Merge branch 'master' of https://gitee.com/wa-lang/wa

waroot/src/bytes/bytealg.wa

+// 版权 @2023 凹语言 作者。保留所有权利。
+
+
+// MaxLen is the maximum length of the string to be searched for (argument b) in Index.
+// If MaxLen is not 0, make sure MaxLen >= 4.
+global bytealg_MaxLen: int
+
+// FIXME: the logic of HashStrBytes, HashStrRevBytes, IndexRabinKarpBytes and HashStr, HashStrRev,
+// IndexRabinKarp are exactly the same, except that the types are different. Can we eliminate
+// three of them without causing allocation?
+
+// PrimeRK is the prime base used in Rabin-Karp algorithm.
+const bytealg_PrimeRK = 16777619
+
+const bytealg_MaxBruteForce = 0
+
+// Index returns the index of the first instance of b in a, or -1 if b is not present in a.
+// Requires 2 <= len(b) <= MaxLen.
+func bytealg_Index(a, b: []byte) => int {
+	panic("unimplemented")
+}
+
+// IndexString returns the index of the first instance of b in a, or -1 if b is not present in a.
+// Requires 2 <= len(b) <= MaxLen.
+func bytealg_IndexString(a, b: string) => int {
+	panic("unimplemented")
+}
+
+// Cutover reports the number of failures of IndexByte we should tolerate
+// before switching over to Index.
+// n is the number of bytes processed so far.
+// See the bytes.Index implementation for details.
+func bytealg_Cutover(n: int) => int {
+	panic("unimplemented")
+}
+
+func bytealg_Count(b: []byte, c: byte) => int {
+	n := 0
+	for _, x := range b {
+		if x == c {
+			n++
+		}
+	}
+	return n
+}
+
+func bytealg_CountString(s: string, c: byte) => int {
+	n := 0
+	for i := 0; i < len(s); i++ {
+		if s[i] == c {
+			n++
+		}
+	}
+	return n
+}

waroot/src/bytes/bytes.wa

 // 版权 @2023 凹语言 作者。保留所有权利。
 
+import (
+	"unicode/utf8"
+)
+
 // Equal reports whether a and b
 // are the same length and contain the same bytes.
 // A nil argument is equivalent to an empty slice.
@@ -24,8 +28,96 @@ func EqualFold(s, t: []byte) => bool {
 	return true
 }
 
-func Index(d: []byte, x: []byte) => int {
-	return 0
+
+// Index returns the index of the first instance of sep in s, or -1 if sep is not present in s.
+func Index(s, sep: []byte) => int {
+	n := len(sep)
+	switch {
+	case n == 0:
+		return 0
+	case n == 1:
+		return IndexByte(s, sep[0])
+	case n == len(s):
+		if Equal(sep, s) {
+			return 0
+		}
+		return -1
+	case n > len(s):
+		return -1
+	case n <= bytealg_MaxLen:
+		// Use brute force when s and sep both are small
+		//if len(s) <= bytealg.MaxBruteForce {
+		//	return bytealg.Index(s, sep)
+		//}
+		c0 := sep[0]
+		c1 := sep[1]
+		i := 0
+		t := len(s) - n + 1
+		fails := 0
+		for i < t {
+			if s[i] != c0 {
+				// IndexByte is faster than bytealg.Index, so use it as long as
+				// we're not getting lots of false positives.
+				o := IndexByte(s[i+1:t], c0)
+				if o < 0 {
+					return -1
+				}
+				i += o + 1
+			}
+			if s[i+1] == c1 && Equal(s[i:i+n], sep) {
+				return i
+			}
+			fails++
+			i++
+			// Switch to bytealg.Index when IndexByte produces too many false positives.
+			//if fails > bytealg.Cutover(i) {
+			//	r := bytealg.Index(s[i:], sep)
+			//	if r >= 0 {
+			//		return r + i
+			//	}
+			//	return -1
+			//}
+		}
+		return -1
+	}
+	c0 := sep[0]
+	c1 := sep[1]
+	i := 0
+	fails := 0
+	t := len(s) - n + 1
+	for i < t {
+		if s[i] != c0 {
+			o := IndexByte(s[i+1:t], c0)
+			if o < 0 {
+				break
+			}
+			i += o + 1
+		}
+		if s[i+1] == c1 && Equal(s[i:i+n], sep) {
+			return i
+		}
+		i++
+		fails++
+		if fails >= 4+i>>4 && i < t {
+			// Give up on IndexByte, it isn't skipping ahead
+			// far enough to be better than Rabin-Karp.
