Archived
Dependency update, new golang.org/x package names
This commit is contained in:
1 parent
5dbaf6ceb0
commit
126c4e9a06
51 files changed
+7847
-7911
No files matched your search
Generated
+20
-20
@@ -5,26 +5,6 @@
|
||||
"./cmd/..."
|
||||
],
|
||||
"Deps": [
|
||||
{
|
||||
"ImportPath": "code.google.com/p/go.crypto/bcrypt",
|
||||
"Comment": "null-216",
|
||||
"Rev": "41cd4647fccc72b0b79ef1bd1fe6735e718257cd"
|
||||
},
|
||||
{
|
||||
"ImportPath": "code.google.com/p/go.crypto/blowfish",
|
||||
"Comment": "null-216",
|
||||
"Rev": "41cd4647fccc72b0b79ef1bd1fe6735e718257cd"
|
||||
},
|
||||
{
|
||||
"ImportPath": "code.google.com/p/go.text/transform",
|
||||
"Comment": "null-90",
|
||||
"Rev": "d65bffbc88a153d23a6d2a864531e6e7c2cde59b"
|
||||
},
|
||||
{
|
||||
"ImportPath": "code.google.com/p/go.text/unicode/norm",
|
||||
"Comment": "null-90",
|
||||
"Rev": "d65bffbc88a153d23a6d2a864531e6e7c2cde59b"
|
||||
},
|
||||
{
|
||||
"ImportPath": "github.com/AudriusButkevicius/lfu-go",
|
||||
"Rev": "164bcecceb92fd6037f4d18a8d97b495ec6ef669"
|
||||
@@ -68,6 +48,26 @@
|
||||
{
|
||||
"ImportPath": "github.com/vitrun/qart/qr",
|
||||
"Rev": "ccb109cf25f0cd24474da73b9fee4e7a3e8a8ce0"
|
||||
},
|
||||
{
|
||||
"ImportPath": "golang.org/x/crypto/bcrypt",
|
||||
"Comment": "null-236",
|
||||
"Rev": "69e2a90ed92d03812364aeb947b7068dc42e561e"
|
||||
},
|
||||
{
|
||||
"ImportPath": "golang.org/x/crypto/blowfish",
|
||||
"Comment": "null-236",
|
||||
"Rev": "69e2a90ed92d03812364aeb947b7068dc42e561e"
|
||||
},
|
||||
{
|
||||
"ImportPath": "golang.org/x/text/transform",
|
||||
"Comment": "null-112",
|
||||
"Rev": "2f707e0ad64637ca1318279be7201f5ed19c4050"
|
||||
},
|
||||
{
|
||||
"ImportPath": "golang.org/x/text/unicode/norm",
|
||||
"Comment": "null-112",
|
||||
"Rev": "2f707e0ad64637ca1318279be7201f5ed19c4050"
|
||||
}
|
||||
]
|
||||
}
|
||||
-45
@@ -1,45 +0,0 @@
|
||||
// Copyright 2011 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// +build ignore
|
||||
|
||||
// Generate test data for trie code.
|
||||
|
||||
package main
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
)
|
||||
|
||||
func main() {
|
||||
printTestTables()
|
||||
}
|
||||
|
||||
// We take the smallest, largest and an arbitrary value for each
|
||||
// of the UTF-8 sequence lengths.
|
||||
var testRunes = []rune{
|
||||
0x01, 0x0C, 0x7F, // 1-byte sequences
|
||||
0x80, 0x100, 0x7FF, // 2-byte sequences
|
||||
0x800, 0x999, 0xFFFF, // 3-byte sequences
|
||||
0x10000, 0x10101, 0x10FFFF, // 4-byte sequences
|
||||
0x200, 0x201, 0x202, 0x210, 0x215, // five entries in one sparse block
|
||||
}
|
||||
|
||||
const fileHeader = `// Generated by running
|
||||
// maketesttables
|
||||
// DO NOT EDIT
|
||||
|
||||
package norm
|
||||
|
||||
`
|
||||
|
||||
func printTestTables() {
|
||||
fmt.Print(fileHeader)
|
||||
fmt.Printf("var testRunes = %#v\n\n", testRunes)
|
||||
t := newNode()
|
||||
for i, r := range testRunes {
|
||||
t.insert(r, uint16(i))
|
||||
}
|
||||
t.printTables("testdata")
|
||||
}
|
||||
-6989
File diff suppressed because it is too large.
