6455 lines
230 KiB
Plaintext
6455 lines
230 KiB
Plaintext
// time — clocks, instants, durations. Mirrors Hare's lib/time
|
||
// (ref/hare/time/duration.ha, instant.ha, arithm.ha,
|
||
// +linux/functions.ha). Calendar / date / strftime / timezone /
|
||
// sleep live in separate Hare modules and graduate when callers /
|
||
// supporting stdlib arrive.
|
||
//
|
||
// `duration` is a NAMED alias of i64 (lib/math/random precedent
|
||
// at lib/math/random/random.ww:8); ww treats NAMED as a newtype,
|
||
// so cross-i64 arithmetic inside this module needs explicit casts.
|
||
// Hare's structural alias semantics let those casts vanish, but
|
||
// our type checker is strict.
|
||
|
||
package time;
|
||
|
||
@symbol("rt_syscall") fn syscall2(num: i64, a: i64, b: i64) i64;
|
||
@symbol("rt_abort") fn abort(msg: str) void;
|
||
|
||
def SYS_CLOCK_GETTIME: i64 = 228;
|
||
|
||
// ref/hare/time/duration.ha:6. 290y representable range.
|
||
export type duration = i64;
|
||
|
||
// ref/hare/time/duration.ha:9-18. Plan-9 naming (lowercase)
|
||
// diverges from Hare's uppercase per project rule 4.
|
||
export def nanosecond: duration = 1i64;
|
||
export def microsecond: duration = 1000i64;
|
||
export def millisecond: duration = 1000000i64;
|
||
export def second: duration = 1000000000i64;
|
||
|
||
// ref/hare/time/instant.ha:9. (sec, nsec) pair — NOT POSIX struct
|
||
// timespec (which uses u32 nsec). Layout matches Linux's struct
|
||
// timespec on 64-bit (i64+i64) so we can pass &instant directly
|
||
// to clock_gettime.
|
||
export type instant = struct {
|
||
sec: i64,
|
||
nsec: i64,
|
||
};
|
||
|
||
// ref/hare/time/+linux/functions.ha:84. First cut exposes only
|
||
// realtime and monotonic; Hare's process_cpu / thread_cpu / boot /
|
||
// realtime_alarm / boot_alarm / tai graduate when a caller needs
|
||
// them (CLAUDE.md rule 9 — Hare-fidelity, no premature surface).
|
||
export type clock = enum i32 {
|
||
realtime = 0,
|
||
monotonic = 1,
|
||
};
|
||
|
||
// ref/hare/time/+linux/functions.ha:138. Hare's now() also aborts
|
||
// on impossible errnos. (instant | oserror) is deliberately not
|
||
// the return shape — EINVAL / EFAULT are programmer errors (bad
|
||
// clock id, bad ptr), and a 1-word-payload sum return walks into
|
||
// task #9's cgen-divergence trap.
|
||
export fn now(c: clock) instant = {
|
||
let i: instant;
|
||
let rc = syscall2(SYS_CLOCK_GETTIME, (c as i32): i64, (&i): i64);
|
||
if (rc != 0i64) { abort("time.now: clock_gettime failed"); };
|
||
return i;
|
||
};
|
||
|
||
// ref/hare/time/arithm.ha:9. Adds duration to instant. The
|
||
// negative-duration branch normalises nsec into [0, second).
|
||
export fn add(i: instant, x: duration) instant = {
|
||
let r: instant;
|
||
let xi: i64 = x: i64;
|
||
let sec: i64 = second: i64;
|
||
let nsec: i64 = nanosecond: i64;
|
||
if (xi == 0i64) {
|
||
r.sec = i.sec;
|
||
r.nsec = i.nsec;
|
||
return r;
|
||
};
|
||
if (xi > 0i64) {
|
||
r.sec = i.sec + (i.nsec + xi) / sec;
|
||
r.nsec = (i.nsec + xi) % sec;
|
||
return r;
|
||
};
|
||
r.sec = i.sec + (i.nsec + xi - sec + nsec) / sec;
|
||
r.nsec = (i.nsec + (xi % sec) + sec) % sec;
|
||
return r;
|
||
};
|
||
|
||
// ref/hare/time/arithm.ha:26. Returns duration from a to b.
|
||
// Sign convention: b - a.
|
||
export fn diff(a: instant, b: instant) duration = {
|
||
let sec: i64 = second: i64;
|
||
let v: i64 = ((b.sec - a.sec) * sec) + (b.nsec - a.nsec);
|
||
return v: duration;
|
||
};
|
||
|
||
// ref/hare/time/arithm.ha:32. -1 if a < b, 0 if equal, +1 if a > b.
|
||
export fn compare(a: instant, b: instant) i8 = {
|
||
if (a.sec < b.sec) { return -1i8; };
|
||
if (a.sec > b.sec) { return 1i8; };
|
||
if (a.nsec < b.nsec) { return -1i8; };
|
||
if (a.nsec > b.nsec) { return 1i8; };
|
||
return 0i8;
|
||
};
|
||
|
||
// 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.
|
||
|
||
package os;
|
||
|
||
import time;
|
||
|
||
@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_free") export 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,
|
||
MKDIR = 83,
|
||
RMDIR = 84,
|
||
UNLINK = 87,
|
||
GETCWD = 79,
|
||
GETDENTS64 = 217,
|
||
NEWFSTATAT = 262,
|
||
};
|
||
|
||
// 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
|
||
EXCL = 128, // 0x80 — pair with CREATE to fail on existing path
|
||
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);
|
||
};
|
||
|
||
// PATH_MAX / pathbuf / kpath — port of Hare's ref/hare/sys/+linux/
|
||
// syscalls.ha:25,27,29-55. Hare's `path` accepts a sum `(str |
|
||
// []u8 | *const u8)`; ww's lib/os public surface narrows to `str`
|
||
// (the Hare-faithful surface at ref/hare/os/os.ha:37,47,50 etc).
|
||
// Internally, [[kpath]] copies the `str` bytes into a single
|
||
// module-level [[pathbuf]] scratch slot and NUL-terminates so the
|
||
// raw Linux syscalls (which require C strings) see a valid
|
||
// terminator. Same precedent as Hare's static `pathbuf`.
|
||
//
|
||
// Non-reentrant: one buffer, every [[stat]] / [[open]] / etc.
|
||
// rewrites it. Same caveat as strconv's `*tos` family (overwritten
|
||
// on next call). Caller must NOT hold a kpath-returned pointer
|
||
// across another lib/os path call. Graduates when ww grows a
|
||
// thread story.
|
||
//
|
||
// `nil`-as-overflow over `(*u8 | oserror)`: wwstage over-allocates
|
||
// 1-word-payload tagged returns to 24B (cstage emits 16B).
|
||
// Task #9; revert at task #10 when fixed. Repro at
|
||
// .ai/probe_tagged_return_pointer_payload.ww.
|
||
export def PATH_MAX: i32 = 4096;
|
||
let pathbuf: [4096]u8;
|
||
|
||
// ref/hare/sys/+linux/types.ha:886-888. ww folds `sys` into `os`, so the
|
||
// std fd NUMBERS live here (the sys role). Typed i32, NOT io.file as in
|
||
// Hare's os::stdout_file (ref/hare/os/+linux/stdfd.ha:28): Hare's `os`
|
||
// imports `io`, but ww's `os` is the import floor and must never import
|
||
// io (lib/CLAUDE.md) — so the io.file/io.handle binding can't live here.
|
||
// Consumers (lib/fmt's stdio wrappers) cast i32→io.file at the use site,
|
||
// where the handle layer is already in scope.
|
||
export def STDIN_FILENO: i32 = 0;
|
||
export def STDOUT_FILENO: i32 = 1;
|
||
export def STDERR_FILENO: i32 = 2;
|
||
|
||
fn kpath(p: str) *u8 = {
|
||
if (p.len + 1 >= PATH_MAX) { return nil: *u8; }; // ENAMETOOLONG
|
||
let i: i32 = 0;
|
||
for (i < p.len) { pathbuf[i] = p[i]; i += 1; };
|
||
pathbuf[p.len] = 0u8;
|
||
return &pathbuf[0];
|
||
};
|
||
|
||
// 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). Returns -errno on failure, fd otherwise.
|
||
// Higher-level callers prefer `tryopen`. Mirrors Hare's os::open
|
||
// (ref/hare/os/os.ha:117); kpath lands the bytes in pathbuf.
|
||
// Returns -ENAMETOOLONG (-36) if the path overflows PATH_MAX.
|
||
export fn open(path: str, flags: flag, mode: i32) i32 = {
|
||
let p: *u8 = kpath(path);
|
||
if (p == nil: *u8) { return -36i32; }; // ENAMETOOLONG
|
||
return syscall3(nr.OPEN, p: i64, (flags as i32): i64, mode: i64): i32;
|
||
};
|
||
|
||
export fn tryopen(path: str, 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;
|
||
|
||
// errno — the raw Linux errno as a positive code (ref/hare/sys/+linux/
|
||
// errno.ha:5, `errno = !int`). ww folds Hare's `sys` role into os
|
||
// (lib/CLAUDE.md), so the sys::errno machinery lands here. Spelled i32
|
||
// rather than int: Linux errnos are kernel ints (32-bit), keeping os's
|
||
// kernel-facing surface uniformly i32. Distinct from [[oserror]] (!i64,
|
||
// the syscall's *negative* raw return) — the two model different
|
||
// things, so they are not unified; the negative→positive normalization
|
||
// lives at the oserror→errors.error boundary in those callers.
|
||
export type errno = !i32;
|
||
|
||
// Mapped errno values, ref/hare/sys/+linux/errno.ha:559-682. Positive,
|
||
// matching Hare's defs (the kernel returns -N; the wrap-to-positive is
|
||
// the caller's concern). Subset: exactly the errnos [[errors.errno]]
|
||
// maps to a named condition; grow as callers surface more.
|
||
export def ENOENT: errno = 2;
|
||
export def EINTR: errno = 4;
|
||
export def EAGAIN: errno = 11;
|
||
export def EACCES: errno = 13;
|
||
export def EBUSY: errno = 16;
|
||
export def EEXIST: errno = 17;
|
||
export def EINVAL: errno = 22;
|
||
export def EOVERFLOW: errno = 75;
|
||
export def ENETUNREACH: errno = 101;
|
||
export def ETIMEDOUT: errno = 110;
|
||
export def ECONNREFUSED: errno = 111;
|
||
export def ECANCELED: errno = 125;
|
||
|
||
// strerror — human-readable text for an [[errno]] (Hare's
|
||
// sys::strerror, ref/hare/sys/+linux/errno.ha:18). FAITHFUL MINIMAL
|
||
// SUBSET: the mapped errnos above plus a generic fallback; grow the
|
||
// switch as callers surface more (lib/CLAUDE.md documented-subset, not
|
||
// a workaround). Messages verbatim from the reference. Hare's
|
||
// unknown_errno formats the numeric value; that is deferred.
|
||
export fn strerror(err: errno) str = {
|
||
switch (err) {
|
||
case ENOENT: return "No such file or directory";
|
||
case EINTR: return "Interrupted system call";
|
||
case EAGAIN: return "Resource temporarily unavailable";
|
||
case EACCES: return "Permission denied";
|
||
case EBUSY: return "Device or resource busy";
|
||
case EEXIST: return "File exists";
|
||
case EINVAL: return "Invalid argument";
|
||
case EOVERFLOW: return "Value too large for defined data type";
|
||
case ENETUNREACH: return "Network is unreachable";
|
||
case ETIMEDOUT: return "Connection timed out";
|
||
case ECONNREFUSED: return "Connection refused";
|
||
case ECANCELED: return "Operation canceled";
|
||
};
|
||
return "Unknown error";
|
||
};
|
||
|
||
// 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).
|
||
// Mirrors Hare's os::access (ref/hare/os/+linux/fs.ha:access).
|
||
// Returns -ENAMETOOLONG (-36) if the path overflows PATH_MAX.
|
||
export fn access(path: str, mode: i32) i32 = {
|
||
let p: *u8 = kpath(path);
|
||
if (p == nil: *u8) { return -36i32; };
|
||
return syscall2(nr.ACCESS, p: i64, mode: i64): i32;
|
||
};
|
||
|
||
// remove — unlink(2). Mirrors Hare's os::remove
|
||
// (ref/hare/os/os.ha:12).
|
||
export fn remove(path: str) i32 = {
|
||
let p: *u8 = kpath(path);
|
||
if (p == nil: *u8) { return -36i32; };
|
||
return syscall1(nr.UNLINK, p: i64): i32;
|
||
};
|
||
|
||
// mkdir — mkdir(2). Mode is the unix permission bitset (e.g. 0o700).
|
||
// Returns 0 on success, negative errno otherwise. Mirrors Hare's
|
||
// os::mkdir (ref/hare/os/os.ha:50).
|
||
export fn mkdir(path: str, mode: i32) i32 = {
|
||
let p: *u8 = kpath(path);
|
||
if (p == nil: *u8) { return -36i32; };
|
||
return syscall2(nr.MKDIR, p: i64, mode: i64): i32;
|
||
};
|
||
|
||
// rmdir — rmdir(2). Mirrors Hare's os::rmdir
|
||
// (ref/hare/os/os.ha:58).
|
||
export fn rmdir(path: str) i32 = {
|
||
let p: *u8 = kpath(path);
|
||
if (p == nil: *u8) { return -36i32; };
|
||
return syscall1(nr.RMDIR, p: i64): i32;
|
||
};
|
||
|
||
// mkdirs — recursive mkdir. Creates `path` and any non-existent
|
||
// parent directories with the given mode. EEXIST is silently
|
||
// accepted (matches Hare's `errors::exists` skip in os::mkdirs);
|
||
// any other syscall failure surfaces as `oserror`.
|
||
//
|
||
// Mirrors Hare's os::mkdirs (ref/hare/os/os.ha:54). The in-place
|
||
// '/' → NUL splice walks the kpath-loaded [[pathbuf]] directly
|
||
// instead of recursing through [[mkdir]] — re-entering kpath would
|
||
// clobber the buffer mid-walk (single static slot, see kpath's
|
||
// non-reentrancy note above).
|
||
export fn mkdirs(path: str, mode: i32) (void | oserror) = {
|
||
let cp: *u8 = kpath(path);
|
||
if (cp == nil: *u8) { return -36i64: oserror; };
|
||
let n: i32 = path.len;
|
||
if (n == 0) { return; };
|
||
|
||
// Walk forward; at each '/' boundary, NUL-terminate the prefix,
|
||
// raw MKDIR syscall on pathbuf, restore the slash, continue.
|
||
// Skip index 0 so a leading '/' on absolute paths doesn't
|
||
// trigger an empty mkdir.
|
||
let i: i32 = 1;
|
||
for (i < n) {
|
||
if (pathbuf[i] == 47u8) { // '/'
|
||
pathbuf[i] = 0u8;
|
||
let r: i32 = syscall2(nr.MKDIR,
|
||
(&pathbuf[0]): i64, mode: i64): i32;
|
||
pathbuf[i] = 47u8;
|
||
if (r < 0) {
|
||
if (r != -17) { return r: i64: oserror; };
|
||
};
|
||
};
|
||
i += 1;
|
||
};
|
||
|
||
let r: i32 = syscall2(nr.MKDIR,
|
||
(&pathbuf[0]): i64, mode: i64): i32;
|
||
if (r < 0) {
|
||
if (r != -17) { return r: i64: oserror; };
|
||
};
|
||
return;
|
||
};
|
||
|
||
// 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. Mirrors Hare's
|
||
// os::exec::exec path arg (str). argv/envp stay `**u8` — the
|
||
// kernel takes a NUL-pointer-terminated table of NUL-terminated
|
||
// C strings, a different shape from a path.
|
||
export fn execve(path: str, argv: **u8, envp: **u8) i32 = {
|
||
let p: *u8 = kpath(path);
|
||
if (p == nil: *u8) { return -36i32; };
|
||
return syscall3(nr.EXECVE, p: 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);
|
||
};
|
||
|
||
// ---- environment ------------------------------------------------------
|
||
|
||
// rt_envp — runtime-side getter. rt/start.s captures envp into a DATAW
|
||
// slot before calling main; this binding lifts the captured pointer
|
||
// into ww. Same FFI shape as rt_syscall / rt_malloc / rt_abort: a TEXT
|
||
// symbol the linker resolves. The returned `**u8` is a NUL-terminated
|
||
// table of `*u8` entries, each pointing at a NUL-terminated
|
||
// "NAME=VALUE" byte sequence.
|
||
//
|
||
// We don't expose `rtenvp` directly; [[getenv]] is the only consumer.
|
||
@symbol("rt_envp") fn rtenvp() **u8;
|
||
|
||
// getenv — POSIX getenv. Returns a borrowed `str` view over the value
|
||
// bytes of the named environment variable, or void if the name is not
|
||
// present. The view is valid for the process lifetime — the bytes
|
||
// live in the kernel-supplied envp table at process entry. A future
|
||
// `setenv` (separate task) that grows the table behind the scenes
|
||
// would invalidate prior views; v1 has no setenv, so callers can
|
||
// hold the view indefinitely.
|
||
//
|
||
// Mirrors Hare's os::tryenv shape (returns void rather than panicking
|
||
// on missing). Hare also ships os::getenv (`(str | void)`) and
|
||
// os::mustenv (panic-on-missing); ww collapses to the single
|
||
// `(str | void)` form for now — consumers wanting "must" semantics
|
||
// abort at the call site.
|
||
//
|
||
// Algorithm: walk the NUL-pointer-terminated `environ` table doing a
|
||
// "name=" prefix match against each entry, byte-wise. NUL inside
|
||
// `name` would never match a real env var (env var names cannot
|
||
// contain '\0'), so we don't filter — POSIX puts that responsibility
|
||
// on the caller.
|
||
export fn getenv(name: str) (str | void) = {
|
||
let envp: **u8 = rtenvp();
|
||
let i: i32 = 0;
|
||
for (true) {
|
||
let entry: *u8 = envp[i];
|
||
if (entry == nil: *u8) { return; };
|
||
let j: i32 = 0;
|
||
let matched: bool = true;
|
||
for (j < name.len) {
|
||
if (entry[j] == 0u8) { matched = false; break; };
|
||
if (entry[j] != name[j]) { matched = false; break; };
|
||
j += 1;
|
||
};
|
||
if (matched) {
|
||
if (entry[name.len] == 61u8) { // '='
|
||
let val: *u8 = entry + ((name.len + 1): u64);
|
||
let n: i32 = 0;
|
||
for (val[n] != 0u8) { n += 1; };
|
||
let r: str;
|
||
r.ptr = val;
|
||
r.len = n;
|
||
return r;
|
||
};
|
||
};
|
||
i += 1;
|
||
};
|
||
return;
|
||
};
|
||
|
||
// ---- stat / lstat / fstat / exists -----------------------------------
|
||
//
|
||
// Ports of Hare's stat family (ref/hare/fs/fs.ha:172,196 +
|
||
// ref/hare/sys/+linux/stat.ha:24-58). The Hare surface returns
|
||
// `filestat` by value; ww's cgreturn ABI tops out at 24B today (see
|
||
// STATUS task #21) and filestat is 80B, so [[stat]] / [[lstat]] /
|
||
// [[fstat]] take an out-parameter and return `(void | oserror)`.
|
||
// Re-evaluate the by-value shape when full sret lands.
|
||
//
|
||
// `filestat`, `mode`, and `stat_mask` live in lib/os because ww has
|
||
// no lib/fs yet; Hare puts them in `fs::`. These types graduate to
|
||
// lib/fs when that module ships — callers should expect a future
|
||
// re-export.
|
||
//
|
||
// Underlying syscall is SYS_newfstatat (262), which unifies
|
||
// stat/lstat/fstat through the `dirfd + flags` triple:
|
||
// stat = newfstatat(AT_FDCWD, path, 0)
|
||
// lstat = newfstatat(AT_FDCWD, path, AT_SYMLINK_NOFOLLOW)
|
||
// fstat = newfstatat(fd, "", AT_EMPTY_PATH)
|
||
// Avoiding SYS_statx — its 256B variable layout would buy btime,
|
||
// but Hare's filestat doesn't expose btime either, so we stay on
|
||
// the simpler 144B kernel struct.
|
||
|
||
// fstatat(2) flag values. Linux constants from <linux/fcntl.h>.
|
||
// Names mirror Hare's ref/hare/sys/+linux/types.ha:45-51 (capital-
|
||
// AT_ prefix, top-level `def`s).
|
||
export def AT_FDCWD: i32 = -100;
|
||
export def AT_SYMLINK_NOFOLLOW: i32 = 256; // 0x100
|
||
export def AT_EMPTY_PATH: i32 = 4096; // 0x1000
|
||
|
||
// mode — file-mode bits. Mirrors Hare's fs::mode (ref/hare/fs/
|
||
// types.ha:63). Permission bits are the standard Unix octal subset;
|
||
// type bits live in the S_IFMT = 0o170000 region. Type-bit test:
|
||
//
|
||
// let t: u32 = (fi.mode as u32) & 61440u32; // 0o170000 mask
|
||
// if (t == os.mode.DIR as u32) { /* directory */ };
|
||
//
|
||
// Numeric values are octal in Hare's source; ww has no octal
|
||
// literals so they're written as decimal with the octal in a
|
||
// trailing comment.
|
||
export type mode = enum u32 {
|
||
// permission bits
|
||
USER_RWX = 448u32, // 0o700
|
||
USER_RW = 384u32, // 0o600
|
||
USER_RX = 320u32, // 0o500
|
||
USER_R = 256u32, // 0o400
|
||
USER_W = 128u32, // 0o200
|
||
USER_X = 64u32, // 0o100
|
||
GROUP_RWX = 56u32, // 0o070
|
||
GROUP_RW = 48u32, // 0o060
|
||
GROUP_RX = 40u32, // 0o050
|
||
GROUP_R = 32u32, // 0o040
|
||
GROUP_W = 16u32, // 0o020
|
||
GROUP_X = 8u32, // 0o010
|
||
OTHER_RWX = 7u32, // 0o007
|
||
OTHER_RW = 6u32, // 0o006
|
||
OTHER_RX = 5u32, // 0o005
|
||
OTHER_R = 4u32, // 0o004
|
||
OTHER_W = 2u32, // 0o002
|
||
OTHER_X = 1u32, // 0o001
|
||
SETUID = 2048u32, // 0o4000
|
||
SETGID = 1024u32, // 0o2000
|
||
STICKY = 512u32, // 0o1000
|
||
// file-type bits (S_IFMT mask = 0o170000 = 61440)
|
||
UNKNOWN = 0u32,
|
||
FIFO = 4096u32, // 0o010000
|
||
CHR = 8192u32, // 0o020000
|
||
DIR = 16384u32, // 0o040000
|
||
BLK = 24576u32, // 0o060000
|
||
REG = 32768u32, // 0o100000
|
||
LINK = 40960u32, // 0o120000
|
||
SOCK = 49152u32, // 0o140000
|
||
};
|
||
|
||
// stat_mask — which filestat fields the call populated. Mirrors
|
||
// Hare's fs::stat_mask (ref/hare/fs/types.ha:129). newfstatat fills
|
||
// every field, so [[stat]] / [[lstat]] / [[fstat]] always set all
|
||
// seven bits OR-folded (see [[fillfilestat]]); per-bit testing is
|
||
// the documented sparse-backend pattern (cf. Hare's fs::fs network
|
||
// backends that only populate mtime+size).
|
||
export type stat_mask = enum u32 {
|
||
UID = 1u32,
|
||
GID = 2u32,
|
||
SIZE = 4u32,
|
||
INODE = 8u32,
|
||
ATIME = 16u32,
|
||
MTIME = 32u32,
|
||
CTIME = 64u32,
|
||
};
|
||
|
||
// filestat — Hare's fs::filestat (ref/hare/fs/types.ha:141). 80
|
||
// bytes. Times are time.instant (ref/hare/time/instant.ha:9) — the
|
||
// canonical Hare shape. See module-header note re: graduation to
|
||
// lib/fs.
|
||
export type filestat = struct {
|
||
mask: stat_mask, // 0 (4)
|
||
mode: mode, // 4 (4)
|
||
uid: u32, // 8 (4)
|
||
gid: u32, // 12 (4)
|
||
sz: u64, // 16 (8)
|
||
inode: u64, // 24 (8)
|
||
atime: time.instant, // 32 (16)
|
||
mtime: time.instant, // 48 (16)
|
||
ctime: time.instant, // 64 (16) — ends at 80
|
||
};
|
||
|
||
// kstat — x86_64 kernel `struct stat` layout. Mirrors
|
||
// arch/x86/include/uapi/asm/stat.h (`__kernel_ulong_t`-keyed
|
||
// fields). 144 bytes. Module-internal; SYS_newfstatat writes into
|
||
// this buffer and the public stat fns then copy the bits into the
|
||
// Hare-shaped [[filestat]].
|
||
type kstat = struct {
|
||
dev: u64, // 0
|
||
ino: u64, // 8
|
||
nlink: u64, // 16
|
||
mode: u32, // 24
|
||
uid: u32, // 28
|
||
gid: u32, // 32
|
||
pad0: u32, // 36
|
||
rdev: u64, // 40
|
||
sz: i64, // 48
|
||
blksize: i64, // 56
|
||
blocks: i64, // 64
|
||
atime_sec: i64, // 72
|
||
atime_nsec: i64, // 80
|
||
mtime_sec: i64, // 88
|
||
mtime_nsec: i64, // 96
|
||
ctime_sec: i64, // 104
|
||
ctime_nsec: i64, // 112
|
||
unused0: i64, // 120
|
||
unused1: i64, // 128
|
||
unused2: i64, // 136 — ends at 144
|
||
};
|
||
|
||
// emptypath — single-NUL byte used as the `pathname` arg to
|
||
// newfstatat with AT_EMPTY_PATH. The kernel requires a non-NULL
|
||
// pointer to a zero-length C string, NOT a null pointer. Bytes are
|
||
// read-only from the kernel's view; ww has no module-level const so
|
||
// this is a writable `let`.
|
||
let emptypath: [1]u8 = [0u8];
|
||
|
||
// fillfilestat — copy a 144B kstat into the 80B Hare-shaped
|
||
// filestat. Internal helper used by all three public entry points.
|
||
// Mirrors Hare's st_to_filestat (ref/hare/os/+linux/dirfdfs.ha:259):
|
||
// newfstatat populates every field, so the mask is the OR-fold of
|
||
// all seven Hare stat_mask bits.
|
||
fn fillfilestat(out: *filestat, k: *kstat) void = {
|
||
out.mask = stat_mask.UID | stat_mask.GID | stat_mask.SIZE
|
||
| stat_mask.INODE | stat_mask.ATIME | stat_mask.MTIME
|
||
| stat_mask.CTIME;
|
||
out.mode = k.mode: mode;
|
||
out.uid = k.uid;
|
||
out.gid = k.gid;
|
||
out.sz = k.sz: u64;
|
||
out.inode = k.ino;
|
||
out.atime.sec = k.atime_sec;
|
||
out.atime.nsec = k.atime_nsec;
|
||
out.mtime.sec = k.mtime_sec;
|
||
out.mtime.nsec = k.mtime_nsec;
|
||
out.ctime.sec = k.ctime_sec;
|
||
out.ctime.nsec = k.ctime_nsec;
|
||
};
|
||
|
||
// stat — fill *out with metadata for `path`. Follows symlinks.
|
||
// Returns ENAMETOOLONG (-36) as `oserror` if the path overflows
|
||
// PATH_MAX.
|
||
//
|
||
// Mirrors Hare's sys::stat (ref/hare/sys/+linux/stat.ha:51) modulo
|
||
// the out-param shape forced by the cgreturn 24B cap. Note: Hare's
|
||
// higher-level fs::stat (ref/hare/fs/fs.ha:172) instead has lstat
|
||
// semantics — we follow sys::stat's POSIX-stat behavior here.
|
||
export fn stat(out: *filestat, path: str) (void | oserror) = {
|
||
let cp: *u8 = kpath(path);
|
||
if (cp == nil: *u8) { return -36i64: oserror; };
|
||
let k: kstat;
|
||
let r: i64 = syscall4(nr.NEWFSTATAT,
|
||
AT_FDCWD: i64, cp: i64, (&k): i64, 0i64);
|
||
if (r < 0) { return r: oserror; };
|
||
fillfilestat(out, &k);
|
||
};
|
||
|
||
// lstat — like [[stat]] but does NOT follow a terminal symlink.
|
||
// Mirrors Hare's sys::lstat (ref/hare/sys/+linux/stat.ha:57).
|
||
export fn lstat(out: *filestat, path: str) (void | oserror) = {
|
||
let cp: *u8 = kpath(path);
|
||
if (cp == nil: *u8) { return -36i64: oserror; };
|
||
let k: kstat;
|
||
let r: i64 = syscall4(nr.NEWFSTATAT,
|
||
AT_FDCWD: i64, cp: i64, (&k): i64,
|
||
AT_SYMLINK_NOFOLLOW: i64);
|
||
if (r < 0) { return r: oserror; };
|
||
fillfilestat(out, &k);
|
||
};
|
||
|
||
// fstat — like [[stat]] but addresses the file by fd. Uses
|
||
// newfstatat(fd, "", AT_EMPTY_PATH); the kernel resolves the fd
|
||
// directly. Mirrors Hare's sys::fstat (ref/hare/sys/+linux/stat.ha:54).
|
||
export fn fstat(out: *filestat, fd: i32) (void | oserror) = {
|
||
let k: kstat;
|
||
let r: i64 = syscall4(nr.NEWFSTATAT,
|
||
fd: i64, (&emptypath[0]): i64, (&k): i64,
|
||
AT_EMPTY_PATH: i64);
|
||
if (r < 0) { return r: oserror; };
|
||
fillfilestat(out, &k);
|
||
};
|
||
|
||
// exists — true if `path` resolves to anything (regular file,
|
||
// directory, symlink, ...). Stat-shaped (Hare's `fs::exists`,
|
||
// ref/hare/fs/fs.ha:196) — no separate syscall. Symlinks are
|
||
// followed; a dangling symlink is `false`. ENAMETOOLONG is
|
||
// swallowed as `false` — Hare's os::exists doc says "true if a
|
||
// node exists at the given path, or false if not."
|
||
//
|
||
// Race warning: prefer "open and handle the error" over "exists
|
||
// then open" in real code (Hare's docstring carries the same
|
||
// note). The race is unavoidable in this shape.
|
||
//
|
||
// Goes through SYS_newfstatat directly rather than match'ing on
|
||
// [[stat]]'s `(void | oserror)` return. Functionally identical;
|
||
// the direct shape sidesteps a cstage/wwstage cgen disagreement
|
||
// on the slot size of `(void | oserror)` (cstage 16B, wwstage 24B
|
||
// — same class as STATUS #22, surfaced first time a match on this
|
||
// shape combined with an 80B local-struct local frame). Use the
|
||
// match shape once #22 lands.
|
||
export fn exists(path: str) bool = {
|
||
let cp: *u8 = kpath(path);
|
||
if (cp == nil: *u8) { return false; };
|
||
let k: kstat;
|
||
let r: i64 = syscall4(nr.NEWFSTATAT,
|
||
AT_FDCWD: i64, cp: i64, (&k): i64, 0i64);
|
||
return r >= 0i64;
|
||
};
|
||
|
||
// strconv — arbitrary-precision decimal engine for float↔string
|
||
// conversion. Mirrors ref/hare/strconv/decimal.ha (Hare in turn ports
|
||
// Go's lib/strconv/decimal.go). Pure integer arithmetic; no f32/f64
|
||
// references (#121 residual-guard SAFE).
|
||
//
|
||
// Spelling divergences from Hare (mechanical, ww-side parser shape):
|
||
// - Hare `let a = X, b = Y;` → two single `let` statements
|
||
// (ww parser doesn't accept comma-separated bindings).
|
||
// - Hare `tbl[lo..]` open-ended slice → direct indexing
|
||
// `tbl[lo + i]` at point-of-use (equivalent algorithm; no
|
||
// allocation, no aliasing). ww `[lo:hi]` uses `:`; `..` form
|
||
// is not parsed.
|
||
// - Hare `0z`/`1z` size literals → ww has no `z` suffix; pre-bind
|
||
// `let SZ_ZERO: size = (0u64: size);` etc. at function entry
|
||
// ("hoisted size casts as local consts" — ww `T: type` casts
|
||
// embedded inside expressions confuse the parser).
|
||
// - Hare `~0u64` typed-suffix literal → ww parser rejects `~` on
|
||
// typed-suffix; route via a named zero local + `~zero`.
|
||
// - Hare `for (cond; afterthought)` 2-clause → ww 3-clause
|
||
// `for (init; cond; post)` (when continue is used; the post
|
||
// must run each iteration) or inline-the-afterthought in body
|
||
// (when no continue exists in the loop).
|
||
// - Hare `fn foo() T = if (cond) {...} else expr;` expression body
|
||
// → ww requires a `{}` block body throughout.
|
||
// - Hare bare `assert(cond)` builtin → `os.assert(cond, msg)`;
|
||
// wwstage cgen has no `assert` intercept (deferred fold).
|
||
// - In-file instances of the above hoist pattern: `i_sz` (line 93)
|
||
// hoists a per-iteration size cast out of a for-loop comparison
|
||
// (bullet 3 sub-case — the size-cast hoist applied inside a loop
|
||
// body, not just at function entry); `lowbit_lit` (line 242)
|
||
// decomposes Hare's `(nd > 0 && d.digits[nd - 1] & 1 != 0)` into
|
||
// a stepwise boolean local to dodge ww parser precedence on mixed
|
||
// `&` / `&&` / `!=` within a single expression.
|
||
//
|
||
// CGEN class closures consumed (post-prereqs):
|
||
// - #131 (4acab6e) — `len(d.digits)` compile-time-folds cs==ww
|
||
// - #134 (36bf603) — `d.digits[nd] >= 5u8` picks JAE (unsigned)
|
||
// - #133 (3986818) — `d.digits[i] += 1u8` load-op-store BOTH
|
||
// stages
|
||
// - #135 (ade6840) — `(*d).digits[i]` read+write N_DOT-base addr
|
||
//
|
||
// Drew CGEN-SAFE invariants:
|
||
// - #129: module-level decls here are integer-literal defs only.
|
||
// - #128: digits is fundamental [800]u8, zero-init only.
|
||
// - #121: zero float ops.
|
||
// - Drew watch-item `*d = decimal{...};` reset (line 110 in Hare):
|
||
// pointer-deref reset to composite-literal probed cs==ww
|
||
// byte-id safe.
|
||
|
||
package strconv;
|
||
|
||
import os;
|
||
|
||
// ref/hare/strconv/decimal.ha:5.
|
||
def maxshift: u8 = 60u8;
|
||
|
||
// ref/hare/strconv/decimal.ha:6.
|
||
def decimal_point_range: u16 = 2047u16;
|
||
|
||
// ref/hare/strconv/decimal.ha:8-26. Field layout 1:1. The 800-digit
|
||
// bound covers subnormal doubles (min exp -1074, max mantissa 4e16
|
||
// → at most 767 digits; 800 leaves headroom).
|
||
export type decimal = struct {
|
||
digits: [800]u8,
|
||
nd: size,
|
||
dp: i32,
|
||
negative: bool,
|
||
truncated: bool,
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:29-33. Strip trailing zeros.
|
||
fn trim(d: *decimal) void = {
|
||
let SZ_ZERO: size = (0u64: size);
|
||
let SZ_ONE: size = (1u64: size);
|
||
for (d.nd > SZ_ZERO && d.digits[d.nd - SZ_ONE] == 0u8) {
|
||
d.nd -= SZ_ONE;
|
||
};
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:35-55. Compute the digit-count
|
||
// increase for a left-shift `shift` (consults left_shift_table +
|
||
// pow5_table from stof_data.ww, bb6f840). Uses `continue` so the
|
||
// loop stays in 3-clause form for byte-id-correct post-increment.
|
||
fn leftshift_newdigits(d: *decimal, shift: u32) u32 = {
|
||
shift &= 63u32;
|
||
let x_a: u32 = (left_shift_table[shift]: u32);
|
||
let x_b: u32 = (left_shift_table[shift + 1u32]: u32);
|
||
let nn: u32 = x_a >> 11u32;
|
||
let pow5_a: u32 = 0x7FFu32 & x_a;
|
||
let pow5_b: u32 = 0x7FFu32 & x_b;
|
||
let n: u32 = pow5_b - pow5_a;
|
||
for (let i: u32 = 0u32; i < n; i += 1u32) {
|
||
let i_sz: size = (i: size);
|
||
if (i_sz >= d.nd) {
|
||
return nn - 1u32;
|
||
} else if (d.digits[i] == pow5_table[pow5_a + i]) {
|
||
continue;
|
||
} else if (d.digits[i] < pow5_table[pow5_a + i]) {
|
||
return nn - 1u32;
|
||
} else {
|
||
return nn;
|
||
};
|
||
};
|
||
return nn;
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:57-91. Shift `d` left by k bits.
|
||
fn leftshift(d: *decimal, k: u32) void = {
|
||
let SZ_ONE: size = (1u64: size);
|
||
let SZ_BOUND: size = (len(d.digits): size);
|
||
let kU64: u64 = (k: u64);
|
||
let MAXSHIFT_U32: u32 = (maxshift: u32);
|
||
os.assert(k <= MAXSHIFT_U32, "strconv.leftshift: k > maxshift");
|
||
if (d.nd == (0u64: size)) { return; };
|
||
let nn: u32 = leftshift_newdigits(d, k);
|
||
let r: int = (d.nd: int) - 1;
|
||
let w: size = (r: size) + (nn: size);
|
||
let n: u64 = 0u64;
|
||
for (r >= 0) {
|
||
n += (d.digits[r]: u64) << kU64;
|
||
let quo: u64 = n / 10u64;
|
||
let rem: u64 = n - 10u64 * quo;
|
||
if (w < SZ_BOUND) {
|
||
d.digits[w] = (rem: u8);
|
||
} else if (rem != 0u64) {
|
||
d.truncated = true;
|
||
};
|
||
n = quo;
|
||
r -= 1;
|
||
w -= SZ_ONE;
|
||
};
|
||
for (n > 0u64) {
|
||
let quo: u64 = n / 10u64;
|
||
let rem: u64 = n - 10u64 * quo;
|
||
if (w < SZ_BOUND) {
|
||
d.digits[w] = (rem: u8);
|
||
} else if (rem != 0u64) {
|
||
d.truncated = true;
|
||
};
|
||
n = quo;
|
||
w -= SZ_ONE;
|
||
};
|
||
d.nd += (nn: size);
|
||
if (d.nd > SZ_BOUND) {
|
||
d.nd = SZ_BOUND;
|
||
};
|
||
d.dp += (nn: i32);
|
||
trim(d);
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:93-134. Shift `d` right by k bits.
