ww/selfhost/test/smoke.combined.ww

// MODULE: os
// os — process and filesystem facade. The body of each call lands
// either in libwwrt.a (rt_syscall trampoline) or libc bindings,
// depending on how the program was linked.

@symbol("rt_syscall") fn syscall0(num: nr) i64;
@symbol("rt_syscall") fn syscall1(num: nr, a: i64) i64;
@symbol("rt_syscall") fn syscall2(num: nr, a: i64, b: i64) i64;
@symbol("rt_syscall") fn syscall3(num: nr, a: i64, b: i64, c: i64) i64;
@symbol("rt_syscall") fn syscall4(num: nr, a: i64, b: i64, c: i64, d: i64) i64;

@symbol("rt_alloc") fn alloc(n: u64) *void;
@symbol("rt_free")  fn free(p: *void, n: u64) void;
@symbol("rt_abort") fn abort(msg: str) void;

// Hare-style runtime check. Caller passes a message that's printed
// to stderr before exit(1).
export fn assert(cond: bool, msg: str) void = {
	if (!cond) { abort(msg); };
};

// Linux amd64 syscall numbers. Internal to this module — passed as
// the first arg of syscall0..4 via libwwrt's rt_syscall trampoline.
// `nr` is the type so the call sites can't accidentally pass an
// arbitrary i64 (`syscall1(0i64, ...)` no longer typechecks).
type nr = enum i64 {
	READ        = 0,
	WRITE       = 1,
	OPEN        = 2,
	CLOSE       = 3,
	LSEEK       = 8,
	ACCESS      = 21,
	DUP2        = 33,
	GETPID      = 39,
	FORK        = 57,
	EXECVE      = 59,
	EXIT        = 60,
	WAIT4       = 61,
	UNLINK      = 87,
	GETCWD      = 79,
	GETDENTS64  = 217,
};

// open(2) flags. Linux values, matching <fcntl.h>. Hare names them
// `fs::flag::RDONLY` etc; we use the same leaf names so callers say
// `os.flag.RDONLY` and `os.flag.WRONLY | os.flag.CREATE`.
export type flag = enum i32 {
	RDONLY  = 0,
	WRONLY  = 1,
	RDWR    = 2,
	CREATE  = 64,	// 0x40
	TRUNC   = 512,	// 0x200
};

// lseek(2) whence. Hare names it `io::whence`.
export type whence = enum i32 {
	SET = 0,
	CUR = 1,
	END = 2,
};

export fn exit(code: i32) void = {
	syscall1(nr.EXIT, code: i64);
};

// Raw, non-fallible primitives. These return Linux's int conventions
// (negative = -errno, non-negative = bytes/fd/etc). Callers wanting a
// Hare-style fallible API use the wrappers below.
export fn write(fd: i32, buf: *u8, n: u64) i64 = {
	return syscall3(nr.WRITE, fd: i64, buf: i64, n: i64);
};

export fn read(fd: i32, buf: *u8, n: u64) i64 = {
	return syscall3(nr.READ, fd: i64, buf: i64, n: i64);
};

export fn close(fd: i32) i32 = {
	return syscall1(nr.CLOSE, fd: i64): i32;
};

// dup2(2): make `newfd` refer to the same description as `oldfd`,
// closing `newfd` first if open. Returns `newfd` on success or a
// negative errno. Used by w6c_ww to redirect stdout into an output
// file without changing the cgen emit path.
export fn dup2(oldfd: i32, newfd: i32) i32 = {
	return syscall2(nr.DUP2, oldfd: i64, newfd: i64): i32;
};

// Fallible wrappers. The error variant is `oserror` (an i64 carrying
// -errno). The sum type makes success/failure explicit and lets
// callers `?` the result up the stack.
export fn tryread(fd: i32, buf: *u8, n: u64) (i64 | oserror) = {
	let r: i64 = read(fd, buf, n);
	if (r < 0) { return r: oserror; };
	return r;
};

export fn trywrite(fd: i32, buf: *u8, n: u64) (i64 | oserror) = {
	let r: i64 = write(fd, buf, n);
	if (r < 0) { return r: oserror; };
	return r;
};

// open — Linux open(2). Path must be NUL-terminated; callers using ww
// `str` must ensure the bytes are followed by a 0 byte (literals are,
// arena-copied paths usually are by construction). Returns -errno on
// failure, fd otherwise. Higher-level callers prefer `tryopen`.
export fn open(path: *u8, flags: flag, mode: i32) i32 = {
	return syscall3(nr.OPEN, path: i64, (flags as i32): i64, mode: i64): i32;
};

export fn tryopen(path: *u8, flags: flag, mode: i32) (i32 | oserror) = {
	let fd: i32 = open(path, flags, mode);
	if (fd < 0) { return fd: i64: oserror; };
	return fd;
};