Load diff
-232
@@ -1,232 +0,0 @@
|
||||
// Copyright 2011 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
package norm
|
||||
|
||||
type valueRange struct {
|
||||
value uint16 // header: value:stride
|
||||
lo, hi byte // header: lo:n
|
||||
}
|
||||
|
||||
type trie struct {
|
||||
index []uint8
|
||||
values []uint16
|
||||
sparse []valueRange
|
||||
sparseOffset []uint16
|
||||
cutoff uint8 // indices >= cutoff are sparse
|
||||
}
|
||||
|
||||
// lookupValue determines the type of block n and looks up the value for b.
|
||||
// For n < t.cutoff, the block is a simple lookup table. Otherwise, the block
|
||||
// is a list of ranges with an accompanying value. Given a matching range r,
|
||||
// the value for b is by r.value + (b - r.lo) * stride.
|
||||
func (t *trie) lookupValue(n uint8, b byte) uint16 {
|
||||
if n < t.cutoff {
|
||||
return t.values[uint16(n)<<6+uint16(b)]
|
||||
}
|
||||
offset := t.sparseOffset[n-t.cutoff]
|
||||
header := t.sparse[offset]
|
||||
lo := offset + 1
|
||||
hi := lo + uint16(header.lo)
|
||||
for lo < hi {
|
||||
m := lo + (hi-lo)/2
|
||||
r := t.sparse[m]
|
||||
if r.lo <= b && b <= r.hi {
|
||||
return r.value + uint16(b-r.lo)*header.value
|
||||
}
|
||||
if b < r.lo {
|
||||
hi = m
|
||||
} else {
|
||||
lo = m + 1
|
||||
}
|
||||
}
|
||||
return 0
|
||||
}
|
||||
|
||||
const (
|
||||
t1 = 0x00 // 0000 0000
|
||||
tx = 0x80 // 1000 0000
|
||||
t2 = 0xC0 // 1100 0000
|
||||
t3 = 0xE0 // 1110 0000
|
||||
t4 = 0xF0 // 1111 0000
|
||||
t5 = 0xF8 // 1111 1000
|
||||
t6 = 0xFC // 1111 1100
|
||||
te = 0xFE // 1111 1110
|
||||
)
|
||||
|
||||
// lookup returns the trie value for the first UTF-8 encoding in s and
|
||||
// the width in bytes of this encoding. The size will be 0 if s does not
|
||||
// hold enough bytes to complete the encoding. len(s) must be greater than 0.
|
||||
func (t *trie) lookup(s []byte) (v uint16, sz int) {
|
||||
c0 := s[0]
|
||||
switch {
|
||||
case c0 < tx:
|
||||
return t.values[c0], 1
|
||||
case c0 < t2:
|
||||
return 0, 1
|
||||
case c0 < t3:
|
||||
if len(s) < 2 {
|
||||
return 0, 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
c1 := s[1]
|
||||
if c1 < tx || t2 <= c1 {
|
||||
return 0, 1
|
||||
}
|
||||
return t.lookupValue(i, c1), 2
|
||||
case c0 < t4:
|
||||
if len(s) < 3 {
|
||||
return 0, 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
c1 := s[1]
|
||||
if c1 < tx || t2 <= c1 {
|
||||
return 0, 1
|
||||
}
|
||||
o := uint16(i)<<6 + uint16(c1)
|
||||
i = t.index[o]
|
||||
c2 := s[2]
|
||||
if c2 < tx || t2 <= c2 {
|
||||
return 0, 2
|
||||
}
|
||||
return t.lookupValue(i, c2), 3
|
||||
case c0 < t5:
|
||||
if len(s) < 4 {
|
||||
return 0, 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
c1 := s[1]
|
||||
if c1 < tx || t2 <= c1 {
|
||||
return 0, 1
|
||||
}
|
||||
o := uint16(i)<<6 + uint16(c1)
|
||||
i = t.index[o]
|
||||
c2 := s[2]
|
||||
if c2 < tx || t2 <= c2 {
|
||||
return 0, 2
|
||||
}
|
||||
o = uint16(i)<<6 + uint16(c2)
|
||||
i = t.index[o]
|
||||
c3 := s[3]
|
||||
if c3 < tx || t2 <= c3 {
|
||||
return 0, 3
|
||||
}
|
||||
return t.lookupValue(i, c3), 4
|
||||
}
|
||||
// Illegal rune
|
||||
return 0, 1
|
||||
}
|
||||
|
||||
// lookupString returns the trie value for the first UTF-8 encoding in s and
|
||||
// the width in bytes of this encoding. The size will be 0 if s does not
|
||||
// hold enough bytes to complete the encoding. len(s) must be greater than 0.
