Fix RegExp literal parsing

JerryScript-DCO-1.0-Signed-off-by: László Langó llango.u-szeged@partner.samsung.com
JerryScript-DCO-1.0-Signed-off-by: Dániel Bátyai dbatyai.u-szeged@partner.samsung.com

Author: dbatyai

jerry-core/parser/js/lexer.cpp

@@ -23,7 +23,7 @@
 #include "lit-strings.h"
 #include "jsp-early-error.h"
 
-static token saved_token, prev_token, sent_token, empty_token;
+static token saved_token, prev_token, sent_token, empty_token, prev_non_lf_token;
 
 static bool allow_dump_lines = false, strict_mode;
 static size_t buffer_size = 0;
@@ -1092,18 +1092,15 @@ lexer_parse_regexp (void)
 
   /* Eat up '/' */
   JERRY_ASSERT (LA (0) == LIT_CHAR_SLASH);
-  consume_char ();
   new_token ();
 
+  consume_char ();
+
   while (true)
   {
     ecma_char_t c = (ecma_char_t) LA (0);
 
-    if (c == LIT_CHAR_NULL)
-    {
-      PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "Unclosed string", token_start_pos);
-    }
-    else if (lit_char_is_line_terminator (c))
+    if (lit_char_is_line_terminator (c))
     {
       PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "RegExp literal shall not contain newline character", token_start_pos);
     }
@@ -1140,16 +1137,15 @@ lexer_parse_regexp (void)
   {
     ecma_char_t c = (ecma_char_t) LA (0);
 
-    if (c == LIT_CHAR_NULL
-        || !lit_char_is_word_char (c)
-        || lit_char_is_line_terminator (c))
+    if (!lit_char_is_word_char (c) || lit_char_is_line_terminator (c))
     {
       break;
     }
+
     consume_char ();
   }
 
-  result = lexer_create_token_for_charset (TOK_REGEXP, TOK_START (), TOK_SIZE ());
+  result = lexer_create_token_for_charset (TOK_REGEXP, TOK_START () + 1, TOK_SIZE () - 1);
 
   is_token_parse_in_progress = false;
   return result;
@@ -1294,16 +1290,16 @@ lexer_parse_token (void)
   }
 
   if (c == LIT_CHAR_SLASH
-      && !(sent_token.type == TOK_NAME
-           || sent_token.type == TOK_NULL
-           || sent_token.type == TOK_BOOL
-           || sent_token.type == TOK_CLOSE_BRACE
-           || sent_token.type == TOK_CLOSE_SQUARE
-           || sent_token.type == TOK_CLOSE_PAREN
-           || sent_token.type == TOK_SMALL_INT
-           || sent_token.type == TOK_NUMBER
-           || sent_token.type == TOK_STRING
-           || sent_token.type == TOK_REGEXP))
+      && !(prev_non_lf_token.type == TOK_NAME
+           || prev_non_lf_token.type == TOK_NULL
+           || prev_non_lf_token.type == TOK_BOOL
+           || prev_non_lf_token.type == TOK_CLOSE_BRACE
+           || prev_non_lf_token.type == TOK_CLOSE_SQUARE
+           || prev_non_lf_token.type == TOK_CLOSE_PAREN
+           || prev_non_lf_token.type == TOK_SMALL_INT
+           || prev_non_lf_token.type == TOK_NUMBER
+           || prev_non_lf_token.type == TOK_STRING
+           || prev_non_lf_token.type == TOK_REGEXP))
   {
     return lexer_parse_regexp ();
   }
@@ -1482,6 +1478,7 @@ lexer_next_token (void)
   if (!is_empty (saved_token))
   {
     sent_token = saved_token;
+    prev_non_lf_token = sent_token;
     saved_token = empty_token;
     goto end;
   }
@@ -1499,6 +1496,11 @@ lexer_next_token (void)
     PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "Unexpected EOF", lit_utf8_iterator_get_pos (&src_iter));
   }
 
+  if (sent_token.type != TOK_NEWLINE)
+  {
+    prev_non_lf_token = sent_token;
+  }
+
   prev_token = sent_token;
   sent_token = lexer_parse_token ();
 

tests/jerry/arithmetics.js

@@ -37,3 +37,5 @@ assert (1 / (-1 % -1) < 0);
 assert (1 / (-1 % 1) < 0);
 assert (1 / (1 % -1) > 0);
 assert (1 / (1 % 1) > 0);
+
+assert (eval ("x\n\n=\n\n6\n\n/\n\n3") === 2)

tests/jerry/regexp-literal.js

@@ -23,3 +23,9 @@ assert ("a"+/x/+"b" == "a/x/b");
 
 t = /\/\[[\]/]/.exec("/[/");
 assert (t == "/[/");
+
+try {
+	eval("/" + String.fromCharCode("0x0000") + "/");
+} catch (e) {
+	assert (false);
+}