// lseek — set/inspect the fd's position. Returns the new offset or
// a negative errno. We use this for fstat-free file-size discovery
// (open ⇒ lseek to end ⇒ lseek back).
export fn lseek(fd: i32, off: i64, w: whence) i64 = {
	return syscall3(nr.LSEEK, fd: i64, off, (w as i32): i64);
};

// oserror — the underlying errno from a failed syscall, as a
// negative i64 (Linux's int convention; e.g. -2 = ENOENT). The
// `!`-flagged alias makes ?-propagation pick this variant as the
// error half of any (T | oserror) shape. Hare's analogue is
// errors::errno carried inside io::error.
export type oserror = !i64;

// filesize — byte length of an open fd via lseek-to-end-and-back.
export fn filesize(fd: i32) (i64 | oserror) = {
	let end: i64 = lseek(fd, 0i64, whence.END);
	if (end < 0) { return end: oserror; };
	let r: i64 = lseek(fd, 0i64, whence.SET);
	if (r < 0) { return r: oserror; };
	return end;
};

// readall — keep reading until `n` bytes have arrived or the fd
// closes early. Hare name (io::readall); the buffer is caller-
// supplied, matching the Plan 9 subset convention.
export fn readall(fd: i32, buf: *u8, n: u64) (i64 | oserror) = {
	let got: u64 = 0u64;
	for (got < n) {
		let r: i64 = read(fd, buf + got, n - got);
		if (r < 0) { return r: oserror; };
		if (r == 0) { return got: i64; };	// short read: caller decides
		got += r: u64;
	};
	return got: i64;
};

// writeall — keep writing until `n` bytes have been accepted or the
// fd refuses progress. Hare name (io::writeall).
export fn writeall(fd: i32, buf: *u8, n: u64) (i64 | oserror) = {
	let sent: u64 = 0u64;
	for (sent < n) {
		let r: i64 = write(fd, buf + sent, n - sent);
		if (r < 0) { return r: oserror; };
		if (r == 0) { return sent: i64; };
		sent += r: u64;
	};
	return sent: i64;
};

// ---- process and filesystem helpers used by the `ww` driver ----------

// access(2): returns 0 if the file is reachable, negative errno
// otherwise. mode is the bitset described in <unistd.h> (F_OK=0).
export fn access(path: *u8, mode: i32) i32 = {
	return syscall2(nr.ACCESS, path: i64, mode: i64): i32;
};

// remove — unlink(2). Hare name; the underlying syscall is unlink(2).
export fn remove(path: *u8) i32 = {
	return syscall1(nr.UNLINK, path: i64): i32;
};

// getpid(2). Used by the driver to mint unique scratch paths.
export fn getpid() i32 = {
	return syscall0(nr.GETPID): i32;
};

// fork(2): 0 in the child, child pid in the parent, negative errno
// on failure.
export fn fork() i32 = {
	return syscall0(nr.FORK): i32;
};

// execve(2): on success, does not return.
export fn execve(path: *u8, argv: **u8, envp: **u8) i32 = {
	return syscall3(nr.EXECVE, path: i64, argv: i64, envp: i64): i32;
};

// wait4(2): wait for `pid` (or any child if -1), store status in
// `*status`, return the pid that ended (or negative errno).
export fn wait4(pid: i32, status: *i32, options: i32, rusage: *void) i32 = {
	return syscall4(nr.WAIT4, pid: i64, status: i64,
	    options: i64, rusage: i64): i32;
};

// getcwd(2) — Linux flavour. Writes the NUL-terminated cwd into `buf`
// and returns the number of bytes written (including the NUL), or a
// negative errno. The driver uses it to expand `.` to the cwd's
// basename for `ww build` / `ww test`.
export fn getcwd(buf: *u8, n: u64) i64 = {
	return syscall2(nr.GETCWD, buf: i64, n: i64);
};

// getdents64(2) — Linux directory enumeration. The fd must be opened
// with O_RDONLY on a directory. `buf` receives a packed sequence of
// linux_dirent64 records:
//
//   struct linux_dirent64 {
//       u64  d_ino;       // 0..7
//       i64  d_off;       // 8..15
//       u16  d_reclen;    // 16..17  — total bytes for this record
//       u8   d_type;      // 18      — DT_REG/DT_DIR/...
//       u8   d_name[];    // 19..    — NUL-terminated name + padding
//   };
//
// Returns bytes written into `buf` (advance by d_reclen to walk),
// 0 at end-of-directory, or a negative errno.
export fn getdents64(fd: i32, buf: *u8, n: u64) i64 = {
	return syscall3(nr.GETDENTS64, fd: i64, buf: i64, n: i64);
};