|
||||
func (t *trie) lookupString(s string) (v uint16, sz int) {
|
||||
c0 := s[0]
|
||||
switch {
|
||||
case c0 < tx:
|
||||
return t.values[c0], 1
|
||||
case c0 < t2:
|
||||
return 0, 1
|
||||
case c0 < t3:
|
||||
if len(s) < 2 {
|
||||
return 0, 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
c1 := s[1]
|
||||
if c1 < tx || t2 <= c1 {
|
||||
return 0, 1
|
||||
}
|
||||
return t.lookupValue(i, c1), 2
|
||||
case c0 < t4:
|
||||
if len(s) < 3 {
|
||||
return 0, 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
c1 := s[1]
|
||||
if c1 < tx || t2 <= c1 {
|
||||
return 0, 1
|
||||
}
|
||||
o := uint16(i)<<6 + uint16(c1)
|
||||
i = t.index[o]
|
||||
c2 := s[2]
|
||||
if c2 < tx || t2 <= c2 {
|
||||
return 0, 2
|
||||
}
|
||||
return t.lookupValue(i, c2), 3
|
||||
case c0 < t5:
|
||||
if len(s) < 4 {
|
||||
return 0, 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
c1 := s[1]
|
||||
if c1 < tx || t2 <= c1 {
|
||||
return 0, 1
|
||||
}
|
||||
o := uint16(i)<<6 + uint16(c1)
|
||||
i = t.index[o]
|
||||
c2 := s[2]
|
||||
if c2 < tx || t2 <= c2 {
|
||||
return 0, 2
|
||||
}
|
||||
o = uint16(i)<<6 + uint16(c2)
|
||||
i = t.index[o]
|
||||
c3 := s[3]
|
||||
if c3 < tx || t2 <= c3 {
|
||||
return 0, 3
|
||||
}
|
||||
return t.lookupValue(i, c3), 4
|
||||
}
|
||||
// Illegal rune
|
||||
return 0, 1
|
||||
}
|
||||
|
||||
// lookupUnsafe returns the trie value for the first UTF-8 encoding in s.
|
||||
// s must hold a full encoding.
|
||||
func (t *trie) lookupUnsafe(s []byte) uint16 {
|
||||
c0 := s[0]
|
||||
if c0 < tx {
|
||||
return t.values[c0]
|
||||
}
|
||||
if c0 < t2 {
|
||||
return 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
if c0 < t3 {
|
||||
return t.lookupValue(i, s[1])
|
||||
}
|
||||
i = t.index[uint16(i)<<6+uint16(s[1])]
|
||||
if c0 < t4 {
|
||||
return t.lookupValue(i, s[2])
|
||||
}
|
||||
i = t.index[uint16(i)<<6+uint16(s[2])]
|
||||
if c0 < t5 {
|
||||
return t.lookupValue(i, s[3])
|
||||
}
|
||||
return 0
|
||||
}
|
||||
|
||||
// lookupStringUnsafe returns the trie value for the first UTF-8 encoding in s.
|
||||
// s must hold a full encoding.
|
||||
func (t *trie) lookupStringUnsafe(s string) uint16 {
|
||||
c0 := s[0]
|
||||
if c0 < tx {
|
||||
return t.values[c0]
|
||||
}
|
||||
if c0 < t2 {
|
||||
return 0
|
||||
}
|
||||
i := t.index[c0]
|
||||
if c0 < t3 {
|
||||
return t.lookupValue(i, s[1])
|
||||
}
|
||||
i = t.index[uint16(i)<<6+uint16(s[1])]
|
||||
if c0 < t4 {
|
||||
return t.lookupValue(i, s[2])
|
||||
}
|
||||
i = t.index[uint16(i)<<6+uint16(s[2])]
|
||||
if c0 < t5 {
|
||||
return t.lookupValue(i, s[3])
|
||||
}
|
||||
return 0
|
||||
}
|
||||
-152
@@ -1,152 +0,0 @@
|
||||
// Copyright 2011 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
package norm
|
||||
|
||||
import (
|
||||
"testing"
|
||||
"unicode/utf8"
|
||||
)
|
||||
|
||||
// Test data is located in triedata_test.go; generated by maketesttables.