+			// Experiments (using IndexPeriodic) suggest
+			// the cutover is about 16 byte skips.
+			// TODO: if large prefixes of sep are matching
+			// we should cutover at even larger average skips,
+			// because Equal becomes that much more expensive.
+			// This code does not take that effect into account.
+
+			//j := bytealg.IndexRabinKarpBytes(s[i:], sep)
+			//if j < 0 {
+			//	return -1
+			//}
+			//return i + j
+
+			return -1
+		}
+	}
+	return -1
 }
 
 func IndexByte(b: []byte, c: byte) => int {
@@ -95,3 +187,187 @@ func Repeat(b: []byte, count: int) => []byte {
 	}
 	return nb
 }
+
+// Compare returns an integer comparing two byte slices lexicographically.
+// The result will be 0 if a==b, -1 if a < b, and +1 if a > b.
+// A nil argument is equivalent to an empty slice.
+func Compare(a, b: []byte) => int {
+	l := len(a)
+	if len(b) < l {
+		l = len(b)
+	}
+	if l == 0 || &a[0] == &b[0] {
+		if len(a) < len(b) {
+			return -1
+		}
+		if len(a) > len(b) {
+			return +1
+		}
+		return 0
+	}
+
+	for i := 0; i < l; i++ {
+		c1, c2 := a[i], b[i]
+		if c1 < c2 {
+			return -1
+		}
+		if c1 > c2 {
+			return +1
+		}
+	}
+
+	if len(a) < len(b) {
+		return -1
+	}
+	if len(a) > len(b) {
+		return +1
+	}
+	return 0
+}
+
+// LastIndexByte returns the index of the last instance of c in s, or -1 if c is not present in s.
+func LastIndexByte(s: []byte, c: byte) => int {
+	for i := len(s) - 1; i >= 0; i-- {
+		if s[i] == c {
+			return i
+		}
+	}
+	return -1
+}
+
+// IndexRune interprets s as a sequence of UTF-8-encoded code points.
+// It returns the byte index of the first occurrence in s of the given rune.
+// It returns -1 if rune is not present in s.
+// If r is utf8.RuneError, it returns the first instance of any
+// invalid UTF-8 byte sequence.
+func IndexRune(s: []byte, r: rune) => int {
+	switch {
+	case 0 <= r && r < utf8.RuneSelf:
+		return IndexByte(s, byte(r))
+	case r == utf8.RuneError:
+		for i := 0; i < len(s); {
+			r1, n := utf8.DecodeRune(s[i:])
+			if r1 == utf8.RuneError {
+				return i
+			}
+			i += n
+		}
+		return -1
+	case !utf8.ValidRune(r):
+		return -1
+	default:
+		b: [utf8.UTFMax]byte
+		n := utf8.EncodeRune(b[:], r)
+		return Index(s, b[:n])
+	}
+}
+
+
+// Count counts the number of non-overlapping instances of sep in s.
+// If sep is an empty slice, Count returns 1 + the number of UTF-8-encoded code points in s.
+func Count(s, sep: []byte) => int {
+	// special case
+	if len(sep) == 0 {
+		return utf8.RuneCount(s) + 1
+	}
+	if len(sep) == 1 {
+		return bytealg_Count(s, sep[0])
+	}
+	n := 0
+	for {
+		i := Index(s, sep)
+		if i == -1 {
+			return n
+		}
+		n++
+		s = s[i+len(sep):]
+	}
+}
+
+// SplitN slices s into subslices separated by sep and returns a slice of
+// the subslices between those separators.
+// If sep is empty, SplitN splits after each UTF-8 sequence.
+// The count determines the number of subslices to return:
+//   n > 0: at most n subslices; the last subslice will be the unsplit remainder.
+//   n == 0: the result is nil (zero subslices)
+//   n < 0: all subslices
+func SplitN(s, sep: []byte, n: int) => [][]byte {
+	return genSplit(s, sep, 0, n)
+}
+
+// Generic split: splits after each instance of sep,
+// including sepSave bytes of sep in the subslices.