// MODULE: strings
// strings — operations over the immutable str type ({ *u8, len }).
// Mirrors Hare's strings::; `len` and `is-empty` aren't functions
// (callers use `s.len` and `s.len == 0` directly).

use os;

// compare — bytewise three-way comparison: negative if a<b, 0 if equal,
// positive if a>b. Matches Hare's strings::compare. ASCII-order, not
// locale-aware. Callers that just need equality use `compare(a, b) == 0`.
export fn compare(a: str, b: str) i32 = {
	let n: i32 = a.len;
	if (b.len < n) { n = b.len; };
	let i: i32 = 0;
	for (i < n) {
		if (a[i] != b[i]) { return (a[i]: i32) - (b[i]: i32); };
		i += 1;
	};
	return a.len - b.len;
};

export fn hasprefix(s: str, p: str) bool = {
	if (p.len > s.len) { return false; };
	let i: i32 = 0;
	for (i < p.len) {
		if (s[i] != p[i]) { return false; };
		i += 1;
	};
	return true;
};

export fn hassuffix(s: str, suf: str) bool = {
	if (suf.len > s.len) { return false; };
	let off: i32 = s.len - suf.len;
	let i: i32 = 0;
	for (i < suf.len) {
		if (s[off + i] != suf[i]) { return false; };
		i += 1;
	};
	return true;
};

// byteindex — first byte position of `needle` in `s`. Mirrors Hare's
// strings::byteindex: a single-codepoint rune scans for the byte that
// encodes it (ASCII only here — multi-byte UTF-8 awaits utf8 encode),
// a str needle scans for the substring. Returns void if absent.
export fn byteindex(s: str, needle: (str | rune)) (i32 | void) = {
	match (needle) {
	case let r: rune => {
		let c: u8 = r: u8;
		let i: i32 = 0;
		for (i < s.len) {
			if (s[i] == c) { return i; };
			i += 1;
		};
		return;
	};
	case let sub: str => {
		if (sub.len == 0) { return 0; };
		if (sub.len > s.len) { return; };
		let last: i32 = s.len - sub.len;
		let i: i32 = 0;
		for (i <= last) {
			let j: i32 = 0;
			let ok: bool = true;
			for (j < sub.len) {
				if (s[i + j] != sub[j]) { ok = false; j = sub.len; }
				else { j += 1; };
			};
			if (ok) { return i; };
			i += 1;
		};
		return;
	};
	};
	return;
};

// contains — true iff `sub` appears in `s`. Mirrors Hare's
// strings::contains shape (byte-wise on the str-needle case).
export fn contains(s: str, sub: str) bool = {
	let r: (i32 | void) = byteindex(s, sub);
	match (r) {
	case let i: i32 => return true;
	case void => return false;
	};
	return false;
};

// concat — joins two strings into a fresh str. Caller owns the
// returned str's storage; release via `os.free(r.ptr, r.len)`. Mirrors
// Hare's strings::concat shape.
export fn concat(a: str, b: str) str = {
	let total: i32 = a.len + b.len;
	let buf: *u8 = os.alloc(total: u64): *u8;
	let i: i32 = 0;
	for (i < a.len) { buf[i] = a[i]; i += 1; };
	let j: i32 = 0;
	for (j < b.len) { buf[a.len + j] = b[j]; j += 1; };
	let r: str;
	r.ptr = buf;
	r.len = total;
	return r;
};

// dup — duplicate a string into a fresh allocation. Caller owns the
// returned str's storage; release via `os.free(r.ptr, r.len)`. Mirrors
// Hare's strings::dup shape — Hare returns `(str | nomem)`, ww doesn't
// have nomem (os.alloc aborts on OOM), so we return plain `str`.
//
// Empty input yields a `{nil, 0}` str — Hare returns the static empty
// string; same observable result.
export fn dup(s: str) str = {
	let r: str;
	r.ptr = nil;
	r.len = 0;
	if (s.len == 0) { return r; };
	let buf: *u8 = os.alloc(s.len: u64): *u8;
	let i: i32 = 0;
	for (i < s.len) { buf[i] = s[i]; i += 1; };
	r.ptr = buf;
	r.len = s.len;
	return r;
};

// rbyteindex — last byte position of `needle` in `s`. Mirrors Hare's
// strings::rbyteindex. Rune needle scans for the byte that encodes it
// (ASCII only); str needle scans for the substring. Empty str needle
// matches at s.len.
export fn rbyteindex(s: str, needle: (str | rune)) (i32 | void) = {
	match (needle) {
	case let r: rune => {
		let c: u8 = r: u8;
		let i: i32 = s.len - 1;
		for (i >= 0) {
			if (s[i] == c) { return i; };
			i -= 1;
		};
		return;
	};
	case let sub: str => {
		if (sub.len == 0) { return s.len; };
		if (sub.len > s.len) { return; };
		let i: i32 = s.len - sub.len;
		for (i >= 0) {
			let j: i32 = 0;
			let ok: bool = true;
			for (j < sub.len) {
				if (s[i + j] != sub[j]) { ok = false; j = sub.len; }
				else { j += 1; };
			};
			if (ok) { return i; };
			i -= 1;
		};
		return;
	};
	};
	return;
};