|
||
// Two outer Hare 2-clause loops (`for (cond; r += 1)`) are inlined
|
||
// as `for (cond) { ... r += SZ_ONE; }` since neither uses continue.
|
||
fn rightshift(d: *decimal, k: u32) void = {
|
||
let SZ_ZERO: size = (0u64: size);
|
||
let SZ_ONE: size = (1u64: size);
|
||
let SZ_BOUND: size = (len(d.digits): size);
|
||
let kU64: u64 = (k: u64);
|
||
let r: size = SZ_ZERO;
|
||
let w: size = SZ_ZERO;
|
||
let n: u64 = 0u64;
|
||
for ((n >> kU64) == 0u64) {
|
||
if (r >= d.nd) {
|
||
if (n == 0u64) {
|
||
d.nd = SZ_ZERO;
|
||
return;
|
||
};
|
||
for ((n >> kU64) == 0u64) {
|
||
n *= 10u64;
|
||
r += SZ_ONE;
|
||
};
|
||
break;
|
||
};
|
||
n = n * 10u64 + (d.digits[r]: u64);
|
||
r += SZ_ONE;
|
||
};
|
||
d.dp -= (r: i32) - 1;
|
||
if (d.dp < -(decimal_point_range: i32)) {
|
||
// Drew-watch-item: pointer-deref reset to composite
|
||
// literal — probed cs==ww byte-id safe in pre-flight.
|
||
*d = decimal { ... };
|
||
return;
|
||
};
|
||
let mask: u64 = (1u64 << kU64) - 1u64;
|
||
for (r < d.nd) {
|
||
let dig: u64 = n >> kU64;
|
||
n &= mask;
|
||
d.digits[w] = (dig: u8);
|
||
w += SZ_ONE;
|
||
n = n * 10u64 + (d.digits[r]: u64);
|
||
r += SZ_ONE;
|
||
};
|
||
for (n > 0u64) {
|
||
let dig: u64 = n >> kU64;
|
||
n &= mask;
|
||
if (w < SZ_BOUND) {
|
||
d.digits[w] = (dig: u8);
|
||
w += SZ_ONE;
|
||
} else if (dig > 0u64) {
|
||
d.truncated = true;
|
||
};
|
||
n *= 10u64;
|
||
};
|
||
d.nd = w;
|
||
trim(d);
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:138-153. Shift right (k < 0) or left
|
||
// (k > 0). Hardware shifts cap at 60 bits without losing top
|
||
// digits, so break large shifts into maxshift-sized chunks.
|
||
fn decimal_shift(d: *decimal, k: int) void = {
|
||
let MAXSHIFT_INT: int = (maxshift: int);
|
||
let MAXSHIFT_U32: u32 = (maxshift: u32);
|
||
if (d.nd == (0u64: size)) { return; };
|
||
if (k > 0) {
|
||
for (k > MAXSHIFT_INT) {
|
||
leftshift(d, MAXSHIFT_U32);
|
||
k -= MAXSHIFT_INT;
|
||
};
|
||
leftshift(d, (k: u32));
|
||
} else if (k < 0) {
|
||
for (k < -MAXSHIFT_INT) {
|
||
rightshift(d, MAXSHIFT_U32);
|
||
k += MAXSHIFT_INT;
|
||
};
|
||
rightshift(d, ((-k): u32));
|
||
};
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:155-160. Banker's rounding decision:
|
||
// at the exact half (digit==5, no more digits) round to even (the
|
||
// preceding digit's low bit decides); past-half rounds up; below-
|
||
// half rounds down. Hare's expression-bodied `if` re-shaped as a
|
||
// block per ww parser.
|
||
fn should_round_up(d: *decimal, nd: uint) bool = {
|
||
let nd_sz: size = (nd: size);
|
||
let SZ_ONE: size = (1u64: size);
|
||
let U_ONE: uint = (1u32: uint);
|
||
let U_ZERO: uint = (0u32: uint);
|
||
if (nd_sz < d.nd) {
|
||
if (d.digits[nd] == 5u8 && (nd_sz + SZ_ONE) == d.nd) {
|
||
let lowbit_lit: bool = false;
|
||
if (nd > U_ZERO) {
|
||
if ((d.digits[nd - U_ONE] & 1u8) != 0u8) {
|
||
lowbit_lit = true;
|
||
};
|
||
};
|
||
return d.truncated || lowbit_lit;
|
||
} else {
|
||
return d.digits[nd] >= 5u8;
|
||
};
|
||
};
|
||
return false;
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:162-166. Round to `nd` digits.
|
||
fn round(d: *decimal, nd: uint) void = {
|
||
if ((nd: size) >= d.nd) { return; };
|
||
if (should_round_up(d, nd)) {
|
||
roundup(d, nd);
|
||
} else {
|
||
rounddown(d, nd);
|
||
};
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:168-172. Truncate to `nd` digits.
|
||
fn rounddown(d: *decimal, nd: uint) void = {
|
||
if ((nd: size) >= d.nd) { return; };
|
||
d.nd = (nd: size);
|
||
trim(d);
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:174-186. Round up to `nd` digits;
|
||
// propagate carry. If all 9s, the result is a single 1 with the
|
||
// decimal point advanced.
|
||
fn roundup(d: *decimal, nd: uint) void = {
|
||
let SZ_ONE: size = (1u64: size);
|
||
if ((nd: size) >= d.nd) { return; };
|
||
for (let i: int = (nd: int) - 1; i >= 0; i -= 1) {
|
||
if (d.digits[i] < 9u8) {
|
||
d.digits[i] += 1u8;
|
||
d.nd = (i: size) + SZ_ONE;
|
||
return;
|
||
};
|
||
};
|
||
d.digits[0] = 1u8;
|
||
d.nd = SZ_ONE;
|
||
d.dp += 1;
|
||
};
|
||
|
||
// ref/hare/strconv/decimal.ha:188-202. Read `d` as the integer
|
||
// rounded to `d.dp` digits. Returns 0 if `d.dp <= 0`; returns
|
||
// ~0u64 if `d.dp > 18` (exceeds u64 range). Hare's two 2-clause
|
||
// loops (`for (cond; i += 1)`) are inlined per the spelling
|
||
// divergence at file top.
|
||
fn decimal_round(d: *decimal) u64 = {
|
||
let SZ_ZERO: size = (0u64: size);
|
||
let SZ_ONE: size = (1u64: size);
|
||
if (d.nd == SZ_ZERO || d.dp < 0) { return 0u64; };
|
||
if (d.dp > 18) {
|
||
// Hare's `~0u64` doesn't parse on a typed-suffix literal
|
||
// in ww; route via a named zero.
|
||
let zero: u64 = 0u64;
|
||
return ~zero;
|
||
};
|
||
let dp_sz: size = ((d.dp: uint): size);
|
||
let i: size = SZ_ZERO;
|
||
let n: u64 = 0u64;
|
||
for (i < dp_sz && i < d.nd) {
|
||
n = n * 10u64 + (d.digits[i]: u64);
|
||
i += SZ_ONE;
|
||
};
|
||
for (i < dp_sz) {
|
||
n *= 10u64;
|
||
i += SZ_ONE;
|
||
};
|
||
if (should_round_up(d, (d.dp: uint))) {
|
||
n += 1u64;
|
||
};
|
||
return n;
|
||
};
|
||
|
||
// floats — f64 classification, sign, bit-reinterpret core, and the f64
|
||
// decompose half (subnormal-normalize + frexp). Ported from
|
||
// ref/hare/math/floats.ha (fold-1: classify/sign/bits; fold-2a:
|
||
// issubnormalf64/normalizef64/frexpf64; strconv-foundation fold-1a:
|
||
// F32 bit-layout + f32bits/f32frombits + floatinfo struct type;
|
||
// fold-1b: NAN_BITS/INF_BITS sentinels; γ-cleanup: f64info/f32info
|
||
// instances re-folded once #149 lowered &math.f64info). frexpf64's
|
||
// zero guard `n == 0f64` rides the #103
|
||
// fix (no-decimal f64 literal now materialized into XMM) and its
|
||
// (f64, i64) tuple return rides the #105 fix (tuple f64-word read).
|
||
// The ldexp/modfrac/nextafter family stays deferred (need f64 DIVIDE).
|
||
|
||
package math;
|
||
|
||
// Returns the binary representation of the given f64.
|
||
// ref/hare/math/floats.ha:5. Parens around &n are load-bearing: ww's `:`
|
||
// cast binds tighter than unary `&`, so Hare's `*(&n: *u64)` would parse
|
||
// as `*(&(n: *u64))`; `(&n): *u64` reinterprets the address as intended.
|
||
export fn f64bits(n: f64) u64 = {
|
||
return *((&n): *u64);
|
||
};
|
||
|
||
// Returns the binary representation of the given f32.
|
||
// ref/hare/math/floats.ha:8
|
||
export fn f32bits(n: f32) u32 = {
|
||
return *((&n): *u32);
|
||
};
|
||
|
||
// Returns f64 with the given binary representation.
|
||
// ref/hare/math/floats.ha:11
|
||
export fn f64frombits(n: u64) f64 = {
|
||
return *((&n): *f64);
|
||
};
|
||
|
||
// Returns f32 with the given binary representation.
|
||
// ref/hare/math/floats.ha:14
|
||
export fn f32frombits(n: u32) f32 = {
|
||
return *((&n): *f32);
|
||
};
|
||
|
||
// ref/hare/math/floats.ha:17,20,23 declare these as untyped int. ww has
|
||
// no untyped def (every def carries a type) and routes shift/bitwise
|
||
// through unify_arith, which rejects mixed operand types (cmd/wcc/
|
||
// check.c:769). The bit-structure consts are used only as u64 shift
|
||
// amounts and mask widths, so they are typed u64 here — the closest
|
||
// stand-in for Hare's untyped-int adapt at those use sites.
|
||
|
||
// The number of bits in the significand of the binary representation of f64.
|
||
export def F64_MANTISSA_BITS: u64 = 52;
|
||
|
||
// The number of bits in the exponent of the binary representation of f64.
|
||
export def F64_EXPONENT_BITS: u64 = 11;
|
||
|
||
// The bias of the exponent of the binary representation of f64. Subtract this
|
||
// from the exponent in the binary representation to get the actual exponent.
|
||
export def F64_EXPONENT_BIAS: u64 = 1023;
|
||
|
||
// Mask with each bit of an f64's mantissa set.
|
||
// ref/hare/math/floats.ha:37
|
||
export def F64_MANTISSA_MASK: u64 = (1 << F64_MANTISSA_BITS) - 1;
|
||
|
||
// Mask with each bit of an f64's exponent set.
|
||
// ref/hare/math/floats.ha:40
|
||
export def F64_EXPONENT_MASK: u64 = (1 << F64_EXPONENT_BITS) - 1;
|
||
|
||
// The mask that gets an f64's sign.
|
||
// ref/hare/math/floats.ha:75
|
||
def F64_SIGN_MASK: u64 = 1u64 << 63;
|
||
|
||
// Mask that clears an f64's exponent field, keeping sign + mantissa.
|
||
// ref/hare/math/floats.ha:77. Hare hardcodes the 0x800FFFFFFFFFFFFF binary
|
||
// literal because its lexer can't const-fold the expression; ww's #88
|
||
// def-const-fold can, so the readable form is kept. floats.ha:79's NOTE
|
||
// expression has an `0u64 &` upstream typo (it would yield 0); the value it
|
||
// documents is exactly ~(F64_EXPONENT_MASK << F64_MANTISSA_BITS).
|
||
def F64_EXP_REMOVAL_MASK: u64 = ~(F64_EXPONENT_MASK << F64_MANTISSA_BITS);
|
||
|
||
// The f64 bit pattern whose exponent field evaluates to zero (0.5 scale).
|
||
// ref/hare/math/floats.ha:84
|
||
def F64_EXP_ZERO: u64 = (F64_EXPONENT_BIAS - 1) << F64_MANTISSA_BITS;
|
||
|
||
// F32 bit-structure constants. ref/hare/math/floats.ha:27,30,33 declare
|
||
// these as untyped int; ww has no untyped def, so they ride u32 (matching
|
||
// the u32 bit container, the same way the F64 family rides u64 — see
|
||
// the note above F64_MANTISSA_BITS).
|
||
|
||
// The number of bits in the significand of the binary representation of f32.
|
||
// ref/hare/math/floats.ha:27
|
||
export def F32_MANTISSA_BITS: u32 = 23u32;
|
||
|
||
// The number of bits in the exponent of the binary representation of f32.
|
||
// ref/hare/math/floats.ha:30
|
||
export def F32_EXPONENT_BITS: u32 = 8u32;
|
||
|
||
// The bias of the exponent of the binary representation of f32. Subtract this
|
||
// from the exponent in the binary representation to get the actual exponent.
|
||
// ref/hare/math/floats.ha:33
|
||
export def F32_EXPONENT_BIAS: u32 = 127u32;
|
||
|
||
// Mask with each bit of an f32's mantissa set.
|
||
// ref/hare/math/floats.ha:43
|
||
export def F32_MANTISSA_MASK: u32 = (1u32 << F32_MANTISSA_BITS) - 1u32;
|
||
|
||
// Mask with each bit of an f32's exponent set.
|
||
// ref/hare/math/floats.ha:46
|
||
export def F32_EXPONENT_MASK: u32 = (1u32 << F32_EXPONENT_BITS) - 1u32;
|
||
|
||
// The mask that gets an f32's sign.
|
||
// ref/hare/math/floats.ha:87
|
||
def F32_SIGN_MASK: u32 = 1u32 << 31;
|
||
|
||
// Mask that clears an f32's exponent field, keeping sign + mantissa.
|
||
// ref/hare/math/floats.ha:92. Hare hardcodes the binary literal (its
|
||
// lexer can't const-fold the expression); ww's #88 def-const-fold can,
|
||
// so the readable form is kept (same call as F64_EXP_REMOVAL_MASK).
|
||
def F32_EXP_REMOVAL_MASK: u32 = ~(F32_EXPONENT_MASK << F32_MANTISSA_BITS);
|
||
|
||
// The f32 bit pattern whose exponent field evaluates to zero (0.5 scale).
|
||
// ref/hare/math/floats.ha:95
|
||
def F32_EXP_ZERO: u32 = (F32_EXPONENT_BIAS - 1u32) << F32_MANTISSA_BITS;
|
||
|
||
// floatinfo — IEEE-754 shape parameters for a binary float type, passed
|
||
// to width-generic helpers in strconv (eisel_lemire, floatbits, hex_to_bits,
|
||
// mkfloat). ref/hare/math/floats.ha:101. Hare's `int` maps to ww's `int`
|
||
// (machine word, 8B; project_int_machine_word_derived_limits), so the
|
||
// expbias field stays `int` — that keeps the fold-4 stof port byte-for-byte
|
||
// against ref/hare/strconv/stof.ha:248,288 (`let e: int = 0` arithmetic
|
||
// against `f.expbias` of the same type, no cast at use site).
|
||
export type floatinfo = struct {
|
||
// Bits in significand.
|
||
mantbits: u64,
|
||
// Bits in exponent.
|
||
expbits: u64,
|
||
// Bias of exponent.
|
||
expbias: int,
|
||
// Mask for mantissa.
|
||
mantmask: u64,
|
||
// Mask for exponent.
|
||
expmask: u64,
|
||
};
|
||
|
||
// floatinfo instances for the f64 / f32 types, consumed by the
|
||
// width-generic strconv helpers via &math.f64info (cross-module
|
||
// address-of, lowered since #149). ref/hare/math/floats.ha:117,126.
|
||
// Hare spells the masks (1 << 52) - 1 / (1 << 23) - 1; the #129 A.2
|
||
// struct-composite static-init path folds only bare-literal field
|
||
// initializers, not const-fold expressions, so the value-identical hex
|
||
// literals are used here (0xFFFFFFFFFFFFF == (1<<52)-1, 0x7FFFFF ==
|
||
// (1<<23)-1 — same hex-literal style as the NAN_BITS/INF_BITS sentinels
|
||
// below). expbias rides `int` (the field type) with no suffix.
|
||
export def f64info: floatinfo = floatinfo {
|
||
mantbits = 52u64,
|
||
expbits = 11u64,
|
||
expbias = 1023,
|
||
mantmask = 0xFFFFFFFFFFFFFu64,
|
||
expmask = 0x7FFu64,
|
||
};
|
||
|
||
export def f32info: floatinfo = floatinfo {
|
||
mantbits = 23u64,
|
||
expbits = 8u64,
|
||
expbias = 127,
|
||
mantmask = 0x7FFFFFu64,
|
||
expmask = 0xFFu64,
|
||
};
|
||
|
||
// IEEE-754 quiet-NaN and positive-Infinity f64 bit sentinels.
|
||
// ref/hare/math/floats.ha:137,141. Hare exports `def NAN = 0.0/0.0;` and
|
||
// `def INF = 1.0/0.0;` (untyped float def-fold); ww's cgen doesn't lower
|
||
// `def: f64 = expr;` (the symbol comes out undefined at link time — see
|
||
// #129). Callers materialize the f64 sentinel via f64frombits(NAN_BITS)
|
||
// / f64frombits(INF_BITS); same bit-exact value, one extra reinterpret.
|
||
// 0x7FF8000000000000 is the IEEE-754 binary64 quiet-NaN (sign=0, exp=
|
||
// all-ones, mantissa MSB=1, rest=0); 0x7FF0000000000000 is +Infinity
|
||
// (sign=0, exp=all-ones, mantissa=0). Re-fold to `def NAN: f64 = ...`
|
||
// when #129 closes (γ-cleanup pattern per amalloc-drop precedent).
|
||
export def NAN_BITS: u64 = 0x7FF8000000000000u64;
|
||
export def INF_BITS: u64 = 0x7FF0000000000000u64;
|
||
|
||
// Returns true if the given floating-point number is NaN.
|
||
// ref/hare/math/floats.ha:144 (Hare's expression body inlined into a
|
||
// block: ww has no expression-bodied fn form, only brace blocks).
|
||
export fn isnan(n: f64) bool = {
|
||
return n != n;
|
||
};
|
||
|
||
// Returns true if the given floating-point number is infinite.
|
||
// ref/hare/math/floats.ha:147
|
||
export fn isinf(n: f64) bool = {
|
||
const bits = f64bits(n);
|
||
const mant = bits & F64_MANTISSA_MASK;
|
||
const exp = bits >> F64_MANTISSA_BITS & F64_EXPONENT_MASK;
|
||
return exp == F64_EXPONENT_MASK && mant == 0;
|
||
};
|
||
|
||
// Returns true if the given f64 is subnormal.
|
||
// ref/hare/math/floats.ha:179
|
||
export fn issubnormalf64(n: f64) bool = {
|
||
const bits = f64bits(n);
|
||
const mant = bits & F64_MANTISSA_MASK;
|
||
const exp = bits >> F64_MANTISSA_BITS & F64_EXPONENT_MASK;
|
||
return exp == 0 && mant != 0;
|
||
};
|
||
|
||
// Returns the absolute value of f64 n.
|
||
// ref/hare/math/floats.ha:195
|
||
export fn absf64(n: f64) f64 = {
|
||
if (isnan(n)) {
|
||
return n;
|
||
};
|
||
return f64frombits(f64bits(n) & ~F64_SIGN_MASK);
|
||
};
|
||
|
||
// Returns 1 if x is positive and -1 if x is negative. Note that zero is also
|
||
// signed.
|
||
// ref/hare/math/floats.ha:212
|
||
export fn signf64(x: f64) i64 = {
|
||
if (f64bits(x) & F64_SIGN_MASK == 0) {
|
||
return 1i64;
|
||
} else {
|
||
return -1i64;
|
||
};
|
||
};
|
||
|
||
// Returns whether or not x is positive.
|
||
// ref/hare/math/floats.ha:231
|
||
export fn ispositivef64(x: f64) bool = {
|
||
return signf64(x) == 1i64;
|
||
};
|
||
|
||
// Returns whether or not x is negative.
|
||
// ref/hare/math/floats.ha:237
|
||
export fn isnegativef64(x: f64) bool = {
|
||
return signf64(x) == -1i64;
|
||
};
|
||
|
||
// Returns x, but with the sign of y.
|
||
// ref/hare/math/floats.ha:243
|
||
export fn copysignf64(x: f64, y: f64) f64 = {
|
||
return f64frombits((f64bits(x) & ~F64_SIGN_MASK) |
|
||
(f64bits(y) & F64_SIGN_MASK));
|
||
};
|
||
|
||
// Takes a potentially subnormal f64 n and returns a normal f64 normal_float
|
||
// and an exponent exp such that n == normal_float * 2^{exp}.
|
||
// ref/hare/math/floats.ha:256
|
||
export fn normalizef64(n: f64) (f64, i64) = {
|
||
if (issubnormalf64(n)) {
|
||
const factor = 1i64 << (F64_MANTISSA_BITS: i64);
|
||
const normal_float = (n * (factor: f64));
|
||
return (normal_float, -(F64_MANTISSA_BITS: i64));
|
||
};
|
||
return (n, 0);
|
||
};
|
||
|
||
// Breaks a f64 down into its mantissa and exponent. The mantissa will be
|
||
// between 0.5 and 1.
|
||
// ref/hare/math/floats.ha:278
|
||
export fn frexpf64(n: f64) (f64, i64) = {
|
||
if (isnan(n) || isinf(n) || n == 0f64) {
|
||
return (n, 0);
|
||
};
|
||
const normalized = normalizef64(n);
|
||
const normal_float = normalized.0;
|
||
const normalization_exp = normalized.1;
|
||
const bits = f64bits(normal_float);
|
||
const raw_exp: u64 = (bits >> F64_MANTISSA_BITS) & F64_EXPONENT_MASK;
|
||
const exp: i64 = normalization_exp +
|
||
(raw_exp: i64) - (F64_EXPONENT_BIAS: i64) + 1;
|
||
const mantissa: f64 =
|
||
f64frombits((bits & F64_EXP_REMOVAL_MASK) | F64_EXP_ZERO);
|
||
return (mantissa, exp);
|
||
};
|
||
|
||
// math — numeric helpers. Subset of Hare's math::; only the absolute-
|
||
// value pair for the signed integer types we currently care about. The
|
||
// return type is unsigned so that abs(I32_MIN) doesn't overflow.
|
||
|
||
package math;
|
||
|
||
export fn absi32(n: i32) u32 = {
|
||
if (n < 0) { return (-n): u32; };
|
||
return n: u32;
|
||
};
|
||
|
||
export fn absi64(n: i64) u64 = {
|
||
if (n < 0) { return (-n): u64; };
|
||
return n: u64;
|
||
};
|
||
|
||
// strconv — float→string via Ryū (shortest round-trippable decimal).
|
||
// Mirrors ref/hare/strconv/ftos_ryu.ha (the algorithm core) +
|
||
// ref/hare/strconv/ftos.ha:432 (the f64tos driver). Ryū: Ulf Adams,
|
||
// https://doi.org/10.1145/3192366.3192369 — Hare translated it from the
|
||
// reference C (https://github.com/ulfjack/ryu); ww follows Hare.
|
||
//
|
||
// SCOPE — the f64tos + f32tos shortest-representation subset (Hare's
|
||
// ffmt::G, prec=void, fflags::NONE). f32tos (ftos.ha:448) + its f32 Ryū
|
||
// sub-path (f32todecf32 + mulpow5inv/pow5_divpow2 + mulshift32 + the *32
|
||
// helpers, reusing the shared u64-core + the f64 SPLIT2 tables — the f32
|
||
// path has no separate tables, matching ftos_ryu.ha) ship here in fold-5b
|
||
// (task #67): the gating #143 f32-arg-push cgen fix landed (aff7725, MOVSS
|
||
// both stages), so f32tos's math.f32bits(n) call — passing an f32 arg — is
|
||
// now byte-id-clean. One deferral remains:
|
||
// - the parametric fftosf/ffmt/fflags/ftosf surface → task #64 (needs
|
||
// io::handle/memio + a `(size|io::error)?` per appendrune (#158);
|
||
// for G/void/NONE the ffmt/fflags/precision/multiprecision-fallback
|
||
// machinery is provably dead code — `ok` is always true → init_dec/
|
||
// compute_round/round unreachable — which bootstrap-coverage rejects).
|
||
// The lib note blesses "a documented subset". This file ships ZERO float
|
||
// literals — Ryū is all bit/integer arithmetic on f64bits(n) — so the
|
||
// wwdump TK_FLOAT embedding concern is moot.
|
||
//
|
||
// Decomposition divergences (the #163-166 tuple/struct-ABI cluster —
|
||
// ww's partial tuple support miscompiles the shapes Hare uses; the
|
||
// WORKING shapes, struct-RETURN + scalar-PARAMS, are this algorithm's
|
||
// own idiom: ftos_ryu.ha:12 already uses `struct r128` not a tuple for
|
||
// u128mul, and fold-4/stof.ww decomposed likewise):
|
||
// - `mulshiftall64`'s tuple param `mul:(u64,u64)` → two scalar params
|
||
// `mul0,mul1` (#163: tuple-as-param reads garbage); its 3-tuple
|
||
// return `(u64,u64,u64)` → 24B struct `ryuv` (#164: 3-tuple return
|
||
// reads 0; struct-RETURN is byte-id-clean — r128 precedent). NO
|
||
// struct-as-PARAM anywhere (#165: 16B struct-param diverges cs≠ww).
|
||
// - `f64computeinvpow5`/`f64computepow5` keep their 2-tuple `(u64,u64)`
|
||
// return (call-return 2-tuple + `.0`/`.1` is byte-id-clean — the
|
||
// math/floats.ww frexpf64 precedent).
|
||
// - dead `mulshift64` (tuple-param, never called) + dead
|
||
// `F32/F64_DECIMAL_DIGITS` dropped.
|
||
//
|
||
// Spelling divergences (mechanical, ww parser/cgen; cite ftos_ryu.ha):
|
||
// - scalar-PARAM mutation (`m<<=1`, `value*=…`) → copy-to-local
|
||
// (stof.ww hex_to_bits precedent).
|
||
// - `&&=` → `x = x && y`. `ibool=if(b)1 else 0` expr-body → block.
|
||
// comma `let a=…, b=…` → split. `if/else` expr-yield → pre-bound
|
||
// local + block. `assert()` → `os.assert(cond,msg)`.
|
||
// - 2D row-bind `mul=TBL[base]` → direct double-index `TBL[base][0/1]`
|
||
// (#155 / #156, stof.ww eisel_lemire precedent).
|
||
// - ibool's u8 result + the u8 BITCOUNT defs cast explicitly to u32/u64
|
||
// at each use (Hare promotes; ww is strict — int-machine-word note).
|
||
// - a `(N: uint)` cast embedded inside an array subscript `[ ]` is
|
||
// rejected by the ww parser → hoist to a named local before the
|
||
// index (decimal.ww "hoist size casts" note); see init_dec_mant_exp
|
||
// + encode_e_dec.
|
||
|
||
package strconv;
|
||
|
||
import math;
|
||
import os;
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:33. (hi:lo) >> s, low 64 bits. Hare's
|
||
// "TODO: use 128-bit integers" — ww has no u128; pure-u64 decomposition.
|
||
// (u128mul + the r128 struct live in stof.ww, fold-4's first consumer;
|
||
// reused in-package here.)
|
||
fn u128rshift(lo: u64, hi: u64, s: u32) u64 = {
|
||
os.assert(s <= 64u32, "strconv.u128rshift: s > 64");
|
||
return (hi << (64u64 - (s: u64))) | (lo >> (s: u64));
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:39. Largest p with 5^p | value.
|
||
fn pow5fac(v: u64) u32 = {
|
||
let value: u64 = v;
|
||
let m_inv_5: u64 = 14757395258967641293u64; // 5 * m_inv_5 == 1 (mod 2^64)
|
||
let n_div_5: u64 = 3689348814741910323u64;
|
||
let count: u32 = 0u32;
|
||
for (true) {
|
||
os.assert(value != 0u64, "strconv.pow5fac: value == 0");
|
||
value *= m_inv_5;
|
||
if (value > n_div_5) { break; };
|
||
count += 1u32;
|
||
};
|
||
return count;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:64.
|
||
fn ibool(b: bool) u8 = {
|
||
if (b) { return 1u8; };
|
||
return 0u8;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:66-67.
|
||
fn pow5multiple(v: u64, p: u32) bool = { return pow5fac(v) >= p; };
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:69.
|
||
fn pow2multiple(v: u64, p: u32) bool = {
|
||
os.assert(v > 0u64, "strconv.pow2multiple: v == 0");
|
||
os.assert(p < 64u32, "strconv.pow2multiple: p >= 64");
|
||
return (v & ((1u64 << (p: u64)) - 1u64)) == 0u64;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:89. The (v+, v-rounded, v-) triple.
|
||
// Decomposed: tuple param → mul0/mul1 scalars (#163); 3-tuple return →
|
||
// this struct (#164). The `mm_shift==1` `if/else`-yield → pre-bound
|
||
// `v_minus` + block.
|
||
type ryuv = struct { vp: u64, vr: u64, vm: u64 };
|
||
|
||
fn mulshiftall64(m: u64, mul0: u64, mul1: u64, j: i32, mm_shift: u32) ryuv = {
|
||
let mm: u64 = m << 1u64;
|
||
let r0: r128 = u128mul(mm, mul0);
|
||
let r1: r128 = u128mul(mm, mul1);
|
||
let lo: u64 = r0.lo;
|
||
let tmp: u64 = r0.hi;
|
||
let mid: u64 = tmp + r1.lo;
|
||
let hi: u64 = r1.hi + (ibool(mid < tmp): u64);
|
||
let lo2: u64 = lo + mul0;
|
||
let mid2: u64 = mid + mul1 + (ibool(lo2 < lo): u64);
|
||
let hi2: u64 = hi + (ibool(mid2 < mid): u64);
|
||
let v_plus: u64 = u128rshift(mid2, hi2, ((j - 64 - 1): u32));
|
||
let v_minus: u64 = 0u64;
|
||
if (mm_shift == 1u32) {
|
||
let lo3: u64 = lo - mul0;
|
||
let mid3: u64 = mid - mul1 - (ibool(lo3 > lo): u64);
|
||
let hi3: u64 = hi - (ibool(mid3 > mid): u64);
|
||
v_minus = u128rshift(mid3, hi3, ((j - 64 - 1): u32));
|
||
} else {
|
||
let lo3: u64 = lo + lo;
|
||
let mid3: u64 = mid + mid + (ibool(lo3 < lo): u64);
|
||
let hi3: u64 = hi + hi + (ibool(mid3 < mid): u64);
|
||
let lo4: u64 = lo3 - mul0;
|
||
let mid4: u64 = mid3 - mul1 - (ibool(lo4 > lo3): u64);
|
||
let hi4: u64 = hi3 - (ibool(mid4 > mid3): u64);
|
||
v_minus = u128rshift(mid4, hi4, ((j - 64): u32));
|
||
};
|
||
let v_rounded: u64 = u128rshift(mid, hi, ((j - 64 - 1): u32));
|
||
return ryuv { vp = v_plus, vr = v_rounded, vm = v_minus };
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:140.
|
||
fn log2pow5(e: u32) u32 = {
|
||
os.assert(e <= 3528u32, "strconv.log2pow5: e > 3528");
|
||
return (e * 1217359u32) >> 19u32;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:145-147.
|
||
fn ceil_log2pow5(e: u32) u32 = { return log2pow5(e) + 1u32; };
|
||
fn pow5bits(e: u32) u32 = { return ceil_log2pow5(e); };
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:149.
|
||
fn log10pow2(e: u32) u32 = {
|
||
os.assert(e <= 1650u32, "strconv.log10pow2: e > 1650");
|
||
return (e * 78913u32) >> 18u32;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:154.
|
||
fn log10pow5(e: u32) u32 = {
|
||
os.assert(e <= 2620u32, "strconv.log10pow5: e > 2620");
|
||
return (e * 732923u32) >> 20u32;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:224. Returns the (low, high) split of the
|
||
// inverse power of five. 2-tuple kept (works); row-bind → double-index.
|
||
fn f64computeinvpow5(i: u32) (u64, u64) = {
|
||
let base: u32 = (i + (POW5_TABLE_SZ: u32) - 1u32) / (POW5_TABLE_SZ: u32);
|
||
let base2: u32 = base * (POW5_TABLE_SZ: u32);
|
||
let off: u32 = base2 - i;
|
||
if (off == 0u32) {
|
||
return (F64_POW5_INV_SPLIT2[base][0], F64_POW5_INV_SPLIT2[base][1]);
|
||
};
|
||
let m: u64 = POW5_TABLE[off];
|
||
let r1: r128 = u128mul(m, F64_POW5_INV_SPLIT2[base][1]);
|
||
let r0: r128 = u128mul(m, F64_POW5_INV_SPLIT2[base][0] - 1u64);
|
||
let high1: u64 = r1.hi;
|
||
let low1: u64 = r1.lo;
|
||
let high0: u64 = r0.hi;
|
||
let low0: u64 = r0.lo;
|
||
let sum: u64 = high0 + low1;
|
||
if (sum < high0) {
|
||
high1 += 1u64;
|
||
};
|
||
let delta: u32 = pow5bits(base2) - pow5bits(i);
|
||
let res0: u64 = u128rshift(low0, sum, delta) + 1u64 +
|
||
(((POW5_INV_OFFSETS[i / 16u32] >> ((i % 16u32) << 1u32)) & 3u32): u64);
|
||
let res1: u64 = u128rshift(sum, high1, delta);
|
||
return (res0, res1);
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:246.
|
||
fn f64computepow5(i: u32) (u64, u64) = {
|
||
let base: u32 = i / (POW5_TABLE_SZ: u32);
|
||
let base2: u32 = base * (POW5_TABLE_SZ: u32);
|
||
let off: u32 = i - base2;
|
||
if (off == 0u32) {
|
||
return (F64_POW5_SPLIT2[base][0], F64_POW5_SPLIT2[base][1]);
|
||
};
|
||
let m: u64 = POW5_TABLE[off];
|
||
let r1: r128 = u128mul(m, F64_POW5_SPLIT2[base][1]);
|
||
let r0: r128 = u128mul(m, F64_POW5_SPLIT2[base][0]);
|
||
let high1: u64 = r1.hi;
|
||
let low1: u64 = r1.lo;
|
||
let high0: u64 = r0.hi;
|
||
let low0: u64 = r0.lo;
|
||
let sum: u64 = high0 + low1;
|
||
if (sum < high0) {
|
||
high1 += 1u64;
|
||
};
|
||
let delta: u32 = pow5bits(i) - pow5bits(base2);
|
||
let res0: u64 = u128rshift(low0, sum, delta) +
|
||
(((POW5_OFFSETS[i / 16u32] >> ((i % 16u32) << 1u32)) & 3u32): u64);
|
||
let res1: u64 = u128rshift(sum, high1, delta);
|
||
return (res0, res1);
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:267. Shortest decimal of an f64:
|
||
// value == mantissa * 10^exponent. `exponent` rides i64 not Hare's i32
|
||
// (ftos_ryu.ha:269): a 16B struct-return with a NARROW (i32) second
|
||
// field unpacks MOVL in wwstage vs MOVQ in cstage (store-width cs≠ww
|
||
// byte-id split, #169); an 8B i64 field unpacks MOVQ in both. The value
|
||
// always fits i32 (cast at the init_dec_mant_exp call site).
|
||
type decf64 = struct { mantissa: u64, exponent: i64 };
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:272. `mantissa`/`exponent` are the raw
|
||
// IEEE-754 fields of an f64.