+func genSplit(s, sep: []byte, sepSave, n: int) => [][]byte {
+	if n == 0 {
+		return nil
+	}
+	if len(sep) == 0 {
+		return explode(s, n)
+	}
+	if n < 0 {
+		n = Count(s, sep) + 1
+	}
+
+	a := make([][]byte, n)
+	n--
+	i := 0
+	for i < n {
+		m := Index(s, sep)
+		if m < 0 {
+			break
+		}
+		a[i] = s[: m+sepSave : m+sepSave]
+		s = s[m+len(sep):]
+		i++
+	}
+	a[i] = s
+	return a[:i+1]
+}
+
+
+// explode splits s into a slice of UTF-8 sequences, one per Unicode code point (still slices of bytes),
+// up to a maximum of n byte slices. Invalid UTF-8 sequences are chopped into individual bytes.
+func explode(s: []byte, n: int) => [][]byte {
+	if n <= 0 {
+		n = len(s)
+	}
+	a := make([][]byte, n)
+	size: int
+	na := 0
+	for len(s) > 0 {
+		if na+1 >= n {
+			a[na] = s
+			na++
+			break
+		}
+		_, size = utf8.DecodeRune(s)
+		a[na] = s[0:size:size]
+		s = s[size:]
+		na++
+	}
+	return a[0:na]
+}
+
+// Join concatenates the elements of s to create a new byte slice. The separator
+// sep is placed between elements in the resulting slice.
+func Join(s: [][]byte, sep: []byte) => []byte {
+	if len(s) == 0 {
+		return []byte{}
+	}
+	if len(s) == 1 {
+		// Just return a copy.
+		return append([]byte(nil), s[0]...)
+	}
+	n := len(sep) * (len(s) - 1)
+	for _, v := range s {
+		n += len(v)
+	}
+
+	b := make([]byte, n)
+	bp := copy(b, s[0])
+	for _, v := range s[1:] {
+		bp += copy(b[bp:], sep)
+		bp += copy(b[bp:], v)
+	}
+	return b
+}

waroot/src/bytes/bytes_test.wa

 // 版权 @2023 凹语言 作者。保留所有权利。
 
+import (
+	"unicode/utf8"
+)
+
+type BinOpTest struct {
+	a: string
+	b: string
+	i: int
+}
+
 func TestEqual {
 	for _, tt := range compareTests {
 		eql := Equal(tt.a, tt.b)
@@ -10,6 +20,471 @@ func TestEqual {
 	}
 }
 
+func TestEqualExhaustive {
+	size := 128
+	//if testing.Short() {
+	size = 32
+	//}
+	a := make([]byte, size)
+	b := make([]byte, size)
+	b_init := make([]byte, size)
+	// randomish but deterministic data
+	for i := 0; i < size; i++ {
+		a[i] = byte(17 * i)
+		b_init[i] = byte(23*i + 100)
+	}
+
+	for len := 0; len <= size; len++ {
+		for x := 0; x <= size-len; x++ {
+			for y := 0; y <= size-len; y++ {
+				copy(b, b_init)
+				copy(b[y:y+len], a[x:x+len])
+				if !Equal(a[x:x+len], b[y:y+len]) || !Equal(b[y:y+len], a[x:x+len]) {
+					assert(false)
+					//t.Errorf("Equal(%d, %d, %d) = false", len, x, y)
+				}
+			}
+		}
+	}
+}
+
+// make sure Equal returns false for minimally different strings. The data
+// is all zeros except for a single one in one location.
+func TestNotEqual {
+	size := 128
+	//if testing.Short() {
+		size = 32
+	//}
+	a := make([]byte, size)
+	b := make([]byte, size)
+
+	for len := 0; len <= size; len++ {
+		for x := 0; x <= size-len; x++ {
+			for y := 0; y <= size-len; y++ {
+				for diffpos := x; diffpos < x+len; diffpos++ {
+					a[diffpos] = 1
+					if Equal(a[x:x+len], b[y:y+len]) || Equal(b[y:y+len], a[x:x+len]) {
+						assert(false)
+						//t.Errorf("NotEqual(%d, %d, %d, %d) = true", len, x, y, diffpos)
+					}
+					a[diffpos] = 0
+				}
+			}
+		}
+	}
+}
+
+
+var indexTests = []BinOpTest{
+	{"", "", 0},
+	{"", "a", -1},
+	{"", "foo", -1},
+	{"fo", "foo", -1},
+	{"foo", "baz", -1},
+	{"foo", "foo", 0},
+	{"oofofoofooo", "f", 2},
+	{"oofofoofooo", "foo", 4},
+	{"barfoobarfoo", "foo", 3},
+	{"foo", "", 0},
+	{"foo", "o", 1},
+	{"abcABCabc", "A", 3},
+	// cases with one byte strings - test IndexByte and special case in Index()
+	{"", "a", -1},
+	{"x", "a", -1},
+	{"x", "x", 0},
+	{"abc", "a", 0},
+	{"abc", "b", 1},
+	{"abc", "c", 2},
+	{"abc", "x", -1},
+	{"barfoobarfooyyyzzzyyyzzzyyyzzzyyyxxxzzzyyy", "x", 33},
+	{"foofyfoobarfoobar", "y", 4},
+	{"oooooooooooooooooooooo", "r", -1},
+	{"oxoxoxoxoxoxoxoxoxoxoxoy", "oy", 22},
+	{"oxoxoxoxoxoxoxoxoxoxoxox", "oy", -1},
+	// test fallback to Rabin-Karp.
