ww/lib/regex/regex.ww

// regex — POSIX extended regular expressions. Port of
// ref/hare/regex/regex.ha. Fold 1 = the data model; fold 2a = the
// compile() literal core (lit/any/match + the leading skip); every
// other metacharacter arm, exec / find / replace are DEFERRED to
// later folds.
//
// Two fold-1 constructs are held back behind filed compiler/fidelity
// gaps (see the charclass_map and finish() sites below):
//   - charclass_map (regex.ha:74-87) — a module-level const slice of
//     (str, *fn(rune) bool) tuples. Blocked on the array-literal→slice
//     element-coercion checker gap (#25; type.c:402-404 #258 borrow
//     uses exact type_eq, no element decay).
//   - finish() (regex.ha:96-102) — frees; ww is a no-free runtime (#27).
package regex;

import strings;
import encoding.utf8;

// ref/hare/regex/regex.ha:14 — an error string describing a compilation
// error.
export type error = !str;

// ref/hare/regex/regex.ha:16-30.
export type inst_lit = rune;
export type inst_charset = struct { idx: size, is_positive: bool };
export type inst_any = void;
export type inst_split = size;
export type inst_jump = size;
export type inst_skip = void;
export type inst_match = bool;
export type inst_groupstart = size;
export type inst_groupend = void;
export type inst_repeat = struct {
	id: size,
	origin: size,
	min: (void | size),
	max: (void | size),
};

// ref/hare/regex/regex.ha:32-35.
export type inst = (inst_lit | inst_any | inst_split | inst_jump |
	inst_skip | inst_match | inst_charset |
	inst_groupstart | inst_groupend |
	inst_repeat);

// The resulting match of a [[regex]] applied to a string.
//
// The first [[capture]] corresponds to the implicit zeroth capture
// group, i.e. the whole expression.
//
// The rest of the [[capture]]s correspond to the rest of the capture
// groups, i.e. the sub-expressions.
// ref/hare/regex/regex.ha:44.
export type result = []capture;

// A (sub)match corresponding to a regular expression's capture group.
// ref/hare/regex/regex.ha:47-53.
export type capture = struct {
	content: str,
	start: size,
	start_bytesize: size,
	end: size,
	end_bytesize: size,
};

// ref/hare/regex/regex.ha:68-72.
export type charset = [](charset_lit_item | charset_range_item |
	charset_class_item);
export type charset_lit_item = rune;
export type charset_range_item = (u32, u32);
export type charset_class_item = (str, *fn(c: rune) bool);

// ref/hare/regex/regex.ha:74-87 — charclass_map: the const
// [](str, *fn(rune) bool) table mapping POSIX class tokens to the
// matching ascii predicate. DEFERRED: the array-literal→slice
// assignability check (type.c:402-404, the #258 borrow) compares
// element types with exact type_eq and applies NO element coercion, so
// the literal `[(":alnum:]", &ascii.isalnum), ...]` (typed
// `[N](untyped_str, *fn(rune) bool)`) is rejected against the declared
// `[](str, *fn(rune) bool)`. Minimal repro: `let xs: [](size, size) =
// [(1, 2)];`. Reshaping to a fixed `[12](...)` array would compile but
// is an unfaithful workaround (CLAUDE.md rule-7), so the table — and
// the `import ascii;` it needs — land with the consuming fold (compile)
// once the checker gap is fixed.

// ref/hare/regex/regex.ha:89-93.
export type regex = struct {
	insts: []inst,
	charsets: []charset,
	n_reps: size,
};

// Frees resources associated with a [[regex]].
//
// ref/hare/regex/regex.ha:96-102 frees re.insts / each charset /
// re.charsets. ww is a no-free runtime (rt/alloc.s:30 — rt_free is a
// no-op; the bump allocator can't reclaim, process-exit does), so the
// faithful ww body drops the frees, matching how the port drops every
// Hare free(). Kept for API parity with the Hare surface. Temporary
// empty body: #27 makes the free() builtin compile to a documented
// no-op, after which this ports the Hare frees VERBATIM (the no-op
// builtin reclaims nothing, same end state).
export fn finish(re: *regex) void = { };

// Compiles a regular expression string into a [[regex]].
//
// ref/hare/regex/regex.ha:227-263. Fold 2a ports the literal core:
// inst_lit / inst_any / inst_match plus the leading inst_skip
// (regex.ha:261-263 — unanchored exec depends on it). Every other
// metacharacter arm is one loud not-yet-ported error — the explicit
// fold boundary; falling to literal would be a silent semantic lie.
// State serving only the deferred arms is dropped with them:
// jump_idxs (ha:241-248), the bracket quad (ha:249-252),
// was_prev_rune_pipe / group_level / capture_idx (ha:253-256).
//
// Hare's `defer if (!ok) free(...)` cleanup (ha:231-237) is omitted:
// ww has no `defer if` (cf lib/strings/strings.ww:85) and is a no-free
// runtime (#27) — the port drops every Hare free().
export fn compile(expr: str) (regex | error | nomem) = {
	// Hare `let insts: []inst = [];` — a bare ww slice declaration
	// zeroes the header (cgen.c:9836 no-rhs multi-word composite
	// zero-fill, symmetric in cgenstmt.ww).
	let insts: []inst;
	let charsets: []charset;	// stays empty until the '[' fold
	let iter: strings.iterator = strings.iter(expr);
	let r_idx: size = 0;
	let n_reps: size = 0;

	for (true) {
		let next: (rune | utf8.done) = strings.next(&iter);

		if (r_idx == 0 && next is rune && (next as rune) != '^') {
			// Bare append: ww append returns void; Hare's
			// `append(...)?` nomem propagation is filed #36.
			// Hare appends the bare type name (`inst_skip`) as
			// the void-variant value; in ww that resolves as a
			// symbol ref (cf the alloc(T) rule), so void
			// variants go through a typed let.
			let sk: inst_skip;
			let v: inst = sk;
			append(insts, v);
		};

		// regex.ha:277-284 minus the group_level check (it rides
		// the deferred '(' arm's state).
		let r: rune = match (next) {
		case utf8.done => break;
		case let x: rune => yield x;
		};

		switch (r) {
		case '.': {		// regex.ha:460-461
			let av: inst_any;
			let v: inst = av;
			append(insts, v);
		};
		case ']':		// regex.ha:315-316 — literal outside a bracket
			append(insts, (r: inst_lit));
		case '\\', '^', '$', '[', '(', ')', '|', '{', '?', '*', '+':
			// fold-2a boundary: regex.ha:286-459 arms deferred.
			return "regex: metacharacter not yet ported": error;
		case:			// regex.ha:462-463
			append(insts, (r: inst_lit));
		};
		r_idx += 1;
	};

	// regex.ha:475-477. `$` appends true: inst_match — deferred, so
	// the guard can only see no-match today; kept verbatim.
	if (insts.len == 0 || !(insts[insts.len - 1] is inst_match)) {
		append(insts, (false: inst_match));
	};
	// regex.ha:479-484. The alternation fixup (ha:470-473) drops with
	// jump_idxs.
	return regex {
		insts = insts,
		charsets = charsets,
		n_reps = n_reps,
	};
};