|
||||
var testdata = testdataTrie
|
||||
|
||||
type rangeTest struct {
|
||||
block uint8
|
||||
lookup byte
|
||||
result uint16
|
||||
table []valueRange
|
||||
offsets []uint16
|
||||
}
|
||||
|
||||
var range1Off = []uint16{0, 2}
|
||||
var range1 = []valueRange{
|
||||
{0, 1, 0},
|
||||
{1, 0x80, 0x80},
|
||||
{0, 2, 0},
|
||||
{1, 0x80, 0x80},
|
||||
{9, 0xff, 0xff},
|
||||
}
|
||||
|
||||
var rangeTests = []rangeTest{
|
||||
{10, 0x80, 1, range1, range1Off},
|
||||
{10, 0x00, 0, range1, range1Off},
|
||||
{11, 0x80, 1, range1, range1Off},
|
||||
{11, 0xff, 9, range1, range1Off},
|
||||
{11, 0x00, 0, range1, range1Off},
|
||||
}
|
||||
|
||||
func TestLookupSparse(t *testing.T) {
|
||||
for i, test := range rangeTests {
|
||||
n := trie{sparse: test.table, sparseOffset: test.offsets, cutoff: 10}
|
||||
v := n.lookupValue(test.block, test.lookup)
|
||||
if v != test.result {
|
||||
t.Errorf("LookupSparse:%d: found %X; want %X", i, v, test.result)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// Test cases for illegal runes.
|
||||
type trietest struct {
|
||||
size int
|
||||
bytes []byte
|
||||
}
|
||||
|
||||
var tests = []trietest{
|
||||
// illegal runes
|
||||
{1, []byte{0x80}},
|
||||
{1, []byte{0xFF}},
|
||||
{1, []byte{t2, tx - 1}},
|
||||
{1, []byte{t2, t2}},
|
||||
{2, []byte{t3, tx, tx - 1}},
|
||||
{2, []byte{t3, tx, t2}},
|
||||
{1, []byte{t3, tx - 1, tx}},
|
||||
{3, []byte{t4, tx, tx, tx - 1}},
|
||||
{3, []byte{t4, tx, tx, t2}},
|
||||
{1, []byte{t4, t2, tx, tx - 1}},
|
||||
{2, []byte{t4, tx, t2, tx - 1}},
|
||||
|
||||
// short runes
|
||||
{0, []byte{t2}},
|
||||
{0, []byte{t3, tx}},
|
||||
{0, []byte{t4, tx, tx}},
|
||||
|
||||
// we only support UTF-8 up to utf8.UTFMax bytes (4 bytes)
|
||||
{1, []byte{t5, tx, tx, tx, tx}},
|
||||
{1, []byte{t6, tx, tx, tx, tx, tx}},
|
||||
}
|
||||
|
||||
func mkUTF8(r rune) ([]byte, int) {
|
||||
var b [utf8.UTFMax]byte
|
||||
sz := utf8.EncodeRune(b[:], r)
|
||||
return b[:sz], sz
|
||||
}
|
||||
|
||||
func TestLookup(t *testing.T) {
|
||||
for i, tt := range testRunes {
|
||||
b, szg := mkUTF8(tt)
|
||||
v, szt := testdata.lookup(b)
|
||||
if int(v) != i {
|
||||
t.Errorf("lookup(%U): found value %#x, expected %#x", tt, v, i)
|
||||
}
|
||||
if szt != szg {
|
||||
t.Errorf("lookup(%U): found size %d, expected %d", tt, szt, szg)
|
||||
}
|
||||
}
|
||||
for i, tt := range tests {
|
||||
v, sz := testdata.lookup(tt.bytes)
|
||||
if v != 0 {
|
||||
t.Errorf("lookup of illegal rune, case %d: found value %#x, expected 0", i, v)
|
||||
}
|
||||
if sz != tt.size {
|
||||
t.Errorf("lookup of illegal rune, case %d: found size %d, expected %d", i, sz, tt.size)
|
||||
}
|
||||
}
|
||||
// Verify defaults.
