about summary refs log tree commit diff
path: root/src/regex
diff options
context:
space:
mode:
authorSzabolcs Nagy <nsz@port70.net>2015-04-18 15:51:16 +0000
committerRich Felker <dalias@aerifal.cx>2016-01-30 20:53:04 -0500
commit7eaa76fc2e7993582989d3838b1ac32dd8abac09 (patch)
tree8d3c0772c1c85b0ba9017fbf919a654a821d8b2d /src/regex
parenta8cc2253843e30dfbdf0bb2954439d9f2b2e8704 (diff)
downloadmusl-7eaa76fc2e7993582989d3838b1ac32dd8abac09.tar.gz
musl-7eaa76fc2e7993582989d3838b1ac32dd8abac09.tar.xz
musl-7eaa76fc2e7993582989d3838b1ac32dd8abac09.zip
regex: reject repetitions in some cases with REG_BADRPT
Previously repetitions were accepted after empty expressions like
in (*|?)|{2}, but in BRE the handling of * and \{\} were not
consistent: they were accepted as literals in some cases and
repetitions in others.

It is better to treat repetitions after an empty expression as an
error (this is allowed by the standard, and glibc mostly does the
same). This is hard to do consistently with the current logic so
the new rule is:

Reject repetitions after empty expressions, except after assertions
^*, $? and empty groups ()+ and never treat them as literals.

Empty alternation (|a) is undefined by the standard, but it can be
useful so that should be accepted.
Diffstat (limited to 'src/regex')
-rw-r--r--src/regex/regcomp.c15
1 files changed, 12 insertions, 3 deletions
diff --git a/src/regex/regcomp.c b/src/regex/regcomp.c
index ac207c89..078f657c 100644
--- a/src/regex/regcomp.c
+++ b/src/regex/regcomp.c
@@ -837,6 +837,10 @@ static reg_errcode_t parse_atom(tre_parse_ctx_t *ctx, const char *s)
 			node = tre_ast_new_literal(ctx->mem, v, v, ctx->position++);
 			s--;
 			break;
+		case '{':
+			/* reject repetitions after empty expression in BRE */
+			if (!ere)
+				return REG_BADRPT;
 		default:
 			if (!ere && (unsigned)*s-'1' < 9) {
 				/* back reference */
@@ -880,10 +884,14 @@ static reg_errcode_t parse_atom(tre_parse_ctx_t *ctx, const char *s)
 		s++;
 		break;
 	case '*':
-	case '|':
+		return REG_BADPAT;
 	case '{':
 	case '+':
 	case '?':
+		/* reject repetitions after empty expression in ERE */
+		if (ere)
+			return REG_BADRPT;
+	case '|':
 		if (!ere)
 			goto parse_literal;
 	case 0:
@@ -964,8 +972,9 @@ static reg_errcode_t tre_parse(tre_parse_ctx_t *ctx)
 		}
 
 	parse_iter:
-		/* extension: repetitions are accepted after an empty node
-		   eg. (+), ^*, a$?, a|{2} */
+		/* extension: repetitions are rejected after an empty node
+		   eg. (+), |*, {2}, but assertions are not treated as empty
+		   so ^* or $? are accepted currently. */
 		switch (*s) {
 		case '+':
 		case '?':