|
||
fn f64todecf64(mantissa: u64, exponent: u32) decf64 = {
|
||
let e2: i32 = (math.F64_EXPONENT_BIAS + math.F64_MANTISSA_BITS + 2u64): i32;
|
||
let m2: u64 = 0u64;
|
||
if (exponent == 0u32) {
|
||
e2 = 1i32 - e2;
|
||
m2 = mantissa;
|
||
} else {
|
||
e2 = (exponent: i32) - e2;
|
||
m2 = (1u64 << math.F64_MANTISSA_BITS) | mantissa;
|
||
};
|
||
let accept_bounds: bool = (m2 & 1u64) == 0u64;
|
||
let mv: u64 = 4u64 * m2;
|
||
let mm_shift: u32 = ibool(mantissa != 0u64 || exponent <= 1u32): u32;
|
||
let vp: u64 = 0u64;
|
||
let vr: u64 = 0u64;
|
||
let vm: u64 = 0u64;
|
||
let e10: i32 = 0i32;
|
||
let vm_trailing_zeros: bool = false;
|
||
let vr_trailing_zeros: bool = false;
|
||
if (e2 >= 0i32) {
|
||
let q: u32 = log10pow2(e2: u32) - (ibool(e2 > 3i32): u32);
|
||
e10 = q: i32;
|
||
let k: u32 = (F64_POW5_INV_BITCOUNT: u32) + pow5bits(q) - 1u32;
|
||
let i: i32 = -e2 + ((q + k): i32);
|
||
let pow5 = f64computeinvpow5(q);
|
||
let res: ryuv = mulshiftall64(m2, pow5.0, pow5.1, i, mm_shift);
|
||
vp = res.vp; vr = res.vr; vm = res.vm;
|
||
if (q <= 21u32) {
|
||
if ((mv - 5u64 * (mv / 5u64)) == 0u64) {
|
||
vr_trailing_zeros = pow5multiple(mv, q);
|
||
} else if (accept_bounds) {
|
||
vm_trailing_zeros = pow5multiple(mv - 1u64 - (mm_shift: u64), q);
|
||
} else {
|
||
vp -= (ibool(pow5multiple(mv + 2u64, q)): u64);
|
||
};
|
||
};
|
||
} else {
|
||
let q: u32 = log10pow5((-e2): u32) - (ibool(-e2 > 1i32): u32);
|
||
e10 = e2 + (q: i32);
|
||
let i: i32 = -e2 - (q: i32);
|
||
let k: i32 = (pow5bits(i: u32): i32) - (F64_POW5_BITCOUNT: i32);
|
||
let j: i32 = (q: i32) - k;
|
||
let pow5 = f64computepow5(i: u32);
|
||
let res: ryuv = mulshiftall64(m2, pow5.0, pow5.1, j, mm_shift);
|
||
vp = res.vp; vr = res.vr; vm = res.vm;
|
||
if (q <= 1u32) {
|
||
vr_trailing_zeros = true;
|
||
if (accept_bounds) {
|
||
vm_trailing_zeros = mm_shift == 1u32;
|
||
} else {
|
||
vp -= 1u64;
|
||
};
|
||
} else if (q < 63u32) {
|
||
vr_trailing_zeros = pow2multiple(mv, q);
|
||
};
|
||
};
|
||
let removed: i32 = 0i32;
|
||
let last_removed_digit: u8 = 0u8;
|
||
let output: u64 = 0u64;
|
||
if (vm_trailing_zeros || vr_trailing_zeros) {
|
||
for (true) {
|
||
let vpby10: u64 = vp / 10u64;
|
||
let vmby10: u64 = vm / 10u64;
|
||
if (vpby10 <= vmby10) { break; };
|
||
let vmmod10: u32 = (vm: u32) - 10u32 * (vmby10: u32);
|
||
let vrby10: u64 = vr / 10u64;
|
||
let vrmod10: u32 = (vr: u32) - 10u32 * (vrby10: u32);
|
||
vm_trailing_zeros = vm_trailing_zeros && (vmmod10 == 0u32);
|
||
vr_trailing_zeros = vr_trailing_zeros && (last_removed_digit == 0u8);
|
||
last_removed_digit = (vrmod10: u8);
|
||
vr = vrby10; vp = vpby10; vm = vmby10;
|
||
removed += 1i32;
|
||
};
|
||
if (vm_trailing_zeros) {
|
||
for (true) {
|
||
let vmby10: u64 = vm / 10u64;
|
||
let vmmod10: u32 = (vm: u32) - 10u32 * (vmby10: u32);
|
||
if (vmmod10 != 0u32) { break; };
|
||
let vpby10: u64 = vp / 10u64;
|
||
let vrby10: u64 = vr / 10u64;
|
||
let vrmod10: u32 = (vr: u32) - 10u32 * (vrby10: u32);
|
||
vr_trailing_zeros = vr_trailing_zeros && (last_removed_digit == 0u8);
|
||
last_removed_digit = (vrmod10: u8);
|
||
vr = vrby10; vp = vpby10; vm = vmby10;
|
||
removed += 1i32;
|
||
};
|
||
};
|
||
if (vr_trailing_zeros && last_removed_digit == 5u8 && (vr & 1u64) == 0u64) {
|
||
last_removed_digit = 4u8; // round to even
|
||
};
|
||
let cond1: bool = (vr == vm) && ((!accept_bounds) || (!vm_trailing_zeros));
|
||
let cond2: bool = last_removed_digit >= 5u8;
|
||
output = vr + (ibool(cond1 || cond2): u64);
|
||
} else {
|
||
let round_up: bool = false;
|
||
let vpby100: u64 = vp / 100u64;
|
||
let vmby100: u64 = vm / 100u64;
|
||
if (vpby100 > vmby100) {
|
||
let vrby100: u64 = vr / 100u64;
|
||
let vrmod100: u32 = (vr: u32) - 100u32 * (vrby100: u32);
|
||
round_up = vrmod100 >= 50u32;
|
||
vr = vrby100; vp = vpby100; vm = vmby100;
|
||
removed += 2i32;
|
||
};
|
||
for (true) {
|
||
let vmby10: u64 = vm / 10u64;
|
||
let vpby10: u64 = vp / 10u64;
|
||
if (vpby10 <= vmby10) { break; };
|
||
let vrby10: u64 = vr / 10u64;
|
||
let vrmod10: u32 = (vr: u32) - 10u32 * (vrby10: u32);
|
||
round_up = vrmod10 >= 5u32;
|
||
vr = vrby10; vp = vpby10; vm = vmby10;
|
||
removed += 1i32;
|
||
};
|
||
output = vr + (ibool(vr == vm || round_up): u64);
|
||
};
|
||
let exp: i32 = e10 + removed;
|
||
return decf64 { exponent = (exp: i64), mantissa = output };
|
||
};
|
||
|
||
// ==== f32 Ryū sub-path (ftos_ryu.ha). The *32 helpers below mirror their
|
||
// u64 siblings at 32-bit width; they reuse the SHARED f64computeinvpow5/
|
||
// f64computepow5 (and thus the f64 SPLIT2 tables) per ftos_ryu.ha — there
|
||
// is no separate f32 table. Same scalar-PARAM-mutation → copy-to-local,
|
||
// comma-split, assert → os.assert, expr-yield → block divergences as the
|
||
// f64 path above. ====
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:52. Largest p with 5^p | value (32-bit).
|
||
fn pow5fac32(v: u32) u32 = {
|
||
let value: u32 = v;
|
||
let count: u32 = 0u32;
|
||
for (true) {
|
||
os.assert(value != 0u32, "strconv.pow5fac32: value == 0");
|
||
let q: u32 = value / 5u32;
|
||
let r: u32 = value % 5u32;
|
||
if (r != 0u32) { break; };
|
||
value = q;
|
||
count += 1u32;
|
||
};
|
||
return count;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:67.
|
||
fn pow5multiple32(v: u32, p: u32) bool = { return pow5fac32(v) >= p; };
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:75.
|
||
fn pow2multiple32(v: u32, p: u32) bool = {
|
||
os.assert(v > 0u32, "strconv.pow2multiple32: v == 0");
|
||
os.assert(p < 32u32, "strconv.pow2multiple32: p >= 32");
|
||
return (v & ((1u32 << p) - 1u32)) == 0u32;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:121. `m * a_lo` etc. carry an explicit
|
||
// (m: u64) cast (Hare promotes the u32 operand; ww is strict). The bound
|
||
// assert inlines U32_MAX's value: ww's types.U32_MAX is package-private
|
||
// (lib/types/types.ww — no `export`), so Hare's `types::U32_MAX` can't be
|
||
// referenced cross-package.
|
||
fn mulshift32(m: u32, a: u64, s: u32) u32 = {
|
||
os.assert(s > 32u32, "strconv.mulshift32: s <= 32");
|
||
let a_lo: u64 = (a: u32): u64;
|
||
let a_hi: u64 = a >> 32u64;
|
||
let b0: u64 = (m: u64) * a_lo;
|
||
let b1: u64 = (m: u64) * a_hi;
|
||
let sum: u64 = (b0 >> 32u64) + b1;
|
||
let ss: u64 = sum >> ((s: u64) - 32u64);
|
||
os.assert(ss <= 4294967295u64, "strconv.mulshift32: ss > U32_MAX");
|
||
return ss: u32;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:130.
|
||
fn mulpow5inv_divpow2(m: u32, q: u32, j: i32) u32 = {
|
||
let pow5 = f64computeinvpow5(q);
|
||
return mulshift32(m, pow5.1 + 1u64, (j: u32));
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:135.
|
||
fn mulpow5_divpow2(m: u32, i: u32, j: i32) u32 = {
|
||
let pow5 = f64computepow5(i);
|
||
return mulshift32(m, pow5.1, (j: u32));
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:387. `exponent` rides i64 not Hare's i32,
|
||
// for the same reason decf64 does: widening the field to a full second
|
||
// eightbyte SIDESTEPS the #169 narrow-i32-field struct-return unpack (a
|
||
// narrow i32 there unpacks MOVL wwstage vs MOVQ cstage). The value always
|
||
// fits i32 (cast at the init_dec_mant_exp call site). `mantissa` stays u32
|
||
// (Hare's width); the {u32, pad, i64} layout's first eightbyte holds
|
||
// mantissa@0 + 4B pad and reads cleanly — byte-id CONFIRMED by the 990-997
|
||
// gate (0-diff cs vs ww), not relied on as an ABI guarantee.
|
||
type decf32 = struct { mantissa: u32, exponent: i64 };
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:392. Shortest decimal of an f32:
|
||
// value == mantissa * 10^exponent. `mantissa`/`exponent` are the raw
|
||
// IEEE-754 fields of an f32.
|
||
fn f32todecf32(mantissa: u32, exponent: u32) decf32 = {
|
||
let e2: i32 = (math.F32_EXPONENT_BIAS + math.F32_MANTISSA_BITS + 2u32): i32;
|
||
let m2: u32 = 0u32;
|
||
if (exponent == 0u32) {
|
||
e2 = 1i32 - e2;
|
||
m2 = mantissa;
|
||
} else {
|
||
e2 = (exponent: i32) - e2;
|
||
m2 = (1u32 << math.F32_MANTISSA_BITS) | mantissa;
|
||
};
|
||
let accept_bounds: bool = (m2 & 1u32) == 0u32;
|
||
let mv: u32 = 4u32 * m2;
|
||
let mp: u32 = mv + 2u32;
|
||
let mm_shift: u32 = ibool(mantissa != 0u32 || exponent <= 1u32): u32;
|
||
let mm: u32 = mv - 1u32 - mm_shift;
|
||
let vr: u32 = 0u32;
|
||
let vp: u32 = 0u32;
|
||
let vm: u32 = 0u32;
|
||
let e10: i32 = 0i32;
|
||
let vm_trailing_zeroes: bool = false;
|
||
let vr_trailing_zeroes: bool = false;
|
||
let last_removed_digit: u8 = 0u8;
|
||
if (e2 >= 0i32) {
|
||
let q: u32 = log10pow2(e2: u32);
|
||
e10 = q: i32;
|
||
let k: u32 = (F32_POW5_INV_BITCOUNT: u32) + pow5bits(q) - 1u32;
|
||
let i: i32 = -e2 + ((q + k): i32);
|
||
vr = mulpow5inv_divpow2(mv, q, i);
|
||
vp = mulpow5inv_divpow2(mp, q, i);
|
||
vm = mulpow5inv_divpow2(mm, q, i);
|
||
if (q != 0u32 && (vp - 1u32) / 10u32 <= vm / 10u32) {
|
||
let l: u32 = (F32_POW5_INV_BITCOUNT: u32) + pow5bits(q - 1u32) - 1u32;
|
||
last_removed_digit = (mulpow5inv_divpow2(mv, q - 1u32,
|
||
-e2 + ((q + l): i32) - 1i32) % 10u32): u8;
|
||
};
|
||
if (q <= 9u32) {
|
||
if (mv % 5u32 == 0u32) {
|
||
vr_trailing_zeroes = pow5multiple32(mv, q);
|
||
} else if (accept_bounds) {
|
||
vm_trailing_zeroes = pow5multiple32(mm, q);
|
||
} else {
|
||
vp -= (ibool(pow5multiple32(mp, q)): u32);
|
||
};
|
||
};
|
||
} else {
|
||
let q: u32 = log10pow5((-e2): u32);
|
||
e10 = (q: i32) + e2;
|
||
let i: u32 = (-e2 - (q: i32)): u32;
|
||
let k: u32 = pow5bits(i) - (F32_POW5_BITCOUNT: u32);
|
||
let j: i32 = (q: i32) - (k: i32);
|
||
vr = mulpow5_divpow2(mv, i, j);
|
||
vp = mulpow5_divpow2(mp, i, j);
|
||
vm = mulpow5_divpow2(mm, i, j);
|
||
if (q != 0u32 && (vp - 1u32) / 10u32 <= vm / 10u32) {
|
||
j = (q: i32) - 1i32 - ((pow5bits(i + 1u32): i32) - (F32_POW5_BITCOUNT: i32));
|
||
last_removed_digit = (mulpow5_divpow2(mv, (i + 1u32), j) % 10u32): u8;
|
||
};
|
||
if (q <= 1u32) {
|
||
vr_trailing_zeroes = true;
|
||
if (accept_bounds) {
|
||
vm_trailing_zeroes = mm_shift == 1u32;
|
||
} else {
|
||
vp -= 1u32;
|
||
};
|
||
} else if (q < 31u32) {
|
||
vr_trailing_zeroes = pow2multiple32(mv, q - 1u32);
|
||
};
|
||
};
|
||
let removed: i32 = 0i32;
|
||
let output: u32 = 0u32;
|
||
if (vm_trailing_zeroes || vr_trailing_zeroes) {
|
||
for ((vp / 10u32) > (vm / 10u32)) {
|
||
vm_trailing_zeroes = vm_trailing_zeroes && ((vm - (vm / 10u32) * 10u32) == 0u32);
|
||
vr_trailing_zeroes = vr_trailing_zeroes && (last_removed_digit == 0u8);
|
||
last_removed_digit = (vr % 10u32): u8;
|
||
vr /= 10u32;
|
||
vp /= 10u32;
|
||
vm /= 10u32;
|
||
removed += 1i32;
|
||
};
|
||
if (vm_trailing_zeroes) {
|
||
for ((vm % 10u32) == 0u32) {
|
||
vr_trailing_zeroes = vr_trailing_zeroes && (last_removed_digit == 0u8);
|
||
last_removed_digit = (vr % 10u32): u8;
|
||
vr /= 10u32;
|
||
vp /= 10u32;
|
||
vm /= 10u32;
|
||
removed += 1i32;
|
||
};
|
||
};
|
||
if (vr_trailing_zeroes && last_removed_digit == 5u8 && vr % 2u32 == 0u32) {
|
||
last_removed_digit = 4u8; // round to even
|
||
};
|
||
let cond1: bool = (vr == vm) && ((!accept_bounds) || (!vm_trailing_zeroes));
|
||
let cond2: bool = last_removed_digit >= 5u8;
|
||
output = vr + (ibool(cond1 || cond2): u32);
|
||
} else {
|
||
for ((vp / 10u32) > (vm / 10u32)) {
|
||
last_removed_digit = (vr % 10u32): u8;
|
||
vr /= 10u32;
|
||
vp /= 10u32;
|
||
vm /= 10u32;
|
||
removed += 1i32;
|
||
};
|
||
output = vr + (ibool(vr == vm || last_removed_digit >= 5u8): u32);
|
||
};
|
||
let exp: i32 = e10 + removed;
|
||
return decf32 { mantissa = output, exponent = (exp: i64) };
|
||
};
|
||
|
||
// ==== G-format encode layer (ftos.ha) — only the ffmt::G / prec=void /
|
||
// fflags::NONE-REACHABLE logic. The SHOW_POINT/precision/E-vs-uppercase
|
||
// arms (ftos.ha:88-105, 127-145, 170-213's zeros/caps) are UNREACHABLE
|
||
// for G/void/NONE (ffpoint(NONE)=false, prec is never uint, f is always
|
||
// G) and are NOT ported — porting them stubbed would be untested dead
|
||
// code. The parametric ftosf/ffmt/fflags surface is deferred (task #64;
|
||
// needs a parametric consumer + io::handle + #158). ====
|
||
|
||
// ref/hare/strconv/ftos.ha:49. Decimal digit-count of n (n <= 1e17).
|
||
fn declen(n: u64) uint = {
|
||
os.assert(n <= 100000000000000000u64, "strconv.declen: n > 1e17");
|
||
if (n >= 100000000000000000u64) { return (18u32: uint); };
|
||
if (n >= 10000000000000000u64) { return (17u32: uint); };
|
||
if (n >= 1000000000000000u64) { return (16u32: uint); };
|
||
if (n >= 100000000000000u64) { return (15u32: uint); };
|
||
if (n >= 10000000000000u64) { return (14u32: uint); };
|
||
if (n >= 1000000000000u64) { return (13u32: uint); };
|
||
if (n >= 100000000000u64) { return (12u32: uint); };
|
||
if (n >= 10000000000u64) { return (11u32: uint); };
|
||
if (n >= 1000000000u64) { return (10u32: uint); };
|
||
if (n >= 100000000u64) { return (9u32: uint); };
|
||
if (n >= 10000000u64) { return (8u32: uint); };
|
||
if (n >= 1000000u64) { return (7u32: uint); };
|
||
if (n >= 100000u64) { return (6u32: uint); };
|
||
if (n >= 10000u64) { return (5u32: uint); };
|
||
if (n >= 1000u64) { return (4u32: uint); };
|
||
if (n >= 100u64) { return (3u32: uint); };
|
||
if (n >= 10u64) { return (2u32: uint); };
|
||
return (1u32: uint);
|
||
};
|
||
|
||
// ref/hare/strconv/ftos.ha:217. Lay the Ryū shortest (mantissa,exponent)
|
||
// into the decimal `d`. `mantissa` is mutated in Hare → local `mant`.
|
||
fn init_dec_mant_exp(d: *decimal, mantissa: u64, exponent: i32) void = {
|
||
// Hoisted uint casts: ww parser rejects a `(N: uint)` cast embedded
|
||
// inside an array subscript (decimal.ww "hoist size casts" note).
|
||
let U_ZERO: uint = (0u32: uint);
|
||
let U_ONE: uint = (1u32: uint);
|
||
let mant: u64 = mantissa;
|
||
let dl: uint = declen(mant);
|
||
let i: uint = U_ZERO;
|
||
for (i < dl) {
|
||
d.digits[dl - i - U_ONE] = (mant % 10u64): u8;
|
||
mant /= 10u64;
|
||
i += U_ONE;
|
||
};
|
||
d.nd = (dl: size);
|
||
d.dp = (dl: i32) + exponent;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos.ha:71. writestr → buffer-cursor adaptation (the
|
||
// *tos static-buffer convention replaces Hare's io::handle sink).
|
||
fn putstr(buf: []u8, out: i32, s: str) i32 = {
|
||
let o: i32 = out;
|
||
let k: i32 = 0i32;
|
||
for (k < s.len) {
|
||
buf[o] = s[k];
|
||
o += 1i32;
|
||
k += 1i32;
|
||
};
|
||
return o;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos.ha:109. Fixed-point render (G/void/NONE-reachable
|
||
// logic only). Writes into `buf` at cursor `out`, returns the new cursor.
|
||
fn encode_f_dec(d: *decimal, buf: []u8, out: i32) i32 = {
|
||
let o: i32 = out;
|
||
let lo: i32 = 0i32;
|
||
if (d.dp <= 0i32) { lo = d.dp - 1i32; };
|
||
let hi: i32 = d.dp;
|
||
if ((d.nd: i32) > d.dp) { hi = (d.nd: i32); };
|
||
if (hi > (d.nd: i32) && d.dp <= 0i32) {
|
||
hi = (d.nd: i32);
|
||
} else if (hi > d.dp && d.dp > 0i32) {
|
||
hi = d.dp;
|
||
if ((d.nd: i32) > d.dp) { hi = (d.nd: i32); };
|
||
};
|
||
let i: i32 = lo;
|
||
for (i < hi) {
|
||
if (i == d.dp) {
|
||
buf[o] = 46u8; // '.'
|
||
o += 1i32;
|
||
};
|
||
if (0i32 <= i && i < (d.nd: i32)) {
|
||
buf[o] = (d.digits[i] + 48u8): u8;
|
||
} else {
|
||
buf[o] = 48u8; // '0'
|
||
};
|
||
o += 1i32;
|
||
i += 1i32;
|
||
};
|
||
return o;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos.ha:160. Scientific render (G/void/NONE-reachable
|
||
// logic only): no precision zeros, lowercase 'e', no '+'/two-digit pad.
|
||
fn encode_e_dec(d: *decimal, buf: []u8, out: i32) i32 = {
|
||
let o: i32 = out;
|
||
os.assert(d.nd > (0u64: size), "strconv.encode_e_dec: nd == 0");
|
||
buf[o] = (d.digits[0] + 48u8): u8;
|
||
o += 1i32;
|
||
if ((d.nd: i32) > 1i32) {
|
||
buf[o] = 46u8; // '.'
|
||
o += 1i32;
|
||
};
|
||
let i: size = (1u64: size);
|
||
for (i < d.nd) {
|
||
buf[o] = (d.digits[i] + 48u8): u8;
|
||
o += 1i32;
|
||
i += (1u64: size);
|
||
};
|
||
buf[o] = 101u8; // 'e'
|
||
o += 1i32;
|
||
let e: i32 = d.dp - 1i32;
|
||
if (e < 0i32) {
|
||
e = -e;
|
||
buf[o] = 45u8; // '-'
|
||
o += 1i32;
|
||
};
|
||
// Hoisted uint casts (ww parser rejects `(N: uint)` inside `[ ]`).
|
||
let U_ONE: uint = (1u32: uint);
|
||
let U_TWO: uint = (2u32: uint);
|
||
let U_THREE: uint = (3u32: uint);
|
||
let ebuf: [3]u8 = [0u8, 0u8, 0u8]; // exponents are at most 3 digits
|
||
let l: uint = declen(e: u64);
|
||
let k: uint = (0u32: uint);
|
||
for (k < l) {
|
||
ebuf[U_TWO - k] = (e % 10i32): u8;
|
||
e /= 10i32;
|
||
k += U_ONE;
|
||
};
|
||
let m: uint = U_THREE - l;
|
||
for (m < U_THREE) {
|
||
buf[o] = (ebuf[m] + 48u8): u8;
|
||
o += 1i32;
|
||
m += U_ONE;
|
||
};
|
||
return o;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos.ha:432. f64 → shortest base-10 str. Returns a
|
||
// view into a static buffer overwritten on the next call (the *tos
|
||
// convention; see strings.dup to retain). Equivalent to Hare's ftosf
|
||
// with format G + precision void. The fftosf G/void/NONE path is inlined
|
||
// (the parametric surface is deferred — task #64).
|
||
//
|
||
// Max output is 24 (ftos.ha:434): sign + digit + '.' + 16 digits + 'e' +
|
||
// exp-sign + 3 exp-digits. Sized 32 not 24: a no-rhs [24]u8 module buffer
|
||
// emits 4 DATAW in wwstage vs 2 in cstage (#43, the size-16/24 emitletdataw
|
||
// split); 32 emits 2 in both (byte-id). The extra 8 bytes are unused.
|
||
let f64tos_buf: [32]u8;
|
||
|
||
export fn f64tos(n: f64) str = {
|
||
let bits: u64 = math.f64bits(n);
|
||
let mantissa: u64 = bits & math.F64_MANTISSA_MASK;
|
||
let exponent: u32 = ((bits >> math.F64_MANTISSA_BITS) & math.F64_EXPONENT_MASK): u32;
|
||
let sign: bool = (bits >> (math.F64_EXPONENT_BITS + math.F64_MANTISSA_BITS)) > 0u64;
|
||
let special: bool = exponent == (math.F64_EXPONENT_MASK: u32);
|
||
|
||
let o: i32 = 0i32;
|
||
let r: str;
|
||
r.ptr = &f64tos_buf[0];
|
||
// NaN carries no sign prefix (ftos.ha:331-333, before sign handling).
|
||
if (special && mantissa != 0u64) {
|
||
o = putstr(f64tos_buf[0:32], o, "nan");
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
if (sign) {
|
||
f64tos_buf[o] = 45u8; // '-'
|
||
o += 1i32;
|
||
};
|
||
if (special) {
|
||
o = putstr(f64tos_buf[0:32], o, "infinity");
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
if (exponent == 0u32 && mantissa == 0u64) {
|
||
f64tos_buf[o] = 48u8; // '0' (encode_zero, G/void/NONE)
|
||
o += 1i32;
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
|
||
let d = decimal { ... };
|
||
// Reads of d.nd / d.dp ride a *decimal pointer: wwstage resolves a
|
||
// scalar-field read of a LOCAL struct (`d.nd`) to a bogus global
|
||
// symbol (`nd(SB)`), but a pointer-deref field read (`pd.nd`) lowers
|
||
// correctly in both stages (the stof.ww/decimal.ww *decimal precedent)
|
||
// — #170. The init/trim/encode calls already took &d; route via pd.
|
||
let pd: *decimal = &d;
|
||
let dd: decf64 = f64todecf64(mantissa, exponent);
|
||
init_dec_mant_exp(pd, dd.mantissa, (dd.exponent: i32));
|
||
// ok = !ffpoint(NONE) || ... is always true → no multiprecision
|
||
// fallback (ftos.ha:365). f == G → trim (ftos.ha:386).
|
||
trim(pd);
|
||
if (pd.nd == (0u64: size)) {
|
||
f64tos_buf[o] = 48u8; // rounded to zero
|
||
o += 1i32;
|
||
} else if (pd.dp < -1i32 || (pd.dp - (pd.nd: i32)) > 2i32) {
|
||
o = encode_e_dec(pd, f64tos_buf[0:32], o);
|
||
} else {
|
||
o = encode_f_dec(pd, f64tos_buf[0:32], o);
|
||
};
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
|
||
// ref/hare/strconv/ftos.ha:448. f32 → shortest base-10 str. Same static-
|
||
// buffer convention + G/void/NONE-inlined path as f64tos. f32bits(n)
|
||
// passes an f32 arg → MOVSS both stages post-#143 (aff7725); this is the
|
||
// piece fold-5b was gated on.
|
||
//
|
||
// Hare sizes this [14]u8 (ftos.ha:451: 1 + 1 + 1 + 7 + 1 + 1 + 2). Sized
|
||
// 32 to reuse f64tos's proven byte-id-clean band: a no-rhs [N]u8 module
|
||
// buffer at the size-16/24 band emits divergent DATAW counts cs≠ww (#43);
|
||
// 32 emits 2 DATAW in both. The unused tail bytes are harmless.
|
||
let f32tos_buf: [32]u8;
|
||
|
||
export fn f32tos(n: f32) str = {
|
||
let bits: u32 = math.f32bits(n);
|
||
let mantissa: u32 = bits & math.F32_MANTISSA_MASK;
|
||
let exponent: u32 = (bits >> math.F32_MANTISSA_BITS) & math.F32_EXPONENT_MASK;
|
||
let sign: bool = (bits >> (math.F32_EXPONENT_BITS + math.F32_MANTISSA_BITS)) > 0u32;
|
||
let special: bool = exponent == math.F32_EXPONENT_MASK;
|
||
|
||
let o: i32 = 0i32;
|
||
let r: str;
|
||
r.ptr = &f32tos_buf[0];
|
||
// NaN carries no sign prefix (ftos.ha:331-333, before sign handling).
|
||
if (special && mantissa != 0u32) {
|
||
o = putstr(f32tos_buf[0:32], o, "nan");
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
if (sign) {
|
||
f32tos_buf[o] = 45u8; // '-'
|
||
o += 1i32;
|
||
};
|
||
if (special) {
|
||
o = putstr(f32tos_buf[0:32], o, "infinity");
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
if (exponent == 0u32 && mantissa == 0u32) {
|
||
f32tos_buf[o] = 48u8; // '0' (encode_zero, G/void/NONE)
|
||
o += 1i32;
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
|
||
let d = decimal { ... };
|
||
// *decimal pointer for the field reads (the #170 dodge; see f64tos).
|
||
let pd: *decimal = &d;
|
||
let dd: decf32 = f32todecf32(mantissa, exponent);
|
||
init_dec_mant_exp(pd, (dd.mantissa: u64), (dd.exponent: i32));
|
||
trim(pd);
|
||
if (pd.nd == (0u64: size)) {
|
||
f32tos_buf[o] = 48u8; // rounded to zero
|
||
o += 1i32;
|
||
} else if (pd.dp < -1i32 || (pd.dp - (pd.nd: i32)) > 2i32) {
|
||
o = encode_e_dec(pd, f32tos_buf[0:32], o);
|
||
} else {
|
||
o = encode_f_dec(pd, f32tos_buf[0:32], o);
|
||
};
|
||
r.len = o;
|
||
return r;
|
||
};
|
||
|
||
// strconv — Ryū float→string lookup tables + bit-count constants.
|
||
// Mirrors ref/hare/strconv/ftos_ryu.ha:159-222 byte-exact. Pure data
|
||
// fold (strconv #106 fold-5): no logic, consumed by ftos.ww's
|
||
// f64computeinvpow5 / f64computepow5 (the Ryū power-of-five cores).
|
||
//
|
||
// File-organisation divergence: Hare keeps these tables INLINE in
|
||
// ftos_ryu.ha. ww splits data from logic into ftos_data.ww (mirroring
|
||
// the stof.ww / stof_data.ww split) — same `package strconv`, so the
|
||
// tables stay visible to ftos.ww with no qualification.
|
||
//
|
||
// Spelling divergences (same as stof_data.ww, candidate #130 + rule-12):
|
||
// - Hare `const TBL = [...]` → ww module-level `let` (ww has no
|
||
// module-`const` keyword; the values are never written).
|
||
// - every literal carries its element-width suffix (`u64`/`u32`):
|
||
// cstage rejects bare integer literals in `[N]uXX` init while
|
||
// wwstage accepts them; the suffixed form is the only shape both
|
||
// stages agree on.
|
||
// - the [N][2]u64 tables stay faithful 2D (rule-12, not flattened);
|
||
// the 2D module-level static-init + double-index read landed in
|
||
// #156 (cbeffea), proven by stof_data.ww's powers_of_ten[596][2]u64.
|
||
|
||
package strconv;
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:159-160. Bit-counts of the split
|
||
// power-of-five tables. Defined u8 (faithful); ftos.ww casts to u32/i32
|
||
// at each use site (Hare promotes a u8 def inside mixed-width arithmetic;
|
||
// ww is strict — explicit cast, project_int_machine_word_derived_limits).
|
||
def F64_POW5_INV_BITCOUNT: u8 = 125u8;
|
||
def F64_POW5_BITCOUNT: u8 = 125u8;
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:162-163. The f32 split-table bit-counts,
|
||
// derived from the f64 siblings (Hare: F64_..._BITCOUNT - 64). Consumed by
|
||
// f32todecf32 (ftos.ww), landed in fold-5b (task #67) — the f32 path reuses
|
||
// the f64 SPLIT2 tables (via f64computeinvpow5/f64computepow5), so no
|
||
// separate F32 tables exist (matches ftos_ryu.ha). u8 like the f64 defs;
|
||
// ftos.ww casts to u32/i32 at each use.
|
||
def F32_POW5_INV_BITCOUNT: u8 = F64_POW5_INV_BITCOUNT - 64u8;
|
||
def F32_POW5_BITCOUNT: u8 = F64_POW5_BITCOUNT - 64u8;
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:165-181.
|
||
let F64_POW5_INV_SPLIT2: [15][2]u64 = [
|
||
[1u64, 2305843009213693952u64],
|
||
[5955668970331000884u64, 1784059615882449851u64],
|
||
[8982663654677661702u64, 1380349269358112757u64],
|
||
[7286864317269821294u64, 2135987035920910082u64],
|
||
[7005857020398200553u64, 1652639921975621497u64],
|
||
[17965325103354776697u64, 1278668206209430417u64],
|
||
[8928596168509315048u64, 1978643211784836272u64],
|
||
[10075671573058298858u64, 1530901034580419511u64],
|
||
[597001226353042382u64, 1184477304306571148u64],
|
||
[1527430471115325346u64, 1832889850782397517u64],
|
||
[12533209867169019542u64, 1418129833677084982u64],
|
||
[5577825024675947042u64, 2194449627517475473u64],
|
||
[11006974540203867551u64, 1697873161311732311u64],
|
||
[10313493231639821582u64, 1313665730009899186u64],
|
||
[12701016819766672773u64, 2032799256770390445u64],
|
||
];
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:183-188.
|
||
let POW5_INV_OFFSETS: [19]u32 = [
|
||
0x54544554u32, 0x04055545u32, 0x10041000u32, 0x00400414u32, 0x40010000u32, 0x41155555u32,
|
||
0x00000454u32, 0x00010044u32, 0x40000000u32, 0x44000041u32, 0x50454450u32, 0x55550054u32,
|
||
0x51655554u32, 0x40004000u32, 0x01000001u32, 0x00010500u32, 0x51515411u32, 0x05555554u32,
|
||
0x00000000u32,
|
||
];
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:190-204.
|
||
let F64_POW5_SPLIT2: [13][2]u64 = [
|
||
[0u64, 1152921504606846976u64],
|
||
[0u64, 1490116119384765625u64],
|
||
[1032610780636961552u64, 1925929944387235853u64],
|
||
[7910200175544436838u64, 1244603055572228341u64],
|
||
[16941905809032713930u64, 1608611746708759036u64],
|
||
[13024893955298202172u64, 2079081953128979843u64],
|
||
[6607496772837067824u64, 1343575221513417750u64],
|
||
[17332926989895652603u64, 1736530273035216783u64],
|
||
[13037379183483547984u64, 2244412773384604712u64],
|
||
[1605989338741628675u64, 1450417759929778918u64],
|
||
[9630225068416591280u64, 1874621017369538693u64],
|
||
[665883850346957067u64, 1211445438634777304u64],
|
||
[14931890668723713708u64, 1565756531257009982u64],
|
||
];
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:206-211.
|
||
let POW5_OFFSETS: [21]u32 = [
|
||
0x00000000u32, 0x00000000u32, 0x00000000u32, 0x00000000u32, 0x40000000u32, 0x59695995u32,
|
||
0x55545555u32, 0x56555515u32, 0x41150504u32, 0x40555410u32, 0x44555145u32, 0x44504540u32,
|
||
0x45555550u32, 0x40004000u32, 0x96440440u32, 0x55565565u32, 0x54454045u32, 0x40154151u32,
|
||
0x55559155u32, 0x51405555u32, 0x00000105u32,
|
||
];
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:213. Divisor/index stride in
|
||
// f64computeinvpow5 / f64computepow5 (ftos.ww). Kept as a def for those
|
||
// arithmetic uses; POW5_TABLE's dimension below must be a literal (cstage
|
||
// rejects a def-named array length — "array length must be an integer
|
||
// literal"; wwstage accepts it but emits an empty DATAW — divergence
|
||
// #167, so the literal `26` is the only shape both stages agree on;
|
||
// matches decimal.ww's `[800]u8` array-dimension-literal precedent).
|
||
def POW5_TABLE_SZ: u8 = 26u8;
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:215-222. 5^0 .. 5^25 (the 5^26 entry is
|
||
// commented out in Hare too — it lives implicitly in the SPLIT2 tables).
|
||
let POW5_TABLE: [26]u64 = [
|
||
1u64, 5u64, 25u64, 125u64, 625u64, 3125u64, 15625u64, 78125u64,
|
||
390625u64, 1953125u64, 9765625u64, 48828125u64, 244140625u64,
|
||
1220703125u64, 6103515625u64, 30517578125u64, 152587890625u64,
|
||
762939453125u64, 3814697265625u64, 19073486328125u64, 95367431640625u64,
|
||
476837158203125u64, 2384185791015625u64, 11920928955078125u64,
|
||
59604644775390625u64, 298023223876953125u64,
|
||
];
|
||
|
||
// types — integer limits. Mirrors Hare's types::limits (I8_MAX, …)
|
||
// platform-fixed for amd64. Numeric helpers live in lib/math, matching
|
||
// Hare's split between types::limits and math::.
|
||
|
||
package types;
|
||
|
||
def I8_MAX: i8 = 127;
|
||
def I16_MAX: i16 = 32767;
|
||
def I32_MAX: i32 = 2147483647;
|
||
def I64_MAX: i64 = 9223372036854775807;
|
||
|
||
def I8_MIN: i8 = -128;
|
||
def I16_MIN: i16 = -32768;
|
||
def I32_MIN: i32 = -2147483648;
|
||
def I64_MIN: i64 = -9223372036854775808;
|
||
|
||
def U8_MAX: u8 = 255;
|
||
def U16_MAX: u16 = 65535;
|
||
def U32_MAX: u32 = 4294967295;
|
||
def U64_MAX: u64 = 18446744073709551615;
|
||
|
||
def U8_MIN: u8 = 0;
|
||
def U16_MIN: u16 = 0;
|
||
def U32_MIN: u32 = 0;
|
||
def U64_MIN: u64 = 0;
|
||
|
||
// int/uint are machine-word (Go-style, type.c:58); limits derived from
|
||
// size(int) per #114 + user ruling; cf Go math.MaxInt; diverges from
|
||
// Hare's per-arch literal (arch+x86_64.ha) because ww's int is 64-bit.
|
||
def INT_MAX: int = (1 << (size(int)*8 - 1)) - 1;
|
||
def INT_MIN: int = -1 << (size(int)*8 - 1);
|
||
def UINT_MIN: uint = 0;
|
||
def UINT_MAX: uint = ~(0: uint);
|
||
|
||
// size is 8B on amd64; no cast needed (size ∈ unsigned class per #113).
|
||
def SIZE_MIN: size = U64_MIN;
|
||
def SIZE_MAX: size = U64_MAX;
|
||
|
||
// uintptr not in the unsigned class, so the cast is required (Hare's form).
|
||
def UINTPTR_MIN: uintptr = U64_MIN: uintptr;
|
||
def UINTPTR_MAX: uintptr = U64_MAX: uintptr;
|
||
|
||
def RUNE_MIN: rune = '\0';
|
||
|
||
// bytes — slice operations over []u8. Mirrors Hare's bytes module
|
||
// (ref/hare/bytes/) for the in-tree subset: search/equality/prefix
|
||
// helpers used by lib/encoding, lib/bufio, lib/memio.
|
||
//
|
||
// Documented divergences from Hare:
|
||
// - index_slice / rindex_slice use naive O(n·m); Hare specialises
|
||
// 2/3/4-byte needles and falls back to two_way (Crochemore-Perrin)
|
||
// for longer (ref/hare/bytes/index.ha:61, ref/hare/bytes/two_way.ha).
|
||
// Correctness equivalent.
|
||
// - peek_token dispatches index/rindex by branching on `reverse`
|
||
// rather than a function-pointer `ifunc` (ref/hare/bytes/tokenize.ha:97).
|
||
// ww has no fn pointers in scope yet — same pattern as lib/strings
|
||
// `move`. Outwardly identical.
|
||
// - tokenize / rtokenize zero the `delim` field on the constructed
|
||
// tokenizer when `in` is empty, rather than mutating the variadic
|
||
// param before the struct write (ref/hare/bytes/tokenize.ha:26-28).
|
||
// Semantically identical; the variadic param is borrowed and
|
||
// captured-by-value into the struct, so mutating either side
|
||
// yields the same observable state.
|
||
|
||
package bytes;
|
||
|
||
import os;
|
||
import types;
|
||
|
||
// done — iteration sentinel returned by next_token / peek_token at
|
||
// end-of-input. ref/hare/bytes/tokenize.ha uses the built-in `done`
|
||
// token; ww spells it per-package the same way lib/encoding/utf8 does
|
||
// (utf8.ww:36). Plain `void` (not `!void`): continuation signal.
|
||
export type done = void;
|
||
|
||
// tokenizer — cursor over an input slice. Layout mirrors
|
||
// ref/hare/bytes/tokenize.ha:6-10. `p` is the cached peek-position;
|
||
// I64_MAX (forward) / I64_MIN (reverse) are the unprimed sentinels.
|
||
// p < 0 also identifies a reverse-direction iterator.
|
||
export type tokenizer = struct {
|
||
in: []u8,
|
||
delim: []u8,
|
||
p: i64,
|
||
};
|
||
|
||
// equal — true iff `a` and `b` have the same length and contents.
|
||
// ref/hare/bytes/equal.ha:9.