|
||||
if v, _ := testdata.lookup([]byte{0xC1, 0x8C}); v != 0 {
|
||||
t.Errorf("lookup of non-existing rune should be 0; found %X", v)
|
||||
}
|
||||
}
|
||||
|
||||
func TestLookupUnsafe(t *testing.T) {
|
||||
for i, tt := range testRunes {
|
||||
b, _ := mkUTF8(tt)
|
||||
v := testdata.lookupUnsafe(b)
|
||||
if int(v) != i {
|
||||
t.Errorf("lookupUnsafe(%U): found value %#x, expected %#x", i, v, i)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func TestLookupString(t *testing.T) {
|
||||
for i, tt := range testRunes {
|
||||
b, szg := mkUTF8(tt)
|
||||
v, szt := testdata.lookupString(string(b))
|
||||
if int(v) != i {
|
||||
t.Errorf("lookup(%U): found value %#x, expected %#x", i, v, i)
|
||||
}
|
||||
if szt != szg {
|
||||
t.Errorf("lookup(%U): found size %d, expected %d", i, szt, szg)
|
||||
}
|
||||
}
|
||||
for i, tt := range tests {
|
||||
v, sz := testdata.lookupString(string(tt.bytes))
|
||||
if int(v) != 0 {
|
||||
t.Errorf("lookup of illegal rune, case %d: found value %#x, expected 0", i, v)
|
||||
}
|
||||
if sz != tt.size {
|
||||
t.Errorf("lookup of illegal rune, case %d: found size %d, expected %d", i, sz, tt.size)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func TestLookupStringUnsafe(t *testing.T) {
|
||||
for i, tt := range testRunes {
|
||||
b, _ := mkUTF8(tt)
|
||||
v := testdata.lookupStringUnsafe(string(b))
|
||||
if int(v) != i {
|
||||
t.Errorf("lookupUnsafe(%U): found value %#x, expected %#x", i, v, i)
|
||||
}
|
||||
}
|
||||
}
|
||||
-85
@@ -1,85 +0,0 @@
|
||||
// Generated by running
|
||||
// maketesttables
|
||||
// DO NOT EDIT
|
||||
|
||||
package norm
|
||||
|
||||
var testRunes = []int32{1, 12, 127, 128, 256, 2047, 2048, 2457, 65535, 65536, 65793, 1114111, 512, 513, 514, 528, 533}
|
||||
|
||||
// testdataValues: 192 entries, 384 bytes
|
||||
// Block 2 is the null block.
|
||||
var testdataValues = [192]uint16{
|
||||
// Block 0x0, offset 0x0
|
||||
0x000c: 0x0001,
|
||||
// Block 0x1, offset 0x40
|
||||
0x007f: 0x0002,
|
||||
// Block 0x2, offset 0x80
|
||||
}
|
||||
|
||||
// testdataSparseOffset: 10 entries, 20 bytes
|
||||
var testdataSparseOffset = []uint16{0x0, 0x2, 0x4, 0x8, 0xa, 0xc, 0xe, 0x10, 0x12, 0x14}
|
||||
|
||||
// testdataSparseValues: 22 entries, 88 bytes
|
||||
var testdataSparseValues = [22]valueRange{
|
||||
// Block 0x0, offset 0x1
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x0003, lo: 0x80, hi: 0x80},
|
||||
// Block 0x1, offset 0x2
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x0004, lo: 0x80, hi: 0x80},
|
||||
// Block 0x2, offset 0x3
|
||||
{value: 0x0001, lo: 0x03},
|
||||
{value: 0x000c, lo: 0x80, hi: 0x82},
|
||||
{value: 0x000f, lo: 0x90, hi: 0x90},
|
||||
{value: 0x0010, lo: 0x95, hi: 0x95},
|
||||
// Block 0x3, offset 0x4
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x0005, lo: 0xbf, hi: 0xbf},
|
||||
// Block 0x4, offset 0x5
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x0006, lo: 0x80, hi: 0x80},
|
||||
// Block 0x5, offset 0x6
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x0007, lo: 0x99, hi: 0x99},
|
||||
// Block 0x6, offset 0x7
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x0008, lo: 0xbf, hi: 0xbf},
|
||||
// Block 0x7, offset 0x8
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x0009, lo: 0x80, hi: 0x80},
|
||||
// Block 0x8, offset 0x9
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x000a, lo: 0x81, hi: 0x81},
|
||||
// Block 0x9, offset 0xa
|
||||
{value: 0x0000, lo: 0x01},
|
||||
{value: 0x000b, lo: 0xbf, hi: 0xbf},
|
||||
}
|
||||
|
||||
// testdataLookup: 640 bytes
|
||||
// Block 0 is the null block.
