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

def SYS_READ:        i64 = 0;
def SYS_WRITE:       i64 = 1;
def SYS_OPEN:        i64 = 2;
def SYS_CLOSE:       i64 = 3;
def SYS_LSEEK:       i64 = 8;
def SYS_ACCESS:      i64 = 21;
def SYS_DUP2:        i64 = 33;
def SYS_GETPID:      i64 = 39;
def SYS_FORK:        i64 = 57;
def SYS_EXECVE:      i64 = 59;
def SYS_EXIT:        i64 = 60;
def SYS_WAIT4:       i64 = 61;
def SYS_UNLINK:      i64 = 87;
def SYS_GETCWD:      i64 = 79;
def SYS_GETDENTS64:  i64 = 217;

// open(2) flags. Linux values, matching <fcntl.h>.
def O_RDONLY: i32 = 0;
def O_WRONLY: i32 = 1;
def O_RDWR:   i32 = 2;
def O_CREAT:  i32 = 64;		// 0x40
def O_TRUNC:  i32 = 512;	// 0x200

// lseek(2) whence.
def SEEK_SET: i32 = 0;
def SEEK_CUR: i32 = 1;
def SEEK_END: i32 = 2;

export fn exit(code: i32) void = {
	syscall1(SYS_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(SYS_WRITE, fd: i64, buf: i64, n: i64);
};

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

export fn close(fd: i32) i32 = {
	return syscall1(SYS_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(SYS_DUP2, oldfd: i64, newfd: i64): i32;
};

// Fallible wrappers. The error variant is a plain str (Plan 9 errstr
// model, see lib/errors); the sum type makes success/failure explicit
// without overloading length-zero.
export fn tryread(fd: i32, buf: *u8, n: u64) (i64 | str) = {
	let r: i64 = read(fd, buf, n);
	if (r < 0) { return "read failed"; };
	return r;
};

export fn trywrite(fd: i32, buf: *u8, n: u64) (i64 | str) = {
	let r: i64 = write(fd, buf, n);
	if (r < 0) { return "write failed"; };
	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: i32, mode: i32) i32 = {
	return syscall3(SYS_OPEN, path: i64, flags: i64, mode: i64): i32;
};

export fn tryopen(path: *u8, flags: i32, mode: i32) (i32 | str) = {
	let fd: i32 = open(path, flags, mode);
	if (fd < 0) { return "open failed"; };
	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, whence: i32) i64 = {
	return syscall3(SYS_LSEEK, fd: i64, off, whence: i64);
};

// filesize — convenience: returns the byte length of an open fd by
// seeking to the end and back. -1 on error.
export fn filesize(fd: i32) i64 = {
	let end: i64 = lseek(fd, 0i64, SEEK_END);
	if (end < 0) { return -1i64; };
	let r: i64 = lseek(fd, 0i64, SEEK_SET);
	if (r < 0) { return -1i64; };
	return end;
};

// readfull — keep reading until `n` bytes have arrived or the fd
// closes early. Returns bytes read (0..=n) or -1 on read error.
export fn readfull(fd: i32, buf: *u8, n: u64) i64 = {
	let got: u64 = 0u64;
	for (got < n) {
		let r: i64 = read(fd, buf + got, n - got);
		if (r < 0) { return -1i64; };
		if (r == 0) { return got: i64; };	// short read: caller decides
		got += r: u64;
	};
	return got: i64;
};

// writefull — keep writing until `n` bytes have been accepted or the
// fd refuses progress. Returns bytes written or -1.
export fn writefull(fd: i32, buf: *u8, n: u64) i64 = {
	let sent: u64 = 0u64;
	for (sent < n) {
		let r: i64 = write(fd, buf + sent, n - sent);
		if (r < 0) { return -1i64; };
		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(SYS_ACCESS, path: i64, mode: i64): i32;
};

// unlink(2).
export fn unlink(path: *u8) i32 = {
	return syscall1(SYS_UNLINK, path: i64): i32;
};

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

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

// execve(2): on success, does not return.
export fn execve(path: *u8, argv: **u8, envp: **u8) i32 = {
	return syscall3(SYS_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(SYS_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(SYS_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(SYS_GETDENTS64, fd: i64, buf: i64, n: i64);
};

// MODULE: strconv
// strconv — number↔string conversions. Decimal i64 to/from a fixed
// buffer. Two error idioms ship side by side:
//   - Plan 9 style (atoi64): tuple `(value, ok)`. Pre-dates the
//     tagged-union work; kept for callers that already use it.
//   - Hare style (parse64/parseu64): `(value | str)`. The error
//     variant carries a short, allocation-free message describing
//     why the parse failed. Prefer this for new code.

// u64toa — write `v` in decimal into `buf` and return the byte count.
// Unsigned-only so callers don't have to think about wraparound when
// printing a u64 that happens to have the high bit set.
export fn u64toa(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 i64toa(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;
};

export fn atoi64(s: str) (i64, bool) = {
	let v: i64 = 0;
	let i: i32 = 0;
	let neg: bool = false;
	if (s.len > 0) {
		if (s[0] == 45u8) { neg = true; i = 1; };
	};
	if (i >= s.len) { return 0, false; };
	for (i < s.len) {
		let c: u8 = s[i];
		if (c < 48u8) { return 0, false; };
		if (c > 57u8) { return 0, false; };
		v = v * 10 + ((c: i64) - 48);
		i += 1;
	};
	if (neg) { v = -v; };
	return v, true;
};