+	{"000000000000000000000000000000000000000000000000000000000000000000000001", "0000000000000000000000000000000000000000000000000000000000000000001", 5},
+}
+
+func TestIndex {
+	// todo
+}
+
+func TestLastIndex {
+	// todo
+}
+func TestIndexAny {
+	// todo
+}
+
+func TestLastIndexAny {
+	// todo
+}
+
+func TestIndexByte{
+	for _, tt := range indexTests {
+		if len(tt.b) != 1 {
+			continue
+		}
+		a := []byte(tt.a)
+		b := tt.b[0]
+		pos := IndexByte(a, b)
+		if pos != tt.i {
+			assert(false)
+			// t.Errorf(`IndexByte(%q, '%c') = %v`, tt.a, b, pos)
+		}
+		posp := indexBytePortable(a, b)
+		if posp != tt.i {
+			assert(false)
+			//t.Errorf(`indexBytePortable(%q, '%c') = %v`, tt.a, b, posp)
+		}
+	}
+}
+
+func indexBytePortable(s: []byte, c: byte) => int {
+	for i, b := range s {
+		if b == c {
+			return i
+		}
+	}
+	return -1
+}
+
+func TestLastIndexByte {
+	testCases := []BinOpTest{
+		{"", "q", -1},
+		{"abcdef", "q", -1},
+		{"abcdefabcdef", "a", len("abcdef")},      // something in the middle
+		{"abcdefabcdef", "f", len("abcdefabcde")}, // last byte
+		{"zabcdefabcdef", "z", 0},                 // first byte
+		{"a☺b☻c☹d", "b", len("a☺")},               // non-ascii
+	}
+	for _, test := range testCases {
+		actual := LastIndexByte([]byte(test.a), test.b[0])
+		if actual != test.i {
+			assert(false)
+			//t.Errorf("LastIndexByte(%q,%c) = %v; want %v", test.a, test.b[0], actual, test.i)
+		}
+	}
+}
+
+// test a larger buffer with different sizes and alignments
+func TestIndexByteBig {
+	n := 1024
+	//if testing.Short() {
+	n = 128
+	//}
+	b := make([]byte, n)
+	for i := 0; i < n; i++ {
+		// different start alignments
+		b1 := b[i:]
+		for j := 0; j < len(b1); j++ {
+			b1[j] = 'x'
+			pos := IndexByte(b1, 'x')
+			if pos != j {
+				assert(false)
+				//t.Errorf("IndexByte(%q, 'x') = %v", b1, pos)
+			}
+			b1[j] = 0
+			pos = IndexByte(b1, 'x')
+			if pos != -1 {
+				assert(false)
+				//t.Errorf("IndexByte(%q, 'x') = %v", b1, pos)
+			}
+		}
+		// different end alignments
+		b1 = b[:i]
+		for j := 0; j < len(b1); j++ {
+			b1[j] = 'x'
+			pos := IndexByte(b1, 'x')
+			if pos != j {
+				assert(false)
+				//t.Errorf("IndexByte(%q, 'x') = %v", b1, pos)
+			}
+			b1[j] = 0
+			pos = IndexByte(b1, 'x')
+			if pos != -1 {
+				assert(false)
+				//t.Errorf("IndexByte(%q, 'x') = %v", b1, pos)
+			}
+		}
+		// different start and end alignments
+		b1 = b[i/2 : n-(i+1)/2]
+		for j := 0; j < len(b1); j++ {
+			b1[j] = 'x'
+			pos := IndexByte(b1, 'x')
+			if pos != j {
+				assert(false)
+				//t.Errorf("IndexByte(%q, 'x') = %v", b1, pos)
+			}
+			b1[j] = 0
+			pos = IndexByte(b1, 'x')
+			if pos != -1 {
+				assert(false)
+				//t.Errorf("IndexByte(%q, 'x') = %v", b1, pos)
+			}
+		}
+	}
+}
+
+// test a small index across all page offsets
+func TestIndexByteSmall {
+	b := make([]byte, 5015) // bigger than a page
+	// Make sure we find the correct byte even when straddling a page.