|
||
export fn equal(a: []u8, b: []u8) bool = {
|
||
if (a.len != b.len) { return false; };
|
||
let i: i32 = 0;
|
||
for (i < a.len) {
|
||
if (a[i] != b[i]) { return false; };
|
||
i += 1;
|
||
};
|
||
return true;
|
||
};
|
||
|
||
// index — first offset of `needle` in `s`. u8 needle scans for the
|
||
// byte; []u8 needle scans for the substring. void if absent.
|
||
// ref/hare/bytes/index.ha:6.
|
||
export fn index(s: []u8, needle: (u8 | []u8)) (i32 | void) = {
|
||
match (needle) {
|
||
case let c: u8 => {
|
||
let i: i32 = 0;
|
||
for (i < s.len) {
|
||
if (s[i] == c) { return i; };
|
||
i += 1;
|
||
};
|
||
return;
|
||
};
|
||
case let sub: []u8 => {
|
||
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;
|
||
};
|
||
|
||
// rindex — last offset of `needle` in `s`. Empty []u8 needle returns
|
||
// s.len (ref/hare/bytes/index.ha:103 — Hare's loop yields r-0 at i=0).
|
||
// ref/hare/bytes/index.ha:86.
|
||
export fn rindex(s: []u8, needle: (u8 | []u8)) (i32 | void) = {
|
||
match (needle) {
|
||
case let c: u8 => {
|
||
let i: i32 = s.len - 1;
|
||
for (i >= 0) {
|
||
if (s[i] == c) { return i; };
|
||
i -= 1;
|
||
};
|
||
return;
|
||
};
|
||
case let sub: []u8 => {
|
||
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;
|
||
};
|
||
|
||
// contains — true iff any of `needles` (byte or sub-slice) appears in `s`.
|
||
// ref/hare/bytes/contains.ha:6.
|
||
export fn contains(s: []u8, needles: (u8 | []u8)...) bool = {
|
||
let i: i32 = 0;
|
||
for (i < needles.len) {
|
||
match (needles[i]) {
|
||
case let b: u8 => {
|
||
match (index(s, b)) {
|
||
case let bo: i32 => return true;
|
||
case void => void;
|
||
};
|
||
};
|
||
case let n: []u8 => {
|
||
match (index(s, n)) {
|
||
case let bo: i32 => return true;
|
||
case void => void;
|
||
};
|
||
};
|
||
};
|
||
i += 1;
|
||
};
|
||
return false;
|
||
};
|
||
|
||
// ltrim — borrowed view of `in` with leading bytes in `trim` stripped.
|
||
// `trim` must be non-empty. ref/hare/bytes/trim.ha:7.
|
||
export fn ltrim(in: []u8, trim: u8...) []u8 = {
|
||
os.assert(trim.len > 0, "bytes.ltrim called with empty trim set");
|
||
let i: i32 = 0;
|
||
for (i < in.len && contains(trim, in[i])) { i += 1; };
|
||
let r: []u8;
|
||
r.ptr = in.ptr + (i: u64);
|
||
r.len = in.len - i;
|
||
r.cap = r.len;
|
||
return r;
|
||
};
|
||
|
||
// rtrim — borrowed view of `in` with trailing bytes in `trim` stripped.
|
||
// `trim` must be non-empty. ref/hare/bytes/trim.ha:17. Hare's loop uses
|
||
// `size` underflow at i==0 to terminate; ww indices are signed i32, so
|
||
// the equivalent termination is spelled `i >= 0` explicitly.
|
||
export fn rtrim(in: []u8, trim: u8...) []u8 = {
|
||
os.assert(trim.len > 0, "bytes.rtrim called with empty trim set");
|
||
let i: i32 = in.len - 1;
|
||
for (i >= 0 && contains(trim, in[i])) { i -= 1; };
|
||
let r: []u8;
|
||
r.ptr = in.ptr;
|
||
r.len = i + 1;
|
||
r.cap = r.len;
|
||
return r;
|
||
};
|
||
|
||
// trim — borrowed view of `in` with both ends in `trim` stripped.
|
||
// ref/hare/bytes/trim.ha:27.
|
||
export fn trim(in: []u8, trim: u8...) []u8 = {
|
||
return ltrim(rtrim(in, trim...), trim...);
|
||
};
|
||
|
||
// hasprefix — true iff `s` starts with `pre`.
|
||
// ref/hare/bytes/contains.ha:21.
|
||
export fn hasprefix(s: []u8, pre: []u8) bool = {
|
||
if (pre.len > s.len) { return false; };
|
||
let i: i32 = 0;
|
||
for (i < pre.len) {
|
||
if (s[i] != pre[i]) { return false; };
|
||
i += 1;
|
||
};
|
||
return true;
|
||
};
|
||
|
||
// hassuffix — true iff `s` ends with `suf`.
|
||
// ref/hare/bytes/contains.ha:35.
|
||
export fn hassuffix(s: []u8, suf: []u8) 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;
|
||
};
|
||
|
||
// reverse — in-place reverse of `s`. ref/hare/bytes/reverse.ha:5.
|
||
export fn reverse(s: []u8) void = {
|
||
let i: i32 = 0;
|
||
let j: i32 = s.len - 1;
|
||
for (i < j) {
|
||
let t: u8 = s[i];
|
||
s[i] = s[j];
|
||
s[j] = t;
|
||
i += 1;
|
||
j -= 1;
|
||
};
|
||
};
|
||
|
||
// zero — set every byte of `s` to 0. ref/hare/bytes/zero.ha:5.
|
||
export fn zero(s: []u8) void = {
|
||
let i: i32 = 0;
|
||
for (i < s.len) {
|
||
s[i] = 0u8;
|
||
i += 1;
|
||
};
|
||
};
|
||
|
||
// tokenize — iterator yielding tokens from `in` separated by any byte
|
||
// in `delim`. Leading / trailing / adjacent delims yield empty tokens.
|
||
// `delim` is borrowed; caller keeps it valid for the tokenizer's
|
||
// lifetime. ref/hare/bytes/tokenize.ha:22.
|
||
export fn tokenize(in: []u8, delim: u8...) tokenizer = {
|
||
os.assert(delim.len > 0, "bytes.tokenize called with empty slice");
|
||
os.assert((in.len: i64) < types.I64_MAX,
|
||
"bytes.tokenize: input length exceeds I64_MAX");
|
||
let t: tokenizer;
|
||
t.in = in;
|
||
t.delim = delim;
|
||
if (in.len == 0) {
|
||
t.delim.len = 0;
|
||
t.delim.cap = 0;
|
||
};
|
||
t.p = types.I64_MAX;
|
||
return t;
|
||
};
|
||
|
||
// rtokenize — reverse-direction tokenize. First next_token yields the
|
||
// last token, last next_token yields the first. ref/hare/bytes/tokenize.ha:40.
|
||
export fn rtokenize(in: []u8, delim: u8...) tokenizer = {
|
||
os.assert(delim.len > 0, "bytes.rtokenize called with empty slice");
|
||
os.assert((in.len: i64) < types.I64_MAX,
|
||
"bytes.rtokenize: input length exceeds I64_MAX");
|
||
let t: tokenizer;
|
||
t.in = in;
|
||
t.delim = delim;
|
||
if (in.len == 0) {
|
||
t.delim.len = 0;
|
||
t.delim.cap = 0;
|
||
};
|
||
t.p = types.I64_MIN;
|
||
return t;
|
||
};
|
||
|
||
// peek_token — next token without advancing the cursor. Returns done
|
||
// once `s.delim` has been zeroed by a prior past-end next_token.
|
||
// ref/hare/bytes/tokenize.ha:91.
|
||
export fn peek_token(s: *tokenizer) ([]u8 | done) = {
|
||
if (s.delim.len == 0) {
|
||
let d: done; return d;
|
||
};
|
||
|
||
let reverse: bool = s.p < 0i64;
|
||
let known: bool = false;
|
||
if (reverse) {
|
||
if (s.p != types.I64_MIN) { known = true; };
|
||
} else {
|
||
if (s.p != types.I64_MAX) { known = true; };
|
||
};
|
||
if (!known) {
|
||
let i: i64 = types.I64_MAX;
|
||
if (reverse) { i = types.I64_MIN; };
|
||
let dlen: i64 = 0i64;
|
||
let slen: i64 = s.in.len: i64;
|
||
|
||
let k: i32 = 0;
|
||
for (k < s.delim.len) {
|
||
let d: u8 = s.delim[k];
|
||
let ix_found: bool = false;
|
||
let ix_val: i32 = 0;
|
||
if (reverse) {
|
||
match (rindex(s.in, d)) {
|
||
case let v: i32 => { ix_found = true; ix_val = v; };
|
||
case void => void;
|
||
};
|
||
} else {
|
||
match (index(s.in, d)) {
|
||
case let v: i32 => { ix_found = true; ix_val = v; };
|
||
case void => void;
|
||
};
|
||
};
|
||
if (ix_found) {
|
||
if (!reverse) {
|
||
if ((ix_val: i64) < i) { i = ix_val: i64; dlen = 1i64; };
|
||
} else {
|
||
if ((ix_val: i64) > i) { i = ix_val: i64; dlen = 1i64; };
|
||
};
|
||
} else {
|
||
if (!reverse) {
|
||
if (slen < i) { i = slen; };
|
||
} else {
|
||
if (0i64 > i) { i = 0i64; };
|
||
};
|
||
};
|
||
k += 1;
|
||
};
|
||
|
||
if (reverse) {
|
||
if (i == slen) {
|
||
s.p = -(slen + 1i64);
|
||
} else {
|
||
s.p = i + dlen - slen - 1i64;
|
||
};
|
||
} else {
|
||
s.p = i;
|
||
};
|
||
};
|
||
|
||
let r: []u8;
|
||
if (reverse) {
|
||
let start: i32 = (s.in.len: i64 + s.p + 1i64): i32;
|
||
r.ptr = s.in.ptr + (start: u64);
|
||
r.len = s.in.len - start;
|
||
r.cap = r.len;
|
||
} else {
|
||
let end: i32 = s.p: i32;
|
||
r.ptr = s.in.ptr;
|
||
r.len = end;
|
||
r.cap = end;
|
||
};
|
||
return r;
|
||
};
|
||
|
||
// next_token — current token, then advance past it and the delim.
|
||
// Once the input is exhausted, returns done and zeros `s.delim` so
|
||
// subsequent peeks short-circuit. ref/hare/bytes/tokenize.ha:59.
|
||
export fn next_token(s: *tokenizer) ([]u8 | done) = {
|
||
let b: []u8;
|
||
match (peek_token(s)) {
|
||
case let v: []u8 => { b = v; };
|
||
case done => { let d: done; return d; };
|
||
};
|
||
|
||
let slen: i64 = s.in.len: i64;
|
||
let reverse: bool = s.p < 0i64;
|
||
if (reverse) {
|
||
if (slen + s.p + 1i64 == 0i64) {
|
||
s.delim.len = 0;
|
||
s.delim.cap = 0;
|
||
s.in.len = 0;
|
||
s.in.cap = 0;
|
||
} else {
|
||
let end: i32 = (slen + s.p + 1i64 - 1i64): i32;
|
||
s.in.len = end;
|
||
s.in.cap = end;
|
||
};
|
||
s.p = types.I64_MIN;
|
||
} else {
|
||
if (s.p == slen) {
|
||
s.delim.len = 0;
|
||
s.delim.cap = 0;
|
||
s.in.len = 0;
|
||
s.in.cap = 0;
|
||
} else {
|
||
let adv: u64 = (s.p: u64) + 1u64;
|
||
let adv_i32: i32 = (s.p: i32) + 1;
|
||
s.in.ptr = s.in.ptr + adv;
|
||
s.in.len = s.in.len - adv_i32;
|
||
s.in.cap = s.in.cap - adv_i32;
|
||
};
|
||
s.p = types.I64_MAX;
|
||
};
|
||
return b;
|
||
};
|
||
|
||
// remaining_tokens — the unconsumed portion of `s.in`. Read-only view.
|
||
// ref/hare/bytes/tokenize.ha:145.
|
||
export fn remaining_tokens(s: *tokenizer) []u8 = {
|
||
return s.in;
|
||
};
|
||
|
||
// rt_ensure is the runtime slice-growth helper invoked by the
|
||
// `append(s, v)` builtin. We bind it directly because the builtin's
|
||
// expansion stores only 8 bytes of the new element (cgen emits a
|
||
// single MOVQ), losing the .len/.cap fields of a []u8 element (24B).
|
||
// Mirrors the same workaround in lib/shlex.shlex (appendstr, 16B) and
|
||
// lib/getopt.getopt (appendoption, 24B); collapses in one go when the
|
||
// append builtin learns to store the full element width.
|
||
@symbol("rt_ensure") fn rtensure(s: *void, membsz: u64) void;
|
||
|
||
// appendslice — grow `*slice` by one and store `item` (24B). Mirror
|
||
// of [[shlex.appendstr]] / [[getopt.appendoption]]. Bypasses the
|
||
// `append` builtin's first-8B-only-store gap for a slice-element.
|
||
fn appendslice(slice: *[][]u8, item: []u8) void = {
|
||
let newlen: i32 = slice.len + 1;
|
||
slice.len = newlen;
|
||
rtensure(slice: *void, 24u64);
|
||
let dst: *[]u8 = &slice.ptr[newlen - 1];
|
||
dst.ptr = item.ptr;
|
||
dst.len = item.len;
|
||
dst.cap = item.cap;
|
||
};
|
||
|
||
// splitn — split `in` on any byte in `delim`, returning up to `n`
|
||
// tokens via forward iteration. The trailing slot (when more than
|
||
// `n - 1` tokens exist) holds the unconsumed remainder.
|
||
//
|
||
// The caller frees the returned slice via
|
||
// `os.free(r.ptr: *void, (r.cap: u64) * 24u64)`. Element bytes are
|
||
// borrowed from `in`.
|
||
//
|
||
// Hare's `([][]u8 | nomem)` collapses to `[][]u8` here: ww os.alloc
|
||
// has no recoverable failure path. Same precedent as
|
||
// shlex.split / getopt.tryparse.
|
||
//
|
||
// ref/hare/bytes/tokenize.ha:156.
|
||
export fn splitn(in: []u8, delim: []u8, n: i32) [][]u8 = {
|
||
os.assert(delim.len > 0,
|
||
"bytes.splitn must not be called with an empty delimiter");
|
||
let toks: [][]u8;
|
||
toks.ptr = nil: *[]u8;
|
||
toks.len = 0;
|
||
toks.cap = 0;
|
||
let tok: tokenizer = tokenize(in, delim...);
|
||
let i: i32 = 0;
|
||
for (i < n - 1) {
|
||
match (next_token(&tok)) {
|
||
case let s: []u8 => { appendslice(&toks, s); };
|
||
case done => { return toks; };
|
||
};
|
||
i += 1;
|
||
};
|
||
match (peek_token(&tok)) {
|
||
case done => void;
|
||
case let pk: []u8 => {
|
||
let r: []u8 = remaining_tokens(&tok);
|
||
appendslice(&toks, r);
|
||
};
|
||
};
|
||
return toks;
|
||
};
|
||
|
||
// rsplitn — reverse-direction counterpart to [[splitn]]: tokens are
|
||
// collected from the end of `in`. The trailing slot holds the
|
||
// unconsumed prefix (everything before the n-th-from-last delim hit).
|
||
//
|
||
// When the input has fewer than n tokens, the `done` short-circuit
|
||
// returns toks UN-reversed (in last-token-first order). Mirrors Hare
|
||
// at ref/hare/bytes/tokenize.ha:196-199 where the in-place reverse
|
||
// step is gated behind the n-1 loop running to completion. Only the
|
||
// "loop ran to completion AND peek saw a remainder" path applies the
|
||
// reverse; both early-exit paths skip it.
|
||
//
|
||
// ref/hare/bytes/tokenize.ha:186.
|
||
export fn rsplitn(in: []u8, delim: []u8, n: i32) [][]u8 = {
|
||
os.assert(delim.len > 0,
|
||
"bytes.rsplitn called with empty delimiter");
|
||
let toks: [][]u8;
|
||
toks.ptr = nil: *[]u8;
|
||
toks.len = 0;
|
||
toks.cap = 0;
|
||
let tok: tokenizer = rtokenize(in, delim...);
|
||
let i: i32 = 0;
|
||
for (i < n - 1) {
|
||
match (next_token(&tok)) {
|
||
case let s: []u8 => { appendslice(&toks, s); };
|
||
case done => { return toks; };
|
||
};
|
||
i += 1;
|
||
};
|
||
match (peek_token(&tok)) {
|
||
case done => void;
|
||
case let pk: []u8 => {
|
||
let r: []u8 = remaining_tokens(&tok);
|
||
appendslice(&toks, r);
|
||
};
|
||
};
|
||
|
||
// In-place reverse so callers see argv-order, matching Hare
|
||
// (ref/hare/bytes/tokenize.ha:207). Element copy is field-wise
|
||
// through `*[]u8` because `toks[i] = toks[j]` (full 24B slice
|
||
// store) lands in the multi-word-store gap noted at
|
||
// cmd/w6c/cgen.c:6515-6523.
|
||
let a: i32 = 0;
|
||
let b: i32 = toks.len - 1;
|
||
for (a < b) {
|
||
let pa: *[]u8 = &toks.ptr[a];
|
||
let pb: *[]u8 = &toks.ptr[b];
|
||
let tp: *u8 = pa.ptr;
|
||
let tl: i32 = pa.len;
|
||
let tc: i32 = pa.cap;
|
||
pa.ptr = pb.ptr;
|
||
pa.len = pb.len;
|
||
pa.cap = pb.cap;
|
||
pb.ptr = tp;
|
||
pb.len = tl;
|
||
pb.cap = tc;
|
||
a += 1;
|
||
b -= 1;
|
||
};
|
||
return toks;
|
||
};
|
||
|
||
// split — full split of `in` on `delim` (no token cap). Mirrors
|
||
// `splitn(in, delim, types::SIZE_MAX)`. ww uses `types.I32_MAX`
|
||
// because the index type is i32 (lib/CLAUDE.md).
|
||
//
|
||
// ref/hare/bytes/tokenize.ha:225.
|
||
export fn split(in: []u8, delim: []u8) [][]u8 = {
|
||
return splitn(in, delim, types.I32_MAX);
|
||
};
|
||
|
||
// cut — split `in` along the first instance of `delim`, returning the
|
||
// portion before and the portion after the delimiter as a borrowed
|
||
// tuple. When `delim` is absent, the whole input is the first half and
|
||
// the second is empty. ref/hare/bytes/tokenize.ha:392.
|
||
//
|
||
// Delim is spelled (u8 | []u8) to match index/rindex (bytes.ww:57/91);
|
||
// the tagged union is an unordered set, so this is the same type as
|
||
// Hare's ([]u8 | u8), not a divergence.
|
||
export fn cut(in: []u8, delim: (u8 | []u8)) ([]u8, []u8) = {
|
||
let ln: i32 = match (delim) {
|
||
case let c: u8 => yield 1i32;
|
||
case let sub: []u8 => {
|
||
os.assert(sub.len > 0,
|
||
"bytes.cut called with empty delimiter");
|
||
yield sub.len;
|
||
};
|
||
};
|
||
match (index(in, delim)) {
|
||
case let i: i32 => {
|
||
let lo: i32 = i + ln;
|
||
return (in[0:i], in[lo:in.len]);
|
||
};
|
||
case void => {
|
||
let empty: []u8;
|
||
empty.ptr = nil; empty.len = 0; empty.cap = 0;
|
||
return (in, empty);
|
||
};
|
||
};
|
||
};
|
||
|
||
// rcut — like [[cut]] but splits along the last instance of `delim`.
|
||
// ref/hare/bytes/tokenize.ha:413.
|
||
export fn rcut(in: []u8, delim: (u8 | []u8)) ([]u8, []u8) = {
|
||
let ln: i32 = match (delim) {
|
||
case let c: u8 => yield 1i32;
|
||
case let sub: []u8 => {
|
||
os.assert(sub.len > 0,
|
||
"bytes.rcut called with empty delimiter");
|
||
yield sub.len;
|
||
};
|
||
};
|
||
match (rindex(in, delim)) {
|
||
case let i: i32 => {
|
||
let lo: i32 = i + ln;
|
||
return (in[0:i], in[lo:in.len]);
|
||
};
|
||
case void => {
|
||
let empty: []u8;
|
||
empty.ptr = nil; empty.len = 0; empty.cap = 0;
|
||
return (in, empty);
|
||
};
|
||
};
|
||
};
|
||
|
||
// encoding/utf8 — UTF-8 encode/decode. Hare port; see
|
||
// ref/hare/encoding/utf8/{types,rune,encode,decode,decodetable}.ha.
|
||
//
|
||
// The decoder is Hoehrmann's branchless DFA, originally published
|
||
// at <https://bjoern.hoehrmann.de/utf-8/decoder/dfa/>. Hare's
|
||
// ref/hare/encoding/utf8/decodetable.ha:4 restructures Hoehrmann's
|
||
// flat table to 2D `[8][256]i8`; we flatten back to 1D `[2048]i8`
|
||
// because ww cgen does not yet ship 2D arrays (task #20).
|
||
//
|
||
// Surface deviation from ref/hare/encoding/utf8:
|
||
//
|
||
// - `encoderune` takes a caller-supplied `out: []u8` and returns
|
||
// the byte count. Hare returns a slice into a `static let buf`;
|
||
// the caller-buffer form mirrors lib/encoding/hex.encode and
|
||
// skips the static-buffer/slice-return pair.
|
||
//
|
||
// Deferred (no in-tree caller, follow-up tasks): `appendrune`,
|
||
// `strencode`, `strdecode`. Hare's string-iteration surface
|
||
// (`strings::iterator`/`strings::next` — ref/hare/strings/iter.ha)
|
||
// lives under lib/strings, not here.
|
||
|
||
// ref/hare/encoding/utf8/types.ha:6 — incomplete trailing sequence.
|
||
// Plain `void` (not `!void`): a truncated tail is a control-flow
|
||
// signal, not an error caller can ignore.
|
||
package utf8;
|
||
|
||
export type more = void;
|
||
|
||
// ref/hare/encoding/utf8/types.ha:9 — invalid UTF-8 sequence.
|
||
export type invalid = !void;
|
||
|
||
// ref/hare/encoding/utf8/types.ha:12 — fixed message; `invalid` carries
|
||
// no payload, so the rendering is constant.
|
||
export fn strerror(err: invalid) str = {
|
||
return "Invalid UTF-8";
|
||
};
|
||
|
||
// `done` is not a built-in singleton in ww (Hare ships it as part of
|
||
// the type system). Plain `void` (not `!void`): end-of-input is a
|
||
// continuation signal, not an error. lib/io spells its EOF the same
|
||
// way (lib/io/io.ww:8-11).
|
||
export type done = void;
|
||
|
||
// ref/hare/encoding/utf8/decodetable.ha:4 — Hoehrmann's UTF-8 DFA,
|
||
// flat 1D `[2048]i8`. Layout: dfa[state*256 + byte] gives the next
|
||
// state (>0), the accept transition (0 — emit rune), or invalid (-1).
|
||
// Values match ref/hare/encoding/utf8/decodetable.ha verbatim.
|
||
let dfa: [2048]i8 = [
|
||
// state 0 — initial byte: ASCII accepts (0), continuation/illegal
|
||
// byte rejects (-1), legal multibyte start emits a state.
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
3i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 4i8, 2i8, 2i8,
|
||
5i8, 6i8, 6i8, 6i8, 7i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
|
||
// state 1 — expecting one continuation byte (0x80..0xBF).
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8, 0i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
|
||
// state 2 — expecting one continuation byte (full 0x80..0xBF range).
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
|
||
// state 3 — first byte was 0xE0; continuation byte must be 0xA0..0xBF
|
||
// (rejects overlong 3-byte encodings).
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
|
||
// state 4 — first byte was 0xED; continuation byte must be 0x80..0x9F
|
||
// (rejects UTF-16 surrogate codepoints U+D800..U+DFFF).
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8, 1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
|
||
// state 5 — first byte was 0xF0; continuation byte must be 0x90..0xBF
|
||
// (rejects overlong 4-byte encodings).
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
|
||
// state 6 — middle continuation byte of a 4-byte sequence (0x80..0xBF).
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
|
||
// state 7 — first byte was 0xF4; continuation byte must be 0x80..0x8F
|
||
// (rejects codepoints above U+10FFFF).
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8, 2i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
-1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8, -1i8,
|
||
];
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:17 — payload-bit masks. Hare's
|
||
// [2][8]u8 flattened to 1D [16]u8; row 0 (offsets 0..7) is the
|
||
// continuation-byte mask (always 0x3F), row 1 (offsets 8..15) is the
|
||
// initial-byte payload mask indexed by the transition class.
|
||
let masks: [16]u8 = [
|
||
0x3fu8, 0x3fu8, 0x3fu8, 0x3fu8, 0x3fu8, 0x3fu8, 0x3fu8, 0x3fu8,
|
||
0x7fu8, 0x1fu8, 0x0fu8, 0x0fu8, 0x0fu8, 0x07u8, 0x07u8, 0x07u8,
|
||
];
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:6 — incremental decoder state.
|
||
export type decoder = struct {
|
||
offs: i32,
|
||
src: []u8,
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:12.
|
||
export fn decode(src: []u8) decoder = {
|
||
let d: decoder;
|
||
d.src = src;
|
||
d.offs = 0;
|
||
return d;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:27. Returns the next rune from a
|
||
// decoder, `done` at end-of-input, `more` on truncated trailing
|
||
// sequence, `invalid` on malformed input (overlong, surrogate,
|
||
// out-of-range, bad continuation).
|
||
//
|
||
// Algorithm is verbatim Hoehrmann (see file header). One structural
|
||
// rewrite: Hare encodes the "initial vs continuation byte" decision
|
||
// as the branchless `(state - 1): uint >> 31`, which assumes a 32-bit
|
||
// uint. ww's uint is 64-bit (cmd/wcc/type.c:58), so the shift answer
|
||
// would be 0x1_ffff_ffff rather than 1. We spell the same predicate
|
||
// with an explicit conditional.
|
||
export fn next(d: *decoder) (rune | done | more | invalid) = {
|
||
if (d.offs == d.src.len) {
|
||
let dn: done; return dn;
|
||
};
|
||
let nx: i32 = 0;
|
||
let state: i32 = 0;
|
||
let r: u32 = 0u32;
|
||
for (d.offs < d.src.len) {
|
||
let b: u8 = d.src[d.offs];
|
||
let bi: i32 = b: i32;
|
||
let row: i32 = state * 256 + bi;
|
||
let cell: i8 = dfa[row];
|
||
nx = cell: i32;
|
||
let mi: i32 = 0;
|
||
if (state == 0) { mi = 1; };
|
||
let m: u8 = masks[mi * 8 + (nx & 7)];
|
||
r = (r << 6u32) | ((b & m): u32);
|
||
if (nx <= 0) {
|
||
d.offs += 1;
|
||
if (nx == 0) { return r: rune; };
|
||
let e: invalid; return e;
|
||
};
|
||
state = nx;
|
||
d.offs += 1;
|
||
};
|
||
let mr: more; return mr;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:207. Strict whole-input check.
|
||
// The hot path: tight DFA loop, no rune assembly. Bails the moment
|
||
// the table returns -1 so malformed inputs don't pay for the rest
|
||
// of the buffer.
|
||
export fn validate(src: []u8) (void | invalid) = {
|
||
let state: i32 = 0;
|
||
let i: i32 = 0;
|
||
for (i < src.len) {
|
||
if (state < 0) { break; };
|
||
let bi: i32 = src[i]: i32;
|
||
let cell: i8 = dfa[state * 256 + bi];
|
||
state = cell: i32;
|
||
i += 1;
|
||
};
|
||
if (state == 0) { return; };
|
||
let e: invalid; return e;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/rune.ha:5. Encoded byte length of `r` as
|
||
// UTF-8. Callers in ww use this to size the buffer they hand to
|
||
// [[encoderune]]; values >0x10FFFF or negative are not legal Unicode
|
||
// codepoints and Hare aborts on them in `encoderune` itself, so we
|
||
// keep `runesz` infallible (matches Hare).
|
||
export fn runesz(r: rune) i32 = {
|
||
let ch: u32 = r: u32;
|
||
if (ch < 128u32) { return 1; };
|
||
if (ch < 2048u32) { return 2; };
|
||
if (ch < 65536u32) { return 3; };
|
||
return 4;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/rune.ha:15. Expected byte length of the
|
||
// codepoint that starts with `c`, or `invalid` if `c` cannot start
|
||
// a legal UTF-8 sequence. Constants written in decimal because ww
|
||
// doesn't accept Hare's `0b1000_0000` binary syntax: 0x80=128,
|
||
// 0xC2=194, 0xE0=224, 0xF0=240, 0xF8=248.
|
||
export fn utf8sz(c: u8) (i32 | invalid) = {
|
||
if (c < 128u8) { return 1; };
|
||
if (c < 194u8) { let e: invalid; return e; };
|
||
if (c >= 248u8) { let e: invalid; return e; };
|
||
if (c < 224u8) { return 2; };
|
||
if (c < 240u8) { return 3; };
|
||
return 4;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/encode.ha:7. Encode `r` into `out` (caller-
|
||
// supplied; must hold at least [[runesz]](r) bytes) and return the
|
||
// byte count. ABORT if `r` is a UTF-16 surrogate or above U+10FFFF —
|
||
// same precondition Hare asserts at ref/hare/encoding/utf8/encode.ha:9.
|
||
//
|
||
// Surface deviation: Hare returns `[]u8` (slice into a static buf).
|
||
// ww uses the caller-buffer form (matches lib/encoding/hex.encode);
|
||
// caller can reuse a [4]u8 stack scratch across encodes.
|
||
export fn encoderune(out: []u8, r: rune) i32 = {
|
||
let ch: u32 = r: u32;
|
||
if (ch >= 0xD800u32) {
|
||
if (ch <= 0xDFFFu32) {
|
||
abort("utf8.encoderune: surrogate codepoint");
|
||
};
|
||
};
|
||
if (ch > 0x10FFFFu32) {
|
||
abort("utf8.encoderune: codepoint > U+10FFFF");
|
||
};
|
||
|
||
let n: i32 = 0;
|
||
let first: u8 = 0u8;
|
||
if (ch < 0x80u32) {
|
||
first = 0u8; n = 1;
|
||
} else if (ch < 0x800u32) {
|
||
first = 0xC0u8; n = 2;
|
||
} else if (ch < 0x10000u32) {
|
||
first = 0xE0u8; n = 3;
|
||
} else {
|
||
first = 0xF0u8; n = 4;
|
||
};
|
||
|
||
let v: u32 = ch;
|
||
let i: i32 = n - 1;
|
||
for (i > 0) {
|
||
out[i] = ((v: u8) & 0x3Fu8) | 0x80u8;
|
||
v = v >> 6u32;
|
||
i -= 1;
|
||
};
|
||
out[0] = (v: u8) | first;
|
||
return n;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:52. Walks back from `d.offs` to a
|
||
// byte that could start a codepoint (state-0 dfa cell != -1), re-decodes
|
||
// forward from there, and confirms the forward decode lands back at the
|
||
// original offset. Returns `done` at start-of-input; `invalid` if no
|
||
// initial byte appears within 4 steps (no legal UTF-8 codepoint exceeds
|
||
// 4 bytes), if the forward decode returns `more`/`invalid`, or if it
|
||
// lands at a different offset than expected. Returns `more` when the
|
||
// walk reaches byte 0 without finding any initial byte.
|
||
//
|
||
// Hare's `for (d.offs < len(d.src); d.offs -= 1)` relies on size_t
|
||
// wrap-around to exit when offs underflows past 0; ww's offs is i32,
|
||
// so we spell the same exit as `d.offs >= 0`. Hare's `defer d.offs = t`
|
||
// is inlined in each match arm — ww has no defer.
|
||
export fn prev(d: *decoder) (rune | done | more | invalid) = {
|
||
if (d.offs == 0) {
|
||
let dn: done; return dn;
|
||
};
|
||
let n: i32 = d.offs;
|
||
d.offs -= 1;
|
||
for (d.offs >= 0) {
|
||
let b: u8 = d.src[d.offs];
|
||
let bi: i32 = b: i32;
|
||
let cell: i8 = dfa[bi];
|
||
if (cell: i32 != -1) {
|
||
let t: i32 = d.offs;
|
||
match (next(d)) {
|
||
case let r: rune => {
|
||
let landed: i32 = d.offs;
|
||
d.offs = t;
|
||
if (landed != n) {
|
||
let e: invalid; return e;
|
||
};
|
||
return r;
|
||
};
|
||
case let dn: done => {
|
||
d.offs = t;
|
||
let e: invalid; return e;
|
||
};
|
||
case let m: more => {
|
||
d.offs = t;
|
||
let e: invalid; return e;
|
||
};
|
||
case let e: invalid => {
|
||
d.offs = t;
|
||
let e2: invalid; return e2;
|
||
};
|
||
};
|
||
};
|
||
if (n - d.offs == 4) {
|
||
let e: invalid; return e;
|
||
};
|
||
d.offs -= 1;
|
||
};
|
||
let mr: more; return mr;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:74. Borrowed view of the bytes from
|
||
// the decoder's current position to the end of its source.
|
||
export fn remaining(d: *decoder) []u8 = {
|
||
let r: []u8;
|
||
r.ptr = d.src.ptr + (d.offs: u64);
|
||
r.len = d.src.len - d.offs;
|
||
r.cap = d.src.len - d.offs;
|
||
return r;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:80. Borrowed view of the bytes
|
||
// between two decoders' positions. Precondition (Hare asserts both):
|
||
// the decoders share the same source, and `begin.offs <= end.offs`.
|
||
export fn slice(begin: *decoder, end: *decoder) []u8 = {
|
||
if (begin.src.ptr != end.src.ptr) {
|
||
abort("utf8.slice: decoders from different sources");
|
||
};
|
||
if (begin.offs > end.offs) {
|
||
abort("utf8.slice: begin past end");
|
||
};
|
||
let r: []u8;
|
||
r.ptr = begin.src.ptr + (begin.offs: u64);
|
||
r.len = end.offs - begin.offs;
|
||
r.cap = end.offs - begin.offs;
|
||
return r;
|
||
};
|
||
|
||
// ref/hare/encoding/utf8/decode.ha:203. Byte position of the decoder
|
||
// in its source.
|
||
export fn position(d: *decoder) i32 = {
|
||
return d.offs;
|
||
};
|
||
|
||
|
||
// rt — runtime primitives exposed to ww programs.
|
||
// Mirrors Hare's rt:: module placement (ref/hare/rt/).
|
||
|
||
package rt;
|
||
|
||
// malloc — mmap-backed page allocator. Untyped: `malloc(n)` returns a
|
||
// `*void`; callers cast to the target type. Diverges from Hare: Hare
|
||
// exposes `alloc` / `free` as typed language builtins that the
|
||
// compiler lowers to rt::malloc/rt::free; ww has no such builtins,
|
||
// so the rt-symbol surface is exposed directly. Stdlib callers that
|
||
// need a typed allocation pattern wrap this with a cast plus a stored
|
||
// capacity (see [[strings.dup]], [[memio.dynamic]]).
|
||
//
|
||
// OOM: rt_malloc is a bare mmap(MAP_ANON|MAP_PRIVATE) wrapper with no
|
||
// error path. The raw Linux mmap syscall returns a negative errno cast
|
||
// to `*void` on failure (e.g. `(void*)-12` for ENOMEM); the
|
||
// `MAP_FAILED` (`(void*)-1`) value is a libc-wrapper convention that
|
||
// rt_malloc doesn't apply. Neither `== nil` nor `== (void*)-1` catches
|
||
// it; any deref of such a return faults. Today the stdlib does not
|
||
// check; OOM faults on first dereference. A typed fallible variant is
|
||
// a future task (task #39). ref/hare/rt/malloc.ha:27.
|
||
@symbol("rt_malloc") export fn malloc(n: u64) *void;
|
||
|
||
// strings — operations over str ({ptr,len}). Hare port; see
|
||
// ref/hare/strings/.
|
||
//
|
||
// Documented divergences from Hare:
|
||
//
|
||
// - `byteindex` / `rbyteindex` rune arms encode via
|
||
// `utf8.encoderune`; the legacy impls scanned for `r: u8` (an
|
||
// undocumented ASCII-only restriction that silently dropped
|
||
// to the wrong byte for U+80..U+7FF and higher).
|
||
// - `dup(s: str) str` — Hare returns `(str | nomem)`. ww's
|
||
// `os.alloc` aborts on OOM (no `nomem` type), so we return plain
|
||
// `str`. Empty input returns `{nil, 0}`; Hare returns the static
|
||
// empty string — same observable result.
|
||
// - `iterator` is flattened (`offs`, `src`, `reverse` fields).
|
||
// Hare uses anonymous-embedded `utf8::decoder`
|
||
// (ref/hare/strings/iter.ha:6-9); ww has no anonymous-embed
|
||
// syntax, so `next`/`prev`/`slice` copy `offs`/`src` into a
|
||
// local `utf8.decoder` for the call (and `next`/`prev` write
|
||
// `offs` back).
|
||
// - Hare's private `move()` helper dispatches on a `forward: bool`
|
||
// using a function-pointer `let fun = if (forward) &utf8::next
|
||
// else &utf8::prev`. ww has no fn-pointers in scope yet, so the
|
||
// dispatch is a branch on `forward` selecting the call site.
|
||
|
||
package strings;
|
||
|
||
import bytes;
|
||
import encoding.utf8;
|
||
import os;
|
||
import rt;
|
||
import types;
|
||
|
||
// toutf8 — borrowed []u8 view of `s`. ref/hare/strings/utf8.ha:29.
|
||
// `cap` equals `len`; the slice does not own a separate allocation.
|
||
export fn toutf8(s: str) []u8 = {
|
||
let r: []u8;
|
||
r.ptr = s.ptr;
|
||
r.len = s.len;
|
||
r.cap = s.len;
|
||
return r;
|
||
};
|
||
|
||
// frombytes — borrowed str view of `in`. Pure reinterpret per
|
||
// CLAUDE.md rule 9 carve-out; ref/hare/strings/utf8.ha:10.
|
||
export fn frombytes(in: []u8) str = {
|
||
let r: str;
|
||
r.ptr = in.ptr;
|
||
r.len = in.len;
|
||
return r;
|
||
};
|
||
|
||
// compare — three-way bytewise codepoint-order comparison. Return is
|
||
// a sign (neg/zero/pos), not an index, so it tracks Hare's `int`
|
||
// rather than the str-index i32 (#8). ref/hare/strings/compare.ha:12.
