hare

commit 9fb0512d6d628c330b3487f8827ec5dd6e051c13
parent 609ef43e6479c9bbc29847a187aeaa2524ae614e
Author: Nolan Prescott <mail@nprescott.com>
Date:   Tue, 26 Apr 2022 23:40:57 -0400

regex: trailing hyphen not a range

Per the POSIX ERE specification[0]:

> The <hyphen-minus> character shall be treated as itself if it occurs
first (after an initial '^', if any) or last in the list, or as an
ending range point in a range expression.

0: https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html#tag_09_03_05

Implements: https://todo.sr.ht/~sircmpwn/hare/624
Signed-off-by: Nolan Prescott <mail@nprescott.com>

Diffstat:
M
regex/+test.ha
 | 
8
++++++++
M
regex/regex.ha
 | 
3
++-

2 files changed, 10 insertions(+), 1 deletion(-)
diff --git a/regex/+test.ha b/regex/+test.ha
@@ -351,6 +351,14 @@ fn run_findall_case(
 			"M15 4QN",
 			matchres::MATCH, 0, -1
 		),
+		(`^[^-a]`, "-bcd", matchres::NOMATCH, 0, 0),
+		(`^[-a]`, "-bcd", matchres::MATCH, 0, 1),
+		(`[^ac-]`, "bde", matchres::MATCH, 0, 1),
+		(`[-ac]`, "foo-de", matchres::MATCH, 3, 4),
+		(`[-ac]`, "def", matchres::NOMATCH, 0, 0),
+		(`foo[-ac]bar`, "foo-bar", matchres::MATCH, 0, 7),
+		(`[ac-]$`, "bde-", matchres::MATCH, 3, 4),
+		(`^[A-Za-z_-]+$`, "foo", matchres::MATCH, 0, 3),
 		// tests from perl
 		(`abc`, "abc", matchres::MATCH, 0, -1),
 		(`abc`, "xbc", matchres::NOMATCH, 0, 0),
diff --git a/regex/regex.ha b/regex/regex.ha
@@ -154,7 +154,8 @@ fn handle_bracket(
 	};
 
 	const is_range = peek1 is rune && peek1 as rune == '-'
-		&& !(peek2 is void) && !(peek3 is void);
+		&& !(peek2 is void) && !(peek3 is void)
+		&& !(peek2 as rune == ']');
 	const range_end = peek2;
 	const is_first_char = *bracket_idx == 0 || *bracket_idx == 1
 		&& !*is_charset_positive;