RepoMirrors
/
musl
mirror of git://git.musl-libc.org/musl
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

regex: treat \| in BRE as alternation 

The standard does not define semantics for \| in BRE, but some code
depends on it meaning alternation. Empty alternative expression is
allowed to be consistent with ERE.

Based on a patch by Rob Landley.
This commit is contained in:
Szabolcs Nagy 2015-04-18 16:47:17 +00:00 committed by Rich Felker
parent 7eaa76fc2e
commit da4cc13b97
1 changed files with 17 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
src/regex/regcomp.c
View File

@ -841,6 +841,14 @@ static reg_errcode_t parse_atom(tre_parse_ctx_t *ctx, const char *s)
/* reject repetitions after empty expression in BRE */ /* reject repetitions after empty expression in BRE */
if (!ere) if (!ere)
return REG_BADRPT; return REG_BADRPT;
case '|':
/* extension: treat \| as alternation in BRE */
if (!ere) {
node = tre_ast_new_literal(ctx->mem, EMPTY, -1, -1);
s--;
goto end;
}
/* fallthrough */
default: default:
if (!ere && (unsigned)*s-'1' < 9) { if (!ere && (unsigned)*s-'1' < 9) {
/* back reference */ /* back reference */
@ -918,6 +926,7 @@ parse_literal:
s += len; s += len;
break; break;
} }
end:
if (!node) if (!node)
return REG_ESPACE; return REG_ESPACE;
ctx->n = node; ctx->n = node;
@ -1016,13 +1025,20 @@ static reg_errcode_t tre_parse(tre_parse_ctx_t *ctx)
if ((ere && *s == '|') || if ((ere && *s == '|') ||
(ere && *s == ')' && depth) || (ere && *s == ')' && depth) ||
(!ere && *s == '\\' && s[1] == ')') || (!ere && *s == '\\' && s[1] == ')') ||
/* extension: treat \| as alternation in BRE */
(!ere && *s == '\\' && s[1] == '|') ||
!*s) { !*s) {
/* extension: empty branch is unspecified (), (|a), (a|) /* extension: empty branch is unspecified (), (|a), (a|)
here they are not rejected but match on empty string */ here they are not rejected but match on empty string */
int c = *s; int c = *s;
nunion = tre_ast_new_union(ctx->mem, nunion, nbranch); nunion = tre_ast_new_union(ctx->mem, nunion, nbranch);
nbranch = 0; nbranch = 0;
if (c != '|') {
if (c == '\\' && s[1] == '|') {
s+=2;
} else if (c == '|') {
s++;
} else {
if (c == '\\') { if (c == '\\') {
if (!depth) return REG_EPAREN; if (!depth) return REG_EPAREN;
s+=2; s+=2;
@ -1042,7 +1058,6 @@ static reg_errcode_t tre_parse(tre_parse_ctx_t *ctx)
nunion = tre_stack_pop_voidptr(stack); nunion = tre_stack_pop_voidptr(stack);
goto parse_iter; goto parse_iter;
} }
s++;
} }
} }
} }