|
||
export fn compare(a: str, b: str) int = {
|
||
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]: int) - (b[i]: int); };
|
||
i += 1;
|
||
};
|
||
return (a.len: int) - (b.len: int);
|
||
};
|
||
|
||
// dup — allocate a fresh copy of `s`. Caller releases with
|
||
// `os.free(r.ptr, r.len: u64)`. ref/hare/strings/dup.ha:7.
|
||
export fn dup(s: str) str = {
|
||
let r: str;
|
||
r.ptr = nil;
|
||
r.len = 0;
|
||
if (s.len == 0) { return r; };
|
||
let buf: []u8 = alloc([], s.len: u64)!;
|
||
let i: i32 = 0;
|
||
for (i < s.len) { buf[i] = s[i]; i += 1; };
|
||
buf.len = s.len;
|
||
return frombytes(buf);
|
||
};
|
||
|
||
// dupall — fresh `[]str` whose elements are independent copies of
|
||
// `s`'s elements. Caller releases via [[freeall]].
|
||
// ref/hare/strings/dup.ha:26 (#6).
|
||
//
|
||
// Hare gates the per-element dup behind `?` and rolls back via
|
||
// `defer if (!ok) freeall(newsl)`. ww has no `defer if`; more
|
||
// importantly, ww's [[dup]] is still unchecked (returns plain `str`,
|
||
// aborts via os.alloc on OOM — see top-of-file divergence note),
|
||
// so the only nomem propagation point is the initial slice alloc.
|
||
// With no inner failure path, the rollback is structurally a no-op
|
||
// and is omitted; it returns once dup graduates to `(str | nomem)`
|
||
// (#46). The pre-allocated slice has `cap == s.len`, so appendstr's
|
||
// rt_ensure call never reaches the grow branch.
|
||
//
|
||
// Empty input bypasses the alloc: rt_malloc(0) is an mmap of 0 bytes
|
||
// which returns -EINVAL, and the alloc-slice `?` shortcut routes
|
||
// that through nomem — Hare's heap allocator hands back a sentinel
|
||
// instead (#47). Return `{nil, 0, 0}` directly so callers get the
|
||
// Hare-observable shape (len==0, freeall is a no-op via cap==0).
|
||
export fn dupall(s: []str) ([]str | nomem) = {
|
||
if (s.len == 0) {
|
||
let r: []str;
|
||
r.ptr = nil: *str;
|
||
r.len = 0;
|
||
r.cap = 0;
|
||
return r;
|
||
};
|
||
let newsl: []str = alloc([], s.len)?;
|
||
let i: i32 = 0;
|
||
for (i < s.len) {
|
||
appendstr(&newsl, dup(s[i]));
|
||
i += 1;
|
||
};
|
||
return newsl;
|
||
};
|
||
|
||
// freeall — release each element + the slice header. The natural
|
||
// disposer for any `[]str` of dup'd elements (e.g. shlex.split).
|
||
// ref/hare/strings/dup.ha:38.
|
||
//
|
||
// Empty elements (`{nil, 0}` from a zero-length dup) are skipped:
|
||
// os.free on a nil pointer at len 0 tickles the rt_free guard. The
|
||
// slice header itself is freed at `cap * size(str)` — the literal
|
||
// would drift under #1's str-layout bump, so route through the
|
||
// typ.ww SSoT. A never-grown slice (cap == 0) skips the header free.
|
||
export fn freeall(s: []str) void = {
|
||
let i: i32 = 0;
|
||
for (i < s.len) {
|
||
if (s[i].len > 0) {
|
||
os.free(s[i].ptr: *void, s[i].len: u64);
|
||
};
|
||
i += 1;
|
||
};
|
||
if (s.cap > 0) {
|
||
os.free(s.ptr: *void, (s.cap: u64) * size(str): u64);
|
||
};
|
||
};
|
||
|
||
// concat — fresh allocation containing each element of `strs` in
|
||
// order. Caller releases with `os.free(r.ptr, r.len: u64)`.
|
||
// ref/hare/strings/concat.ha:5. Hare's `nomem` return is dropped:
|
||
// `os.alloc` aborts on OOM.
|
||
export fn concat(strs: str...) str = {
|
||
let total: i32 = 0;
|
||
let i: i32 = 0;
|
||
for (i < strs.len) { total += strs[i].len; i += 1; };
|
||
let r: str;
|
||
r.ptr = nil;
|
||
r.len = 0;
|
||
if (total == 0) { return r; };
|
||
let buf: []u8 = alloc([], total: u64)!;
|
||
let off: i32 = 0;
|
||
i = 0;
|
||
for (i < strs.len) {
|
||
let j: i32 = 0;
|
||
for (j < strs[i].len) {
|
||
buf[off + j] = strs[i][j];
|
||
j += 1;
|
||
};
|
||
off += strs[i].len;
|
||
i += 1;
|
||
};
|
||
buf.len = total;
|
||
return frombytes(buf);
|
||
};
|
||
|
||
// join — fresh allocation with `delim` placed between each element of
|
||
// `strs`. Caller releases with `os.free(r.ptr, r.len: u64)`.
|
||
// ref/hare/strings/concat.ha:46. Hare's `nomem` return is dropped:
|
||
// `os.alloc` aborts on OOM.
|
||
export fn join(delim: str, strs: str...) str = {
|
||
let total: i32 = 0;
|
||
let i: i32 = 0;
|
||
for (i < strs.len) {
|
||
total += strs[i].len;
|
||
if (i + 1 < strs.len) { total += delim.len; };
|
||
i += 1;
|
||
};
|
||
let r: str;
|
||
r.ptr = nil;
|
||
r.len = 0;
|
||
if (total == 0) { return r; };
|
||
let buf: []u8 = alloc([], total: u64)!;
|
||
let off: i32 = 0;
|
||
i = 0;
|
||
for (i < strs.len) {
|
||
let j: i32 = 0;
|
||
for (j < strs[i].len) {
|
||
buf[off + j] = strs[i][j];
|
||
j += 1;
|
||
};
|
||
off += strs[i].len;
|
||
if (i + 1 < strs.len) {
|
||
j = 0;
|
||
for (j < delim.len) {
|
||
buf[off + j] = delim[j];
|
||
j += 1;
|
||
};
|
||
off += delim.len;
|
||
};
|
||
i += 1;
|
||
};
|
||
buf.len = total;
|
||
return frombytes(buf);
|
||
};
|
||
|
||
// utf8bytelenbounded — walk `it` forward `end` runes and return the
|
||
// resulting byte offset. ref/hare/strings/sub.ha:10. Aborts on
|
||
// short input per Hare's contract for the rune-wise [[sub]].
|
||
fn utf8bytelenbounded(it: *iterator, end: i32) i32 = {
|
||
let i: i32 = 0;
|
||
for (i < end) {
|
||
match (next(it)) {
|
||
case let r: rune => void;
|
||
case utf8.done => abort("strings.sub: index exceeds string length");
|
||
};
|
||
i += 1;
|
||
};
|
||
return it.offs;
|
||
};
|
||
|
||
// sub — borrowed substring [start, end) where start/end are rune
|
||
// indices. ref/hare/strings/sub.ha:30. Hare's 2-arg `sub(s, start)`
|
||
// defaulting end=END is omitted: ww has no default-parameter syntax
|
||
// (filed as #37). Byte-indexed counterpart: [[bytesub]].
|
||
export fn sub(s: str, start: i32, end: i32) str = {
|
||
os.assert(start <= end, "strings.sub: start is higher than end");
|
||
let it: iterator = iter(s);
|
||
let starti: i32 = utf8bytelenbounded(&it, start);
|
||
let endi: i32 = utf8bytelenbounded(&it, end - start);
|
||
let r: str;
|
||
r.ptr = s.ptr + (starti: u64);
|
||
r.len = endi - starti;
|
||
return r;
|
||
};
|
||
|
||
// bytesub — borrowed substring [start, end) where start/end are byte
|
||
// offsets. ref/hare/strings/sub.ha:59 (#7). Returns `utf8.invalid` if
|
||
// either endpoint lands on a continuation byte (would split a
|
||
// codepoint); the equivalent Hare predicate is `s[i] & 0xc0 == 0x80`
|
||
// at ref/hare/strings/sub.ha:72-73.
|
||
export fn bytesub(s: str, start: i32, end: i32) (str | utf8.invalid) = {
|
||
os.assert(start <= end, "strings.bytesub: start is higher than end");
|
||
os.assert(end <= s.len, "strings.bytesub: end exceeds string length");
|
||
if (start < s.len) {
|
||
if ((s[start] & 0xC0u8) == 0x80u8) {
|
||
let e: utf8.invalid; return e;
|
||
};
|
||
};
|
||
if (end < s.len) {
|
||
if ((s[end] & 0xC0u8) == 0x80u8) {
|
||
let e: utf8.invalid; return e;
|
||
};
|
||
};
|
||
let r: str;
|
||
r.ptr = s.ptr + (start: u64);
|
||
r.len = end - start;
|
||
return r;
|
||
};
|
||
|
||
// runebytes — encode `r` into caller's `scratch` (must hold 4 bytes)
|
||
// and return the borrowed slice trimmed to the encoded length. Hare
|
||
// inlines the same shape at ref/hare/strings/index.ha:132.
|
||
fn runebytes(scratch: []u8, r: rune) []u8 = {
|
||
let n: i32 = utf8.encoderune(scratch, r);
|
||
let s: []u8;
|
||
s.ptr = scratch.ptr;
|
||
s.len = n;
|
||
s.cap = n;
|
||
return s;
|
||
};
|
||
|
||
// hasprefix — true iff `in` begins with `prefix`.
|
||
// ref/hare/strings/suffix.ha:8.
|
||
export fn hasprefix(in: str, prefix: (str | rune)) bool = {
|
||
let scratch: [4]u8;
|
||
let p: []u8 = match (prefix) {
|
||
case let s: str => yield toutf8(s);
|
||
case let r: rune => yield runebytes(scratch[0:4], r);
|
||
};
|
||
return bytes.hasprefix(toutf8(in), p);
|
||
};
|
||
|
||
// hassuffix — true iff `in` ends with `suff`.
|
||
// ref/hare/strings/suffix.ha:26.
|
||
export fn hassuffix(in: str, suff: (str | rune)) bool = {
|
||
let scratch: [4]u8;
|
||
let s: []u8 = match (suff) {
|
||
case let v: str => yield toutf8(v);
|
||
case let r: rune => yield runebytes(scratch[0:4], r);
|
||
};
|
||
return bytes.hassuffix(toutf8(in), s);
|
||
};
|
||
|
||
// byteindex — byte-wise offset of `needle` in `haystack`, or void if
|
||
// absent. ref/hare/strings/index.ha:127. Rune arm encodes via
|
||
// utf8.encoderune (Hare passes the encoded slice straight to
|
||
// bytes::index).
|
||
export fn byteindex(haystack: str, needle: (str | rune)) (i32 | void) = {
|
||
let scratch: [4]u8;
|
||
let n: []u8 = match (needle) {
|
||
case let s: str => yield toutf8(s);
|
||
case let r: rune => yield runebytes(scratch[0:4], r);
|
||
};
|
||
return bytes.index(toutf8(haystack), n);
|
||
};
|
||
|
||
// rbyteindex — byte-wise offset of the last `needle` in `haystack`.
|
||
// ref/hare/strings/index.ha:138.
|
||
export fn rbyteindex(haystack: str, needle: (str | rune)) (i32 | void) = {
|
||
let scratch: [4]u8;
|
||
let n: []u8 = match (needle) {
|
||
case let s: str => yield toutf8(s);
|
||
case let r: rune => yield runebytes(scratch[0:4], r);
|
||
};
|
||
return bytes.rindex(toutf8(haystack), n);
|
||
};
|
||
|
||
// indexstring — str-arm of [[index]]. Dual-rune-iterator walk: at each
|
||
// candidate rune index `i`, compare `haystack` from that position
|
||
// against `needle` rune-by-rune until needle is exhausted (match) or
|
||
// a mismatch / haystack-exhaustion breaks the inner loop. Mirrors
|
||
// ref/hare/strings/index.ha:59 (#10). Hare copies `rest_iter = s_iter`
|
||
// directly via struct assignment; ww re-seats `rest_iter` field-wise
|
||
// because the let-init struct-copy form diverges between cstage and
|
||
// wwstage on this iterator type (993_ww_ww + 995_self_rebuild fail,
|
||
// filed as #41) and rule #10 (CLAUDE.md) forbids stage asymmetry.
|
||
fn indexstring(haystack: str, needle: str) (i32 | void) = {
|
||
let s_iter: iterator = iter(haystack);
|
||
let i: i32 = 0;
|
||
for (true) {
|
||
let rest_iter: iterator;
|
||
rest_iter.src = s_iter.src;
|
||
rest_iter.offs = s_iter.offs;
|
||
rest_iter.reverse = s_iter.reverse;
|
||
let needle_iter: iterator = iter(needle);
|
||
let matched: bool = false;
|
||
for (true) {
|
||
let rest_done: bool = false;
|
||
let rest_r: rune;
|
||
match (next(&rest_iter)) {
|
||
case let r: rune => rest_r = r;
|
||
case utf8.done => rest_done = true;
|
||
};
|
||
let needle_done: bool = false;
|
||
let needle_r: rune;
|
||
match (next(&needle_iter)) {
|
||
case let r: rune => needle_r = r;
|
||
case utf8.done => needle_done = true;
|
||
};
|
||
if (rest_done && !needle_done) { break; };
|
||
if (needle_done) { matched = true; break; };
|
||
if (rest_r != needle_r) { break; };
|
||
};
|
||
if (matched) { return i; };
|
||
match (next(&s_iter)) {
|
||
case let r: rune => i += 1;
|
||
case utf8.done => return;
|
||
};
|
||
};
|
||
return;
|
||
};
|
||
|
||
// index — rune-wise offset of `needle`'s first occurrence in
|
||
// `haystack`, or void if absent. ref/hare/strings/index.ha:10. The
|
||
// str-arm delegates to [[indexstring]] (dual-iterator rune-by-rune
|
||
// walk per Hare's `index_string`, #10); the rune-arm mirrors Hare's
|
||
// `index_rune` (ref/hare/strings/index.ha:31).
|
||
export fn index(haystack: str, needle: (str | rune)) (i32 | void) = {
|
||
match (needle) {
|
||
case let s: str => return indexstring(haystack, s);
|
||
case let r: rune => {
|
||
let it: iterator = iter(haystack);
|
||
let i: i32 = 0;
|
||
for (true) {
|
||
match (next(&it)) {
|
||
case let n: rune => {
|
||
if (n == r) { return i; };
|
||
i += 1;
|
||
};
|
||
case utf8.done => return;
|
||
};
|
||
};
|
||
};
|
||
};
|
||
return;
|
||
};
|
||
|
||
// rindex — rune-wise offset of `needle`'s last occurrence in
|
||
// `haystack`, or void if absent. ref/hare/strings/index.ha:22. The
|
||
// str-arm reuses `rbyteindex`; the rune-arm walks forward tracking
|
||
// the most recent matching rune index (Hare's `rindex_rune` with
|
||
// `riter` returns a byte-offset value for multibyte strings, which
|
||
// disagrees with the rune-wise docstring; we keep the docstring's
|
||
// contract).
|
||
export fn rindex(haystack: str, needle: (str | rune)) (i32 | void) = {
|
||
match (needle) {
|
||
case let s: str => {
|
||
match (rbyteindex(haystack, s)) {
|
||
case void => return;
|
||
case let bo: i32 => {
|
||
let it: iterator = iter(haystack);
|
||
let i: i32 = 0;
|
||
for (position(&it) < bo) {
|
||
match (next(&it)) {
|
||
case let r: rune => i += 1;
|
||
case utf8.done => break;
|
||
};
|
||
};
|
||
return i;
|
||
};
|
||
};
|
||
};
|
||
case let r: rune => {
|
||
let it: iterator = iter(haystack);
|
||
let i: i32 = 0;
|
||
let last: i32 = -1;
|
||
for (true) {
|
||
match (next(&it)) {
|
||
case let n: rune => {
|
||
if (n == r) { last = i; };
|
||
i += 1;
|
||
};
|
||
case utf8.done => break;
|
||
};
|
||
};
|
||
if (last < 0) { return; };
|
||
return last;
|
||
};
|
||
};
|
||
return;
|
||
};
|
||
|
||
// contains — true iff any of `needles` occurs in `haystack`.
|
||
// ref/hare/strings/contains.ha:9.
|
||
export fn contains(haystack: str, needles: (str | rune)...) bool = {
|
||
let i: i32 = 0;
|
||
for (i < needles.len) {
|
||
match (needles[i]) {
|
||
case let s: str => {
|
||
match (byteindex(haystack, s)) {
|
||
case let bo: i32 => return true;
|
||
case void => void;
|
||
};
|
||
};
|
||
case let r: rune => {
|
||
match (byteindex(haystack, r)) {
|
||
case let bo: i32 => return true;
|
||
case void => void;
|
||
};
|
||
};
|
||
};
|
||
i += 1;
|
||
};
|
||
return false;
|
||
};
|
||
|
||
// trimprefix — `s` with `prefix` stripped from the front, or `s`
|
||
// unchanged if it doesn't start with `prefix`. Borrowed view.
|
||
// ref/hare/strings/trim.ha:60.
|
||
export fn trimprefix(input: str, prefix: str) str = {
|
||
if (!hasprefix(input, prefix)) { return input; };
|
||
let r: str;
|
||
r.ptr = input.ptr + (prefix.len: u64);
|
||
r.len = input.len - prefix.len;
|
||
return r;
|
||
};
|
||
|
||
// trimsuffix — symmetric. ref/hare/strings/trim.ha:69.
|
||
export fn trimsuffix(input: str, suffix: str) str = {
|
||
if (!hassuffix(input, suffix)) { return input; };
|
||
let r: str;
|
||
r.ptr = input.ptr;
|
||
r.len = input.len - suffix.len;
|
||
return r;
|
||
};
|
||
|
||
// whitespace — ASCII whitespace set used by the 0-arg ltrim/rtrim/trim
|
||
// branches (#9). ref/hare/strings/trim.ha:6.
|
||
let whitespace: [4]u8 = [0x20u8, 0x0Au8, 0x09u8, 0x0Du8];
|
||
|
||
// ltrim — strip leading runes that occur in `trim`. Borrowed view.
|
||
// 0-arg strips ASCII whitespace via [[bytes.ltrim]] (#9).
|
||
// ref/hare/strings/trim.ha:11. The spread expression is inlined
|
||
// because `let ws: []u8 = whitespace[0:4]` produces a slice whose
|
||
// ptr doesn't track the module-level array storage (filed as #40);
|
||
// `b.flush = flushdefault[0:1]` in lib/bufio is the same shape via
|
||
// the working field-assign path.
|
||
export fn ltrim(input: str, trim: rune...) str = {
|
||
if (trim.len == 0) {
|
||
return frombytes(bytes.ltrim(toutf8(input), whitespace[0:4]...));
|
||
};
|
||
let it: iterator = iter(input);
|
||
for (true) {
|
||
match (next(&it)) {
|
||
case let r: rune => {
|
||
let j: i32 = 0;
|
||
let found: bool = false;
|
||
for (j < trim.len) {
|
||
if (r == trim[j]) { found = true; j = trim.len; }
|
||
else { j += 1; };
|
||
};
|
||
if (!found) {
|
||
match (prev(&it)) {
|
||
case let r2: rune => void;
|
||
case utf8.done => void;
|
||
};
|
||
break;
|
||
};
|
||
};
|
||
case utf8.done => break;
|
||
};
|
||
};
|
||
return iterstr(&it);
|
||
};
|
||
|
||
// rtrim — strip trailing runes that occur in `trim`. Borrowed view.
|
||
// 0-arg strips ASCII whitespace via [[bytes.rtrim]] (#9). Spread is
|
||
// inlined to dodge #40 — see [[ltrim]].
|
||
// ref/hare/strings/trim.ha:32.
|
||
export fn rtrim(input: str, trim: rune...) str = {
|
||
if (trim.len == 0) {
|
||
return frombytes(bytes.rtrim(toutf8(input), whitespace[0:4]...));
|
||
};
|
||
let it: iterator = riter(input);
|
||
for (true) {
|
||
match (next(&it)) {
|
||
case let r: rune => {
|
||
let j: i32 = 0;
|
||
let found: bool = false;
|
||
for (j < trim.len) {
|
||
if (r == trim[j]) { found = true; j = trim.len; }
|
||
else { j += 1; };
|
||
};
|
||
if (!found) {
|
||
match (prev(&it)) {
|
||
case let r2: rune => void;
|
||
case utf8.done => void;
|
||
};
|
||
break;
|
||
};
|
||
};
|
||
case utf8.done => break;
|
||
};
|
||
};
|
||
return iterstr(&it);
|
||
};
|
||
|
||
// trim — strip from both ends. ref/hare/strings/trim.ha:54.
|
||
export fn trim(input: str, trim: rune...) str = {
|
||
return ltrim(rtrim(input, trim...), trim...);
|
||
};
|
||
|
||
// iterator — UTF-8 rune cursor over a `str`. Layout flattens Hare's
|
||
// anonymous-embedded `utf8::decoder` (ref/hare/strings/iter.ha:6-9) to
|
||
// explicit fields. `reverse` selects walk direction: forward iterators
|
||
// (`iter`) advance through utf8.next; reverse iterators (`riter`) advance
|
||
// through utf8.prev. May be copied to save state.
|
||
export type iterator = struct {
|
||
offs: i32,
|
||
src: []u8,
|
||
reverse: bool,
|
||
};
|
||
|
||
// iter — initialize a forward iterator at the start of `src`.
|
||
// ref/hare/strings/iter.ha:24.
|
||
export fn iter(src: str) iterator = {
|
||
let r: iterator;
|
||
r.src = toutf8(src);
|
||
r.offs = 0;
|
||
r.reverse = false;
|
||
return r;
|
||
};
|
||
|
||
// riter — initialize a reverse iterator at the end of `src`. `next`
|
||
// on a reverse iterator walks back through the string.
|
||
// ref/hare/strings/iter.ha:32.
|
||
export fn riter(src: str) iterator = {
|
||
let r: iterator;
|
||
r.src = toutf8(src);
|
||
r.offs = src.len;
|
||
r.reverse = true;
|
||
return r;
|
||
};
|
||
|
||
// move — private dispatch shared by next/prev. `forward` selects
|
||
// utf8.next vs utf8.prev. Aborts on more/invalid per Hare's
|
||
// ref/hare/strings/iter.ha:51-58 ("Invalid UTF-8 string (this should
|
||
// not happen)"). Hare picks the utf8 function via a fn-pointer; ww
|
||
// branches on `forward` at each call site instead.
|
||
fn move(forward: bool, it: *iterator) (rune | utf8.done) = {
|
||
let d: utf8.decoder;
|
||
d.src = it.src;
|
||
d.offs = it.offs;
|
||
if (forward) {
|
||
match (utf8.next(&d)) {
|
||
case let r: rune => { it.offs = d.offs; return r; };
|
||
case let dn: utf8.done => return dn;
|
||
case let m: utf8.more => abort("strings.move: invalid UTF-8");
|
||
case let e: utf8.invalid => abort("strings.move: invalid UTF-8");
|
||
};
|
||
} else {
|
||
match (utf8.prev(&d)) {
|
||
case let r: rune => { it.offs = d.offs; return r; };
|
||
case let dn: utf8.done => return dn;
|
||
case let m: utf8.more => abort("strings.move: invalid UTF-8");
|
||
case let e: utf8.invalid => abort("strings.move: invalid UTF-8");
|
||
};
|
||
};
|
||
};
|
||
|
||
// next — advance the iterator one rune. Forward iterators step
|
||
// through utf8.next; reverse iterators (riter) step backward through
|
||
// utf8.prev. Returns utf8.done at end-of-walk. ref/hare/strings/iter.ha:45.
|
||
export fn next(it: *iterator) (rune | utf8.done) = {
|
||
return move(!it.reverse, it);
|
||
};
|
||
|
||
// prev — step back one rune. Dual to next: on a forward iterator
|
||
// this walks utf8.prev; on a reverse iterator (riter) it walks
|
||
// utf8.next. ref/hare/strings/iter.ha:49.
|
||
export fn prev(it: *iterator) (rune | utf8.done) = {
|
||
return move(it.reverse, it);
|
||
};
|
||
|
||
// iterstr — borrowed view of the bytes remaining in the iterator's
|
||
// walk direction. Forward iter: bytes from offs to end; reverse iter:
|
||
// bytes from start to offs. ref/hare/strings/iter.ha:63.
|
||
export fn iterstr(it: *iterator) str = {
|
||
let r: []u8;
|
||
if (it.reverse) {
|
||
r = it.src[0:it.offs];
|
||
} else {
|
||
r = it.src[it.offs:it.src.len];
|
||
};
|
||
return frombytes(r);
|
||
};
|
||
|
||
// slice — borrowed substring between two iterator positions.
|
||
// ref/hare/strings/iter.ha:75. Hare passes `*iterator` directly where
|
||
// `*utf8::decoder` is expected via anonymous-embed coercion; ww has
|
||
// no anonymous embed, so we reconstruct a local utf8.decoder for each
|
||
// endpoint and forward — same pattern as `move` above.
|
||
export fn slice(begin: *iterator, end: *iterator) str = {
|
||
let b: utf8.decoder;
|
||
b.src = begin.src;
|
||
b.offs = begin.offs;
|
||
let e: utf8.decoder;
|
||
e.src = end.src;
|
||
e.offs = end.offs;
|
||
return frombytes(utf8.slice(&b, &e));
|
||
};
|
||
|
||
// position — byte-wise offset of the iterator in its source.
|
||
// ref/hare/strings/iter.ha:82.
|
||
export fn position(it: *iterator) i32 = {
|
||
return it.offs;
|
||
};
|
||
|
||
// tokenizer — re-export of bytes.tokenizer. ref/hare/strings/tokenize.ha:7.
|
||
// First cross-module type alias in tree; needs #22's transitive
|
||
// alias-chain unwrap (cstage type_chase_named + wwstage
|
||
// structlookupchain) to walk struct fields through the chain.
|
||
export type tokenizer = bytes.tokenizer;
|
||
|
||
// tokenize — yield substrings of `s` split on any byte in `delim`.
|
||
// Leading / trailing / adjacent delims yield empty tokens. `s` and
|
||
// `delim` are borrowed; caller keeps them live for the tokenizer's
|
||
// lifetime. ref/hare/strings/tokenize.ha:32. ASCII-only delim
|
||
// asserted per Hare lines 35-37: a multibyte rune in delim would
|
||
// split on a single continuation byte and yield invalid UTF-8.
|
||
export fn tokenize(s: str, delim: str) tokenizer = {
|
||
let d: []u8 = toutf8(delim);
|
||
let i: i32 = 0;
|
||
for (i < d.len) {
|
||
os.assert((d[i] & 0x80u8) == 0u8,
|
||
"strings.tokenize cannot tokenize on non-ASCII delimiters");
|
||
i += 1;
|
||
};
|
||
return bytes.tokenize(toutf8(s), d...);
|
||
};
|
||
|
||
// rtokenize — reverse-direction counterpart to [[tokenize]]. First
|
||
// next_token yields the last token, last yields the first.
|
||
// ref/hare/strings/tokenize.ha:44.
|
||
export fn rtokenize(s: str, delim: str) tokenizer = {
|
||
let d: []u8 = toutf8(delim);
|
||
let i: i32 = 0;
|
||
for (i < d.len) {
|
||
os.assert((d[i] & 0x80u8) == 0u8,
|
||
"strings.rtokenize cannot tokenize on non-ASCII delimiters");
|
||
i += 1;
|
||
};
|
||
return bytes.rtokenize(toutf8(s), d...);
|
||
};
|
||
|
||
// next_token — current token, advancing the cursor.
|
||
// ref/hare/strings/tokenize.ha:62.
|
||
export fn next_token(s: *tokenizer) (str | bytes.done) = {
|
||
let b: *bytes.tokenizer = s: *bytes.tokenizer;
|
||
match (bytes.next_token(b)) {
|
||
case let v: []u8 => return frombytes(v);
|
||
case bytes.done => { let d: bytes.done; return d; };
|
||
};
|
||
};
|
||
|
||
// peek_token — current token without advancing.
|
||
// ref/hare/strings/tokenize.ha:71.
|
||
export fn peek_token(s: *tokenizer) (str | bytes.done) = {
|
||
let b: *bytes.tokenizer = s: *bytes.tokenizer;
|
||
match (bytes.peek_token(b)) {
|
||
case let v: []u8 => return frombytes(v);
|
||
case bytes.done => { let d: bytes.done; return d; };
|
||
};
|
||
};
|
||
|
||
// remaining_tokens — unconsumed portion of the input ahead of the
|
||
// cursor. ref/hare/strings/tokenize.ha:79.
|
||
export fn remaining_tokens(s: *tokenizer) str = {
|
||
let b: *bytes.tokenizer = s: *bytes.tokenizer;
|
||
return frombytes(bytes.remaining_tokens(b));
|
||
};
|
||
|
||
// cut — split `in` along the first instance of `delim`, returning the
|
||
// portions before and after it. When `delim` is absent the whole input
|
||
// is the first half and the second is empty. Both halves are borrowed
|
||
// from `in`; caller ensures `delim` is non-empty.
|
||
// ref/hare/strings/tokenize.ha:288.
|
||
export fn cut(in: str, delim: str) (str, str) = {
|
||
let (a, b) = bytes.cut(toutf8(in), toutf8(delim));
|
||
return (frombytes(a), frombytes(b));
|
||
};
|
||
|
||
// rcut — like [[cut]] but split along the LAST instance of `delim`.
|
||
// ref/hare/strings/tokenize.ha:302.
|
||
export fn rcut(in: str, delim: str) (str, str) = {
|
||
let (a, b) = bytes.rcut(toutf8(in), toutf8(delim));
|
||
return (frombytes(a), frombytes(b));
|
||
};
|
||
|
||
// rt_ensure is the runtime slice-growth helper invoked by the
|
||
// `append(s, v)` builtin. Direct bind for the same reason as
|
||
// lib/shlex.shlex (appendstr, 16B): the builtin's expansion stores
|
||
// only 8B of the new element, losing the `.len` half of a `str`.
|
||
@symbol("rt_ensure") fn rtensure(s: *void, membsz: u64) void;
|
||
|
||
// appendstr — grow `*slice` by one and store `item` (16B). Mirror of
|
||
// lib/shlex.shlex appendstr. Collapses when the append builtin learns
|
||
// to store the full element width.
|
||
fn appendstr(slice: *[]str, item: str) void = {
|
||
let newlen: i32 = slice.len + 1;
|
||
slice.len = newlen;
|
||
rtensure(slice: *void, size(str): u64);
|
||
let dst: *str = &slice.ptr[newlen - 1];
|
||
dst.ptr = item.ptr;
|
||
dst.len = item.len;
|
||
};
|
||
|
||
// splitn — split `in` on any byte in `delim`, returning up to `n`
|
||
// tokens via forward iteration. The trailing slot (when more than
|
||
// `n - 1` tokens exist) holds the unconsumed remainder. Strings
|
||
// within the result are borrowed from `in`.
|
||
//
|
||
// The caller frees the returned slice via
|
||
// `os.free(r.ptr: *void, (r.cap: u64) * size(str): u64)`.
|
||
//
|
||
// Hare's `([]str | nomem)` collapses to `[]str` here: ww os.alloc
|
||
// has no recoverable failure path. Same precedent as
|
||
// shlex.split / bytes.splitn.
|
||
//
|
||
// ref/hare/strings/tokenize.ha:172.
|
||
export fn splitn(in: str, delim: str, n: i32) []str = {
|
||
let toks: []str;
|
||
toks.ptr = nil: *str;
|
||
toks.len = 0;
|
||
toks.cap = 0;
|
||
let tok: tokenizer = tokenize(in, delim);
|
||
let i: i32 = 0;
|
||
for (i < n - 1) {
|
||
match (next_token(&tok)) {
|
||
case let s: str => { appendstr(&toks, s); };
|
||
case bytes.done => { return toks; };
|
||
};
|
||
i += 1;
|
||
};
|
||
match (peek_token(&tok)) {
|
||
case bytes.done => void;
|
||
case let pk: str => {
|
||
let r: str = remaining_tokens(&tok);
|
||
appendstr(&toks, r);
|
||
};
|
||
};
|
||
return toks;
|
||
};
|
||
|
||
// rsplitn — reverse-direction counterpart to [[splitn]]: tokens are
|
||
// collected from the end of `in`. The trailing slot holds the
|
||
// unconsumed prefix (everything before the n-th-from-last delim hit).
|
||
//
|
||
// When the input has fewer than n tokens, the `done` short-circuit
|
||
// returns toks UN-reversed (in last-token-first order). Mirrors Hare
|
||
// at ref/hare/strings/tokenize.ha:219-224 where the in-place reverse
|
||
// step is gated behind the n-1 loop running to completion.
|
||
//
|
||
// ref/hare/strings/tokenize.ha:200.
|
||
export fn rsplitn(in: str, delim: str, n: i32) []str = {
|
||
let toks: []str;
|
||
toks.ptr = nil: *str;
|
||
toks.len = 0;
|
||
toks.cap = 0;
|
||
let tok: tokenizer = rtokenize(in, delim);
|
||
let i: i32 = 0;
|
||
for (i < n - 1) {
|
||
match (next_token(&tok)) {
|
||
case let s: str => { appendstr(&toks, s); };
|
||
case bytes.done => { return toks; };
|
||
};
|
||
i += 1;
|
||
};
|
||
match (peek_token(&tok)) {
|
||
case bytes.done => void;
|
||
case let pk: str => {
|
||
let r: str = remaining_tokens(&tok);
|
||
appendstr(&toks, r);
|
||
};
|
||
};
|
||
|
||
// In-place reverse so callers see argv-order, matching Hare
|
||
// (ref/hare/strings/tokenize.ha:220). Element copy is field-wise
|
||
// through `*str` because `toks[i] = toks[j]` (full 16B str store)
|
||
// lands in the multi-word-store gap noted at cmd/w6c/cgen.c:6515.
|
||
let a: i32 = 0;
|
||
let b: i32 = toks.len - 1;
|
||
for (a < b) {
|
||
let pa: *str = &toks.ptr[a];
|
||
let pb: *str = &toks.ptr[b];
|
||
let tp: *u8 = pa.ptr;
|
||
let tl: i32 = pa.len;
|
||
pa.ptr = pb.ptr;
|
||
pa.len = pb.len;
|
||
pb.ptr = tp;
|
||
pb.len = tl;
|
||
a += 1;
|
||
b -= 1;
|
||
};
|
||
return toks;
|
||
};
|
||
|
||
// split — full split of `in` on `delim` (no token cap). Mirrors
|
||
// `splitn(in, delim, types::SIZE_MAX)`. ww uses `types.I32_MAX`
|
||
// because the index type is i32 (lib/CLAUDE.md).
|
||
//
|
||
// ref/hare/strings/tokenize.ha:242.
|
||
export fn split(in: str, delim: str) []str = {
|
||
return splitn(in, delim, types.I32_MAX);
|
||
};
|
||
|
||
// lpad — left-pad `s` with `p` rune until the result reaches `maxlen`
|
||
// bytes. Length comparison is BYTES, mirroring Hare's `len(s) >= maxlen`
|
||
// at ref/hare/strings/pad.ha:9. A multibyte `p` whose encoded width
|
||
// doesn't divide `maxlen - s.len` evenly leaves a trailing pad byte
|
||
// pair sliced mid-codepoint at byte `maxlen-1`, exactly as Hare's
|
||
// `res[..maxlen]` does (ref/hare/strings/pad.ha:20). When
|
||
// `(maxlen - s.len) * pad.len >= maxlen` (multibyte pad overflows the
|
||
// budget), `s` is entirely sliced off — same as Hare. Caller releases
|
||
// with `os.free(r.ptr, r.len: u64)`. Hare's `nomem` return is dropped:
|
||
// `os.alloc` aborts on OOM. Buf size == r.len keeps the free-contract
|
||
// shape of [[dup]] / [[concat]] / [[join]]; Hare's `alloc([], maxlen)!`
|
||
// over-allocs via append then slices, but Hare's slice-free recovers
|
||
// the true capacity from the heap allocator (rt/ensure.ha:24), which
|
||
// ww's munmap-based `os.free` cannot do.
|
||
export fn lpad(s: str, p: rune, maxlen: i32) str = {
|
||
if (s.len >= maxlen) { return dup(s); };
|
||
let scratch: [4]u8;
|
||
let pad: []u8 = runebytes(scratch[0:4], p);
|
||
let buf: []u8 = alloc([], maxlen: u64)!;
|
||
let padwrite: i32 = (maxlen - s.len) * pad.len;
|
||
if (padwrite > maxlen) { padwrite = maxlen; };
|
||
let off: i32 = 0;
|
||
for (off < padwrite) {
|
||
buf[off] = pad.ptr[off % pad.len];
|
||
off += 1;
|
||
};
|
||
let k: i32 = 0;
|
||
let srem: i32 = maxlen - off;
|
||
if (srem > s.len) { srem = s.len; };
|
||
for (k < srem) {
|
||
buf[off + k] = s[k];
|
||
k += 1;
|
||
};
|
||
buf.len = maxlen;
|
||
return frombytes(buf);
|
||
};
|
||
|
||
// replace — fresh allocation of `s` with every non-overlapping
|
||
// occurrence of `needle` replaced by `target`. Caller releases with
|
||
// `os.free(r.ptr, r.len: u64)`. ref/hare/strings/replace.ha:8 (#4).
|
||
//
|
||
// Hare delegates to [[multireplace]] with a single pair; ww has no
|
||
// `(str, str)` variadic shape today (#39), so this is a standalone
|
||
// two-pass implementation: pass 1 counts matches to size the result,
|
||
// pass 2 copies chunks and `target` into a single fresh buffer.
|
||
// Single nomem path (the `alloc([], total)?`) preserves Hare's
|
||
// signature without a per-write `append(...)?` (ww's append builtin
|
||
// aborts on OOM, #11). Empty `needle` would hasprefix-match every
|
||
// position with a zero stride — same infinite loop Hare exhibits at
|
||
// ref/hare/strings/replace.ha:31; not gated.