// sub — borrowed substring `s[start..end]`. Mirrors Hare's
// strings::sub. Caller must ensure 0 <= start <= end <= s.len; out-of-
// range indices are clamped silently here, where Hare aborts.
export fn sub(s: str, start: i32, end: i32) str = {
	let lo: i32 = start;
	let hi: i32 = end;
	if (lo < 0)     { lo = 0; };
	if (hi > s.len) { hi = s.len; };
	if (hi < lo)    { hi = lo; };
	let r: str;
	r.ptr = s.ptr + (lo: u64);
	r.len = hi - lo;
	return r;
};

// trimprefix — `s` with `pre` stripped from the front, or `s`
// unchanged if it doesn't start with `pre`. Returns a borrowed view.
// Mirrors Hare's strings::trimprefix.
export fn trimprefix(s: str, pre: str) str = {
	if (!hasprefix(s, pre)) { return s; };
	let r: str;
	r.ptr = s.ptr + (pre.len: u64);
	r.len = s.len - pre.len;
	return r;
};

// trimsuffix — `s` with `suf` stripped from the end, or `s` unchanged
// if it doesn't end with `suf`. Returns a borrowed view. Mirrors
// Hare's strings::trimsuffix.
export fn trimsuffix(s: str, suf: str) str = {
	if (!hassuffix(s, suf)) { return s; };
	let r: str;
	r.ptr = s.ptr;
	r.len = s.len - suf.len;
	return r;
};

// ltrimbyte / rtrimbyte / trimbyte — strip occurrences of a single
// byte from the left, right, or both ends. Returns a borrowed view.
// Hare's strings::ltrim / rtrim / trim take a rune varargs set; ww's
// subset takes a single byte (the common ASCII case).
export fn ltrimbyte(s: str, c: u8) str = {
	let i: i32 = 0;
	for (i < s.len) {
		if (s[i] != c) { break; };
		i += 1;
	};
	let r: str;
	r.ptr = s.ptr + (i: u64);
	r.len = s.len - i;
	return r;
};

export fn rtrimbyte(s: str, c: u8) str = {
	let n: i32 = s.len;
	for (n > 0) {
		if (s[n - 1] != c) { break; };
		n -= 1;
	};
	let r: str;
	r.ptr = s.ptr;
	r.len = n;
	return r;
};

export fn trimbyte(s: str, c: u8) str = {
	return rtrimbyte(ltrimbyte(s, c), c);
};

// MODULE: strconv
// strconv — number↔string conversions.
//
// Mirrors Hare's strconv:: surface. The *tos functions return a
// `const str` view into a module-level buffer that is overwritten on
// the next call to the same function; callers must copy the bytes if
// they need to outlive the next invocation. See [[strings.dup]] to
// duplicate. Matches Hare's strconv::*tos semantics.

use os;
use strings;

// 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.
export type invalid = !i32;

// overflow — input was valid but doesn't fit the target type.
// Mirrors Hare's strconv::overflow = !void.
export type overflow = !void;

// error — any error from a strconv call. Mirrors Hare's strconv::error.
export type error = !(invalid | overflow);

// base — numeric base for parsing/formatting. Mirrors Hare's
// `strconv::base` (Hare uses `enum uint`; we pick `enum i32` since
// the underlying parse/format loops index with i32).
//
// HEX is an alias for HEX_UPPER; HEX_LOWER is a pseudo-base that
// produces lowercase a-f digits.
export type base = enum i32 {
	DEFAULT   = 0,
	BIN       = 2,
	OCT       = 8,
	DEC       = 10,
	HEX_UPPER = 16,
	HEX       = 16,
	HEX_LOWER = 17,
};

fn basenum(b: base) i64 = {
	if (b == base.BIN)       { return 2;  };
	if (b == base.OCT)       { return 8;  };
	if (b == base.HEX)       { return 16; };
	if (b == base.HEX_UPPER) { return 16; };
	if (b == base.HEX_LOWER) { return 16; };
	return 10;	// DEC and DEFAULT
};

fn basedigit(d: i64, b: base) u8 = {
	if (d < 10) { return (d + 48): u8; };
	let off: i64 = d - 10;
	if (b == base.HEX_LOWER) { return (off + 97): u8; };
	return (off + 65): u8;
};