// parse64 — Hare-style fallible signed decimal parser. The value
// variant is i64; the error variant is a short str describing the
// reason. No locale, no whitespace, no underscores: a leading '-' is
// the only non-digit accepted, and only at position 0.
export fn parse64(s: str) (i64 | str) = {
	if (s.len == 0) { return "parse: empty"; };
	let i: i32 = 0;
	let neg: bool = false;
	if (s[0] == 45u8) { neg = true; i = 1; };
	if (i >= s.len) { return "parse: lone sign"; };
	let v: i64 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		if (c < 48u8) { return "parse: invalid digit"; };
		if (c > 57u8) { return "parse: invalid digit"; };
		v = v * 10 + ((c: i64) - 48);
		i += 1;
	};
	if (neg) { v = -v; };
	return v;
};

// parseu64 — fallible unsigned decimal parser. No leading sign.
export fn parseu64(s: str) (u64 | str) = {
	if (s.len == 0) { return "parse: empty"; };
	let v: u64 = 0u64;
	let i: i32 = 0;
	for (i < s.len) {
		let c: u8 = s[i];
		if (c < 48u8) { return "parse: invalid digit"; };
		if (c > 57u8) { return "parse: invalid digit"; };
		v = v * 10u64 + ((c: u64) - 48u64);
		i += 1;
	};
	return v;
};

// MODULE: ascii
// ascii — byte-class predicates and case folding for the ASCII range.
// Matches Hare's ascii::isdigit family. Bytes 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: u8) bool = {
	if (c < 48u8) { return false; };
	if (c > 57u8) { return false; };
	return true;
};

export fn isupper(c: u8) bool = {
	if (c < 65u8) { return false; };
	if (c > 90u8) { return false; };
	return true;
};

export fn islower(c: u8) bool = {
	if (c < 97u8) { return false; };
	if (c > 122u8) { return false; };
	return true;
};

export fn isalpha(c: u8) bool = {
	if (isupper(c)) { return true; };
	return islower(c);
};

export fn isalnum(c: u8) bool = {
	if (isalpha(c)) { return true; };
	return isdigit(c);
};

// isspace — the C/Hare set: space, tab, NL, VT, FF, CR.
export fn isspace(c: u8) bool = {
	if (c == 32u8) { return true; };	// ' '
	if (c == 9u8)  { return true; };	// '\t'
	if (c == 10u8) { return true; };	// '\n'
	if (c == 11u8) { return true; };	// '\v'
	if (c == 12u8) { return true; };	// '\f'
	if (c == 13u8) { return true; };	// '\r'
	return false;
};

export fn ishex(c: u8) bool = {
	if (isdigit(c)) { return true; };
	if (c >= 65u8) {
		if (c <= 70u8) { return true; };	// 'A'..'F'
	};
	if (c >= 97u8) {
		if (c <= 102u8) { return true; };	// 'a'..'f'
	};
	return false;
};

// digitval — value of `c` as a hex/decimal digit, or -1 if not one.
// Useful when scanning numeric literals.
export fn digitval(c: u8) i32 = {
	if (isdigit(c)) { return (c - 48u8): i32; };
	if (c >= 65u8) {
		if (c <= 70u8) { return ((c - 65u8) + 10u8): i32; };
	};
	if (c >= 97u8) {
		if (c <= 102u8) { return ((c - 97u8) + 10u8): i32; };
	};
	return -1;
};

// isidstart / isidpart — identifier classes used by the lexer.
// Alpha or '_' starts; alnum or '_' continues.
export fn isidstart(c: u8) bool = {
	if (isalpha(c)) { return true; };
	if (c == 95u8) { return true; };	// '_'
	return false;
};

export fn isidpart(c: u8) bool = {
	if (isalnum(c)) { return true; };
	if (c == 95u8) { return true; };
	return false;
};

// tolower / toupper — fold ASCII case. Non-letters pass through.
export fn tolower(c: u8) u8 = {
	if (isupper(c)) { return c + 32u8; };
	return c;
};

export fn toupper(c: u8) u8 = {
	if (islower(c)) { return c - 32u8; };
	return c;
};

// 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 outbuf: [32]u8;
	let nb: i32 = strconv.i64toa(outbuf[0:32], 4242i64);
	if (nb != 4) { return 11; };
	if (outbuf[0] != 52u8) { return 12; };	// '4'
	if (outbuf[3] != 50u8) { return 13; };	// '2'

	// Probe 6 — ascii classifications.
	if (!ascii.isdigit(53u8)) { return 14; };	// '5'
	if (ascii.isdigit(65u8))  { return 15; };	// 'A' is not a digit
	if (!ascii.isalpha(122u8)) { return 16; };	// 'z'
	if (!ascii.isidstart(95u8)) { return 17; };	// '_'
	if (!ascii.isidpart(48u8)) { return 18; };	// '0' is part
	if (ascii.digitval(70u8) != 15) { return 19; };	// 'F' = 15
	if (ascii.tolower(65u8) != 97u8) { return 20; };	// 'A' -> 'a'

	// 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";
	let fd_or_err: (i32 | str) = os.tryopen(path.ptr, os.O_RDONLY, 0i32);
	let fd: i32 = 0;
	match (fd_or_err) {
		case let v: i32 => fd = v;
		case let e: str => return 21;
	};
	let rbuf: [128]u8;
	let n: i64 = os.readfull(fd, rbuf.ptr, 128u64);
	os.close(fd);
	if (n <= 0i64) { return 22; };

	return 42;
};