Improved compilation of heredocs and interpolated strings. (Matt)

3 files changed, 37 insertions, 38 deletions

diff --git a/ext/tokenizer/tests/001.phpt b/ext/tokenizer/tests/001.phpt
index a8ab6bd523..7bac6b96e5 100644
--- a/ext/tokenizer/tests/001.phpt
+++ b/ext/tokenizer/tests/001.phpt
@@ -57,8 +57,6 @@ echo token_name(T_STRING_VARNAME), "\n";
 echo token_name(T_VARIABLE), "\n";
 echo token_name(T_NUM_STRING), "\n";
 echo token_name(T_INLINE_HTML), "\n";
-echo token_name(T_CHARACTER), "\n";
-echo token_name(T_BAD_CHARACTER), "\n";
 echo token_name(T_ENCAPSED_AND_WHITESPACE), "\n";
 echo token_name(T_CONSTANT_ENCAPSED_STRING), "\n";
 echo token_name(T_ECHO), "\n";
@@ -185,8 +183,6 @@ T_STRING_VARNAME
 T_VARIABLE
 T_NUM_STRING
 T_INLINE_HTML
-T_CHARACTER
-T_BAD_CHARACTER
 T_ENCAPSED_AND_WHITESPACE
 T_CONSTANT_ENCAPSED_STRING
 T_ECHO
@@ -314,8 +310,6 @@ T_STRING_VARNAME
 T_VARIABLE
 T_NUM_STRING
 T_INLINE_HTML
-T_CHARACTER
-T_BAD_CHARACTER
 T_ENCAPSED_AND_WHITESPACE
 T_CONSTANT_ENCAPSED_STRING
 T_ECHO
diff --git a/ext/tokenizer/tests/bug26463.phpt b/ext/tokenizer/tests/bug26463.phpt
index c72d478403..de32769617 100644
--- a/ext/tokenizer/tests/bug26463.phpt
+++ b/ext/tokenizer/tests/bug26463.phpt
@@ -15,12 +15,12 @@ DDDD;
 ?>';
 var_dump(token_get_all($str));
 ?>
---EXPECT--
+--EXPECTF--
 array(19) {
   [0]=>
   array(3) {
     [0]=>
-    int(370)
+    int(%d)
     [1]=>
     string(6) "<?php
 "
@@ -30,7 +30,7 @@ array(19) {
   [1]=>
   array(3) {
     [0]=>
-    int(311)
+    int(%d)
     [1]=>
     string(2) "$x"
     [2]=>
@@ -41,7 +41,7 @@ array(19) {
   [3]=>
   array(3) {
     [0]=>
-    int(374)
+    int(%d)
     [1]=>
     string(6) "<<<DD
 "
@@ -51,7 +51,7 @@ array(19) {
   [4]=>
   array(3) {
     [0]=>
-    int(309)
+    int(%d)
     [1]=>
     string(13) "jhdsjkfhjdsh
 "
@@ -61,7 +61,7 @@ array(19) {
   [5]=>
   array(3) {
     [0]=>
-    int(375)
+    int(%d)
     [1]=>
     string(2) "DD"
     [2]=>
@@ -70,7 +70,7 @@ array(19) {
   [6]=>
   array(3) {
     [0]=>
-    int(373)
+    int(%d)
     [1]=>
     string(1) "
 "
@@ -82,7 +82,7 @@ array(19) {
   [8]=>
   array(3) {
     [0]=>
-    int(317)
+    int(%d)
     [1]=>
     string(2) """"
     [2]=>
@@ -93,7 +93,7 @@ array(19) {
   [10]=>
   array(3) {
     [0]=>
-    int(373)
+    int(%d)
     [1]=>
     string(1) "
 "
@@ -103,7 +103,7 @@ array(19) {
   [11]=>
   array(3) {
     [0]=>
-    int(311)
+    int(%d)
     [1]=>
     string(2) "$a"
     [2]=>
@@ -114,7 +114,7 @@ array(19) {
   [13]=>
   array(3) {
     [0]=>
-    int(374)
+    int(%d)
     [1]=>
     string(8) "<<<DDDD
 "
@@ -124,7 +124,7 @@ array(19) {
   [14]=>
   array(3) {
     [0]=>
-    int(309)
+    int(%d)
     [1]=>
     string(13) "jhdsjkfhjdsh
 "
@@ -134,7 +134,7 @@ array(19) {
   [15]=>
   array(3) {
     [0]=>
-    int(375)
+    int(%d)
     [1]=>
     string(4) "DDDD"
     [2]=>
@@ -145,7 +145,7 @@ array(19) {
   [17]=>
   array(3) {
     [0]=>
-    int(373)
+    int(%d)
     [1]=>
     string(1) "
 "
@@ -155,7 +155,7 @@ array(19) {
   [18]=>
   array(3) {
     [0]=>
-    int(372)
+    int(%d)
     [1]=>
     string(2) "?>"
     [2]=>
@@ -167,7 +167,7 @@ array(19) {
   [0]=>
   array(3) {
     [0]=>
-    int(370)
+    int(%d)
