARKER marker
+#define YYFILL(n) { return 0; }
 
-%option 8bit
-%option batch
-%option case-insensitive
-%option header-file="scanner.h"
-%option outfile="scanner.c"
-%option prefix="css_"
-%option reentrant
-%option never-interactive
-%option noyywrap
-%option yylineno
 
-/* see CSS2 Specification, chapter 4
-   http://www.w3.org/TR/REC-CSS2/syndata.html,
-   and errata
-   http://www.w3.org/Style/css2-updates/REC-CSS2-19980512-errata */
+/**
+ * Identify a CSS source token.
+ *
+ * \param  buffer      source to tokenise, updated to new position
+ * \param  end         end of source
+ * \param  token_text  updated to start of recognized token
+ * \return  token number
+ */
+
+int css_tokenise(unsigned char **buffer, unsigned char *end,
+		unsigned char **token_text)
+{
+	unsigned char *marker;
 
-ident		{nmstart}{nmchar}*
-name		{nmchar}+
-nmstart		[a-zA-Z_]|{nonascii}|{escape}
-nonascii	[\200-\377]
-unicode		\\[0-9a-f]{1,6}[ \n\r\t\f]?
-escape		{unicode}|\\[ -~\200-\377]
-nmchar		[-a-zA-Z0-9_]|{nonascii}|{escape}
-num		[+-]?[0-9]+|[0-9]*\.[0-9]+
-string		{string1}|{string2}
-string1		\"([\t !#$%&(-~]|\\{nl}|\'|{nonascii}|{escape})*\"
-string2		\'([\t !#$%&(-~]|\\{nl}|\"|{nonascii}|{escape})*\'
-nl		\n|\r\n|\r|\f
-w		[ \t\r\n\f]*
+start:
+	*token_text = YYCURSOR;
 
-%%
+/*!re2c
+nonascii	=	[\200-\377];
+unicode		=	"\\" [0-9a-f]+ [ \n\r\t\f]?;
+escape		=	unicode | "\\" [ -~\200-\377];
+nmchar		=	[-a-zA-Z0-9_] | nonascii | escape;
+nmstart		=	[a-zA-Z_] | nonascii | escape;
+ident		=	nmstart nmchar*;
+name		=	nmchar+;
+num		=	[+-]? [0-9]+ | [0-9]* "." [0-9]+;
+nl		=	"\n" | "\r\n" | "\r" | "\f";
+string1		=	"\"" ([\t !#$%&(-~] | "\\" nl | "'" | nonascii | escape)* "\"";
+string2		=	"'"  ([\t !#$%&(-~] | "\\" nl | "\""| nonascii | escape)* "'";
+string		=	string1 | string2;
+w		=	[ \t\r\n\f]*;
+any		=	[\000-\377];
 
-{ident}		{ return IDENT; }
-@{ident}	{ return ATKEYWORD; }
-{string}	{ return STRING; }
-#{name}		{ return HASH; }
-{num}		{ return NUMBER; }
-{num}%		{ return PERCENTAGE; }
-{num}{ident}	{ return DIMENSION; }
-url\({w}{string}{w}\)|url\({w}([!#$%&*-~]|{nonascii}|{escape})*{w}\)	{
-		  return URI; }
-U\+[0-9A-F?]{1,6}(-[0-9A-F]{1,6})?	{
-		  return UNICODE_RANGE; }
-"<!--"		/* ignore CDO */
-"-->"		/* ignore CDC */
-;		{ return SEMI; }
-\{		{ return LBRACE; }
-\}		{ return RBRACE; }
-\(		{ return LPAREN; }
-\)		{ return RPAREN; }
-\[		{ return LBRAC; }
-\]		{ return RBRAC; }
-[ \t\r\n\f]+	/* ignore whitespace */
-\/\*[^*]*\*+([^/][^*]*\*+)*\/	/* ignore comments */
-{ident}\(	{ return FUNCTION; }
-=		{ return EQUALS; }
-~=		{ return INCLUDES; }
+ident		{ return IDENT; }
+"@" ident	{ return ATKEYWORD; }
+string		{ return STRING; }
+"#" name	{ return HASH; }
+num		{ return NUMBER; }
+num "%"		{ return PERCENTAGE; }
+num ident	{ return DIMENSION; }
+"url(" w string w ")" | "url(" w ([!#$%&*-~]|nonascii|escape)* w ")"
+		{ return URI; }
+"U+" [0-9A-F?]+ ("-" [0-9A-F]+ )?
+		{ return UNICODE_RANGE; }
+"<!--"		{ goto start; /* ignore CDO */ }
+"-->"		{ goto start; /* ignore CDC */ }
+";"		{ return SEMI; }
+"{"		{ return LBRACE; }
+"}"		{ return RBRACE; }
+"("		{ return LPAREN; }
+")"		{ return RPAREN; }
+"["		{ return LBRAC; }
+"]"		{ return RBRAC; }
+[ \t\r\n\f]+	{ goto start; /* ignore whitespace */ }
+"/*" (any\[*])* "*"+ ((any\[/]) (any\[*])* "*"+)* "/"
+		{ goto start; /* ignore comments */ }
+ident "("	{ return FUNCTION; }
+"="		{ return EQUALS; }
+"~="		{ return INCLUDES; }
 "|="		{ return DASHMATCH; }
-:		{ return COLON; }
-,		{ return COMMA; }
+":"		{ return COLON; }
+","		{ return COMMA; }
 "+"		{ return PLUS; }
->		{ return GT; }
+">"		{ return GT; }
 "."		{ return DOT; }
 "*"		{ return ASTERISK; }
-.		{ return DELIM; }
-
-%%
+any		{ return DELIM; }
+*/
 
+}
-- 
cgit v1.2.3