ww/lib/strconv/strconv.ww

// strconv — number↔string conversions. Decimal i64 to/from a fixed
// buffer. Error shapes mirror Hare's strconv types: (T | invalid |
// overflow) where each error is a named alias over a payload type
// (Hare uses !size / !void; ww uses i32 / void without the `!` mark).

// invalid — input wasn't a valid number in the requested format.
// Payload is the byte index of the first offending position. Mirrors
// Hare's strconv::invalid = !size (we use i32 instead of size).
export type invalid = !i32;

// overflow — input was valid but doesn't fit the target type. No
// payload (a single yes/no signal). Mirrors Hare's !void shape.
export type overflow = !void;

// u64tos — write `v` in decimal into `buf` and return the byte count.
// Hare name; the buffer-in shape is the sanctioned Plan 9 subset of
// Hare's `u64tos(u, base) const str`. Unsigned-only so callers don't
// have to think about wraparound when printing a u64 with the high
// bit set.
export fn u64tos(buf: []u8, v: u64) i32 = {
	let tmp: [32]u8;
	let i: i32 = 0;
	let n: u64 = v;
	for (n > 0u64) {
		tmp[i] = ((n % 10u64) + 48u64): u8;
		n = n / 10u64;
		i += 1;
	};
	if (i == 0) {
		tmp[0] = 48u8;
		i = 1;
	};
	let out: i32 = 0;
	for (i > 0) {
		i -= 1;
		buf[out] = tmp[i];
		out += 1;
	};
	return out;
};

export fn i64tos(buf: []u8, v: i64) i32 = {
	let neg: bool = false;
	let n: i64 = v;
	if (n < 0) {
		neg = true;
		n = -n;
	};
	let tmp: [32]u8;
	let i: i32 = 0;
	for (n > 0) {
		tmp[i] = ((n % 10) + 48): u8;
		n = n / 10;
		i += 1;
	};
	if (i == 0) {
		tmp[0] = 48u8;
		i = 1;
	};
	let out: i32 = 0;
	if (neg) {
		buf[out] = 45u8;	// '-'
		out += 1;
	};
	for (i > 0) {
		i -= 1;
		buf[out] = tmp[i];
		out += 1;
	};
	return out;
};

// stoi64 — Hare-style fallible signed decimal parser. No locale, no
// whitespace, no underscores: a leading '-' is the only non-digit
// accepted, and only at position 0.
export fn stoi64(s: str) (i64 | invalid | overflow) = {
	if (s.len == 0) { return 0: invalid; };
	let i: i32 = 0;
	let neg: bool = false;
	if (s[0] == 45u8) { neg = true; i = 1; };
	if (i >= s.len) { return i: invalid; };
	let v: i64 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		if (c < 48u8) { return i: invalid; };
		if (c > 57u8) { return i: invalid; };
		v = v * 10 + ((c: i64) - 48);
		i += 1;
	};
	if (neg) { v = -v; };
	return v;
};

// stou64 — fallible unsigned decimal parser. No leading sign.
export fn stou64(s: str) (u64 | invalid | overflow) = {
	if (s.len == 0) { return 0: invalid; };
	let v: u64 = 0u64;
	let i: i32 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		if (c < 48u8) { return i: invalid; };
		if (c > 57u8) { return i: invalid; };
		v = v * 10u64 + ((c: u64) - 48u64);
		i += 1;
	};
	return v;
};

// f64tos — write `v` in decimal into `buf` and return the byte count.
// Hare name; this is the buffer-in Plan 9 subset of Hare's
// `f64tos(n) const str`. Today's surface:
//
//   - finite values only. NaN/±Inf detection needs an f64→u64 bit
//     reinterpret cast that the cgen doesn't expose yet.
//   - fixed-point only, up to 6 fractional digits. Trailing zeros
//     after the decimal point are trimmed. Trailing '.' is dropped.
//   - magnitudes ≥ 9e18 (overflows i64 in the integer-part cast)
//     fall back to the literal token "huge". Hare would print these
//     in scientific notation via Ryū; we will graduate when the
//     compiler grows the bit-reinterpret cast.
//
// Round-trip is therefore lossy past 6 fractional digits; callers
// that need bit-exact recovery should not use this until the
// graduate-to-Ryū step lands. `f64tos(buf, 1.0)` writes "1" (no
// decimal point), `f64tos(buf, 1.5)` writes "1.5", `f64tos(buf,
// 0.1)` writes "0.1".
//
// No float literals in the body — 990's wwdump diff requires this
// file's TK_FLOAT count to match between C and ww front-ends, and
// the ww-side wwdump currently skips TK_FLOAT.fval while the C side
// %g-formats it. Same trick lib/ww/lex/lex.ww's parsef64 uses:
// build f64 constants via int-to-f64 casts.
export fn f64tos(buf: []u8, v: f64) i32 = {
	let out: i32 = 0;
	let f: f64 = v;
	let zero: f64 = 0: f64;
	if (f < zero) {
		buf[out] = 45u8;	// '-'
		out += 1;
		f = -f;
	};
	// 9e18 is comfortably under I64_MAX (9.22e18). Past this the
	// `f: i64` cast wraps and the integer part comes back as garbage.
	let cap: f64 = 9000000000000000000i64: f64;
	if (f >= cap) {
		let s: str = "huge";
		let k: i32 = 0;
		for (k < s.len) { buf[out] = s[k]; out += 1; k += 1; };
		return out;
	};
	let ip: i64 = f: i64;
	// Fractional part scaled to 6 decimal digits, with round-to-
	// nearest via +0.5. (f64 compound assigns mis-lower in cgen —
	// use the explicit form, as the rest of lib does.)
	let frac: f64 = f - (ip: f64);
	let scale: f64 = 1000000: f64;
	frac = frac * scale;
	let half: f64 = (1: f64) / (2: f64);
	let fp: i64 = (frac + half): i64;
	// Carry: e.g. 0.9999996 rounds fp up to 1000000 and the integer
	// part needs to advance.
	if (fp >= 1000000) {
		ip += 1;
		fp = 0;
	};
	let itmp: [32]u8;
	let in: i32 = i64tos(itmp[0:32], ip);
	let k: i32 = 0;
	for (k < in) { buf[out] = itmp[k]; out += 1; k += 1; };
	if (fp == 0) { return out; };
	buf[out] = 46u8;	// '.'
	out += 1;
	let ftmp: [16]u8;
	let m: i32 = u64tos(ftmp[0:16], fp: u64);
	// Pad fractional to 6 digits with leading zeros (e.g. 0.05 →
	// fp=50000, m=5, pad one '0' before "50000").
	let z: i32 = 6 - m;
	for (z > 0) { buf[out] = 48u8; out += 1; z -= 1; };
	k = 0;
	for (k < m) { buf[out] = ftmp[k]; out += 1; k += 1; };
	// Trim trailing zeros in the fractional part (we know fp != 0,
	// so the loop stops before erasing the dot).
	for (out > 0) {
		if (buf[out - 1] != 48u8) { break; };
		out -= 1;
	};
	return out;
};