+	for i := 0; i <= len(b)-15; i++ {
+		for j := 0; j < 15; j++ {
+			b[i+j] = byte(100 + j)
+		}
+		for j := 0; j < 15; j++ {
+			p := IndexByte(b[i:i+15], byte(100+j))
+			if p != j {
+				assert(false)
+				//t.Errorf("IndexByte(%q, %d) = %d", b[i:i+15], 100+j, p)
+			}
+		}
+		for j := 0; j < 15; j++ {
+			b[i+j] = 0
+		}
+	}
+	// Make sure matches outside the slice never trigger.
+	for i := 0; i <= len(b)-15; i++ {
+		for j := 0; j < 15; j++ {
+			b[i+j] = 1
+		}
+		for j := 0; j < 15; j++ {
+			p := IndexByte(b[i:i+15], byte(0))
+			if p != -1 {
+				assert(false)
+				//t.Errorf("IndexByte(%q, %d) = %d", b[i:i+15], 0, p)
+			}
+		}
+		for j := 0; j < 15; j++ {
+			b[i+j] = 0
+		}
+	}
+}
+
+func TestIndexRune {
+	if true {
+		//x := IndexRune([]byte("foo"), 'o')
+		//if x != 1 {
+		//	println(x)
+		//	assert(false)
+		//}
+	}
+	tests := []struct {
+		in:   string
+		rune: rune
+		want: int
+	}{
+		{"", 'a', -1},
+		{"", '☺', -1},
+		{"foo", '☹', -1},
+		{"foo", 'o', 1},
+		{"foo☺bar", '☺', 3},
+		//{"foo☺☻☹bar", '☹', 9},
+		{"a A x", 'A', 2},
+		{"some_text=some_value", '=', 9},
+		{"☺a", 'a', 3},
+		{"a☻☺b", '☺', 4},
+
+		// RuneError should match any invalid UTF-8 byte sequence.
+		{"�", '�', 0},
+		//{"\xff", '�', 0},
+		{"☻x�", '�', len("☻x")},
+		{"☻x\xe2\x98", '�', len("☻x")},
+		{"☻x\xe2\x98�", '�', len("☻x")},
+		{"☻x\xe2\x98x", '�', len("☻x")},
+
+		// Invalid rune values should never match.
+		{"a☺b☻c☹d\xe2\x98�\xff�\xed\xa0\x80", -1, -1},
+		{"a☺b☻c☹d\xe2\x98�\xff�\xed\xa0\x80", 0xD800, -1}, // Surrogate pair
+		{"a☺b☻c☹d\xe2\x98�\xff�\xed\xa0\x80", utf8.MaxRune + 1, -1},
+	}
+	for i, tt := range tests {
+		if got := IndexRune([]byte(tt.in), tt.rune); got != tt.want {
+			println(i, tt.in, tt.rune, got, tt.want)
+			assert(false)
+			//t.Errorf("IndexRune(%q, %d) = %v; want %v", tt.in, tt.rune, got, tt.want)
+		}
+	}
+
+	haystack := []byte("test世界")
+	{
+		if i := IndexRune(haystack, 's'); i != 2 {
+			assert(false)
+			//t.Fatalf("'s' at %d; want 2", i)
+		}
+		if i := IndexRune(haystack, '世'); i != 4 {
+			assert(false)
+			//t.Fatalf("'世' at %d; want 4", i)
+		}
+	}
+}
+
+// test count of a single byte across page offsets
+func TestCountByte {
+	b := make([]byte, 5015) // bigger than a page
+	windows := []int{1, 2, 3, 4, 15, 16, 17, 31, 32, 33, 63, 64, 65, 128}
+	testCountWindow := func(i, window: int) {
+		for j := 0; j < window; j++ {