|
||
export fn replace(s: str, needle: str, target: str) (str | nomem) = {
|
||
let sb: []u8 = toutf8(s);
|
||
let nb: []u8 = toutf8(needle);
|
||
let tb: []u8 = toutf8(target);
|
||
let count: i32 = 0;
|
||
let i: i32 = 0;
|
||
for (i < sb.len) {
|
||
if (bytes.hasprefix(sb[i:sb.len], nb)) {
|
||
count += 1;
|
||
i += nb.len;
|
||
} else {
|
||
i += 1;
|
||
};
|
||
};
|
||
let total: i32 = sb.len + count * (tb.len - nb.len);
|
||
if (total == 0) {
|
||
let r: str;
|
||
r.ptr = nil;
|
||
r.len = 0;
|
||
return r;
|
||
};
|
||
let res: []u8 = alloc([], total)?;
|
||
let off: i32 = 0;
|
||
i = 0;
|
||
for (i < sb.len) {
|
||
if (bytes.hasprefix(sb[i:sb.len], nb)) {
|
||
let j: i32 = 0;
|
||
for (j < tb.len) {
|
||
res.ptr[off + j] = tb.ptr[j];
|
||
j += 1;
|
||
};
|
||
off += tb.len;
|
||
i += nb.len;
|
||
} else {
|
||
res.ptr[off] = sb.ptr[i];
|
||
off += 1;
|
||
i += 1;
|
||
};
|
||
};
|
||
res.len = total;
|
||
return frombytes(res);
|
||
};
|
||
|
||
// rpad — right-pad `s` with `p` rune until the result reaches `maxlen`
|
||
// bytes. Symmetric with [[lpad]]. ref/hare/strings/pad.ha:39.
|
||
export fn rpad(s: str, p: rune, maxlen: i32) str = {
|
||
if (s.len >= maxlen) { return dup(s); };
|
||
let scratch: [4]u8;
|
||
let pad: []u8 = runebytes(scratch[0:4], p);
|
||
let buf: []u8 = alloc([], maxlen: u64)!;
|
||
let k: i32 = 0;
|
||
for (k < s.len) {
|
||
buf[k] = s[k];
|
||
k += 1;
|
||
};
|
||
let padwrite: i32 = maxlen - s.len;
|
||
let i: i32 = 0;
|
||
for (i < padwrite) {
|
||
buf[s.len + i] = pad.ptr[i % pad.len];
|
||
i += 1;
|
||
};
|
||
buf.len = maxlen;
|
||
return frombytes(buf);
|
||
};
|
||
|
||
// 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.
|
||
|
||
package ascii;
|
||
|
||
import strings;
|
||
|
||
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;
|
||
};
|
||
|
||
// strlower — ASCII-lowercased copy of s, newly allocated.
|
||
// ref/hare/ascii/string.ha:11.
|
||
export fn strlower(s: str) (str | nomem) = {
|
||
// empty bypass: ww alloc([],0) routes through nomem; Hare allocs 0
|
||
// and zero-loops (ref/hare/ascii/string.ha:12).
|
||
if (s.len == 0) {
|
||
let r: str;
|
||
r.ptr = nil;
|
||
r.len = 0;
|
||
return r;
|
||
};
|
||
let buf: []u8 = alloc([], s.len: u64)?;
|
||
return strlower_buf(s, buf);
|
||
};
|
||
|
||
// strlower_buf — ASCII-lowercase s into buf (overwrites). nomem if buf
|
||
// too small. ref/hare/ascii/string.ha:21.
|
||
// Byte-wise fold: ASCII case-fold only touches bytes <0x80; UTF-8
|
||
// multibyte bytes are >=0x80 and pass through unchanged, so byte-wise
|
||
// equals Hare's rune fold and is length-preserving.
|
||
// ww uses an explicit `buf.cap < s.len` check + `let nm: nomem` value
|
||
// because it has no static-append builtin; Hare reaches the same
|
||
// nomem-on-too-small via `static append(buf, ...)?` (string.ha:25).
|
||
export fn strlower_buf(s: str, buf: []u8) (str | nomem) = {
|
||
if (buf.cap < s.len) {
|
||
let nm: nomem;
|
||
return nm;
|
||
};
|
||
let i: i32 = 0;
|
||
for (i < s.len) {
|
||
buf.ptr[i] = tolower(s[i]: rune): u8;
|
||
i += 1;
|
||
};
|
||
buf.len = s.len;
|
||
return strings.frombytes(buf);
|
||
};
|
||
|
||
// strupper — ASCII-uppercased copy of s, newly allocated.
|
||
// ref/hare/ascii/string.ha:33.
|
||
export fn strupper(s: str) (str | nomem) = {
|
||
if (s.len == 0) {
|
||
let r: str;
|
||
r.ptr = nil;
|
||
r.len = 0;
|
||
return r;
|
||
};
|
||
let buf: []u8 = alloc([], s.len: u64)?;
|
||
return strupper_buf(s, buf);
|
||
};
|
||
|
||
// strupper_buf — see strlower_buf. ref/hare/ascii/string.ha:43.
|
||
export fn strupper_buf(s: str, buf: []u8) (str | nomem) = {
|
||
if (buf.cap < s.len) {
|
||
let nm: nomem;
|
||
return nm;
|
||
};
|
||
let i: i32 = 0;
|
||
for (i < s.len) {
|
||
buf.ptr[i] = toupper(s[i]: rune): u8;
|
||
i += 1;
|
||
};
|
||
buf.len = s.len;
|
||
return strings.frombytes(buf);
|
||
};
|
||
|
||
// strconv — string-to-float. Mirrors ref/hare/strconv/stof.ha
|
||
// (Hare in turn adapts Go): Eisel-Lemire fast path [1] with the
|
||
// Simple-Decimal-Conversion slow path [2] (decimal.ww) as fallback.
|
||
// [1]: https://nigeltao.github.io/blog/2020/eisel-lemire.html
|
||
// [2]: https://nigeltao.github.io/blog/2020/parse-number-f64-simple.html
|
||
//
|
||
// The Eisel-Lemire fast path (`eisel_lemire` + the `powers_of_ten`
|
||
// table in stof_data.ww + the three call sites: floatbits's d.nd<=19
|
||
// block, stof64/stof32's !truncated block) is a pure speed
|
||
// optimisation — it returns the same correctly-rounded value the
|
||
// decimal slow path (decimal_parse → floatbits) computes, or void to
|
||
// defer. Its prereqs landed: the 2D `[596][2]u64` static-init +
|
||
// double-index read (#156) and the tagged float-variant return-pack
|
||
// (#157, which the public `(f64|invalid|overflow)` return needs).
|
||
//
|
||
// Spelling divergences from Hare (mechanical, ww parser/cgen shape):
|
||
// - str scan index rides `i32` (ww `str.len: i32` + `invalid = !i32`
|
||
// payload), not Hare's `size`/`len(s)`. lib CLAUDE.md str-index note.
|
||
// - char literals kept faithful (`buf[i] == '.'`, `c - '0'`); probed
|
||
// byte-id + value-correct both stages.
|
||
// - Hare `?` error-propagation → nested statement-`match` with all-
|
||
// return arms + a `case void => void` continuation. ww's `?`
|
||
// lowering and a bound `match`-expression with mixed yield/return
|
||
// arms both diverge cs≠ww (the latter wwstage-checker-rejected);
|
||
// strconv.ww's stoi32 set the explicit-match precedent.
|
||
// - Hare `for (cond; afterthought)` 2-clause + `continue` → ww
|
||
// 2-clause `for (cond)` with the afterthought inlined at body end
|
||
// AND before each `continue` (ww has no empty-init 3-clause
|
||
// `for (; c; p)`; #138 post-skip is dodged since 2-clause has no
|
||
// post). decimal.ww set the inline-afterthought precedent.
|
||
// - Hare `if`/`switch`-expression yield → explicit if-statements +
|
||
// pre-bound scalar locals (ww has no expression-bodied if).
|
||
// - Hare fn-pointer-in-tuple + `switch yield` selecting the digit
|
||
// predicate in fast_parse → a `base==HEX` bool + an `isdigitbase`
|
||
// helper that branches to ascii.isdigit/isxdigit (no fn-ptr, no
|
||
// tuple, no switch).
|
||
// - struct-param field MUTATION (hex_to_bits mutates its by-value
|
||
// `p`) → copy p's fields to scalar locals at entry; ww miscompiles
|
||
// + diverges on writing a by-value struct param's fields (filed).
|
||
// - default arg dropped: Hare `b: base = base::DEC` → callers pass
|
||
// base explicitly (no lib fn ships a default arg; strconv.ww
|
||
// stoi64 precedent). The base param is normalised through a local
|
||
// `bb` (param reassignment avoided).
|
||
// - `math::NAN`/`math::INF` (f32) absent in ww math → materialised
|
||
// via f32frombits of the IEEE-754 f32 bit patterns (same honest
|
||
// construction as math/floats.ww's NAN_BITS/INF_BITS).
|
||
// - narrowing int→i32 assignments carry explicit casts (ww `int` is
|
||
// an 8B machine word; project_int_machine_word_derived_limits).
|
||
// - `r128`/`u128mul` live here (fold-4 is first consumer); fold-5
|
||
// ftos (Ryū) shares them in-package.
|
||
|
||
package strconv;
|
||
|
||
import ascii;
|
||
import math;
|
||
import os;
|
||
import strings;
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:12. 64×64→128 result halves.
|
||
type r128 = struct {
|
||
hi: u64,
|
||
lo: u64,
|
||
};
|
||
|
||
// ref/hare/strconv/ftos_ryu.ha:18. 64×64→128 via 32-bit decomposition
|
||
// (Hare's own "TODO: use 128-bit integers when implemented" — ww has
|
||
// no u128; the decomposition is the portable shape both stages agree
|
||
// on). Comma let-bindings split per decimal.ww divergence.
|
||
fn u128mul(a: u64, b: u64) r128 = {
|
||
let a0: u64 = (a: u32): u64;
|
||
let a1: u64 = a >> 32u64;
|
||
let b0: u64 = (b: u32): u64;
|
||
let b1: u64 = b >> 32u64;
|
||
let p00: u64 = a0 * b0;
|
||
let p01: u64 = a0 * b1;
|
||
let p10: u64 = a1 * b0;
|
||
let p11: u64 = a1 * b1;
|
||
let p00_lo: u64 = (p00: u32): u64;
|
||
let p00_hi: u64 = p00 >> 32u64;
|
||
let mid1: u64 = p10 + p00_hi;
|
||
let mid1_lo: u64 = (mid1: u32): u64;
|
||
let mid1_hi: u64 = mid1 >> 32u64;
|
||
let mid2: u64 = p01 + mid1_lo;
|
||
let mid2_lo: u64 = (mid2: u32): u64;
|
||
let mid2_hi: u64 = mid2 >> 32u64;
|
||
let r_hi: u64 = p11 + mid1_hi + mid2_hi;
|
||
let r_lo: u64 = (mid2_lo << 32u64) | p00_lo;
|
||
return r128 { hi = r_hi, lo = r_lo };
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:14.
|
||
fn todig(c: u8) u8 = {
|
||
if ('0' <= c && c <= '9') { return c - '0'; };
|
||
if ('a' <= c && c <= 'f') { return c - 'a' + 10u8; };
|
||
if ('A' <= c && c <= 'F') { return c - 'A' + 10u8; };
|
||
abort("strconv.todig: unreachable");
|
||
return 0u8; // unreachable; rt_abort is void-typed (path-cov)
|
||
};
|
||
|
||
@symbol("rt_abort") fn abort(msg: str) void;
|
||
|
||
// ref/hare/strconv/stof.ha:25.
|
||
type fast_parsed_float = struct {
|
||
mantissa: u64,
|
||
exponent: i32,
|
||
negative: bool,
|
||
truncated: bool,
|
||
};
|
||
|
||
// Digit-class predicate selector for fast_parse — replaces Hare's
|
||
// fn-pointer-in-tuple (`&ascii::isdigit` / `&ascii::isxdigit`).
|
||
fn isdigitbase(c: rune, ishex: bool) bool = {
|
||
if (ishex) { return ascii.isxdigit(c); };
|
||
return ascii.isdigit(c);
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:32.
|
||
fn fast_parse(s: str, b: base) (fast_parsed_float | invalid) = {
|
||
let buf: []u8 = strings.toutf8(s);
|
||
let i: i32 = 0;
|
||
let neg: bool = false;
|
||
let trunc: bool = false;
|
||
if (buf[i] == '-') {
|
||
neg = true;
|
||
i += 1;
|
||
} else if (buf[i] == '+') {
|
||
i += 1;
|
||
};
|
||
|
||
let ishex: bool = (b == base.HEX);
|
||
let expchr: rune = 'e';
|
||
let max_ndmant: int = 19;
|
||
if (ishex) {
|
||
expchr = 'p';
|
||
max_ndmant = 16;
|
||
};
|
||
let bnum: u64 = (b: i32): u64;
|
||
|
||
let sawdot: bool = false;
|
||
let sawdigits: bool = false;
|
||
let nd: int = 0;
|
||
let ndmant: int = 0;
|
||
let dp: int = 0;
|
||
let mant: u64 = 0u64;
|
||
let exp: i32 = 0i32;
|
||
for (i < s.len) {
|
||
if (buf[i] == '.') {
|
||
if (sawdot) { return i: invalid; };
|
||
sawdot = true;
|
||
dp = nd;
|
||
} else if (isdigitbase(buf[i]: rune, ishex)) {
|
||
sawdigits = true;
|
||
if (buf[i] == '0' && nd == 0) {
|
||
dp -= 1;
|
||
i += 1;
|
||
continue;
|
||
};
|
||
nd += 1;
|
||
if (ndmant < max_ndmant) {
|
||
mant = mant * bnum + (todig(buf[i]): u64);
|
||
ndmant += 1;
|
||
} else if (buf[i] != '0') {
|
||
trunc = true;
|
||
};
|
||
} else {
|
||
break;
|
||
};
|
||
i += 1;
|
||
};
|
||
if (!sawdigits) { return i: invalid; };
|
||
if (!sawdot) {
|
||
dp = nd;
|
||
};
|
||
if (b == base.HEX) {
|
||
dp *= 4;
|
||
ndmant *= 4;
|
||
};
|
||
if (i < s.len && ascii.tolower(buf[i]: rune) == expchr) {
|
||
i += 1;
|
||
if (i >= s.len) { return i: invalid; };
|
||
let expsign: int = 1;
|
||
if (buf[i] == '+') {
|
||
i += 1;
|
||
} else if (buf[i] == '-') {
|
||
expsign = -1;
|
||
i += 1;
|
||
};
|
||
if (i >= s.len || !ascii.isdigit(buf[i]: rune)) {
|
||
return i: invalid;
|
||
};
|
||
let e: int = 0;
|
||
for (i < s.len && ascii.isdigit(buf[i]: rune)) {
|
||
if (e < 10000) {
|
||
e = e * 10 + ((buf[i] - '0'): int);
|
||
};
|
||
i += 1;
|
||
};
|
||
dp += e * expsign;
|
||
} else if (b == base.HEX) {
|
||
return i: invalid; // hex floats must have an exponent
|
||
};
|
||
if (i != s.len) { return i: invalid; };
|
||
if (mant != 0u64) {
|
||
exp = (dp - ndmant): i32;
|
||
};
|
||
return fast_parsed_float {
|
||
mantissa = mant,
|
||
exponent = exp,
|
||
negative = neg,
|
||
truncated = trunc,
|
||
};
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:115. Fills the slow-path decimal `d`.
|
||
fn decimal_parse(d: *decimal, s: str) (void | invalid) = {
|
||
let i: i32 = 0;
|
||
let buf: []u8 = strings.toutf8(s);
|
||
d.negative = false;
|
||
d.truncated = false;
|
||
if (buf[0] == '+') {
|
||
i += 1;
|
||
} else if (buf[0] == '-') {
|
||
d.negative = true;
|
||
i += 1;
|
||
};
|
||
let sawdot: bool = false;
|
||
let sawdigits: bool = false;
|
||
for (i < s.len) {
|
||
if (buf[i] == '.') {
|
||
if (sawdot) { return i: invalid; };
|
||
sawdot = true;
|
||
d.dp = (d.nd: i32);
|
||
} else if (ascii.isdigit(buf[i]: rune)) {
|
||
sawdigits = true;
|
||
if (buf[i] == '0' && d.nd == (0u64: size)) {
|
||
d.dp -= 1;
|
||
i += 1;
|
||
continue;
|
||
};
|
||
if (d.nd < (len(d.digits): size)) {
|
||
d.digits[d.nd] = buf[i] - '0';
|
||
d.nd += (1u64: size);
|
||
} else if (buf[i] != '0') {
|
||
d.truncated = true;
|
||
};
|
||
} else {
|
||
break;
|
||
};
|
||
i += 1;
|
||
};
|
||
if (!sawdigits) { return i: invalid; };
|
||
if (!sawdot) {
|
||
d.dp = (d.nd: i32);
|
||
};
|
||
if (i < s.len && (buf[i] == 'e' || buf[i] == 'E')) {
|
||
i += 1;
|
||
if (i >= s.len) { return i: invalid; };
|
||
let expsign: int = 1;
|
||
if (buf[i] == '+') {
|
||
i += 1;
|
||
} else if (buf[i] == '-') {
|
||
expsign = -1;
|
||
i += 1;
|
||
};
|
||
if (i >= s.len || !ascii.isdigit(buf[i]: rune)) {
|
||
return i: invalid;
|
||
};
|
||
let e: int = 0;
|
||
for (i < s.len && ascii.isdigit(buf[i]: rune)) {
|
||
if (e < 10000) {
|
||
e = e * 10 + ((buf[i] - '0'): int);
|
||
};
|
||
i += 1;
|
||
};
|
||
d.dp += (e * expsign): i32;
|
||
};
|
||
if (i != s.len) { return i: invalid; };
|
||
return;
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:173. Count of leading zero bits in n>0.
|
||
fn leading_zeroes(n: u64) uint = {
|
||
os.assert(n > 0u64, "strconv.leading_zeroes: n == 0");
|
||
let b: u64 = 0u64;
|
||
if ((n & 0xFFFFFFFF00000000u64) > 0u64) {
|
||
n >>= 32u64;
|
||
b |= 32u64;
|
||
};
|
||
if ((n & 0xFFFF0000u64) > 0u64) {
|
||
n >>= 16u64;
|
||
b |= 16u64;
|
||
};
|
||
if ((n & 0xFF00u64) > 0u64) {
|
||
n >>= 8u64;
|
||
b |= 8u64;
|
||
};
|
||
if ((n & 0xF0u64) > 0u64) {
|
||
n >>= 4u64;
|
||
b |= 4u64;
|
||
};
|
||
if ((n & 0xCu64) > 0u64) {
|
||
n >>= 2u64;
|
||
b |= 2u64;
|
||
};
|
||
if ((n & 0x2u64) > 0u64) {
|
||
n >>= 1u64;
|
||
b |= 1u64;
|
||
};
|
||
return ((63u64 - b): uint);
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:203. Eisel-Lemire fast path: a correctly-
|
||
// rounded f64/f32 from (mantissa, exp10) when the 128-bit product is
|
||
// unambiguous, else void → caller falls to the decimal slow path.
|
||
// Divergences at-site: `mantissa <<= clz` (scalar-param mutate) → local
|
||
// `mnt`; whole-struct local reassign `x = merged` copies only the first
|
||
// word in cgen → per-field `x.hi = …; x.lo = …` (#155); `po10 =
|
||
// powers_of_ten[i]` row-bind → direct double-index (#155, A2); bitwise-
|
||
// vs-compare fully parenthesised; comma let-bindings split.
|
||
fn eisel_lemire(
|
||
mantissa: u64,
|
||
exp10: i32,
|
||
neg: bool,
|
||
f: *math.floatinfo,
|
||
) (u64 | void) = {
|
||
if (mantissa == 0u64 || exp10 > 288 || exp10 < -307) {
|
||
return;
|
||
};
|
||
let idx: i32 = exp10 + 307;
|
||
let clz: uint = leading_zeroes(mantissa);
|
||
let mnt: u64 = mantissa << (clz: u64);
|
||
let shift: u64 = 64u64 - f.mantbits - 3u64;
|
||
let mask: u64 = (1u64 << shift) - 1u64;
|
||
// log(10)/log(2) ≈ 217706 / 65536; x / 65536 = x >> 16.
|
||
let exp: int = (217706 * (exp10: int)) >> 16;
|
||
let e2: u64 = ((exp + f.expbias + 64): u64) - (clz: u64);
|
||
let x: r128 = u128mul(mnt, powers_of_ten[idx][1]);
|
||
if ((x.hi & mask) == mask && (x.lo + mnt) < mnt) {
|
||
let y: r128 = u128mul(mnt, powers_of_ten[idx][0]);
|
||
let merged: r128 = r128 { hi = x.hi, lo = x.lo + y.hi };
|
||
if (merged.lo < x.lo) {
|
||
// local-struct-field compound-assign drops the load in
|
||
// wwstage (sets =1, not +=1) — explicit form, byte-id.
|
||
merged.hi = merged.hi + 1u64;
|
||
};
|
||
if ((merged.hi & mask) == mask && (merged.lo + 1u64) == 0u64 &&
|
||
(y.lo + mnt) < mnt) {
|
||
return;
|
||
};
|
||
x.hi = merged.hi;
|
||
x.lo = merged.lo;
|
||
};
|
||
let msb: u64 = x.hi >> 63u64;
|
||
let mant: u64 = x.hi >> (msb + shift);
|
||
e2 -= 1u64 ^ msb;
|
||
if (x.lo == 0u64 && (x.hi & mask) == 0u64 && (mant & 3u64) == 1u64) {
|
||
return;
|
||
};
|
||
mant += mant & 1u64;
|
||
mant >>= 1u64;
|
||
if ((mant >> (f.mantbits + 1u64)) > 0u64) {
|
||
mant >>= 1u64;
|
||
e2 += 1u64;
|
||
};
|
||
if (e2 <= 0u64 || e2 >= (1u64 << f.expbits) - 1u64) {
|
||
return;
|
||
};
|
||
return mkfloat(mant, (e2: uint), neg, f);
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:247. Slow-path: decimal `d` → IEEE bits.
|
||
fn floatbits(d: *decimal, f: *math.floatinfo) (u64 | overflow) = {
|
||
let e: int = 0;
|
||
let m: u64 = 0u64;
|
||
let powtab: [19]i8 = [
|
||
0i8, 3i8, 6i8, 9i8, 13i8, 16i8, 19i8, 23i8, 26i8, 29i8,
|
||
33i8, 36i8, 39i8, 43i8, 46i8, 49i8, 53i8, 56i8, 59i8,
|
||
];
|
||
if (d.nd == (0u64: size) || d.dp < -326) {
|
||
if (d.negative) {
|
||
return mkfloat(0u64, (0u32: uint), d.negative, f);
|
||
};
|
||
return 0u64;
|
||
} else if (d.dp > 310) {
|
||
return overflow{};
|
||
};
|
||
if (d.nd <= (19u64: size)) {
|
||
let dmant: u64 = 0u64;
|
||
let i: size = (0u64: size);
|
||
for (i < d.nd) {
|
||
dmant = 10u64 * dmant + (d.digits[i]: u64);
|
||
i += (1u64: size);
|
||
};
|
||
let exp10: i32 = d.dp - (d.nd: i32);
|
||
match (eisel_lemire(dmant, exp10, d.negative, f)) {
|
||
case let r: u64 => { return r; };
|
||
case void => void;
|
||
};
|
||
};
|
||
for (d.dp > 0) {
|
||
let n: int = 0;
|
||
if ((d.dp: uint) >= (len(powtab): uint)) {
|
||
n = (maxshift: int);
|
||
} else {
|
||
n = (powtab[d.dp]: int);
|
||
};
|
||
decimal_shift(d, -n);
|
||
e += n;
|
||
};
|
||
for (d.dp <= 0) {
|
||
let n: int = 0;
|
||
if (d.dp == 0) {
|
||
if (d.digits[0] >= 5u8) { break; };
|
||
if (d.digits[0] < 2u8) { n = 2; } else { n = 1; };
|
||
} else if ((-d.dp) >= (len(powtab): i32)) {
|
||
n = (maxshift: int);
|
||
} else {
|
||
n = (powtab[-d.dp]: int);
|
||
};
|
||
decimal_shift(d, n);
|
||
e -= n;
|
||
};
|
||
e -= 1;
|
||
if (e <= -f.expbias + 1) {
|
||
let nn: int = -f.expbias - e + 1;
|
||
decimal_shift(d, -nn);
|
||
e += nn;
|
||
};
|
||
if (e + f.expbias >= ((1u64 << f.expbits): int) - 1) {
|
||
return overflow{};
|
||
};
|
||
decimal_shift(d, (f.mantbits: int) + 1);
|
||
m = decimal_round(d);
|
||
if (m == (2u64 << f.mantbits)) {
|
||
m >>= 1u64;
|
||
e += 1;
|
||
if (e + f.expbias >= ((1u64 << f.expbits): int) - 1) {
|
||
return overflow{};
|
||
};
|
||
};
|
||
if ((m & (1u64 << f.mantbits)) == 0u64) {
|
||
e = -f.expbias;
|
||
};
|
||
return mkfloat(m, ((e + f.expbias): uint), d.negative, f);
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:311. Assemble sign|exp|mantissa.
|
||
fn mkfloat(m: u64, e: uint, negative: bool, f: *math.floatinfo) u64 = {
|
||
let n: u64 = m & ((1u64 << f.mantbits) - 1u64);
|
||
n |= ((e: u64) & ((1u64 << f.expbits) - 1u64)) << f.mantbits;
|
||
if (negative) {
|
||
n |= 1u64 << (f.mantbits + f.expbits);
|
||
};
|
||
return n;
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:320. Exact f64 powers of ten 1e0..1e22 (all
|
||
// exactly representable; see stof64exact).
|
||
let f64pow10: [23]f64 = [
|
||
1.0e0, 1.0e1, 1.0e2, 1.0e3, 1.0e4, 1.0e5, 1.0e6, 1.0e7, 1.0e8, 1.0e9,
|
||
1.0e10, 1.0e11, 1.0e12, 1.0e13, 1.0e14, 1.0e15, 1.0e16, 1.0e17, 1.0e18,
|
||
1.0e19, 1.0e20, 1.0e21, 1.0e22,
|
||
];
|
||
|
||
// ref/hare/strconv/stof.ha:326.
|
||
fn stof64exact(mant: u64, exp: i32, neg: bool) (f64 | void) = {
|
||
if (mant >> math.F64_MANTISSA_BITS != 0u64) { return; };
|
||
let n: f64 = (mant: i64): f64;
|
||
if (neg) {
|
||
n = -n;
|
||
};
|
||
if (exp == 0i32) {
|
||
return n;
|
||
};
|
||
if (-22i32 <= exp && exp <= 22i32) {
|
||
if (exp >= 0i32) {
|
||
// f64 compound-assign mis-lowers in cgen — explicit
|
||
// form (strconv.ww f64tos precedent).
|
||
n = n * f64pow10[exp];
|
||
} else {
|
||
n = n / f64pow10[-exp];
|
||
};
|
||
} else {
|
||
return;
|
||
};
|
||
return n;
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:345. Exact f32 powers of ten 1e0..1e10.
|
||
let f32pow10: [11]f32 = [
|
||
1.0e0f32, 1.0e1f32, 1.0e2f32, 1.0e3f32, 1.0e4f32, 1.0e5f32, 1.0e6f32,
|
||
1.0e7f32, 1.0e8f32, 1.0e9f32, 1.0e10f32,
|
||
];
|
||
|
||
// ref/hare/strconv/stof.ha:349.
|
||
fn stof32exact(mant: u64, exp: i32, neg: bool) (f32 | void) = {
|
||
if (mant >> (math.F32_MANTISSA_BITS: u64) != 0u64) { return; };
|
||
let n: f32 = (mant: i32): f32;
|
||
if (neg) {
|
||
n = -n;
|
||
};
|
||
if (exp == 0i32) {
|
||
return n;
|
||
};
|
||
if (-10i32 <= exp && exp <= 10i32) {
|
||
if (exp >= 0i32) {
|
||
// f32 compound-assign mis-lowers in cgen — explicit form.
|
||
n = n * f32pow10[exp];
|
||
} else {
|
||
n = n / (f64pow10[-exp]: f32);
|
||
};
|
||
} else {
|
||
return;
|
||
};
|
||
return n;
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:369. Adapted from Go's atofHex. The by-value
|
||
// `p` is mutated in Hare; ww copies its fields to scalar locals (struct
|
||
// param field-write miscompiles + diverges — filed).
|
||
fn hex_to_bits(p: fast_parsed_float, info: *math.floatinfo) (u64 | overflow) = {
|
||
let pmant: u64 = p.mantissa;
|
||
let pexp: i32 = p.exponent;
|
||
let pneg: bool = p.negative;
|
||
let ptrunc: bool = p.truncated;
|
||
let max_exp: int = ((1u64 << info.expbits): int) - info.expbias - 2;
|
||
let min_exp: int = -info.expbias + 1;
|
||
pexp += (info.mantbits: i32);
|
||
|
||
// Shift left until a leading 1 bit followed by mantbits + 2 rounding.
|
||
for (pmant != 0u64 && pmant >> (info.mantbits + 2u64) == 0u64) {
|
||
pmant <<= 1u64;
|
||
pexp -= 1;
|
||
};
|
||
if (ptrunc) {
|
||
pmant |= 1u64;
|
||
};
|
||
// Too many bits: shift right (sticky-or the dropped bit).
|
||
for (pmant >> (3u64 + info.mantbits) != 0u64) {
|
||
pmant = (pmant >> 1u64) | (pmant & 1u64);
|
||
pexp += 1;
|
||
};
|
||
// Denormalise if the exponent is small.
|
||
for (pmant > 1u64 && pexp < (min_exp: i32) - 2) {
|
||
pmant = (pmant >> 1u64) | (pmant & 1u64);
|
||
pexp += 1;
|
||
};
|
||
// Round to even.
|
||
let round: u64 = pmant & 3u64;
|
||
pmant >>= 2u64;
|
||
round |= pmant & 1u64;
|
||
pexp += 2;
|
||
if (round == 3u64) {
|
||
pmant += 1u64;
|
||
if (pmant == 1u64 << (1u64 + info.mantbits)) {
|
||
pmant >>= 1u64;
|
||
pexp += 1;
|
||
};
|
||
};
|
||
// Denormal or zero.
|
||
if (pmant >> info.mantbits == 0u64) {
|
||
pexp = (-info.expbias): i32;
|
||
};
|
||
if (pexp > (max_exp: i32)) {
|
||
return overflow{};
|
||
};
|
||
let bits: u64 = pmant & info.mantmask;
|
||
bits |= (((pexp + (info.expbias: i32)): u64) & info.expmask) << info.mantbits;
|
||
if (pneg) {
|
||
bits |= 1u64 << (info.mantbits + info.expbits);
|
||
};
|
||
return bits;
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:425. "nan"/"infinity"/±"infinity",
|
||
// case-insensitive. ww math has no f32 NAN/INF consts → f32frombits of
|
||
// the IEEE-754 f32 bit patterns (qNaN 0x7FC00000, ±Inf 0x7F800000 /
|
||
// 0xFF800000).
|
||
fn special(s: str) (f32 | void) = {
|
||
if (ascii.strcasecmp(s, "nan") == 0) {
|
||
return math.f32frombits(0x7FC00000u32);
|
||
} else if (ascii.strcasecmp(s, "infinity") == 0) {
|
||
return math.f32frombits(0x7F800000u32);
|
||
} else if (ascii.strcasecmp(s, "+infinity") == 0) {
|
||
return math.f32frombits(0x7F800000u32);
|
||
} else if (ascii.strcasecmp(s, "-infinity") == 0) {
|
||
return math.f32frombits(0xFF800000u32);
|
||
};
|
||
return;
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:445. Parse `s` as f64 (base DEC or HEX). See
|
||
// the module note: the EL fast path is HELD; the decimal fallback gives
|
||
// correct results meanwhile.
|
||
export fn stof64(s: str, b: base) (f64 | invalid | overflow) = {
|
||
let bb: base = b;
|
||
if (bb == base.DEFAULT) {
|
||
bb = base.DEC;
|
||
} else if (bb == base.HEX_LOWER) {
|
||
bb = base.HEX;
|
||
};
|
||
os.assert(bb == base.DEC || bb == base.HEX,
|
||
"strconv.stof64: base must be DEC or HEX");
|
||
|
||
if (s.len == 0) {
|
||
return 0: invalid;
|
||
};
|
||
|
||
match (special(s)) {
|
||
case let f: f32 => { return (f: f64); };
|
||
case void => void;
|
||
};
|
||
|
||
match (fast_parse(s, bb)) {
|
||
case let p: fast_parsed_float => {
|
||
if (bb == base.HEX) {
|
||
match (hex_to_bits(p, &math.f64info)) {
|
||
case let bits: u64 => { return math.f64frombits(bits); };
|
||
case let eo: overflow => { return eo; };
|
||
};
|
||
} else if (!p.truncated) {
|
||
match (stof64exact(p.mantissa, p.exponent, p.negative)) {
|
||
case let n: f64 => { return n; };
|
||
case void => void;
|
||
};
|
||
match (eisel_lemire(p.mantissa, p.exponent, p.negative,
|
||
&math.f64info)) {
|
||
case let n: u64 => { return math.f64frombits(n); };
|
||
case void => void;
|
||
};
|
||
};
|
||
let d = decimal { ... };
|
||
match (decimal_parse(&d, s)) {
|
||
case let ei: invalid => { return ei; };
|
||
case void => void;
|
||
};
|
||
match (floatbits(&d, &math.f64info)) {
|
||
case let n: u64 => { return math.f64frombits(n); };
|
||
case let eo: overflow => { return eo; };
|
||
};
|
||
};
|
||
case let ei: invalid => { return ei; };
|
||
};
|
||
return 0: invalid; // unreachable (path-cov)
|
||
};
|
||
|
||
// ref/hare/strconv/stof.ha:491. Parse `s` as f32 (base DEC or HEX).
|
||
export fn stof32(s: str, b: base) (f32 | invalid | overflow) = {
|
||
let bb: base = b;
|
||
if (bb == base.DEFAULT) {
|
||
bb = base.DEC;
|
||
} else if (bb == base.HEX_LOWER) {
|
||
bb = base.HEX;
|
||
};
|
||
os.assert(bb == base.DEC || bb == base.HEX,
|
||
"strconv.stof32: base must be DEC or HEX");
|
||
|
||
if (s.len == 0) {
|
||
return 0: invalid;
|
||
};
|
||
|
||
match (special(s)) {
|
||
case let f: f32 => { return f; };
|
||
case void => void;
|
||
};
|
||
|
||
match (fast_parse(s, bb)) {
|
||
case let p: fast_parsed_float => {
|
||
if (bb == base.HEX) {
|
||
match (hex_to_bits(p, &math.f32info)) {
|
||
case let bits: u64 => {
|
||
return math.f32frombits(bits: u32);
|
||
};
|
||
case let eo: overflow => { return eo; };
|
||
};
|
||
} else if (!p.truncated) {
|
||
match (stof32exact(p.mantissa, p.exponent, p.negative)) {
|
||
case let n: f32 => { return n; };
|
||
case void => void;
|
||
};
|
||
match (eisel_lemire(p.mantissa, p.exponent, p.negative,
|
||
&math.f32info)) {
|
||
case let n: u64 => { return math.f32frombits(n: u32); };
|
||
case void => void;
|
||
};
|
||
};
|
||
let d = decimal { ... };
|
||
match (decimal_parse(&d, s)) {
|
||
case let ei: invalid => { return ei; };
|
||
case void => void;
|
||
};
|
||
match (floatbits(&d, &math.f32info)) {
|
||
case let n: u64 => { return math.f32frombits(n: u32); };
|
||
case let eo: overflow => { return eo; };
|
||
};
|
||
};
|
||
case let ei: invalid => { return ei; };
|
||
};
|
||
return 0: invalid; // unreachable (path-cov)
|
||
};
|
||
|
||
// strconv — stof/ftos lookup tables. Mirrors ref/hare/strconv/stof_data.ha
|
||
// byte-exact. Pure-data fold (strconv #106 fold-2, was fold-3 before drew
|
||
// re-sequenced 2026-05-26): no logic, exercised transitively when fold-3's
|
||
// `leftshift_newdigits` lands (ref/hare/strconv/decimal.ha:35).
|
||
//
|
||
// ww uses module-level `let` for compile-time array data (ref/hare/strconv
|
||
// `const` has no ww keyword equivalent; lib/encoding/utf8/utf8.ww:48 sets
|
||
// the precedent with [2048]i8 dfa). Literal suffixes (`u16`, `u8`) are
|
||
// required because cstage rejects bare integer literals in `[N]u8`/`[N]u16`
|
||
// init while wwstage accepts them; the suffixed form is the only shape
|
||
// both stages agree on (candidate #130).
|
||
//
|
||
// `powers_of_ten: [596][2]u64` (ref/hare/strconv/stof_data.ha:73) is the
|
||
// Eisel-Lemire fast-path table (consumed by stof.ww's eisel_lemire); it
|
||
// lands here in fold-4 alongside its consumer, indexed `[exp10 + 307]`
|
||
// for exp10 in [-307, 288]. Faithful 2D `[596][2]u64` (the {hi,lo} pair
|
||
// IS the 128-bit truncated power-of-ten; rule-12, not flattened) — the
|
||
// 2D module-level static-init + double-index read it needs landed in
|
||
// #156 (cbeffea). See the table at the foot of this file.
|
||
|
||
package strconv;
|
||
|
||
// ref/hare/strconv/stof_data.ha:4. Powers-of-five decimal-expansion
|
||
// metadata for `leftshift_newdigits` (decimal.ha:35). The top 5 bits
|
||
// of each entry are the new-digit-count `nn`; the low 11 bits index
|
||
// `pow5_table` for the digits themselves.
|
||
let left_shift_table: [65]u16 = [
|
||
0x0000u16, 0x0800u16, 0x0801u16, 0x0803u16, 0x1006u16, 0x1009u16, 0x100Du16, 0x1812u16, 0x1817u16,
|
||
0x181Du16, 0x2024u16, 0x202Bu16, 0x2033u16, 0x203Cu16, 0x2846u16, 0x2850u16, 0x285Bu16, 0x3067u16,
|
||
0x3073u16, 0x3080u16, 0x388Eu16, 0x389Cu16, 0x38ABu16, 0x38BBu16, 0x40CCu16, 0x40DDu16, 0x40EFu16,
|
||
0x4902u16, 0x4915u16, 0x4929u16, 0x513Eu16, 0x5153u16, 0x5169u16, 0x5180u16, 0x5998u16, 0x59B0u16,
|
||
0x59C9u16, 0x61E3u16, 0x61FDu16, 0x6218u16, 0x6A34u16, 0x6A50u16, 0x6A6Du16, 0x6A8Bu16, 0x72AAu16,
|
||
0x72C9u16, 0x72E9u16, 0x7B0Au16, 0x7B2Bu16, 0x7B4Du16, 0x8370u16, 0x8393u16, 0x83B7u16, 0x83DCu16,
|
||
0x8C02u16, 0x8C28u16, 0x8C4Fu16, 0x9477u16, 0x949Fu16, 0x94C8u16, 0x9CF2u16, 0x051Cu16, 0x051Cu16,
|
||
0x051Cu16, 0x051Cu16,
|
||
];
|
||
|
||
// ref/hare/strconv/stof_data.ha:15. Decimal digits of 5^k for k=1..60,
|
||
// concatenated. Indexed via `left_shift_table` (above); each shift k
|
||
// reads its `pow5_b - pow5_a` digits starting at `pow5_a`.
