A pattern anchoring some alternatives with ^ and not others is refused with why, not a bare bad regular expression

`^def|x` and `foo$|^zèbre` were refused with a bare "bad regular expression",
the same words as a syntax error, so a script could not tell what to change.
They now say "an alternation anchors every branch with ^ or none (^a|^b, not
^a|b)", through addr and Edit alike. fs.md states the rule and that a `$`
does not count toward it (`foo$|bar` is fine).

Co-Authored-By: Claude Opus 5.5 <[email protected]>

1 files changed, 6 insertions, 4 deletions

diff --git a/src/regexp.zig b/src/regexp.zig
index ddb5eb28..575e6861 100644
--- a/src/regexp.zig
+++ b/src/regexp.zig
@@ -56,6 +56,7 @@ pub const Regex = struct {
     pub const e_anchor = "bad regular expression: in a pattern with \\n, ^ can only come first and $ only just before a \\n";
     pub const e_wide = std.fmt.comptimePrint("bad regular expression: a range of runes wider than {d} in [...] is not supported", .{max_range});
     pub const e_negated = "bad regular expression: a [^...] with non-ASCII runes is not supported (mvzr's classes hold bytes)";
+    pub const e_mixed = "bad regular expression: an alternation anchors every branch with ^ or none (^a|^b, not ^a|b)";
 
     /// The most runes a non-ASCII range in a class is spelled out as.
     pub const max_range = 256;
@@ -79,7 +80,7 @@ pub const Regex = struct {
         return @intCast(n);
     }
 
-    pub const Error = error{ Bad, Anchor, TooLong, WideRange, NegatedRunes };
+    pub const Error = error{ Bad, Anchor, TooLong, WideRange, NegatedRunes, Mixed };
 
     pub fn compile(pat: []const u8) Error!Regex {
         if (pat.len == 0) return error.Bad;
@@ -273,7 +274,7 @@ pub const Regex = struct {
     /// `^a|^b` as `^(a|b)` in `out`, when the pattern is an alternation at
     /// its top level and every branch starts with `^`; null when it is not
     /// one, or no branch does.
-    fn anchoredAlternation(pat: []const u8, out: *[5 * max_ops + 2]u8) error{Bad}!?[]const u8 {
+    fn anchoredAlternation(pat: []const u8, out: *[5 * max_ops + 2]u8) error{ Bad, Mixed }!?[]const u8 {
         var bars: [16]usize = undefined;
         var n: usize = 0;
         var depth: usize = 0;
@@ -306,7 +307,8 @@ pub const Regex = struct {
             from = to + 1;
         }
         if (anchored == 0) return null;
-        if (anchored != n + 1) return error.Bad;
+        // Only `^` counts: `$` is mvzr's own at a branch's end (`a$|b`).
+        if (anchored != n + 1) return error.Mixed;
         var w = std.Io.Writer.fixed(out);
         w.writeAll("^(") catch return error.Bad;
         from = 0;
@@ -481,7 +483,7 @@ test "a ^ after | anchors that branch: ^def|^  finds a line that starts either w
     var both = try Regex.compile("^def|^ ");
     try std.testing.expectEqual(@as(usize, 6), (try both.find(text, 0, text.len, text.len)).?.start);
     try std.testing.expectEqual(@as(usize, 9), (try both.find(text, 7, text.len, text.len)).?.start);
-    try std.testing.expectError(error.Bad, Regex.compile("^def|x"));
+    try std.testing.expectError(error.Mixed, Regex.compile("^def|x"));
     var plain = try Regex.compile("a|b");
     try std.testing.expectEqual(@as(usize, 7), (try plain.find(text, 0, text.len, text.len)).?.start);
 }