+			b[i+j] = byte(100)
+			p := Count(b[i:i+window], []byte{100})
+			if p != j+1 {
+				assert(false)
+				//t.Errorf("TestCountByte.Count(%q, 100) = %d", b[i:i+window], p)
+			}
+		}
+	}
+
+	maxWnd := windows[len(windows)-1]
+
+	for i := 0; i <= 2*maxWnd; i++ {
+		for _, window := range windows {
+			if window > len(b[i:]) {
+				window = len(b[i:])
+			}
+			testCountWindow(i, window)
+			for j := 0; j < window; j++ {
+				b[i+j] = byte(0)
+			}
+		}
+	}
+	for i := 4096 - (maxWnd + 1); i < len(b); i++ {
+		for _, window := range windows {
+			if window > len(b[i:]) {
+				window = len(b[i:])
+			}
+			testCountWindow(i, window)
+			for j := 0; j < window; j++ {
+				b[i+j] = byte(0)
+			}
+		}
+	}
+}
+
+// Make sure we don't count bytes outside our window
+func TestCountByteNoMatch {
+	b := make([]byte, 5015)
+	windows := []int{1, 2, 3, 4, 15, 16, 17, 31, 32, 33, 63, 64, 65, 128}
+	for i := 0; i <= len(b); i++ {
+		for _, window := range windows {
+			if window > len(b[i:]) {
+				window = len(b[i:])
+			}
+			// Fill the window with non-match
+			for j := 0; j < window; j++ {
+				b[i+j] = byte(100)
+			}
+			// Try to find something that doesn't exist
+			p := Count(b[i:i+window], []byte{0})
+			if p != 0 {
+				assert(false)
+				//t.Errorf("TestCountByteNoMatch(%q, 0) = %d", b[i:i+window], p)
+			}
+			for j := 0; j < window; j++ {
+				b[i+j] = byte(0)
+			}
+		}
+	}
+}
+
+type SplitTest struct {
+	s:   string
+	sep: string
+	n:   int
+	a:   []string
+}
+
+const abcd = "abcd"
+const faces = "☺☻☹"
+const commas = "1,2,3,4"
+const dots = "1....2....3....4"
+
+global splittests = []SplitTest{
+	{"", "", -1, []string{}},
+	{abcd, "a", 0, nil},
+	{abcd, "", 2, []string{"a", "bcd"}},
+	{abcd, "a", -1, []string{"", "bcd"}},
+	{abcd, "z", -1, []string{"abcd"}},
+	{abcd, "", -1, []string{"a", "b", "c", "d"}},
+	{commas, ",", -1, []string{"1", "2", "3", "4"}},
+	{dots, "...", -1, []string{"1", ".2", ".3", ".4"}},
+	{faces, "☹", -1, []string{"☺☻", ""}},
+	{faces, "~", -1, []string{faces}},
+	{faces, "", -1, []string{"☺", "☻", "☹"}},
+	{"1 2 3 4", " ", 3, []string{"1", "2", "3 4"}},
+	{"1 2", " ", 3, []string{"1", "2"}},
+	{"123", "", 2, []string{"1", "23"}},
+	{"123", "", 17, []string{"1", "2", "3"}},
+}
+
+func TestSplit {
+	for _, tt := range splittests {
+		a := SplitN([]byte(tt.s), []byte(tt.sep), tt.n)
+
+		// Appending to the results should not change future results.
