hare

commit 413a5e43a6bfa6f7d7e0fca27a367e92c0ede2da
parent d1cf40891bc75c036e9ce60ab5bc9cfad32baf74
Author: Max Schillinger <max@mxsr.de>
Date:   Sun, 14 Apr 2024 22:44:24 +0200

regex: implement whole-expression alternation

When a pattern contains a `|` outside of a capture group, make this the
jump position of a whole-expression alternation.

Signed-off-by: Max Schillinger <max@mxsr.de>
Implements: https://todo.sr.ht/~sircmpwn/hare/695

Diffstat:
M
regex/+test.ha
 | 
7
++++---
M
regex/regex.ha
 | 
14
+++++++++++++-

2 files changed, 17 insertions(+), 4 deletions(-)
diff --git a/regex/+test.ha b/regex/+test.ha
@@ -577,9 +577,10 @@ fn run_rawreplace_case(
 		(`(a|ab)(bcd|c)(d|.*)`, "abcd", matchres::MATCH, 0, -1), // POSIX: (0,4)(0,2)(2,3)(3,4)
 		(`(ab|a)(c|bcd)(d|.*)`, "abcd", matchres::MATCH, 0, -1), // POSIX: (0,4)(0,2)(2,3)(3,4)
 		(`(ab|a)(bcd|c)(d|.*)`, "abcd", matchres::MATCH, 0, -1), // POSIX: (0,4)(0,2)(2,3)(3,4)
-		// TODO: whole-expression alternation
-		// (`ab|cd`, "abc", matchres::MATCH, 0, -1),
-		// (`ab|cd`, "abcd", matchres::MATCH, 0, -1),
+		// whole-expression alternation
+		(`ab|cd`, "cd", matchres::MATCH, 0, 2),
+		(`ab|cd`, "abc", matchres::MATCH, 0, 2),
+		(`ab|cd`, "abcd", matchres::MATCH, 0, 2),
 		// TODO: multiple alternation
 		// (`a|b|c|d|e`, "e", matchres::MATCH, 0, -1),
 		// (`(a|b|c|d|e)f`, "ef", matchres::MATCH, 0, -1),
diff --git a/regex/regex.ha b/regex/regex.ha
@@ -300,7 +300,12 @@ export fn compile(expr: str) (regex | error) = {
 			};
 		case '|' =>
 			append(insts, types::SIZE_MAX: inst_jump);
-			const origin = find_last_groupstart(&insts)? + 1;
+			const origin = match (find_last_groupstart(&insts)) {
+			case error =>
+				yield 0z;
+			case let sz: size =>
+				yield sz + 1;
+			};
 			const newinst = (len(insts) + 1): inst_split;
 			insert(insts[origin], newinst);
 			curr_alt_jump_idx = (len(insts) - 1): int;
@@ -394,6 +399,13 @@ export fn compile(expr: str) (regex | error) = {
 		r_idx += 1;
 	};
 
+	// handle whole expression alternation
+	if (curr_alt_jump_idx != -1) {
+		assert(insts[curr_alt_jump_idx] is inst_jump);
+		insts[curr_alt_jump_idx] = len(insts): inst_jump;
+		curr_alt_jump_idx = -1;
+	};
+
 	append(insts, anchored: inst_match);
 
 	return regex {