// u64tos — convert v to a base-b numeric string. Returns a view into
// `u64tos_buf` which is overwritten on the next call. Matches Hare's
// strconv::u64tos.
let u64tos_buf: [65]u8;

export fn u64tos(v: u64, b: base) str = {
	let nb: u64 = basenum(b): u64;
	let tmp: [65]u8;
	let i: i32 = 0;
	let n: u64 = v;
	if (n == 0u64) { tmp[0] = 48u8; i = 1; };
	for (n > 0u64) {
		let d: i64 = (n % nb): i64;
		tmp[i] = basedigit(d, b);
		n = n / nb;
		i += 1;
	};
	let out: i32 = 0;
	for (i > 0) {
		i -= 1;
		u64tos_buf[out] = tmp[i];
		out += 1;
	};
	let r: str;
	r.ptr = &u64tos_buf[0];
	r.len = out;
	return r;
};

// i64tos — convert v to a base-b numeric string. Returns a view into
// `i64tos_buf` which is overwritten on the next call. Independent
// buffer from u64tos so i64tos's own call to u64tos doesn't clobber
// the in-flight result. Matches Hare's strconv::i64tos.
let i64tos_buf: [66]u8;

export fn i64tos(v: i64, b: base) str = {
	let neg: bool = false;
	let n: i64 = v;
	if (n < 0) { neg = true; n = -n; };
	let nb: i64 = basenum(b);
	let tmp: [65]u8;
	let i: i32 = 0;
	if (n == 0) { tmp[0] = 48u8; i = 1; };
	for (n > 0) {
		let d: i64 = n % nb;
		tmp[i] = basedigit(d, b);
		n = n / nb;
		i += 1;
	};
	let out: i32 = 0;
	if (neg) { i64tos_buf[out] = 45u8; out += 1; };	// '-'
	for (i > 0) {
		i -= 1;
		i64tos_buf[out] = tmp[i];
		out += 1;
	};
	let r: str;
	r.ptr = &i64tos_buf[0];
	r.len = out;
	return r;
};

export fn i32tos(v: i32, b: base) str = { return i64tos(v: i64, b); };
export fn i16tos(v: i16, b: base) str = { return i64tos(v: i64, b); };
export fn i8tos(v: i8,   b: base) str = { return i64tos(v: i64, b); };

export fn u32tos(v: u32, b: base) str = { return u64tos(v: u64, b); };
export fn u16tos(v: u16, b: base) str = { return u64tos(v: u64, b); };
export fn u8tos(v: u8,   b: base) str = { return u64tos(v: u64, b); };

// digval — value of digit byte `c` under base `b`, or -1 if not a
// valid digit. Letters are accepted case-insensitively under HEX /
// HEX_UPPER; only lowercase under HEX_LOWER.
fn digval(c: u8, b: base) i32 = {
	if (c >= 48u8) { if (c <= 57u8) { return (c - 48u8): i32; }; };
	if (b == base.HEX_LOWER) {
		if (c >= 97u8) { if (c <= 102u8) { return ((c - 97u8) + 10u8): i32; }; };
		return -1;
	};
	if (c >= 65u8) { if (c <= 70u8)  { return ((c - 65u8) + 10u8): i32; }; };
	if (c >= 97u8) { if (c <= 102u8) { return ((c - 97u8) + 10u8): i32; }; };
	return -1;
};

// stoi64 — parse signed base-b number. Mirrors Hare's strconv::stoi64.
// No locale, no whitespace, no underscores: optional leading '-' then
// digits. Returns invalid with the offending index or overflow on
// out-of-range.
export fn stoi64(s: str, b: base) (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 nb: i32 = basenum(b): i32;
	let v: i64 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		let d: i32 = digval(c, b);
		if (d < 0)   { return i: invalid; };
		if (d >= nb) { return i: invalid; };
		v = v * (nb: i64) + (d: i64);
		i += 1;
	};
	if (neg) { v = -v; };
	return v;
};

// stou64 — parse unsigned base-b number. Mirrors Hare's strconv::stou64.
export fn stou64(s: str, b: base) (u64 | invalid | overflow) = {
	if (s.len == 0) { return 0: invalid; };
	let nb: u64 = basenum(b): u64;
	let v: u64 = 0u64;
	let i: i32 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		let d: i32 = digval(c, b);
		if (d < 0)         { return i: invalid; };
		if ((d: u64) >= nb) { return i: invalid; };
		v = v * nb + (d: u64);
		i += 1;
	};
	return v;
};

export fn stoi32(s: str, b: base) (i32 | invalid | overflow) = {
	let r = stoi64(s, b);
	match (r) {
	case let v: i64 => {
		if (v > 2147483647i64)  { return overflow{}; };
		if (v < -2147483648i64) { return overflow{}; };
		return v: i32;
	};
	case let e: invalid  => return e;
	case let e: overflow => return e;
	};
	return 0: invalid;	// unreachable; appeases the path-cov checker
};