|
||||
var testdataLookup = [640]uint8{
|
||||
// Block 0x0, offset 0x0
|
||||
// Block 0x1, offset 0x40
|
||||
// Block 0x2, offset 0x80
|
||||
// Block 0x3, offset 0xc0
|
||||
0x0c2: 0x01, 0x0c4: 0x02,
|
||||
0x0c8: 0x03,
|
||||
0x0df: 0x04,
|
||||
0x0e0: 0x02,
|
||||
0x0ef: 0x03,
|
||||
0x0f0: 0x05, 0x0f4: 0x07,
|
||||
// Block 0x4, offset 0x100
|
||||
0x120: 0x05, 0x126: 0x06,
|
||||
// Block 0x5, offset 0x140
|
||||
0x17f: 0x07,
|
||||
// Block 0x6, offset 0x180
|
||||
0x180: 0x08, 0x184: 0x09,
|
||||
// Block 0x7, offset 0x1c0
|
||||
0x1d0: 0x04,
|
||||
// Block 0x8, offset 0x200
|
||||
0x23f: 0x0a,
|
||||
// Block 0x9, offset 0x240
|
||||
0x24f: 0x06,
|
||||
}
|
||||
|
||||
var testdataTrie = trie{testdataLookup[:], testdataValues[:], testdataSparseValues[:], testdataSparseOffset[:], 1}
|
||||
-317
@@ -1,317 +0,0 @@
|
||||
// Copyright 2011 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// +build ignore
|
||||
|
||||
// Trie table generator.
|
||||
// Used by make*tables tools to generate a go file with trie data structures
|
||||
// for mapping UTF-8 to a 16-bit value. All but the last byte in a UTF-8 byte
|
||||
// sequence are used to lookup offsets in the index table to be used for the
|
||||
// next byte. The last byte is used to index into a table with 16-bit values.
|
||||
|
||||
package main
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"hash/crc32"
|
||||
"log"
|
||||
"unicode/utf8"
|
||||
)
|
||||
|
||||
const (
|
||||
blockSize = 64
|
||||
blockOffset = 2 // Subtract two blocks to compensate for the 0x80 added to continuation bytes.
|
||||
maxSparseEntries = 16
|
||||
)
|
||||
|
||||
// Intermediate trie structure
|
||||
type trieNode struct {
|
||||
table [256]*trieNode
|
||||
value int
|
||||
b byte
|
||||
leaf bool
|
||||
}
|
||||
|
||||
func newNode() *trieNode {
|
||||
return new(trieNode)
|
||||
}
|
||||
|
||||
func (n trieNode) String() string {
|
||||
s := fmt.Sprint("trieNode{table: { non-nil at index: ")
|
||||
for i, v := range n.table {
|
||||
if v != nil {
|
||||
s += fmt.Sprintf("%d, ", i)
|
||||
}
|
||||
}
|
||||
s += fmt.Sprintf("}, value:%#x, b:%#x leaf:%v}", n.value, n.b, n.leaf)
|
||||
return s
|
||||
}
|
||||
|
||||
func (n trieNode) isInternal() bool {
|
||||
internal := true
|
||||
for i := 0; i < 256; i++ {
|
||||
if nn := n.table[i]; nn != nil {
|
||||
if !internal && !nn.leaf {
|
||||
log.Fatalf("triegen: isInternal: node contains both leaf and non-leaf children (%v)", n)
|
||||
}
|
||||
internal = internal && !nn.leaf
|
||||
}
|
||||
}
|
||||
return internal
|
||||
}
|
||||
|
||||
func (n trieNode) mostFrequentStride() int {
|
||||
counts := make(map[int]int)
|
||||
v := 0
|
||||
for _, t := range n.table[0x80 : 0x80+blockSize] {
|
||||
if t != nil {
|
||||
if stride := t.value - v; v != 0 && stride >= 0 {
|
||||
counts[stride]++
|
||||
}
|
||||
v = t.value
|
||||
} else {
|
||||
v = 0
|
||||
}
|
||||
}
|
||||
var maxs, maxc int
|
