From 0c470af3144de7931cd88cf3238d7cd87413134e Mon Sep 17 00:00:00 2001
From: Zeev Suraski <zeev@php.net>
Date: Mon, 7 Feb 2000 18:39:14 +0000
Subject: [PATCH] Syntax highlighting was erronously emitting more than one
 semicolon and/or garbage with heredocs

---
 Zend/zend-parser.y    |  7 ++++---
 Zend/zend-scanner.l   |  6 ++++--
 Zend/zend_compile.c   |  3 +++
 Zend/zend_highlight.c | 11 +++++++++++
 4 files changed, 22 insertions(+), 5 deletions(-)

diff --git a/Zend/zend-parser.y b/Zend/zend-parser.y
index d5f851e1b7..e358e0b9db 100644
--- a/Zend/zend-parser.y
+++ b/Zend/zend-parser.y
@@ -47,7 +47,7 @@
 %}
 
 %pure_parser
-%expect 4
+%expect 7
 
 %left T_INCLUDE T_EVAL T_IMPORT
 %left ','
@@ -129,7 +129,8 @@
 %token T_OPEN_TAG_WITH_ECHO
 %token T_CLOSE_TAG
 %token T_WHITESPACE
-%token T_HEREDOC
+%token T_START_HEREDOC
+%token T_END_HEREDOC
 %token T_DOLLAR_OPEN_CURLY_BRACES
 %token T_CURLY_OPEN
 %token T_PAAMAYIM_NEKUDOTAYIM
@@ -538,7 +539,7 @@ scalar:
 	|	common_scalar			{ $$ = $1; }
 	|	'"' encaps_list '"' 	{ $$ = $2; }
 	|	'\'' encaps_list '\''	{ $$ = $2; }
-	|	T_HEREDOC encaps_list T_HEREDOC { $$ = $2; do_end_heredoc(CLS_C); }
+	|	T_START_HEREDOC encaps_list T_END_HEREDOC { $$ = $2; do_end_heredoc(CLS_C); }
 ;
 
 
diff --git a/Zend/zend-scanner.l b/Zend/zend-scanner.l
index eb5a4bebdd..23ad787306 100644
--- a/Zend/zend-scanner.l
+++ b/Zend/zend-scanner.l
@@ -1440,7 +1440,7 @@ ESCAPED_AND_WHITESPACE [\n\t\r #'.:;,()|^&+-/*=%!~<>?@]+
 	}
 	CG(heredoc) = estrndup(s, CG(heredoc_len));
 	BEGIN(ST_HEREDOC);
-	return T_HEREDOC;
+	return T_START_HEREDOC;
 }
 	
 
@@ -1475,6 +1475,8 @@ ESCAPED_AND_WHITESPACE [\n\t\r #'.:;,()|^&+-/*=%!~<>?@]+
 	}
 
 	if (label_len==CG(heredoc_len) && !memcmp(yytext, CG(heredoc), label_len)) {
+		zendlval->value.str.val = estrndup(yytext, yyleng); /* unput destroys yytext */
+		zendlval->value.str.len = yyleng;
 		if (unput_semicolon) {
 			unput(';');
 		}
@@ -1482,7 +1484,7 @@ ESCAPED_AND_WHITESPACE [\n\t\r #'.:;,()|^&+-/*=%!~<>?@]+
 		CG(heredoc)=NULL;
 		CG(heredoc_len)=0;
 		BEGIN(ST_IN_SCRIPTING);
-		return T_HEREDOC;
+		return T_END_HEREDOC;
 	} else {
 		zendlval->value.str.val = (char *)estrndup(yytext, yyleng);
 		zendlval->value.str.len = yyleng;
diff --git a/Zend/zend_compile.c b/Zend/zend_compile.c
index ff98273866..9019850c45 100644
--- a/Zend/zend_compile.c
+++ b/Zend/zend_compile.c
@@ -2289,6 +2289,9 @@ int zendlex(znode *zendlval CLS_DC)
 		case T_OPEN_TAG_WITH_ECHO:
 			retval = T_ECHO;
 			break;
+		case T_END_HEREDOC:
+			efree(zendlval->u.constant.value.str.val);
+			break;
 	}
 		
 	INIT_PZVAL(&zendlval->u.constant);
diff --git a/Zend/zend_highlight.c b/Zend/zend_highlight.c
index c68b4cd91f..841adb9271 100644
--- a/Zend/zend_highlight.c
+++ b/Zend/zend_highlight.c
@@ -137,6 +137,9 @@ ZEND_API void zend_highlight(zend_syntax_highlighter_ini *syntax_highlighter_ini
 			case T_DOLLAR_OPEN_CURLY_BRACES:
 				html_puts("{", 1);
 				break;
+			case T_END_HEREDOC:
+				html_puts(token.value.str.val, token.value.str.len);
+				break;
 			default:
 				html_puts(zendtext, zendleng);
 				break;
@@ -152,6 +155,14 @@ ZEND_API void zend_highlight(zend_syntax_highlighter_ini *syntax_highlighter_ini
 					efree(token.value.str.val);
 					break;
 			}
+		} else if (token_type == T_END_HEREDOC) {
+			zend_bool has_semicolon=(strchr(token.value.str.val, ';')?1:0);
+
+			efree(token.value.str.val);
+			if (has_semicolon) {
+				/* the following semicolon was unput(), ignore it */
+				lex_scan(&token CLS_CC);
+			}
 		}
 		token.type = 0;
 	}
-- 
2.40.0