+		x: []byte
+		for _, v := range a {
+			x = append(v, 'z')
+		}
+
+		result := sliceOfString(a)
+		if !eq(result, tt.a) {
+			print("result:[")
+			for i, x := range result {
+				if i > 0 {
+					print(", ")
+				}
+				print(x)
+			}
+			println("]")
+
+			print("tt.a:[")
+			for i, x := range tt.a {
+				if i > 0 {
+					print(", ")
+				}
+				print(x)
+			}
+			println("]")
+
+			assert(false)
+			//t.Errorf(`Split(%q, %q, %d) = %v; want %v`, tt.s, tt.sep, tt.n, result, tt.a)
+			continue
+		}
+		if tt.n == 0 || len(a) == 0 {
+			continue
+		}
+
+		if want := tt.a[len(tt.a)-1] + "z"; string(x) != want {
+			println("x:", string(x))
+			println("want:", want)
+			assert(false)
+			//t.Errorf("last appended result was %s; want %s", x, want)
+		}
+
+		s := Join(a, []byte(tt.sep))
+		if string(s) != tt.s {
+			assert(false)
+			//t.Errorf(`Join(Split(%q, %q, %d), %q) = %q`, tt.s, tt.sep, tt.n, tt.sep, s)
+		}
+		if tt.n < 0 {
+			//b := Split([]byte(tt.s), []byte(tt.sep))
+			//if !reflect.DeepEqual(a, b) {
+				assert(false)
+				//t.Errorf("Split disagrees withSplitN(%q, %q, %d) = %v; want %v", tt.s, tt.sep, tt.n, b, a)
+			//}
+		}
+		if len(a) > 0 {
+			in, out := a[0], s
+			if cap(in) == cap(out) && &in[:1][0] == &out[:1][0] {
+				assert(false)
+				//t.Errorf("Join(%#v, %q) didn't copy", a, tt.sep)
+			}
+		}
+	}
+}
+
 func TestEqualFold {
 	for _, tt := range EqualFoldTests {
 		if out := EqualFold([]byte(tt.s), []byte(tt.t)); out != tt.out {
@@ -35,31 +510,21 @@ global EqualFoldTests = []struct {
 	{"abcdefghijk", "abcdefghijX", false},
 }
 
-global compareTests = []struct {
-	a, b []byte
-	i    int
-}{
-	{[]byte(""), []byte(""), 0},
-	{[]byte("a"), []byte(""), 1},
-	{[]byte(""), []byte("a"), -1},
-	{[]byte("abc"), []byte("abc"), 0},
-	{[]byte("abd"), []byte("abc"), 1},
-	{[]byte("abc"), []byte("abd"), -1},
-	{[]byte("ab"), []byte("abc"), -1},
-	{[]byte("abc"), []byte("ab"), 1},
-	{[]byte("x"), []byte("ab"), 1},
-	{[]byte("ab"), []byte("x"), -1},
-	{[]byte("x"), []byte("a"), 1},
-	{[]byte("b"), []byte("x"), -1},
-	// test runtime·memeq's chunked implementation
-	{[]byte("abcdefgh"), []byte("abcdefgh"), 0},
-	{[]byte("abcdefghi"), []byte("abcdefghi"), 0},
-	{[]byte("abcdefghi"), []byte("abcdefghj"), -1},
-	{[]byte("abcdefghj"), []byte("abcdefghi"), 1},
-	// nil tests
-	{nil, nil, 0},
-	{[]byte(""), nil, 0},
-	{nil, []byte(""), 0},
-	{[]byte("a"), nil, 1},
-	{nil, []byte("a"), -1},
+func eq(a, b: []string) => bool {
+	if len(a) != len(b) {
+		return false
+	}
+	for i := 0; i < len(a); i++ {
+		if a[i] != b[i] {
+			return false
+		}
+	}
+	return true
+}
+func sliceOfString(s: [][]byte) => []string {
+	result := make([]string, len(s))
+	for i, v := range s {
+		result[i] = string(v)
+	}
+	return result
 }

waroot/src/bytes/compare_test.wa

+// 版权 @2023 凹语言 作者。保留所有权利。
+
+global compareTests = []struct {
+	a, b: []byte
+	i:    int
+}{
+	{[]byte(""), []byte(""), 0},
+	{[]byte("a"), []byte(""), 1},
+	{[]byte(""), []byte("a"), -1},
+	{[]byte("abc"), []byte("abc"), 0},
+	{[]byte("abd"), []byte("abc"), 1},
+	{[]byte("abc"), []byte("abd"), -1},
+	{[]byte("ab"), []byte("abc"), -1},
+	{[]byte("abc"), []byte("ab"), 1},
+	{[]byte("x"), []byte("ab"), 1},
+	{[]byte("ab"), []byte("x"), -1},
+	{[]byte("x"), []byte("a"), 1},
+	{[]byte("b"), []byte("x"), -1},
+	// test runtime·memeq's chunked implementation
+	{[]byte("abcdefgh"), []byte("abcdefgh"), 0},
+	{[]byte("abcdefghi"), []byte("abcdefghi"), 0},
+	{[]byte("abcdefghi"), []byte("abcdefghj"), -1},
+	{[]byte("abcdefghj"), []byte("abcdefghi"), 1},
+	// nil tests
+	{nil, nil, 0},
+	{[]byte(""), nil, 0},
+	{nil, []byte(""), 0},
+	{[]byte("a"), nil, 1},
+	{nil, []byte("a"), -1},
+}
+
+func TestCompare {
+	for _, tt := range compareTests {
+		numShifts := 16
+		buffer := make([]byte, len(tt.b)+numShifts)
+		// vary the input alignment of tt.b
+		for offset := 0; offset <= numShifts; offset++ {
+			shiftedB := buffer[offset : len(tt.b)+offset]
+			copy(shiftedB, tt.b)
+			cmp := Compare(tt.a, shiftedB)
+			if cmp != tt.i {
+				assert(false)
+				// t.Errorf(`Compare(%q, %q), offset %d = %v; want %v`, tt.a, tt.b, offset, cmp, tt.i)
+			}
+		}
+	}
+}
+
+func TestCompareIdenticalSlice {
+	b := []byte("Hello Gophers!")