     [1]=>
     string(6) "<?php
 "
@@ -177,7 +177,7 @@ array(19) {
   [1]=>
   array(3) {
     [0]=>
-    int(311)
+    int(%d)
     [1]=>
     string(2) "$x"
     [2]=>
@@ -188,7 +188,7 @@ array(19) {
   [3]=>
   array(3) {
     [0]=>
-    int(374)
+    int(%d)
     [1]=>
     string(6) "<<<DD
 "
@@ -198,7 +198,7 @@ array(19) {
   [4]=>
   array(3) {
     [0]=>
-    int(309)
+    int(%d)
     [1]=>
     string(13) "jhdsjkfhjdsh
 "
@@ -208,7 +208,7 @@ array(19) {
   [5]=>
   array(3) {
     [0]=>
-    int(375)
+    int(%d)
     [1]=>
     string(2) "DD"
     [2]=>
@@ -217,7 +217,7 @@ array(19) {
   [6]=>
   array(3) {
     [0]=>
-    int(373)
+    int(%d)
     [1]=>
     string(1) "
 "
@@ -229,7 +229,7 @@ array(19) {
   [8]=>
   array(3) {
     [0]=>
-    int(317)
+    int(%d)
     [1]=>
     string(2) """"
     [2]=>
@@ -240,7 +240,7 @@ array(19) {
   [10]=>
   array(3) {
     [0]=>
-    int(373)
+    int(%d)
     [1]=>
     string(1) "
 "
@@ -250,7 +250,7 @@ array(19) {
   [11]=>
   array(3) {
     [0]=>
-    int(311)
+    int(%d)
     [1]=>
     string(2) "$a"
     [2]=>
@@ -261,7 +261,7 @@ array(19) {
   [13]=>
   array(3) {
     [0]=>
-    int(374)
+    int(%d)
     [1]=>
     string(8) "<<<DDDD
 "
@@ -271,7 +271,7 @@ array(19) {
   [14]=>
   array(3) {
     [0]=>
-    int(309)
+    int(%d)
     [1]=>
     string(13) "jhdsjkfhjdsh
 "
@@ -281,7 +281,7 @@ array(19) {
   [15]=>
   array(3) {
     [0]=>
-    int(375)
+    int(%d)
     [1]=>
     string(4) "DDDD"
     [2]=>
@@ -292,7 +292,7 @@ array(19) {
   [17]=>
   array(3) {
     [0]=>
-    int(373)
+    int(%d)
     [1]=>
     string(1) "
 "
@@ -302,7 +302,7 @@ array(19) {
   [18]=>
   array(3) {
     [0]=>
-    int(372)
+    int(%d)
     [1]=>
     string(2) "?>"
     [2]=>
diff --git a/ext/tokenizer/tokenizer.c b/ext/tokenizer/tokenizer.c
index 3727ec0a49..bd60a970fe 100644
--- a/ext/tokenizer/tokenizer.c
+++ b/ext/tokenizer/tokenizer.c
@@ -282,12 +282,15 @@ static void tokenize(zval *return_value TSRMLS_DC)
 	while ((token_type = lex_scan(&token TSRMLS_CC))) {
 		destroy = 1;
 		switch (token_type) {
+			case T_CLOSE_TAG:
+				if (zendtext[zendleng - 1] != '>') {
+					CG(zend_lineno)++;
+				}
 			case T_OPEN_TAG:
 			case T_OPEN_TAG_WITH_ECHO:
 			case T_WHITESPACE:
 			case T_COMMENT:
 			case T_DOC_COMMENT:
-			case T_CLOSE_TAG:
 				destroy = 0;
 				break;
 		}
@@ -297,6 +300,10 @@ static void tokenize(zval *return_value TSRMLS_DC)
 			array_init(keyword);
 			add_next_index_long(keyword, token_type);
 			if (token_type == T_END_HEREDOC) {
+				if (CG(increment_lineno)) {
+					token_line = ++CG(zend_lineno);
+					CG(increment_lineno) = 0;
+				}
 				add_next_index_stringl(keyword, Z_STRVAL(token), Z_STRLEN(token), 1);
 				efree(Z_STRVAL(token));
 			} else {
@@ -372,8 +379,6 @@ get_token_type_name(int token_type)
 		case T_VARIABLE: return "T_VARIABLE";
 		case T_NUM_STRING: return "T_NUM_STRING";
 		case T_INLINE_HTML: return "T_INLINE_HTML";
-		case T_CHARACTER: return "T_CHARACTER";
-		case T_BAD_CHARACTER: return "T_BAD_CHARACTER";
 		case T_ENCAPSED_AND_WHITESPACE: return "T_ENCAPSED_AND_WHITESPACE";
 		case T_CONSTANT_ENCAPSED_STRING: return "T_CONSTANT_ENCAPSED_STRING";
 		case T_ECHO: return "T_ECHO";