export fn stoi16(s: str, b: base) (i16 | invalid | overflow) = {
	let r = stoi64(s, b);
	match (r) {
	case let v: i64 => {
		if (v > 32767i64)  { return overflow{}; };
		if (v < -32768i64) { return overflow{}; };
		return v: i16;
	};
	case let e: invalid  => return e;
	case let e: overflow => return e;
	};
	return 0: invalid;
};

export fn stoi8(s: str, b: base) (i8 | invalid | overflow) = {
	let r = stoi64(s, b);
	match (r) {
	case let v: i64 => {
		if (v > 127i64)  { return overflow{}; };
		if (v < -128i64) { return overflow{}; };
		return v: i8;
	};
	case let e: invalid  => return e;
	case let e: overflow => return e;
	};
	return 0: invalid;
};

export fn stou32(s: str, b: base) (u32 | invalid | overflow) = {
	let r = stou64(s, b);
	match (r) {
	case let v: u64 => {
		if (v > 4294967295u64) { return overflow{}; };
		return v: u32;
	};
	case let e: invalid  => return e;
	case let e: overflow => return e;
	};
	return 0: invalid;
};

export fn stou16(s: str, b: base) (u16 | invalid | overflow) = {
	let r = stou64(s, b);
	match (r) {
	case let v: u64 => {
		if (v > 65535u64) { return overflow{}; };
		return v: u16;
	};
	case let e: invalid  => return e;
	case let e: overflow => return e;
	};
	return 0: invalid;
};

export fn stou8(s: str, b: base) (u8 | invalid | overflow) = {
	let r = stou64(s, b);
	match (r) {
	case let v: u64 => {
		if (v > 255u64) { return overflow{}; };
		return v: u8;
	};
	case let e: invalid  => return e;
	case let e: overflow => return e;
	};
	return 0: invalid;
};

// f64tos — convert v to a decimal string. Returns owned str; release
// via os.free. Mirrors Hare's strconv::f64tos (current ww impl is
// fixed-point only, max 6 fractional digits, no NaN/Inf support —
// see graduate-to-Ryū note below).
//
// 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.
//
// 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.
let f64tos_buf: [64]u8;

export fn f64tos(v: f64) str = {
	let out: i32 = 0;
	let f: f64 = v;
	let zero: f64 = 0: f64;
	if (f < zero) {
		f64tos_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) { f64tos_buf[out] = s[k]; out += 1; k += 1; };
		let r: str;
		r.ptr = &f64tos_buf[0];
		r.len = out;
		return r;
	};
	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 intstr: str = i64tos(ip, base.DEC);
	let k: i32 = 0;
	for (k < intstr.len) { f64tos_buf[out] = intstr.ptr[k]; out += 1; k += 1; };
	if (fp != 0) {
		f64tos_buf[out] = 46u8;	// '.'
		out += 1;
		let fracstr: str = u64tos(fp: u64, base.DEC);
		// Pad fractional to 6 digits with leading zeros (e.g. 0.05 →
		// fp=50000, fracstr="50000", pad one '0' before).
		let z: i32 = 6 - fracstr.len;
		for (z > 0) { f64tos_buf[out] = 48u8; out += 1; z -= 1; };
		k = 0;
		for (k < fracstr.len) { f64tos_buf[out] = fracstr.ptr[k]; out += 1; k += 1; };
		// Trim trailing zeros in the fractional part.
		for (out > 0) {
			if (f64tos_buf[out - 1] != 48u8) { break; };
			out -= 1;
		};
	};
	let r: str;
	r.ptr = &f64tos_buf[0];
	r.len = out;
	return r;
};

// strerror — convert an strconv error to a user-readable string.
// Returns owned str; release via os.free. Mirrors Hare's
// strconv::strerror.
export fn strerror(e: error) str = {
	match (e) {
	case let v: invalid  => return strings.dup("input is not a valid number");
	case let v: overflow => return strings.dup("input number doesn't fit target type");
	};
	return strings.dup("");
};

// MODULE: ascii
// ascii — rune-class predicates and case folding for the ASCII range.
// Matches Hare's ascii::isdigit family (rune-taking signature). Runes
// outside 0..127 always answer `false`. The lexer hot path uses these
// inline; they are expected to inline to a couple of compares.

export fn isdigit(c: rune) bool = {
	if (c < 48) { return false; };
	if (c > 57) { return false; };
	return true;
};

export fn isupper(c: rune) bool = {
	if (c < 65) { return false; };
	if (c > 90) { return false; };
	return true;
};

export fn islower(c: rune) bool = {
	if (c < 97) { return false; };
	if (c > 122) { return false; };
	return true;
};

export fn isalpha(c: rune) bool = {
	if (isupper(c)) { return true; };
	return islower(c);
};

export fn isalnum(c: rune) bool = {
	if (isalpha(c)) { return true; };
	return isdigit(c);
};