|
||
let pow5_table: [0x051C]u8 = [
|
||
5u8, 2u8, 5u8, 1u8, 2u8, 5u8, 6u8, 2u8, 5u8, 3u8, 1u8, 2u8, 5u8, 1u8, 5u8, 6u8, 2u8, 5u8, 7u8, 8u8, 1u8, 2u8, 5u8, 3u8,
|
||
9u8, 0u8, 6u8, 2u8, 5u8, 1u8, 9u8, 5u8, 3u8, 1u8, 2u8, 5u8, 9u8, 7u8, 6u8, 5u8, 6u8, 2u8, 5u8, 4u8, 8u8, 8u8, 2u8, 8u8,
|
||
1u8, 2u8, 5u8, 2u8, 4u8, 4u8, 1u8, 4u8, 0u8, 6u8, 2u8, 5u8, 1u8, 2u8, 2u8, 0u8, 7u8, 0u8, 3u8, 1u8, 2u8, 5u8, 6u8, 1u8,
|
||
0u8, 3u8, 5u8, 1u8, 5u8, 6u8, 2u8, 5u8, 3u8, 0u8, 5u8, 1u8, 7u8, 5u8, 7u8, 8u8, 1u8, 2u8, 5u8, 1u8, 5u8, 2u8, 5u8, 8u8,
|
||
7u8, 8u8, 9u8, 0u8, 6u8, 2u8, 5u8, 7u8, 6u8, 2u8, 9u8, 3u8, 9u8, 4u8, 5u8, 3u8, 1u8, 2u8, 5u8, 3u8, 8u8, 1u8, 4u8, 6u8,
|
||
9u8, 7u8, 2u8, 6u8, 5u8, 6u8, 2u8, 5u8, 1u8, 9u8, 0u8, 7u8, 3u8, 4u8, 8u8, 6u8, 3u8, 2u8, 8u8, 1u8, 2u8, 5u8, 9u8, 5u8,
|
||
3u8, 6u8, 7u8, 4u8, 3u8, 1u8, 6u8, 4u8, 0u8, 6u8, 2u8, 5u8, 4u8, 7u8, 6u8, 8u8, 3u8, 7u8, 1u8, 5u8, 8u8, 2u8, 0u8, 3u8,
|
||
1u8, 2u8, 5u8, 2u8, 3u8, 8u8, 4u8, 1u8, 8u8, 5u8, 7u8, 9u8, 1u8, 0u8, 1u8, 5u8, 6u8, 2u8, 5u8, 1u8, 1u8, 9u8, 2u8, 0u8,
|
||
9u8, 2u8, 8u8, 9u8, 5u8, 5u8, 0u8, 7u8, 8u8, 1u8, 2u8, 5u8, 5u8, 9u8, 6u8, 0u8, 4u8, 6u8, 4u8, 4u8, 7u8, 7u8, 5u8, 3u8,
|
||
9u8, 0u8, 6u8, 2u8, 5u8, 2u8, 9u8, 8u8, 0u8, 2u8, 3u8, 2u8, 2u8, 3u8, 8u8, 7u8, 6u8, 9u8, 5u8, 3u8, 1u8, 2u8, 5u8, 1u8,
|
||
4u8, 9u8, 0u8, 1u8, 1u8, 6u8, 1u8, 1u8, 9u8, 3u8, 8u8, 4u8, 7u8, 6u8, 5u8, 6u8, 2u8, 5u8, 7u8, 4u8, 5u8, 0u8, 5u8, 8u8,
|
||
0u8, 5u8, 9u8, 6u8, 9u8, 2u8, 3u8, 8u8, 2u8, 8u8, 1u8, 2u8, 5u8, 3u8, 7u8, 2u8, 5u8, 2u8, 9u8, 0u8, 2u8, 9u8, 8u8, 4u8,
|
||
6u8, 1u8, 9u8, 1u8, 4u8, 0u8, 6u8, 2u8, 5u8, 1u8, 8u8, 6u8, 2u8, 6u8, 4u8, 5u8, 1u8, 4u8, 9u8, 2u8, 3u8, 0u8, 9u8, 5u8,
|
||
7u8, 0u8, 3u8, 1u8, 2u8, 5u8, 9u8, 3u8, 1u8, 3u8, 2u8, 2u8, 5u8, 7u8, 4u8, 6u8, 1u8, 5u8, 4u8, 7u8, 8u8, 5u8, 1u8, 5u8,
|
||
6u8, 2u8, 5u8, 4u8, 6u8, 5u8, 6u8, 6u8, 1u8, 2u8, 8u8, 7u8, 3u8, 0u8, 7u8, 7u8, 3u8, 9u8, 2u8, 5u8, 7u8, 8u8, 1u8, 2u8,
|
||
5u8, 2u8, 3u8, 2u8, 8u8, 3u8, 0u8, 6u8, 4u8, 3u8, 6u8, 5u8, 3u8, 8u8, 6u8, 9u8, 6u8, 2u8, 8u8, 9u8, 0u8, 6u8, 2u8, 5u8,
|
||
1u8, 1u8, 6u8, 4u8, 1u8, 5u8, 3u8, 2u8, 1u8, 8u8, 2u8, 6u8, 9u8, 3u8, 4u8, 8u8, 1u8, 4u8, 4u8, 5u8, 3u8, 1u8, 2u8, 5u8,
|
||
5u8, 8u8, 2u8, 0u8, 7u8, 6u8, 6u8, 0u8, 9u8, 1u8, 3u8, 4u8, 6u8, 7u8, 4u8, 0u8, 7u8, 2u8, 2u8, 6u8, 5u8, 6u8, 2u8, 5u8,
|
||
2u8, 9u8, 1u8, 0u8, 3u8, 8u8, 3u8, 0u8, 4u8, 5u8, 6u8, 7u8, 3u8, 3u8, 7u8, 0u8, 3u8, 6u8, 1u8, 3u8, 2u8, 8u8, 1u8, 2u8,
|
||
5u8, 1u8, 4u8, 5u8, 5u8, 1u8, 9u8, 1u8, 5u8, 2u8, 2u8, 8u8, 3u8, 6u8, 6u8, 8u8, 5u8, 1u8, 8u8, 0u8, 6u8, 6u8, 4u8, 0u8,
|
||
6u8, 2u8, 5u8, 7u8, 2u8, 7u8, 5u8, 9u8, 5u8, 7u8, 6u8, 1u8, 4u8, 1u8, 8u8, 3u8, 4u8, 2u8, 5u8, 9u8, 0u8, 3u8, 3u8, 2u8,
|
||
0u8, 3u8, 1u8, 2u8, 5u8, 3u8, 6u8, 3u8, 7u8, 9u8, 7u8, 8u8, 8u8, 0u8, 7u8, 0u8, 9u8, 1u8, 7u8, 1u8, 2u8, 9u8, 5u8, 1u8,
|
||
6u8, 6u8, 0u8, 1u8, 5u8, 6u8, 2u8, 5u8, 1u8, 8u8, 1u8, 8u8, 9u8, 8u8, 9u8, 4u8, 0u8, 3u8, 5u8, 4u8, 5u8, 8u8, 5u8, 6u8,
|
||
4u8, 7u8, 5u8, 8u8, 3u8, 0u8, 0u8, 7u8, 8u8, 1u8, 2u8, 5u8, 9u8, 0u8, 9u8, 4u8, 9u8, 4u8, 7u8, 0u8, 1u8, 7u8, 7u8, 2u8,
|
||
9u8, 2u8, 8u8, 2u8, 3u8, 7u8, 9u8, 1u8, 5u8, 0u8, 3u8, 9u8, 0u8, 6u8, 2u8, 5u8, 4u8, 5u8, 4u8, 7u8, 4u8, 7u8, 3u8, 5u8,
|
||
0u8, 8u8, 8u8, 6u8, 4u8, 6u8, 4u8, 1u8, 1u8, 8u8, 9u8, 5u8, 7u8, 5u8, 1u8, 9u8, 5u8, 3u8, 1u8, 2u8, 5u8, 2u8, 2u8, 7u8,
|
||
3u8, 7u8, 3u8, 6u8, 7u8, 5u8, 4u8, 4u8, 3u8, 2u8, 3u8, 2u8, 0u8, 5u8, 9u8, 4u8, 7u8, 8u8, 7u8, 5u8, 9u8, 7u8, 6u8, 5u8,
|
||
6u8, 2u8, 5u8, 1u8, 1u8, 3u8, 6u8, 8u8, 6u8, 8u8, 3u8, 7u8, 7u8, 2u8, 1u8, 6u8, 1u8, 6u8, 0u8, 2u8, 9u8, 7u8, 3u8, 9u8,
|
||
3u8, 7u8, 9u8, 8u8, 8u8, 2u8, 8u8, 1u8, 2u8, 5u8, 5u8, 6u8, 8u8, 4u8, 3u8, 4u8, 1u8, 8u8, 8u8, 6u8, 0u8, 8u8, 0u8, 8u8,
|
||
0u8, 1u8, 4u8, 8u8, 6u8, 9u8, 6u8, 8u8, 9u8, 9u8, 4u8, 1u8, 4u8, 0u8, 6u8, 2u8, 5u8, 2u8, 8u8, 4u8, 2u8, 1u8, 7u8, 0u8,
|
||
9u8, 4u8, 3u8, 0u8, 4u8, 0u8, 4u8, 0u8, 0u8, 7u8, 4u8, 3u8, 4u8, 8u8, 4u8, 4u8, 9u8, 7u8, 0u8, 7u8, 0u8, 3u8, 1u8, 2u8,
|
||
5u8, 1u8, 4u8, 2u8, 1u8, 0u8, 8u8, 5u8, 4u8, 7u8, 1u8, 5u8, 2u8, 0u8, 2u8, 0u8, 0u8, 3u8, 7u8, 1u8, 7u8, 4u8, 2u8, 2u8,
|
||
4u8, 8u8, 5u8, 3u8, 5u8, 1u8, 5u8, 6u8, 2u8, 5u8, 7u8, 1u8, 0u8, 5u8, 4u8, 2u8, 7u8, 3u8, 5u8, 7u8, 6u8, 0u8, 1u8, 0u8,
|
||
0u8, 1u8, 8u8, 5u8, 8u8, 7u8, 1u8, 1u8, 2u8, 4u8, 2u8, 6u8, 7u8, 5u8, 7u8, 8u8, 1u8, 2u8, 5u8, 3u8, 5u8, 5u8, 2u8, 7u8,
|
||
1u8, 3u8, 6u8, 7u8, 8u8, 8u8, 0u8, 0u8, 5u8, 0u8, 0u8, 9u8, 2u8, 9u8, 3u8, 5u8, 5u8, 6u8, 2u8, 1u8, 3u8, 3u8, 7u8, 8u8,
|
||
9u8, 0u8, 6u8, 2u8, 5u8, 1u8, 7u8, 7u8, 6u8, 3u8, 5u8, 6u8, 8u8, 3u8, 9u8, 4u8, 0u8, 0u8, 2u8, 5u8, 0u8, 4u8, 6u8, 4u8,
|
||
6u8, 7u8, 7u8, 8u8, 1u8, 0u8, 6u8, 6u8, 8u8, 9u8, 4u8, 5u8, 3u8, 1u8, 2u8, 5u8, 8u8, 8u8, 8u8, 1u8, 7u8, 8u8, 4u8, 1u8,
|
||
9u8, 7u8, 0u8, 0u8, 1u8, 2u8, 5u8, 2u8, 3u8, 2u8, 3u8, 3u8, 8u8, 9u8, 0u8, 5u8, 3u8, 3u8, 4u8, 4u8, 7u8, 2u8, 6u8, 5u8,
|
||
6u8, 2u8, 5u8, 4u8, 4u8, 4u8, 0u8, 8u8, 9u8, 2u8, 0u8, 9u8, 8u8, 5u8, 0u8, 0u8, 6u8, 2u8, 6u8, 1u8, 6u8, 1u8, 6u8, 9u8,
|
||
4u8, 5u8, 2u8, 6u8, 6u8, 7u8, 2u8, 3u8, 6u8, 3u8, 2u8, 8u8, 1u8, 2u8, 5u8, 2u8, 2u8, 2u8, 0u8, 4u8, 4u8, 6u8, 0u8, 4u8,
|
||
9u8, 2u8, 5u8, 0u8, 3u8, 1u8, 3u8, 0u8, 8u8, 0u8, 8u8, 4u8, 7u8, 2u8, 6u8, 3u8, 3u8, 3u8, 6u8, 1u8, 8u8, 1u8, 6u8, 4u8,
|
||
0u8, 6u8, 2u8, 5u8, 1u8, 1u8, 1u8, 0u8, 2u8, 2u8, 3u8, 0u8, 2u8, 4u8, 6u8, 2u8, 5u8, 1u8, 5u8, 6u8, 5u8, 4u8, 0u8, 4u8,
|
||
2u8, 3u8, 6u8, 3u8, 1u8, 6u8, 6u8, 8u8, 0u8, 9u8, 0u8, 8u8, 2u8, 0u8, 3u8, 1u8, 2u8, 5u8, 5u8, 5u8, 5u8, 1u8, 1u8, 1u8,
|
||
5u8, 1u8, 2u8, 3u8, 1u8, 2u8, 5u8, 7u8, 8u8, 2u8, 7u8, 0u8, 2u8, 1u8, 1u8, 8u8, 1u8, 5u8, 8u8, 3u8, 4u8, 0u8, 4u8, 5u8,
|
||
4u8, 1u8, 0u8, 1u8, 5u8, 6u8, 2u8, 5u8, 2u8, 7u8, 7u8, 5u8, 5u8, 5u8, 7u8, 5u8, 6u8, 1u8, 5u8, 6u8, 2u8, 8u8, 9u8, 1u8,
|
||
3u8, 5u8, 1u8, 0u8, 5u8, 9u8, 0u8, 7u8, 9u8, 1u8, 7u8, 0u8, 2u8, 2u8, 7u8, 0u8, 5u8, 0u8, 7u8, 8u8, 1u8, 2u8, 5u8, 1u8,
|
||
3u8, 8u8, 7u8, 7u8, 7u8, 8u8, 7u8, 8u8, 0u8, 7u8, 8u8, 1u8, 4u8, 4u8, 5u8, 6u8, 7u8, 5u8, 5u8, 2u8, 9u8, 5u8, 3u8, 9u8,
|
||
5u8, 8u8, 5u8, 1u8, 1u8, 3u8, 5u8, 2u8, 5u8, 3u8, 9u8, 0u8, 6u8, 2u8, 5u8, 6u8, 9u8, 3u8, 8u8, 8u8, 9u8, 3u8, 9u8, 0u8,
|
||
3u8, 9u8, 0u8, 7u8, 2u8, 2u8, 8u8, 3u8, 7u8, 7u8, 6u8, 4u8, 7u8, 6u8, 9u8, 7u8, 9u8, 2u8, 5u8, 5u8, 6u8, 7u8, 6u8, 2u8,
|
||
6u8, 9u8, 5u8, 3u8, 1u8, 2u8, 5u8, 3u8, 4u8, 6u8, 9u8, 4u8, 4u8, 6u8, 9u8, 5u8, 1u8, 9u8, 5u8, 3u8, 6u8, 1u8, 4u8, 1u8,
|
||
8u8, 8u8, 8u8, 2u8, 3u8, 8u8, 4u8, 8u8, 9u8, 6u8, 2u8, 7u8, 8u8, 3u8, 8u8, 1u8, 3u8, 4u8, 7u8, 6u8, 5u8, 6u8, 2u8, 5u8,
|
||
1u8, 7u8, 3u8, 4u8, 7u8, 2u8, 3u8, 4u8, 7u8, 5u8, 9u8, 7u8, 6u8, 8u8, 0u8, 7u8, 0u8, 9u8, 4u8, 4u8, 1u8, 1u8, 9u8, 2u8,
|
||
4u8, 4u8, 8u8, 1u8, 3u8, 9u8, 1u8, 9u8, 0u8, 6u8, 7u8, 3u8, 8u8, 2u8, 8u8, 1u8, 2u8, 5u8, 8u8, 6u8, 7u8, 3u8, 6u8, 1u8,
|
||
7u8, 3u8, 7u8, 9u8, 8u8, 8u8, 4u8, 0u8, 3u8, 5u8, 4u8, 7u8, 2u8, 0u8, 5u8, 9u8, 6u8, 2u8, 2u8, 4u8, 0u8, 6u8, 9u8, 5u8,
|
||
9u8, 5u8, 3u8, 3u8, 6u8, 9u8, 1u8, 4u8, 0u8, 6u8, 2u8, 5u8,
|
||
];
|
||
|
||
// ref/hare/strconv/stof_data.ha:73. Eisel-Lemire 128-bit power-of-ten
|
||
// table (see header note). 596 rows, {hi, lo} u64 pair per row.
|
||
let powers_of_ten: [596][2]u64 = [
|
||
[0xA5D3B6D479F8E056u64, 0x8FD0C16206306BABu64],
|
||
[0x8F48A4899877186Cu64, 0xB3C4F1BA87BC8696u64],
|
||
[0x331ACDABFE94DE87u64, 0xE0B62E2929ABA83Cu64],
|
||
[0x9FF0C08B7F1D0B14u64, 0x8C71DCD9BA0B4925u64],
|
||
[0x07ECF0AE5EE44DD9u64, 0xAF8E5410288E1B6Fu64],
|
||
[0xC9E82CD9F69D6150u64, 0xDB71E91432B1A24Au64],
|
||
[0xBE311C083A225CD2u64, 0x892731AC9FAF056Eu64],
|
||
[0x6DBD630A48AAF406u64, 0xAB70FE17C79AC6CAu64],
|
||
[0x092CBBCCDAD5B108u64, 0xD64D3D9DB981787Du64],
|
||
[0x25BBF56008C58EA5u64, 0x85F0468293F0EB4Eu64],
|
||
[0xAF2AF2B80AF6F24Eu64, 0xA76C582338ED2621u64],
|
||
[0x1AF5AF660DB4AEE1u64, 0xD1476E2C07286FAAu64],
|
||
[0x50D98D9FC890ED4Du64, 0x82CCA4DB847945CAu64],
|
||
[0xE50FF107BAB528A0u64, 0xA37FCE126597973Cu64],
|
||
[0x1E53ED49A96272C8u64, 0xCC5FC196FEFD7D0Cu64],
|
||
[0x25E8E89C13BB0F7Au64, 0xFF77B1FCBEBCDC4Fu64],
|
||
[0x77B191618C54E9ACu64, 0x9FAACF3DF73609B1u64],
|
||
[0xD59DF5B9EF6A2417u64, 0xC795830D75038C1Du64],
|
||
[0x4B0573286B44AD1Du64, 0xF97AE3D0D2446F25u64],
|
||
[0x4EE367F9430AEC32u64, 0x9BECCE62836AC577u64],
|
||
[0x229C41F793CDA73Fu64, 0xC2E801FB244576D5u64],
|
||
[0x6B43527578C1110Fu64, 0xF3A20279ED56D48Au64],
|
||
[0x830A13896B78AAA9u64, 0x9845418C345644D6u64],
|
||
[0x23CC986BC656D553u64, 0xBE5691EF416BD60Cu64],
|
||
[0x2CBFBE86B7EC8AA8u64, 0xEDEC366B11C6CB8Fu64],
|
||
[0x7BF7D71432F3D6A9u64, 0x94B3A202EB1C3F39u64],
|
||
[0xDAF5CCD93FB0CC53u64, 0xB9E08A83A5E34F07u64],
|
||
[0xD1B3400F8F9CFF68u64, 0xE858AD248F5C22C9u64],
|
||
[0x23100809B9C21FA1u64, 0x91376C36D99995BEu64],
|
||
[0xABD40A0C2832A78Au64, 0xB58547448FFFFB2Du64],
|
||
[0x16C90C8F323F516Cu64, 0xE2E69915B3FFF9F9u64],
|
||
[0xAE3DA7D97F6792E3u64, 0x8DD01FAD907FFC3Bu64],
|
||
[0x99CD11CFDF41779Cu64, 0xB1442798F49FFB4Au64],
|
||
[0x40405643D711D583u64, 0xDD95317F31C7FA1Du64],
|
||
[0x482835EA666B2572u64, 0x8A7D3EEF7F1CFC52u64],
|
||
[0xDA3243650005EECFu64, 0xAD1C8EAB5EE43B66u64],
|
||
[0x90BED43E40076A82u64, 0xD863B256369D4A40u64],
|
||
[0x5A7744A6E804A291u64, 0x873E4F75E2224E68u64],
|
||
[0x711515D0A205CB36u64, 0xA90DE3535AAAE202u64],
|
||
[0x0D5A5B44CA873E03u64, 0xD3515C2831559A83u64],
|
||
[0xE858790AFE9486C2u64, 0x8412D9991ED58091u64],
|
||
[0x626E974DBE39A872u64, 0xA5178FFF668AE0B6u64],
|
||
[0xFB0A3D212DC8128Fu64, 0xCE5D73FF402D98E3u64],
|
||
[0x7CE66634BC9D0B99u64, 0x80FA687F881C7F8Eu64],
|
||
[0x1C1FFFC1EBC44E80u64, 0xA139029F6A239F72u64],
|
||
[0xA327FFB266B56220u64, 0xC987434744AC874Eu64],
|
||
[0x4BF1FF9F0062BAA8u64, 0xFBE9141915D7A922u64],
|
||
[0x6F773FC3603DB4A9u64, 0x9D71AC8FADA6C9B5u64],
|
||
[0xCB550FB4384D21D3u64, 0xC4CE17B399107C22u64],
|
||
[0x7E2A53A146606A48u64, 0xF6019DA07F549B2Bu64],
|
||
[0x2EDA7444CBFC426Du64, 0x99C102844F94E0FBu64],
|
||
[0xFA911155FEFB5308u64, 0xC0314325637A1939u64],
|
||
[0x793555AB7EBA27CAu64, 0xF03D93EEBC589F88u64],
|
||
[0x4BC1558B2F3458DEu64, 0x96267C7535B763B5u64],
|
||
[0x9EB1AAEDFB016F16u64, 0xBBB01B9283253CA2u64],
|
||
[0x465E15A979C1CADCu64, 0xEA9C227723EE8BCBu64],
|
||
[0x0BFACD89EC191EC9u64, 0x92A1958A7675175Fu64],
|
||
[0xCEF980EC671F667Bu64, 0xB749FAED14125D36u64],
|
||
[0x82B7E12780E7401Au64, 0xE51C79A85916F484u64],
|
||
[0xD1B2ECB8B0908810u64, 0x8F31CC0937AE58D2u64],
|
||
[0x861FA7E6DCB4AA15u64, 0xB2FE3F0B8599EF07u64],
|
||
[0x67A791E093E1D49Au64, 0xDFBDCECE67006AC9u64],
|
||
[0xE0C8BB2C5C6D24E0u64, 0x8BD6A141006042BDu64],
|
||
[0x58FAE9F773886E18u64, 0xAECC49914078536Du64],
|
||
[0xAF39A475506A899Eu64, 0xDA7F5BF590966848u64],
|
||
[0x6D8406C952429603u64, 0x888F99797A5E012Du64],
|
||
[0xC8E5087BA6D33B83u64, 0xAAB37FD7D8F58178u64],
|
||
[0xFB1E4A9A90880A64u64, 0xD5605FCDCF32E1D6u64],
|
||
[0x5CF2EEA09A55067Fu64, 0x855C3BE0A17FCD26u64],
|
||
[0xF42FAA48C0EA481Eu64, 0xA6B34AD8C9DFC06Fu64],
|
||
[0xF13B94DAF124DA26u64, 0xD0601D8EFC57B08Bu64],
|
||
[0x76C53D08D6B70858u64, 0x823C12795DB6CE57u64],
|
||
[0x54768C4B0C64CA6Eu64, 0xA2CB1717B52481EDu64],
|
||
[0xA9942F5DCF7DFD09u64, 0xCB7DDCDDA26DA268u64],
|
||
[0xD3F93B35435D7C4Cu64, 0xFE5D54150B090B02u64],
|
||
[0xC47BC5014A1A6DAFu64, 0x9EFA548D26E5A6E1u64],
|
||
[0x359AB6419CA1091Bu64, 0xC6B8E9B0709F109Au64],
|
||
[0xC30163D203C94B62u64, 0xF867241C8CC6D4C0u64],
|
||
[0x79E0DE63425DCF1Du64, 0x9B407691D7FC44F8u64],
|
||
[0x985915FC12F542E4u64, 0xC21094364DFB5636u64],
|
||
[0x3E6F5B7B17B2939Du64, 0xF294B943E17A2BC4u64],
|
||
[0xA705992CEECF9C42u64, 0x979CF3CA6CEC5B5Au64],
|
||
[0x50C6FF782A838353u64, 0xBD8430BD08277231u64],
|
||
[0xA4F8BF5635246428u64, 0xECE53CEC4A314EBDu64],
|
||
[0x871B7795E136BE99u64, 0x940F4613AE5ED136u64],
|
||
[0x28E2557B59846E3Fu64, 0xB913179899F68584u64],
|
||
[0x331AEADA2FE589CFu64, 0xE757DD7EC07426E5u64],
|
||
[0x3FF0D2C85DEF7621u64, 0x9096EA6F3848984Fu64],
|
||
[0x0FED077A756B53A9u64, 0xB4BCA50B065ABE63u64],
|
||
[0xD3E8495912C62894u64, 0xE1EBCE4DC7F16DFBu64],
|
||
[0x64712DD7ABBBD95Cu64, 0x8D3360F09CF6E4BDu64],
|
||
[0xBD8D794D96AACFB3u64, 0xB080392CC4349DECu64],
|
||
[0xECF0D7A0FC5583A0u64, 0xDCA04777F541C567u64],
|
||
[0xF41686C49DB57244u64, 0x89E42CAAF9491B60u64],
|
||
[0x311C2875C522CED5u64, 0xAC5D37D5B79B6239u64],
|
||
[0x7D633293366B828Bu64, 0xD77485CB25823AC7u64],
|
||
[0xAE5DFF9C02033197u64, 0x86A8D39EF77164BCu64],
|
||
[0xD9F57F830283FDFCu64, 0xA8530886B54DBDEBu64],
|
||
[0xD072DF63C324FD7Bu64, 0xD267CAA862A12D66u64],
|
||
[0x4247CB9E59F71E6Du64, 0x8380DEA93DA4BC60u64],
|
||
[0x52D9BE85F074E608u64, 0xA46116538D0DEB78u64],
|
||
[0x67902E276C921F8Bu64, 0xCD795BE870516656u64],
|
||
[0x00BA1CD8A3DB53B6u64, 0x806BD9714632DFF6u64],
|
||
[0x80E8A40ECCD228A4u64, 0xA086CFCD97BF97F3u64],
|
||
[0x6122CD128006B2CDu64, 0xC8A883C0FDAF7DF0u64],
|
||
[0x796B805720085F81u64, 0xFAD2A4B13D1B5D6Cu64],
|
||
[0xCBE3303674053BB0u64, 0x9CC3A6EEC6311A63u64],
|
||
[0xBEDBFC4411068A9Cu64, 0xC3F490AA77BD60FCu64],
|
||
[0xEE92FB5515482D44u64, 0xF4F1B4D515ACB93Bu64],
|
||
[0x751BDD152D4D1C4Au64, 0x991711052D8BF3C5u64],
|
||
[0xD262D45A78A0635Du64, 0xBF5CD54678EEF0B6u64],
|
||
[0x86FB897116C87C34u64, 0xEF340A98172AACE4u64],
|
||
[0xD45D35E6AE3D4DA0u64, 0x9580869F0E7AAC0Eu64],
|
||
[0x8974836059CCA109u64, 0xBAE0A846D2195712u64],
|
||
[0x2BD1A438703FC94Bu64, 0xE998D258869FACD7u64],
|
||
[0x7B6306A34627DDCFu64, 0x91FF83775423CC06u64],
|
||
[0x1A3BC84C17B1D542u64, 0xB67F6455292CBF08u64],
|
||
[0x20CABA5F1D9E4A93u64, 0xE41F3D6A7377EECAu64],
|
||
[0x547EB47B7282EE9Cu64, 0x8E938662882AF53Eu64],
|
||
[0xE99E619A4F23AA43u64, 0xB23867FB2A35B28Du64],
|
||
[0x6405FA00E2EC94D4u64, 0xDEC681F9F4C31F31u64],
|
||
[0xDE83BC408DD3DD04u64, 0x8B3C113C38F9F37Eu64],
|
||
[0x9624AB50B148D445u64, 0xAE0B158B4738705Eu64],
|
||
[0x3BADD624DD9B0957u64, 0xD98DDAEE19068C76u64],
|
||
[0xE54CA5D70A80E5D6u64, 0x87F8A8D4CFA417C9u64],
|
||
[0x5E9FCF4CCD211F4Cu64, 0xA9F6D30A038D1DBCu64],
|
||
[0x7647C3200069671Fu64, 0xD47487CC8470652Bu64],
|
||
[0x29ECD9F40041E073u64, 0x84C8D4DFD2C63F3Bu64],
|
||
[0xF468107100525890u64, 0xA5FB0A17C777CF09u64],
|
||
[0x7182148D4066EEB4u64, 0xCF79CC9DB955C2CCu64],
|
||
[0xC6F14CD848405530u64, 0x81AC1FE293D599BFu64],
|
||
[0xB8ADA00E5A506A7Cu64, 0xA21727DB38CB002Fu64],
|
||
[0xA6D90811F0E4851Cu64, 0xCA9CF1D206FDC03Bu64],
|
||
[0x908F4A166D1DA663u64, 0xFD442E4688BD304Au64],
|
||
[0x9A598E4E043287FEu64, 0x9E4A9CEC15763E2Eu64],
|
||
[0x40EFF1E1853F29FDu64, 0xC5DD44271AD3CDBAu64],
|
||
[0xD12BEE59E68EF47Cu64, 0xF7549530E188C128u64],
|
||
[0x82BB74F8301958CEu64, 0x9A94DD3E8CF578B9u64],
|
||
[0xE36A52363C1FAF01u64, 0xC13A148E3032D6E7u64],
|
||
[0xDC44E6C3CB279AC1u64, 0xF18899B1BC3F8CA1u64],
|
||
[0x29AB103A5EF8C0B9u64, 0x96F5600F15A7B7E5u64],
|
||
[0x7415D448F6B6F0E7u64, 0xBCB2B812DB11A5DEu64],
|
||
[0x111B495B3464AD21u64, 0xEBDF661791D60F56u64],
|
||
[0xCAB10DD900BEEC34u64, 0x936B9FCEBB25C995u64],
|
||
[0x3D5D514F40EEA742u64, 0xB84687C269EF3BFBu64],
|
||
[0x0CB4A5A3112A5112u64, 0xE65829B3046B0AFAu64],
|
||
[0x47F0E785EABA72ABu64, 0x8FF71A0FE2C2E6DCu64],
|
||
[0x59ED216765690F56u64, 0xB3F4E093DB73A093u64],
|
||
[0x306869C13EC3532Cu64, 0xE0F218B8D25088B8u64],
|
||
[0x1E414218C73A13FBu64, 0x8C974F7383725573u64],
|
||
[0xE5D1929EF90898FAu64, 0xAFBD2350644EEACFu64],
|
||
[0xDF45F746B74ABF39u64, 0xDBAC6C247D62A583u64],
|
||
[0x6B8BBA8C328EB783u64, 0x894BC396CE5DA772u64],
|
||
[0x066EA92F3F326564u64, 0xAB9EB47C81F5114Fu64],
|
||
[0xC80A537B0EFEFEBDu64, 0xD686619BA27255A2u64],
|
||
[0xBD06742CE95F5F36u64, 0x8613FD0145877585u64],
|
||
[0x2C48113823B73704u64, 0xA798FC4196E952E7u64],
|
||
[0xF75A15862CA504C5u64, 0xD17F3B51FCA3A7A0u64],
|
||
[0x9A984D73DBE722FBu64, 0x82EF85133DE648C4u64],
|
||
[0xC13E60D0D2E0EBBAu64, 0xA3AB66580D5FDAF5u64],
|
||
[0x318DF905079926A8u64, 0xCC963FEE10B7D1B3u64],
|
||
[0xFDF17746497F7052u64, 0xFFBBCFE994E5C61Fu64],
|
||
[0xFEB6EA8BEDEFA633u64, 0x9FD561F1FD0F9BD3u64],
|
||
[0xFE64A52EE96B8FC0u64, 0xC7CABA6E7C5382C8u64],
|
||
[0x3DFDCE7AA3C673B0u64, 0xF9BD690A1B68637Bu64],
|
||
[0x06BEA10CA65C084Eu64, 0x9C1661A651213E2Du64],
|
||
[0x486E494FCFF30A62u64, 0xC31BFA0FE5698DB8u64],
|
||
[0x5A89DBA3C3EFCCFAu64, 0xF3E2F893DEC3F126u64],
|
||
[0xF89629465A75E01Cu64, 0x986DDB5C6B3A76B7u64],
|
||
[0xF6BBB397F1135823u64, 0xBE89523386091465u64],
|
||
[0x746AA07DED582E2Cu64, 0xEE2BA6C0678B597Fu64],
|
||
[0xA8C2A44EB4571CDCu64, 0x94DB483840B717EFu64],
|
||
[0x92F34D62616CE413u64, 0xBA121A4650E4DDEBu64],
|
||
[0x77B020BAF9C81D17u64, 0xE896A0D7E51E1566u64],
|
||
[0x0ACE1474DC1D122Eu64, 0x915E2486EF32CD60u64],
|
||
[0x0D819992132456BAu64, 0xB5B5ADA8AAFF80B8u64],
|
||
[0x10E1FFF697ED6C69u64, 0xE3231912D5BF60E6u64],
|
||
[0xCA8D3FFA1EF463C1u64, 0x8DF5EFABC5979C8Fu64],
|
||
[0xBD308FF8A6B17CB2u64, 0xB1736B96B6FD83B3u64],
|
||
[0xAC7CB3F6D05DDBDEu64, 0xDDD0467C64BCE4A0u64],
|
||
[0x6BCDF07A423AA96Bu64, 0x8AA22C0DBEF60EE4u64],
|
||
[0x86C16C98D2C953C6u64, 0xAD4AB7112EB3929Du64],
|
||
[0xE871C7BF077BA8B7u64, 0xD89D64D57A607744u64],
|
||
[0x11471CD764AD4972u64, 0x87625F056C7C4A8Bu64],
|
||
[0xD598E40D3DD89BCFu64, 0xA93AF6C6C79B5D2Du64],
|
||
[0x4AFF1D108D4EC2C3u64, 0xD389B47879823479u64],
|
||
[0xCEDF722A585139BAu64, 0x843610CB4BF160CBu64],
|
||
[0xC2974EB4EE658828u64, 0xA54394FE1EEDB8FEu64],
|
||
[0x733D226229FEEA32u64, 0xCE947A3DA6A9273Eu64],
|
||
[0x0806357D5A3F525Fu64, 0x811CCC668829B887u64],
|
||
[0xCA07C2DCB0CF26F7u64, 0xA163FF802A3426A8u64],
|
||
[0xFC89B393DD02F0B5u64, 0xC9BCFF6034C13052u64],
|
||
[0xBBAC2078D443ACE2u64, 0xFC2C3F3841F17C67u64],
|
||
[0xD54B944B84AA4C0Du64, 0x9D9BA7832936EDC0u64],
|
||
[0x0A9E795E65D4DF11u64, 0xC5029163F384A931u64],
|
||
[0x4D4617B5FF4A16D5u64, 0xF64335BCF065D37Du64],
|
||
[0x504BCED1BF8E4E45u64, 0x99EA0196163FA42Eu64],
|
||
[0xE45EC2862F71E1D6u64, 0xC06481FB9BCF8D39u64],
|
||
[0x5D767327BB4E5A4Cu64, 0xF07DA27A82C37088u64],
|
||
[0x3A6A07F8D510F86Fu64, 0x964E858C91BA2655u64],
|
||
[0x890489F70A55368Bu64, 0xBBE226EFB628AFEAu64],
|
||
[0x2B45AC74CCEA842Eu64, 0xEADAB0ABA3B2DBE5u64],
|
||
[0x3B0B8BC90012929Du64, 0x92C8AE6B464FC96Fu64],
|
||
[0x09CE6EBB40173744u64, 0xB77ADA0617E3BBCBu64],
|
||
[0xCC420A6A101D0515u64, 0xE55990879DDCAABDu64],
|
||
[0x9FA946824A12232Du64, 0x8F57FA54C2A9EAB6u64],
|
||
[0x47939822DC96ABF9u64, 0xB32DF8E9F3546564u64],
|
||
[0x59787E2B93BC56F7u64, 0xDFF9772470297EBDu64],
|
||
[0x57EB4EDB3C55B65Au64, 0x8BFBEA76C619EF36u64],
|
||
[0xEDE622920B6B23F1u64, 0xAEFAE51477A06B03u64],
|
||
[0xE95FAB368E45ECEDu64, 0xDAB99E59958885C4u64],
|
||
[0x11DBCB0218EBB414u64, 0x88B402F7FD75539Bu64],
|
||
[0xD652BDC29F26A119u64, 0xAAE103B5FCD2A881u64],
|
||
[0x4BE76D3346F0495Fu64, 0xD59944A37C0752A2u64],
|
||
[0x6F70A4400C562DDBu64, 0x857FCAE62D8493A5u64],
|
||
[0xCB4CCD500F6BB952u64, 0xA6DFBD9FB8E5B88Eu64],
|
||
[0x7E2000A41346A7A7u64, 0xD097AD07A71F26B2u64],
|
||
[0x8ED400668C0C28C8u64, 0x825ECC24C873782Fu64],
|
||
[0x728900802F0F32FAu64, 0xA2F67F2DFA90563Bu64],
|
||
[0x4F2B40A03AD2FFB9u64, 0xCBB41EF979346BCAu64],
|
||
[0xE2F610C84987BFA8u64, 0xFEA126B7D78186BCu64],
|
||
[0x0DD9CA7D2DF4D7C9u64, 0x9F24B832E6B0F436u64],
|
||
[0x91503D1C79720DBBu64, 0xC6EDE63FA05D3143u64],
|
||
[0x75A44C6397CE912Au64, 0xF8A95FCF88747D94u64],
|
||
[0xC986AFBE3EE11ABAu64, 0x9B69DBE1B548CE7Cu64],
|
||
[0xFBE85BADCE996168u64, 0xC24452DA229B021Bu64],
|
||
[0xFAE27299423FB9C3u64, 0xF2D56790AB41C2A2u64],
|
||
[0xDCCD879FC967D41Au64, 0x97C560BA6B0919A5u64],
|
||
[0x5400E987BBC1C920u64, 0xBDB6B8E905CB600Fu64],
|
||
[0x290123E9AAB23B68u64, 0xED246723473E3813u64],
|
||
[0xF9A0B6720AAF6521u64, 0x9436C0760C86E30Bu64],
|
||
[0xF808E40E8D5B3E69u64, 0xB94470938FA89BCEu64],
|
||
[0xB60B1D1230B20E04u64, 0xE7958CB87392C2C2u64],
|
||
[0xB1C6F22B5E6F48C2u64, 0x90BD77F3483BB9B9u64],
|
||
[0x1E38AEB6360B1AF3u64, 0xB4ECD5F01A4AA828u64],
|
||
[0x25C6DA63C38DE1B0u64, 0xE2280B6C20DD5232u64],
|
||
[0x579C487E5A38AD0Eu64, 0x8D590723948A535Fu64],
|
||
[0x2D835A9DF0C6D851u64, 0xB0AF48EC79ACE837u64],
|
||
[0xF8E431456CF88E65u64, 0xDCDB1B2798182244u64],
|
||
[0x1B8E9ECB641B58FFu64, 0x8A08F0F8BF0F156Bu64],
|
||
[0xE272467E3D222F3Fu64, 0xAC8B2D36EED2DAC5u64],
|
||
[0x5B0ED81DCC6ABB0Fu64, 0xD7ADF884AA879177u64],
|
||
[0x98E947129FC2B4E9u64, 0x86CCBB52EA94BAEAu64],
|
||
[0x3F2398D747B36224u64, 0xA87FEA27A539E9A5u64],
|
||
[0x8EEC7F0D19A03AADu64, 0xD29FE4B18E88640Eu64],
|
||
[0x1953CF68300424ACu64, 0x83A3EEEEF9153E89u64],
|
||
[0x5FA8C3423C052DD7u64, 0xA48CEAAAB75A8E2Bu64],
|
||
[0x3792F412CB06794Du64, 0xCDB02555653131B6u64],
|
||
[0xE2BBD88BBEE40BD0u64, 0x808E17555F3EBF11u64],
|
||
[0x5B6ACEAEAE9D0EC4u64, 0xA0B19D2AB70E6ED6u64],
|
||
[0xF245825A5A445275u64, 0xC8DE047564D20A8Bu64],
|
||
[0xEED6E2F0F0D56712u64, 0xFB158592BE068D2Eu64],
|
||
[0x55464DD69685606Bu64, 0x9CED737BB6C4183Du64],
|
||
[0xAA97E14C3C26B886u64, 0xC428D05AA4751E4Cu64],
|
||
[0xD53DD99F4B3066A8u64, 0xF53304714D9265DFu64],
|
||
[0xE546A8038EFE4029u64, 0x993FE2C6D07B7FABu64],
|
||
[0xDE98520472BDD033u64, 0xBF8FDB78849A5F96u64],
|
||
[0x963E66858F6D4440u64, 0xEF73D256A5C0F77Cu64],
|
||