+	if Compare(b, b) != 0 {
+		assert(false)
+		//t.Error("b != b")
+	}
+	if Compare(b, b[:1]) != 1 {
+		assert(false)
+		//t.Error("b > b[:1] failed")
+	}
+}
+
+func TestCompareBytes {
+	lengths := make([]int, 0) // lengths to test in ascending order
+	for i := 0; i <= 128; i++ {
+		lengths = append(lengths, i)
+	}
+	lengths = append(lengths, 256, 512, 1024, 1333, 4095, 4096, 4097)
+
+	//if !testing.Short() || testenv.Builder() != "" {
+	//	lengths = append(lengths, 65535, 65536, 65537, 99999)
+	//}
+
+	n := lengths[len(lengths)-1]
+	a := make([]byte, n+1)
+	b := make([]byte, n+1)
+	for _, len := range lengths {
+		// randomish but deterministic data. No 0 or 255.
+		for i := 0; i < len; i++ {
+			a[i] = byte(1 + 31*i%254)
+			b[i] = byte(1 + 31*i%254)
+		}
+		// data past the end is different
+		for i := len; i <= n; i++ {
+			a[i] = 8
+			b[i] = 9
+		}
+		cmp := Compare(a[:len], b[:len])
+		if cmp != 0 {
+			assert(false)
+			//t.Errorf(`CompareIdentical(%d) = %d`, len, cmp)
+		}
+		if len > 0 {
+			cmp = Compare(a[:len-1], b[:len])
+			if cmp != -1 {
+				assert(false)
+				//t.Errorf(`CompareAshorter(%d) = %d`, len, cmp)
+			}
+			cmp = Compare(a[:len], b[:len-1])
+			if cmp != 1 {
+				assert(false)
+				//t.Errorf(`CompareBshorter(%d) = %d`, len, cmp)
+			}
+		}
+		for k := 0; k < len; k++ {
+			b[k] = a[k] - 1
+			cmp = Compare(a[:len], b[:len])
+			if cmp != 1 {
+				assert(false)
+				//t.Errorf(`CompareAbigger(%d,%d) = %d`, len, k, cmp)
+			}
+			b[k] = a[k] + 1
+			cmp = Compare(a[:len], b[:len])
+			if cmp != -1 {
+				assert(false)
+				//t.Errorf(`CompareBbigger(%d,%d) = %d`, len, k, cmp)
+			}
+			b[k] = a[k]
+		}
+	}
+}
+
+func TestEndianBaseCompare {
+	// This test compares byte slices that are almost identical, except one
+	// difference that for some j, a[j]>b[j] and a[j+1]<b[j+1]. If the implementation
+	// compares large chunks with wrong endianness, it gets wrong result.
+	// no vector register is larger than 512 bytes for now
+	const maxLength = 512
+	a := make([]byte, maxLength)
+	b := make([]byte, maxLength)
+	// randomish but deterministic data. No 0 or 255.
+	for i := 0; i < maxLength; i++ {
+		a[i] = byte(1 + 31*i%254)
+		b[i] = byte(1 + 31*i%254)
+	}
+	for i := 2; i <= maxLength; i <<= 1 {
+		for j := 0; j < i-1; j++ {
+			a[j] = b[j] - 1
+			a[j+1] = b[j+1] + 1
+			cmp := Compare(a[:i], b[:i])
+			if cmp != -1 {
+				assert(false)
+				//t.Errorf(`CompareBbigger(%d,%d) = %d`, i, j, cmp)
+			}
+			a[j] = b[j] + 1
+			a[j+1] = b[j+1] - 1
+			cmp = Compare(a[:i], b[:i])
+			if cmp != 1 {
+				assert(false)
+				//t.Errorf(`CompareAbigger(%d,%d) = %d`, i, j, cmp)
+			}
+			a[j] = b[j]
+			a[j+1] = b[j+1]
+		}
+	}
+}