// isspace — the C/Hare set: space, tab, NL, VT, FF, CR.
export fn isspace(c: rune) bool = {
	if (c == 32) { return true; };	// ' '
	if (c == 9)  { return true; };	// '\t'
	if (c == 10) { return true; };	// '\n'
	if (c == 11) { return true; };	// '\v'
	if (c == 12) { return true; };	// '\f'
	if (c == 13) { return true; };	// '\r'
	return false;
};

export fn isxdigit(c: rune) bool = {
	if (isdigit(c)) { return true; };
	if (c >= 65) {
		if (c <= 70) { return true; };	// 'A'..'F'
	};
	if (c >= 97) {
		if (c <= 102) { return true; };	// 'a'..'f'
	};
	return false;
};

// valid — `c` is in the 0..127 ASCII range.
export fn valid(c: rune) bool = {
	if (c < 0)   { return false; };
	if (c > 127) { return false; };
	return true;
};

// validstr — every byte in `s` is ASCII (0..127).
export fn validstr(s: str) bool = {
	let i: i32 = 0;
	for (i < s.len) {
		// High-bit test rather than `> 127u8`; both cgens lower
		// the bitwise form identically. The `> u8` form picks
		// JA vs JG depending on signed/unsigned dispatch.
		if ((s[i] & 128u8) != 0u8) { return false; };
		i += 1;
	};
	return true;
};

// iscntrl — control chars: 0..31 and 127.
export fn iscntrl(c: rune) bool = {
	if (c >= 0) { if (c <= 31) { return true; }; };
	if (c == 127) { return true; };
	return false;
};

// isblank — space and tab.
export fn isblank(c: rune) bool = {
	if (c == 32) { return true; };	// ' '
	if (c == 9)  { return true; };	// '\t'
	return false;
};

// isprint — printable: space through '~'.
export fn isprint(c: rune) bool = {
	if (c < 32)  { return false; };
	if (c > 126) { return false; };
	return true;
};

// isgraph — printable, non-space.
export fn isgraph(c: rune) bool = {
	if (c < 33)  { return false; };
	if (c > 126) { return false; };
	return true;
};

// ispunct — printable, non-alnum, non-space.
export fn ispunct(c: rune) bool = {
	if (!isgraph(c)) { return false; };
	if (isalnum(c))  { return false; };
	return true;
};

// tolower / toupper — fold ASCII case. Non-letters pass through.
export fn tolower(c: rune) rune = {
	if (isupper(c)) { return c + 32; };
	return c;
};

export fn toupper(c: rune) rune = {
	if (islower(c)) { return c - 32; };
	return c;
};

// strcasecmp — three-way ASCII case-insensitive compare.
export fn strcasecmp(a: str, b: str) i32 = {
	let n: i32 = a.len;
	if (b.len < n) { n = b.len; };
	let i: i32 = 0;
	for (i < n) {
		let ca: rune = tolower(a[i]: rune);
		let cb: rune = tolower(b[i]: rune);
		if (ca != cb) { return (ca - cb): i32; };
		i += 1;
	};
	return a.len - b.len;
};

// MODULE: test
// selfhost/test/smoke.ww — end-to-end smoke for the selfhost path.
//
// Exercises the patterns the real ww-side compiler port will use:
//   - bump arena allocator (mem.ww shape)
//   - error idiom (T | str)
//   - struct of fn pointers + ctx pointer (the io.stream-style
//     polymorphism we use instead of interfaces)
//   - byte-level scanning that mirrors the hot path inside lex.ww
//   - strconv round-trip via the real stdlib
//
// `main` returns 42 when every check passes, 1..N on failure
// indicating which probe broke. The 990_selfhost test asserts 42.
//
// Note: only stack-local mutable state. Top-level `let` mutation
// requires a writable .data segment in w6l, which is a separate
// task; until then we exercise polymorphism via ctx pointers, which
// is what the real port wants anyway.

use os;
use strconv;
use ascii;

// --- bump arena ---------------------------------------------------------

type arena = struct {
	buf: *u8,
	off: u64,
	cap: u64,
};

// In-place init. Returning a 24-byte struct by value isn't yet
// supported in w6c (SysV requires a hidden return-slot pointer for
// structs >16 bytes), so we initialize through a pointer like the
// real compiler does today.
fn arena_init(a: *arena, buf: *u8, cap: u64) void = {
	a.buf = buf;
	a.off = 0u64;
	a.cap = cap;
};

fn arena_alloc(a: *arena, n: u64) *u8 = {
	if (n > a.cap - a.off) { return nil; };
	let p: *u8 = a.buf + a.off;
	a.off += n;
	return p;
};

