From 7eaa76fc2e7993582989d3838b1ac32dd8abac09 Mon Sep 17 00:00:00 2001 From: Szabolcs Nagy Date: Sat, 18 Apr 2015 15:51:16 +0000 Subject: 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 ^*, $? and empty groups ()+ and never treat them as literals. Empty alternation (|a) is undefined by the standard, but it can be useful so that should be accepted. --- src/regex/regcomp.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'src/regex') diff --git a/src/regex/regcomp.c b/src/regex/regcomp.c index ac207c89..078f657c 100644 --- a/src/regex/regcomp.c +++ b/src/regex/regcomp.c @@ -837,6 +837,10 @@ static reg_errcode_t parse_atom(tre_parse_ctx_t *ctx, const char *s) node = tre_ast_new_literal(ctx->mem, v, v, ctx->position++); s--; break; + case '{': + /* reject repetitions after empty expression in BRE */ + if (!ere) + return REG_BADRPT; default: if (!ere && (unsigned)*s-'1' < 9) { /* back reference */ @@ -880,10 +884,14 @@ static reg_errcode_t parse_atom(tre_parse_ctx_t *ctx, const char *s) s++; break; case '*': - case '|': + return REG_BADPAT; case '{': case '+': case '?': + /* reject repetitions after empty expression in ERE */ + if (ere) + return REG_BADRPT; + case '|': if (!ere) goto parse_literal; case 0: @@ -964,8 +972,9 @@ static reg_errcode_t tre_parse(tre_parse_ctx_t *ctx) } parse_iter: - /* extension: repetitions are accepted after an empty node - eg. (+), ^*, a$?, 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 '?': -- cgit v1.2.1