[0xDDE7001379A44AA8u64, 0x95A8637627989AADu64],
|
||
[0x5560C018580D5D52u64, 0xBB127C53B17EC159u64],
|
||
[0xAAB8F01E6E10B4A6u64, 0xE9D71B689DDE71AFu64],
|
||
[0xCAB3961304CA70E8u64, 0x9226712162AB070Du64],
|
||
[0x3D607B97C5FD0D22u64, 0xB6B00D69BB55C8D1u64],
|
||
[0x8CB89A7DB77C506Au64, 0xE45C10C42A2B3B05u64],
|
||
[0x77F3608E92ADB242u64, 0x8EB98A7A9A5B04E3u64],
|
||
[0x55F038B237591ED3u64, 0xB267ED1940F1C61Cu64],
|
||
[0x6B6C46DEC52F6688u64, 0xDF01E85F912E37A3u64],
|
||
[0x2323AC4B3B3DA015u64, 0x8B61313BBABCE2C6u64],
|
||
[0xABEC975E0A0D081Au64, 0xAE397D8AA96C1B77u64],
|
||
[0x96E7BD358C904A21u64, 0xD9C7DCED53C72255u64],
|
||
[0x7E50D64177DA2E54u64, 0x881CEA14545C7575u64],
|
||
[0xDDE50BD1D5D0B9E9u64, 0xAA242499697392D2u64],
|
||
[0x955E4EC64B44E864u64, 0xD4AD2DBFC3D07787u64],
|
||
[0xBD5AF13BEF0B113Eu64, 0x84EC3C97DA624AB4u64],
|
||
[0xECB1AD8AEACDD58Eu64, 0xA6274BBDD0FADD61u64],
|
||
[0x67DE18EDA5814AF2u64, 0xCFB11EAD453994BAu64],
|
||
[0x80EACF948770CED7u64, 0x81CEB32C4B43FCF4u64],
|
||
[0xA1258379A94D028Du64, 0xA2425FF75E14FC31u64],
|
||
[0x096EE45813A04330u64, 0xCAD2F7F5359A3B3Eu64],
|
||
[0x8BCA9D6E188853FCu64, 0xFD87B5F28300CA0Du64],
|
||
[0x775EA264CF55347Du64, 0x9E74D1B791E07E48u64],
|
||
[0x95364AFE032A819Du64, 0xC612062576589DDAu64],
|
||
[0x3A83DDBD83F52204u64, 0xF79687AED3EEC551u64],
|
||
[0xC4926A9672793542u64, 0x9ABE14CD44753B52u64],
|
||
[0x75B7053C0F178293u64, 0xC16D9A0095928A27u64],
|
||
[0x5324C68B12DD6338u64, 0xF1C90080BAF72CB1u64],
|
||
[0xD3F6FC16EBCA5E03u64, 0x971DA05074DA7BEEu64],
|
||
[0x88F4BB1CA6BCF584u64, 0xBCE5086492111AEAu64],
|
||
[0x2B31E9E3D06C32E5u64, 0xEC1E4A7DB69561A5u64],
|
||
[0x3AFF322E62439FCFu64, 0x9392EE8E921D5D07u64],
|
||
[0x09BEFEB9FAD487C2u64, 0xB877AA3236A4B449u64],
|
||
[0x4C2EBE687989A9B3u64, 0xE69594BEC44DE15Bu64],
|
||
[0x0F9D37014BF60A10u64, 0x901D7CF73AB0ACD9u64],
|
||
[0x538484C19EF38C94u64, 0xB424DC35095CD80Fu64],
|
||
[0x2865A5F206B06FB9u64, 0xE12E13424BB40E13u64],
|
||
[0xF93F87B7442E45D3u64, 0x8CBCCC096F5088CBu64],
|
||
[0xF78F69A51539D748u64, 0xAFEBFF0BCB24AAFEu64],
|
||
[0xB573440E5A884D1Bu64, 0xDBE6FECEBDEDD5BEu64],
|
||
[0x31680A88F8953030u64, 0x89705F4136B4A597u64],
|
||
[0xFDC20D2B36BA7C3Du64, 0xABCC77118461CEFCu64],
|
||
[0x3D32907604691B4Cu64, 0xD6BF94D5E57A42BCu64],
|
||
[0xA63F9A49C2C1B10Fu64, 0x8637BD05AF6C69B5u64],
|
||
[0x0FCF80DC33721D53u64, 0xA7C5AC471B478423u64],
|
||
[0xD3C36113404EA4A8u64, 0xD1B71758E219652Bu64],
|
||
[0x645A1CAC083126E9u64, 0x83126E978D4FDF3Bu64],
|
||
[0x3D70A3D70A3D70A3u64, 0xA3D70A3D70A3D70Au64],
|
||
[0xCCCCCCCCCCCCCCCCu64, 0xCCCCCCCCCCCCCCCCu64],
|
||
[0x0000000000000000u64, 0x8000000000000000u64],
|
||
[0x0000000000000000u64, 0xA000000000000000u64],
|
||
[0x0000000000000000u64, 0xC800000000000000u64],
|
||
[0x0000000000000000u64, 0xFA00000000000000u64],
|
||
[0x0000000000000000u64, 0x9C40000000000000u64],
|
||
[0x0000000000000000u64, 0xC350000000000000u64],
|
||
[0x0000000000000000u64, 0xF424000000000000u64],
|
||
[0x0000000000000000u64, 0x9896800000000000u64],
|
||
[0x0000000000000000u64, 0xBEBC200000000000u64],
|
||
[0x0000000000000000u64, 0xEE6B280000000000u64],
|
||
[0x0000000000000000u64, 0x9502F90000000000u64],
|
||
[0x0000000000000000u64, 0xBA43B74000000000u64],
|
||
[0x0000000000000000u64, 0xE8D4A51000000000u64],
|
||
[0x0000000000000000u64, 0x9184E72A00000000u64],
|
||
[0x0000000000000000u64, 0xB5E620F480000000u64],
|
||
[0x0000000000000000u64, 0xE35FA931A0000000u64],
|
||
[0x0000000000000000u64, 0x8E1BC9BF04000000u64],
|
||
[0x0000000000000000u64, 0xB1A2BC2EC5000000u64],
|
||
[0x0000000000000000u64, 0xDE0B6B3A76400000u64],
|
||
[0x0000000000000000u64, 0x8AC7230489E80000u64],
|
||
[0x0000000000000000u64, 0xAD78EBC5AC620000u64],
|
||
[0x0000000000000000u64, 0xD8D726B7177A8000u64],
|
||
[0x0000000000000000u64, 0x878678326EAC9000u64],
|
||
[0x0000000000000000u64, 0xA968163F0A57B400u64],
|
||
[0x0000000000000000u64, 0xD3C21BCECCEDA100u64],
|
||
[0x0000000000000000u64, 0x84595161401484A0u64],
|
||
[0x0000000000000000u64, 0xA56FA5B99019A5C8u64],
|
||
[0x0000000000000000u64, 0xCECB8F27F4200F3Au64],
|
||
[0x4000000000000000u64, 0x813F3978F8940984u64],
|
||
[0x5000000000000000u64, 0xA18F07D736B90BE5u64],
|
||
[0xA400000000000000u64, 0xC9F2C9CD04674EDEu64],
|
||
[0x4D00000000000000u64, 0xFC6F7C4045812296u64],
|
||
[0xF020000000000000u64, 0x9DC5ADA82B70B59Du64],
|
||
[0x6C28000000000000u64, 0xC5371912364CE305u64],
|
||
[0xC732000000000000u64, 0xF684DF56C3E01BC6u64],
|
||
[0x3C7F400000000000u64, 0x9A130B963A6C115Cu64],
|
||
[0x4B9F100000000000u64, 0xC097CE7BC90715B3u64],
|
||
[0x1E86D40000000000u64, 0xF0BDC21ABB48DB20u64],
|
||
[0x1314448000000000u64, 0x96769950B50D88F4u64],
|
||
[0x17D955A000000000u64, 0xBC143FA4E250EB31u64],
|
||
[0x5DCFAB0800000000u64, 0xEB194F8E1AE525FDu64],
|
||
[0x5AA1CAE500000000u64, 0x92EFD1B8D0CF37BEu64],
|
||
[0xF14A3D9E40000000u64, 0xB7ABC627050305ADu64],
|
||
[0x6D9CCD05D0000000u64, 0xE596B7B0C643C719u64],
|
||
[0xE4820023A2000000u64, 0x8F7E32CE7BEA5C6Fu64],
|
||
[0xDDA2802C8A800000u64, 0xB35DBF821AE4F38Bu64],
|
||
[0xD50B2037AD200000u64, 0xE0352F62A19E306Eu64],
|
||
[0x4526F422CC340000u64, 0x8C213D9DA502DE45u64],
|
||
[0x9670B12B7F410000u64, 0xAF298D050E4395D6u64],
|
||
[0x3C0CDD765F114000u64, 0xDAF3F04651D47B4Cu64],
|
||
[0xA5880A69FB6AC800u64, 0x88D8762BF324CD0Fu64],
|
||
[0x8EEA0D047A457A00u64, 0xAB0E93B6EFEE0053u64],
|
||
[0x72A4904598D6D880u64, 0xD5D238A4ABE98068u64],
|
||
[0x47A6DA2B7F864750u64, 0x85A36366EB71F041u64],
|
||
[0x999090B65F67D924u64, 0xA70C3C40A64E6C51u64],
|
||
[0xFFF4B4E3F741CF6Du64, 0xD0CF4B50CFE20765u64],
|
||
[0xBFF8F10E7A8921A4u64, 0x82818F1281ED449Fu64],
|
||
[0xAFF72D52192B6A0Du64, 0xA321F2D7226895C7u64],
|
||
[0x9BF4F8A69F764490u64, 0xCBEA6F8CEB02BB39u64],
|
||
[0x02F236D04753D5B4u64, 0xFEE50B7025C36A08u64],
|
||
[0x01D762422C946590u64, 0x9F4F2726179A2245u64],
|
||
[0x424D3AD2B7B97EF5u64, 0xC722F0EF9D80AAD6u64],
|
||
[0xD2E0898765A7DEB2u64, 0xF8EBAD2B84E0D58Bu64],
|
||
[0x63CC55F49F88EB2Fu64, 0x9B934C3B330C8577u64],
|
||
[0x3CBF6B71C76B25FBu64, 0xC2781F49FFCFA6D5u64],
|
||
[0x8BEF464E3945EF7Au64, 0xF316271C7FC3908Au64],
|
||
[0x97758BF0E3CBB5ACu64, 0x97EDD871CFDA3A56u64],
|
||
[0x3D52EEED1CBEA317u64, 0xBDE94E8E43D0C8ECu64],
|
||
[0x4CA7AAA863EE4BDDu64, 0xED63A231D4C4FB27u64],
|
||
[0x8FE8CAA93E74EF6Au64, 0x945E455F24FB1CF8u64],
|
||
[0xB3E2FD538E122B44u64, 0xB975D6B6EE39E436u64],
|
||
[0x60DBBCA87196B616u64, 0xE7D34C64A9C85D44u64],
|
||
[0xBC8955E946FE31CDu64, 0x90E40FBEEA1D3A4Au64],
|
||
[0x6BABAB6398BDBE41u64, 0xB51D13AEA4A488DDu64],
|
||
[0xC696963C7EED2DD1u64, 0xE264589A4DCDAB14u64],
|
||
[0xFC1E1DE5CF543CA2u64, 0x8D7EB76070A08AECu64],
|
||
[0x3B25A55F43294BCBu64, 0xB0DE65388CC8ADA8u64],
|
||
[0x49EF0EB713F39EBEu64, 0xDD15FE86AFFAD912u64],
|
||
[0x6E3569326C784337u64, 0x8A2DBF142DFCC7ABu64],
|
||
[0x49C2C37F07965404u64, 0xACB92ED9397BF996u64],
|
||
[0xDC33745EC97BE906u64, 0xD7E77A8F87DAF7FBu64],
|
||
[0x69A028BB3DED71A3u64, 0x86F0AC99B4E8DAFDu64],
|
||
[0xC40832EA0D68CE0Cu64, 0xA8ACD7C0222311BCu64],
|
||
[0xF50A3FA490C30190u64, 0xD2D80DB02AABD62Bu64],
|
||
[0x792667C6DA79E0FAu64, 0x83C7088E1AAB65DBu64],
|
||
[0x577001B891185938u64, 0xA4B8CAB1A1563F52u64],
|
||
[0xED4C0226B55E6F86u64, 0xCDE6FD5E09ABCF26u64],
|
||
[0x544F8158315B05B4u64, 0x80B05E5AC60B6178u64],
|
||
[0x696361AE3DB1C721u64, 0xA0DC75F1778E39D6u64],
|
||
[0x03BC3A19CD1E38E9u64, 0xC913936DD571C84Cu64],
|
||
[0x04AB48A04065C723u64, 0xFB5878494ACE3A5Fu64],
|
||
[0x62EB0D64283F9C76u64, 0x9D174B2DCEC0E47Bu64],
|
||
[0x3BA5D0BD324F8394u64, 0xC45D1DF942711D9Au64],
|
||
[0xCA8F44EC7EE36479u64, 0xF5746577930D6500u64],
|
||
[0x7E998B13CF4E1ECBu64, 0x9968BF6ABBE85F20u64],
|
||
[0x9E3FEDD8C321A67Eu64, 0xBFC2EF456AE276E8u64],
|
||
[0xC5CFE94EF3EA101Eu64, 0xEFB3AB16C59B14A2u64],
|
||
[0xBBA1F1D158724A12u64, 0x95D04AEE3B80ECE5u64],
|
||
[0x2A8A6E45AE8EDC97u64, 0xBB445DA9CA61281Fu64],
|
||
[0xF52D09D71A3293BDu64, 0xEA1575143CF97226u64],
|
||
[0x593C2626705F9C56u64, 0x924D692CA61BE758u64],
|
||
[0x6F8B2FB00C77836Cu64, 0xB6E0C377CFA2E12Eu64],
|
||
[0x0B6DFB9C0F956447u64, 0xE498F455C38B997Au64],
|
||
[0x4724BD4189BD5EACu64, 0x8EDF98B59A373FECu64],
|
||
[0x58EDEC91EC2CB657u64, 0xB2977EE300C50FE7u64],
|
||
[0x2F2967B66737E3EDu64, 0xDF3D5E9BC0F653E1u64],
|
||
[0xBD79E0D20082EE74u64, 0x8B865B215899F46Cu64],
|
||
[0xECD8590680A3AA11u64, 0xAE67F1E9AEC07187u64],
|
||
[0xE80E6F4820CC9495u64, 0xDA01EE641A708DE9u64],
|
||
[0x3109058D147FDCDDu64, 0x884134FE908658B2u64],
|
||
[0xBD4B46F0599FD415u64, 0xAA51823E34A7EEDEu64],
|
||
[0x6C9E18AC7007C91Au64, 0xD4E5E2CDC1D1EA96u64],
|
||
[0x03E2CF6BC604DDB0u64, 0x850FADC09923329Eu64],
|
||
[0x84DB8346B786151Cu64, 0xA6539930BF6BFF45u64],
|
||
[0xE612641865679A63u64, 0xCFE87F7CEF46FF16u64],
|
||
[0x4FCB7E8F3F60C07Eu64, 0x81F14FAE158C5F6Eu64],
|
||
[0xE3BE5E330F38F09Du64, 0xA26DA3999AEF7749u64],
|
||
[0x5CADF5BFD3072CC5u64, 0xCB090C8001AB551Cu64],
|
||
[0x73D9732FC7C8F7F6u64, 0xFDCB4FA002162A63u64],
|
||
[0x2867E7FDDCDD9AFAu64, 0x9E9F11C4014DDA7Eu64],
|
||
[0xB281E1FD541501B8u64, 0xC646D63501A1511Du64],
|
||
[0x1F225A7CA91A4226u64, 0xF7D88BC24209A565u64],
|
||
[0x3375788DE9B06958u64, 0x9AE757596946075Fu64],
|
||
[0x0052D6B1641C83AEu64, 0xC1A12D2FC3978937u64],
|
||
[0xC0678C5DBD23A49Au64, 0xF209787BB47D6B84u64],
|
||
[0xF840B7BA963646E0u64, 0x9745EB4D50CE6332u64],
|
||
[0xB650E5A93BC3D898u64, 0xBD176620A501FBFFu64],
|
||
[0xA3E51F138AB4CEBEu64, 0xEC5D3FA8CE427AFFu64],
|
||
[0xC66F336C36B10137u64, 0x93BA47C980E98CDFu64],
|
||
[0xB80B0047445D4184u64, 0xB8A8D9BBE123F017u64],
|
||
[0xA60DC059157491E5u64, 0xE6D3102AD96CEC1Du64],
|
||
[0x87C89837AD68DB2Fu64, 0x9043EA1AC7E41392u64],
|
||
[0x29BABE4598C311FBu64, 0xB454E4A179DD1877u64],
|
||
[0xF4296DD6FEF3D67Au64, 0xE16A1DC9D8545E94u64],
|
||
[0x1899E4A65F58660Cu64, 0x8CE2529E2734BB1Du64],
|
||
[0x5EC05DCFF72E7F8Fu64, 0xB01AE745B101E9E4u64],
|
||
[0x76707543F4FA1F73u64, 0xDC21A1171D42645Du64],
|
||
[0x6A06494A791C53A8u64, 0x899504AE72497EBAu64],
|
||
[0x0487DB9D17636892u64, 0xABFA45DA0EDBDE69u64],
|
||
[0x45A9D2845D3C42B6u64, 0xD6F8D7509292D603u64],
|
||
[0x0B8A2392BA45A9B2u64, 0x865B86925B9BC5C2u64],
|
||
[0x8E6CAC7768D7141Eu64, 0xA7F26836F282B732u64],
|
||
[0x3207D795430CD926u64, 0xD1EF0244AF2364FFu64],
|
||
[0x7F44E6BD49E807B8u64, 0x8335616AED761F1Fu64],
|
||
[0x5F16206C9C6209A6u64, 0xA402B9C5A8D3A6E7u64],
|
||
[0x36DBA887C37A8C0Fu64, 0xCD036837130890A1u64],
|
||
[0xC2494954DA2C9789u64, 0x802221226BE55A64u64],
|
||
[0xF2DB9BAA10B7BD6Cu64, 0xA02AA96B06DEB0FDu64],
|
||
[0x6F92829494E5ACC7u64, 0xC83553C5C8965D3Du64],
|
||
[0xCB772339BA1F17F9u64, 0xFA42A8B73ABBF48Cu64],
|
||
[0xFF2A760414536EFBu64, 0x9C69A97284B578D7u64],
|
||
[0xFEF5138519684ABAu64, 0xC38413CF25E2D70Du64],
|
||
[0x7EB258665FC25D69u64, 0xF46518C2EF5B8CD1u64],
|
||
[0xEF2F773FFBD97A61u64, 0x98BF2F79D5993802u64],
|
||
[0xAAFB550FFACFD8FAu64, 0xBEEEFB584AFF8603u64],
|
||
[0x95BA2A53F983CF38u64, 0xEEAABA2E5DBF6784u64],
|
||
[0xDD945A747BF26183u64, 0x952AB45CFA97A0B2u64],
|
||
[0x94F971119AEEF9E4u64, 0xBA756174393D88DFu64],
|
||
[0x7A37CD5601AAB85Du64, 0xE912B9D1478CEB17u64],
|
||
[0xAC62E055C10AB33Au64, 0x91ABB422CCB812EEu64],
|
||
[0x577B986B314D6009u64, 0xB616A12B7FE617AAu64],
|
||
[0xED5A7E85FDA0B80Bu64, 0xE39C49765FDF9D94u64],
|
||
[0x14588F13BE847307u64, 0x8E41ADE9FBEBC27Du64],
|
||
[0x596EB2D8AE258FC8u64, 0xB1D219647AE6B31Cu64],
|
||
[0x6FCA5F8ED9AEF3BBu64, 0xDE469FBD99A05FE3u64],
|
||
[0x25DE7BB9480D5854u64, 0x8AEC23D680043BEEu64],
|
||
[0xAF561AA79A10AE6Au64, 0xADA72CCC20054AE9u64],
|
||
[0x1B2BA1518094DA04u64, 0xD910F7FF28069DA4u64],
|
||
[0x90FB44D2F05D0842u64, 0x87AA9AFF79042286u64],
|
||
[0x353A1607AC744A53u64, 0xA99541BF57452B28u64],
|
||
[0x42889B8997915CE8u64, 0xD3FA922F2D1675F2u64],
|
||
[0x69956135FEBADA11u64, 0x847C9B5D7C2E09B7u64],
|
||
[0x43FAB9837E699095u64, 0xA59BC234DB398C25u64],
|
||
[0x94F967E45E03F4BBu64, 0xCF02B2C21207EF2Eu64],
|
||
[0x1D1BE0EEBAC278F5u64, 0x8161AFB94B44F57Du64],
|
||
[0x6462D92A69731732u64, 0xA1BA1BA79E1632DCu64],
|
||
[0x7D7B8F7503CFDCFEu64, 0xCA28A291859BBF93u64],
|
||
[0x5CDA735244C3D43Eu64, 0xFCB2CB35E702AF78u64],
|
||
[0x3A0888136AFA64A7u64, 0x9DEFBF01B061ADABu64],
|
||
[0x088AAA1845B8FDD0u64, 0xC56BAEC21C7A1916u64],
|
||
[0x8AAD549E57273D45u64, 0xF6C69A72A3989F5Bu64],
|
||
[0x36AC54E2F678864Bu64, 0x9A3C2087A63F6399u64],
|
||
[0x84576A1BB416A7DDu64, 0xC0CB28A98FCF3C7Fu64],
|
||
[0x656D44A2A11C51D5u64, 0xF0FDF2D3F3C30B9Fu64],
|
||
[0x9F644AE5A4B1B325u64, 0x969EB7C47859E743u64],
|
||
[0x873D5D9F0DDE1FEEu64, 0xBC4665B596706114u64],
|
||
[0xA90CB506D155A7EAu64, 0xEB57FF22FC0C7959u64],
|
||
[0x09A7F12442D588F2u64, 0x9316FF75DD87CBD8u64],
|
||
[0x0C11ED6D538AEB2Fu64, 0xB7DCBF5354E9BECEu64],
|
||
[0x8F1668C8A86DA5FAu64, 0xE5D3EF282A242E81u64],
|
||
[0xF96E017D694487BCu64, 0x8FA475791A569D10u64],
|
||
[0x37C981DCC395A9ACu64, 0xB38D92D760EC4455u64],
|
||
[0x85BBE253F47B1417u64, 0xE070F78D3927556Au64],
|
||
[0x93956D7478CCEC8Eu64, 0x8C469AB843B89562u64],
|
||
[0x387AC8D1970027B2u64, 0xAF58416654A6BABBu64],
|
||
[0x06997B05FCC0319Eu64, 0xDB2E51BFE9D0696Au64],
|
||
[0x441FECE3BDF81F03u64, 0x88FCF317F22241E2u64],
|
||
[0xD527E81CAD7626C3u64, 0xAB3C2FDDEEAAD25Au64],
|
||
[0x8A71E223D8D3B074u64, 0xD60B3BD56A5586F1u64],
|
||
[0xF6872D5667844E49u64, 0x85C7056562757456u64],
|
||
[0xB428F8AC016561DBu64, 0xA738C6BEBB12D16Cu64],
|
||
[0xE13336D701BEBA52u64, 0xD106F86E69D785C7u64],
|
||
[0xECC0024661173473u64, 0x82A45B450226B39Cu64],
|
||
[0x27F002D7F95D0190u64, 0xA34D721642B06084u64],
|
||
[0x31EC038DF7B441F4u64, 0xCC20CE9BD35C78A5u64],
|
||
[0x7E67047175A15271u64, 0xFF290242C83396CEu64],
|
||
[0x0F0062C6E984D386u64, 0x9F79A169BD203E41u64],
|
||
[0x52C07B78A3E60868u64, 0xC75809C42C684DD1u64],
|
||
[0xA7709A56CCDF8A82u64, 0xF92E0C3537826145u64],
|
||
[0x88A66076400BB691u64, 0x9BBCC7A142B17CCBu64],
|
||
[0x6ACFF893D00EA435u64, 0xC2ABF989935DDBFEu64],
|
||
[0x0583F6B8C4124D43u64, 0xF356F7EBF83552FEu64],
|
||
[0xC3727A337A8B704Au64, 0x98165AF37B2153DEu64],
|
||
[0x744F18C0592E4C5Cu64, 0xBE1BF1B059E9A8D6u64],
|
||
[0x1162DEF06F79DF73u64, 0xEDA2EE1C7064130Cu64],
|
||
[0x8ADDCB5645AC2BA8u64, 0x9485D4D1C63E8BE7u64],
|
||
[0x6D953E2BD7173692u64, 0xB9A74A0637CE2EE1u64],
|
||
[0xC8FA8DB6CCDD0437u64, 0xE8111C87C5C1BA99u64],
|
||
[0x1D9C9892400A22A2u64, 0x910AB1D4DB9914A0u64],
|
||
[0x2503BEB6D00CAB4Bu64, 0xB54D5E4A127F59C8u64],
|
||
[0x2E44AE64840FD61Du64, 0xE2A0B5DC971F303Au64],
|
||
[0x5CEAECFED289E5D2u64, 0x8DA471A9DE737E24u64],
|
||
[0x7425A83E872C5F47u64, 0xB10D8E1456105DADu64],
|
||
[0xD12F124E28F77719u64, 0xDD50F1996B947518u64],
|
||
[0x82BD6B70D99AAA6Fu64, 0x8A5296FFE33CC92Fu64],
|
||
[0x636CC64D1001550Bu64, 0xACE73CBFDC0BFB7Bu64],
|
||
[0x3C47F7E05401AA4Eu64, 0xD8210BEFD30EFA5Au64],
|
||
[0x65ACFAEC34810A71u64, 0x8714A775E3E95C78u64],
|
||
[0x7F1839A741A14D0Du64, 0xA8D9D1535CE3B396u64],
|
||
[0x1EDE48111209A050u64, 0xD31045A8341CA07Cu64],
|
||
[0x934AED0AAB460432u64, 0x83EA2B892091E44Du64],
|
||
[0xF81DA84D5617853Fu64, 0xA4E4B66B68B65D60u64],
|
||
[0x36251260AB9D668Eu64, 0xCE1DE40642E3F4B9u64],
|
||
[0xC1D72B7C6B426019u64, 0x80D2AE83E9CE78F3u64],
|
||
[0xB24CF65B8612F81Fu64, 0xA1075A24E4421730u64],
|
||
[0xDEE033F26797B627u64, 0xC94930AE1D529CFCu64],
|
||
[0x169840EF017DA3B1u64, 0xFB9B7CD9A4A7443Cu64],
|
||
[0x8E1F289560EE864Eu64, 0x9D412E0806E88AA5u64],
|
||
[0xF1A6F2BAB92A27E2u64, 0xC491798A08A2AD4Eu64],
|
||
[0xAE10AF696774B1DBu64, 0xF5B5D7EC8ACB58A2u64],
|
||
[0xACCA6DA1E0A8EF29u64, 0x9991A6F3D6BF1765u64],
|
||
[0x17FD090A58D32AF3u64, 0xBFF610B0CC6EDD3Fu64],
|
||
[0xDDFC4B4CEF07F5B0u64, 0xEFF394DCFF8A948Eu64],
|
||
[0x4ABDAF101564F98Eu64, 0x95F83D0A1FB69CD9u64],
|
||
[0x9D6D1AD41ABE37F1u64, 0xBB764C4CA7A4440Fu64],
|
||
[0x84C86189216DC5EDu64, 0xEA53DF5FD18D5513u64],
|
||
[0x32FD3CF5B4E49BB4u64, 0x92746B9BE2F8552Cu64],
|
||
[0x3FBC8C33221DC2A1u64, 0xB7118682DBB66A77u64],
|
||
[0x0FABAF3FEAA5334Au64, 0xE4D5E82392A40515u64],
|
||
[0x29CB4D87F2A7400Eu64, 0x8F05B1163BA6832Du64],
|
||
[0x743E20E9EF511012u64, 0xB2C71D5BCA9023F8u64],
|
||
[0x914DA9246B255416u64, 0xDF78E4B2BD342CF6u64],
|
||
[0x1AD089B6C2F7548Eu64, 0x8BAB8EEFB6409C1Au64],
|
||
[0xA184AC2473B529B1u64, 0xAE9672ABA3D0C320u64],
|
||
[0xC9E5D72D90A2741Eu64, 0xDA3C0F568CC4F3E8u64],
|
||
[0x7E2FA67C7A658892u64, 0x8865899617FB1871u64],
|
||
[0xDDBB901B98FEEAB7u64, 0xAA7EEBFB9DF9DE8Du64],
|
||
[0x552A74227F3EA565u64, 0xD51EA6FA85785631u64],
|
||
[0xD53A88958F87275Fu64, 0x8533285C936B35DEu64],
|
||
[0x8A892ABAF368F137u64, 0xA67FF273B8460356u64],
|
||
[0x2D2B7569B0432D85u64, 0xD01FEF10A657842Cu64],
|
||
[0x9C3B29620E29FC73u64, 0x8213F56A67F6B29Bu64],
|
||
[0x8349F3BA91B47B8Fu64, 0xA298F2C501F45F42u64],
|
||
[0x241C70A936219A73u64, 0xCB3F2F7642717713u64],
|
||
[0xED238CD383AA0110u64, 0xFE0EFB53D30DD4D7u64],
|
||
[0xF4363804324A40AAu64, 0x9EC95D1463E8A506u64],
|
||
[0xB143C6053EDCD0D5u64, 0xC67BB4597CE2CE48u64],
|
||
[0xDD94B7868E94050Au64, 0xF81AA16FDC1B81DAu64],
|
||
[0xCA7CF2B4191C8326u64, 0x9B10A4E5E9913128u64],
|
||
[0xFD1C2F611F63A3F0u64, 0xC1D4CE1F63F57D72u64],
|
||
[0xBC633B39673C8CECu64, 0xF24A01A73CF2DCCFu64],
|
||
[0xD5BE0503E085D813u64, 0x976E41088617CA01u64],
|
||
[0x4B2D8644D8A74E18u64, 0xBD49D14AA79DBC82u64],
|
||
[0xDDF8E7D60ED1219Eu64, 0xEC9C459D51852BA2u64],
|
||
[0xCABB90E5C942B503u64, 0x93E1AB8252F33B45u64],
|
||
[0x3D6A751F3B936243u64, 0xB8DA1662E7B00A17u64],
|
||
[0x0CC512670A783AD4u64, 0xE7109BFBA19C0C9Du64],
|
||
[0x27FB2B80668B24C5u64, 0x906A617D450187E2u64],
|
||
[0xB1F9F660802DEDF6u64, 0xB484F9DC9641E9DAu64],
|
||
[0x5E7873F8A0396973u64, 0xE1A63853BBD26451u64],
|
||
[0xDB0B487B6423E1E8u64, 0x8D07E33455637EB2u64],
|
||
[0x91CE1A9A3D2CDA62u64, 0xB049DC016ABC5E5Fu64],
|
||
[0x7641A140CC7810FBu64, 0xDC5C5301C56B75F7u64],
|
||
[0xA9E904C87FCB0A9Du64, 0x89B9B3E11B6329BAu64],
|
||
[0x546345FA9FBDCD44u64, 0xAC2820D9623BF429u64],
|
||
[0xA97C177947AD4095u64, 0xD732290FBACAF133u64],
|
||
[0x49ED8EABCCCC485Du64, 0x867F59A9D4BED6C0u64],
|
||
[0x5C68F256BFFF5A74u64, 0xA81F301449EE8C70u64],
|
||
[0x73832EEC6FFF3111u64, 0xD226FC195C6A2F8Cu64],
|
||
];
|
||
|
||
// 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.
|
||
|
||
package strconv;
|
||
|
||
import os;
|
||
import 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 — graduated to the Ryū shortest-round-trippable implementation
|
||
// in ftos.ww (strconv #106 fold-5). The old lossy fixed-point version
|
||
// (6 fractional digits, "huge" fallback ≥9e18, no NaN/Inf) was deleted
|
||
// here per the lib-note graduation rule ("replace in one go, don't keep
|
||
// both"); ftos.ww's f64tos is the live one. f32tos follows in fold-5b
|
||
// (task #67, gated on the #143 f32-arg-push cgen fix).
|
||
|
||
// 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("");
|
||
};
|
||
|
||
// 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.
|
||
|
||
package test;
|
||
|
||
import os;
|
||
import strconv;
|
||
import 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;
|
||
};
|
||
|
||
// --- size/align/offset typed-builtin fixtures (#42) --------------------
|
||
|
||
type point = struct {
|
||
x: i32,
|
||
y: i32,
|
||
};
|
||
|
||
// Mixed-alignment struct: i8 lays at 0, then i64 needs to skip to
|
||
// offset 8 (the i64's natural align). Probe asserts both ends.
|
||
type mixalign = struct {
|
||
tag: i8,
|
||
val: i64,
|
||
};
|
||
|
||
// --- 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, 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; };
|
||
|
||
// Probe 8 — size(T) / align(T) / offset(e.f) typed-builtin folds
|
||
// (#42). Each call folds to an N_INTLIT at check time; cgen
|
||
// materialises the literal as a plain `MOVQ $N, AX`. Mirrors
|
||
// cstage cmd/wcc/check.c:907-960 byte-for-byte on this corpus.
|
||
if (size(str) != 24) { return 23; }; // str IS []u8: {ptr,len,cap} 24B (#1/Phase 3)
|
||
if (size(i64) != 8) { return 24; };
|
||
if (size(i32) != 4) { return 25; };
|
||
if (align(i64) != 8) { return 26; };
|
||
if (align(i32) != 4) { return 27; };
|
||
// Initialize struct locals explicitly so the cgen path doesn't
|
||
// drift from cstage on bare `let X: T;` zero-init (pre-existing
|
||
// wwstage divergence outside #42).
|
||
let pt: point = point { x = 0, y = 0 };
|
||
if (offset(pt.x) != 0) { return 28; };
|
||
if (offset(pt.y) != 4) { return 29; };
|
||
let mx: mixalign = mixalign { tag = 0i8, val = 0i64 };
|
||
if (offset(mx.tag) != 0) { return 30; };
|
||
if (offset(mx.val) != 8) { return 31; }; // align-padded to 8
|
||
// Width breadth: smallest prim, ptr, slice, struct (8B + padded),
|
||
// covering astsize's TPTR/TSLICE/TNAME-resolve-to-struct arms.
|
||
if (size(i8) != 1) { return 32; };
|
||
if (align(i8) != 1) { return 33; };
|
||
if (size(*i32) != 8) { return 34; };
|
||
if (size([]i32) != 24) { return 35; };
|
||
if (size(point) != 8) { return 36; };
|
||
if (size(mixalign) != 16) { return 37; };
|
||
|
||
return 42;
|
||
};
|
||
|