// --- (i32 | str) error idiom -------------------------------------------

fn checked_div(num: i32, den: i32) (i32 | str) = {
	if (den == 0) { return "div by zero"; };
	return num / den;
};

// --- struct-of-fn-pointer polymorphism ---------------------------------
//
// A trivial "writer" abstraction: a function pointer plus a context.
// This mirrors how io.stream / Plan 9 Bio work. The ctx pointer lets
// the implementation own its own state without a global.

type counter = struct {
	n: i32,
};

type writer = struct {
	ctx:  *void,
	emit: fn(ctx: *void, b: u8) void,
};

fn count_emit(ctx: *void, b: u8) void = {
	let c: *counter = ctx: *counter;
	c.n += 1;
};

// --- byte scanner like lex.ww's hot path -------------------------------

fn count_digits(s: str) i32 = {
	let i: i32 = 0;
	let n: i32 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		if (c >= 48u8) {
			if (c <= 57u8) { n += 1; };
		};
		i += 1;
	};
	return n;
};

// --- entry --------------------------------------------------------------

export fn main() i32 = {
	// Probe 1 — arena hands out distinct pointers, refuses oversize.
	let buf: [256]u8;
	let a: arena;
	arena_init(&a, buf.ptr, 256u64);
	let p1: *u8 = arena_alloc(&a, 32u64);
	let p2: *u8 = arena_alloc(&a, 32u64);
	if (p1 == nil) { return 1; };
	if (p2 == nil) { return 2; };
	if (p1 == p2) { return 3; };
	let p3: *u8 = arena_alloc(&a, 1024u64);
	if (p3 != nil) { return 4; };

	// Probe 2 — error union both ways.
	let r_ok: (i32 | str) = checked_div(84, 2);
	let r_bad: (i32 | str) = checked_div(1, 0);
	let acc: i32 = 0;
	match (r_ok) {
		case let v: i32 => acc = v;
		case let e: str => return 5;
	};
	if (acc != 42) { return 6; };
	match (r_bad) {
		case let v: i32 => return 7;
		case let e: str => acc = e.len: i32;
	};
	if (acc != 11) { return 8; };	// len("div by zero") == 11

	// Probe 3 — struct-of-fn-pointer dispatch via ctx pointer.
	let c: counter = counter { n = 0 };
	let w: writer = writer { ctx = (&c): *void, emit = count_emit };
	w.emit(w.ctx, 65u8);
	w.emit(w.ctx, 66u8);
	w.emit(w.ctx, 67u8);
	if (c.n != 3) { return 9; };

	// Probe 4 — byte scan over a literal.
	let dn: i32 = count_digits("ww123abc");
	if (dn != 3) { return 10; };

	// Probe 5 — strconv round-trip via the real stdlib.
	let s: str = strconv.i64tos(4242i64, strconv.base.DEC);
	if (s.len != 4) { return 11; };
	if (s.ptr[0] != 52u8) { return 12; };	// '4'
	if (s.ptr[3] != 50u8) { return 13; };	// '2'

	// Probe 6 — ascii classifications (rune-taking, Hare-shaped).
	if (!ascii.isdigit(53)) { return 14; };		// '5'
	if (ascii.isdigit(65))  { return 15; };		// 'A' is not a digit
	if (!ascii.isalpha(122)) { return 16; };	// 'z'
	if (!ascii.isxdigit(70)) { return 17; };	// 'F'
	if (ascii.isxdigit(71))  { return 18; };	// 'G' is not hex
	if (ascii.tolower(65) != 97) { return 19; };	// 'A' -> 'a'
	if (ascii.toupper(122) != 90) { return 20; };	// 'z' -> 'Z'

	// Probe 7 — file open/read via the new os APIs. /proc/self/cmdline
	// always exists on Linux, no write side, and is non-empty.
	let path: str = "/proc/self/cmdline";
	// Use raw os.open here (returns i32 with -errno) for the same
	// reason as os.read below: probe 6 in 990_selfhost compiles
	// smoke.ww standalone (no `use` expansion), so cross-module type
	// references like `os.oserror` and `os.flag` don't resolve at
	// that step. RDONLY is 0; passing the literal keeps the call
	// site standalone-compilable to byte-identical asm on both
	// compilers.
	let fd: i32 = os.open(path.ptr, 0, 0i32);
	if (fd < 0) { return 21; };
	let rbuf: [128]u8;
	// Use raw os.read here (single syscall, plain i64) instead of
	// os.readall: the 990 cgen-match probe compiles smoke.ww
	// standalone without `use os;` expansion, so cross-module type
	// references like `os.oserror` can't be resolved.
	let n: i64 = os.read(fd, rbuf.ptr, 128u64);
	os.close(fd);
	if (n <= 0i64) { return 22; };

	return 42;
};