||||
for stride, cnt := range counts {
|
||||
if cnt > maxc || (cnt == maxc && stride < maxs) {
|
||||
maxs, maxc = stride, cnt
|
||||
}
|
||||
}
|
||||
return maxs
|
||||
}
|
||||
|
||||
func (n trieNode) countSparseEntries() int {
|
||||
stride := n.mostFrequentStride()
|
||||
var count, v int
|
||||
for _, t := range n.table[0x80 : 0x80+blockSize] {
|
||||
tv := 0
|
||||
if t != nil {
|
||||
tv = t.value
|
||||
}
|
||||
if tv-v != stride {
|
||||
if tv != 0 {
|
||||
count++
|
||||
}
|
||||
}
|
||||
v = tv
|
||||
}
|
||||
return count
|
||||
}
|
||||
|
||||
func (n *trieNode) insert(r rune, value uint16) {
|
||||
var p [utf8.UTFMax]byte
|
||||
sz := utf8.EncodeRune(p[:], r)
|
||||
|
||||
for i := 0; i < sz; i++ {
|
||||
if n.leaf {
|
||||
log.Fatalf("triegen: insert: node (%#v) should not be a leaf", n)
|
||||
}
|
||||
nn := n.table[p[i]]
|
||||
if nn == nil {
|
||||
nn = newNode()
|
||||
nn.b = p[i]
|
||||
n.table[p[i]] = nn
|
||||
}
|
||||
n = nn
|
||||
}
|
||||
n.value = int(value)
|
||||
n.leaf = true
|
||||
}
|
||||
|
||||
type nodeIndex struct {
|
||||
lookupBlocks []*trieNode
|
||||
valueBlocks []*trieNode
|
||||
sparseBlocks []*trieNode
|
||||
sparseOffset []uint16
|
||||
sparseCount int
|
||||
|
||||
lookupBlockIdx map[uint32]int
|
||||
valueBlockIdx map[uint32]int
|
||||
}
|
||||
|
||||
func newIndex() *nodeIndex {
|
||||
index := &nodeIndex{}
|
||||
index.lookupBlocks = make([]*trieNode, 0)
|
||||
index.valueBlocks = make([]*trieNode, 0)
|
||||
index.sparseBlocks = make([]*trieNode, 0)
|
||||
index.sparseOffset = make([]uint16, 1)
|
||||
index.lookupBlockIdx = make(map[uint32]int)
|
||||
index.valueBlockIdx = make(map[uint32]int)
|
||||
return index
|
||||
}
|
||||
|
||||
func computeOffsets(index *nodeIndex, n *trieNode) int {
|
||||
if n.leaf {
|
||||
return n.value
|
||||
}
|
||||
hasher := crc32.New(crc32.MakeTable(crc32.IEEE))
|
||||
// We only index continuation bytes.
|
||||
for i := 0; i < blockSize; i++ {
|
||||
v := 0
|
||||
if nn := n.table[0x80+i]; nn != nil {
|
||||
v = computeOffsets(index, nn)
|
||||
}
|
||||
hasher.Write([]byte{uint8(v >> 8), uint8(v)})
|
||||
}
|
||||
h := hasher.Sum32()
|
||||
if n.isInternal() {
|
||||
v, ok := index.lookupBlockIdx[h]
|
||||
if !ok {
|
||||
v = len(index.lookupBlocks) - blockOffset
|
||||
index.lookupBlocks = append(index.lookupBlocks, n)
|
||||
index.lookupBlockIdx[h] = v
|
||||
}
|
||||
n.value = v
|
||||
} else {
|
||||
v, ok := index.valueBlockIdx[h]
|
||||
if !ok {
|
||||
if c := n.countSparseEntries(); c > maxSparseEntries {
|
||||
v = len(index.valueBlocks) - blockOffset
|
||||
index.valueBlocks = append(index.valueBlocks, n)
|
||||
index.valueBlockIdx[h] = v
|
||||
} else {
|
||||
v = -len(index.sparseOffset)
|
||||
index.sparseBlocks = append(index.sparseBlocks, n)
|
||||
index.sparseOffset = append(index.sparseOffset, uint16(index.sparseCount))
|
||||
index.sparseCount += c + 1
|
||||
index.valueBlockIdx[h] = v
|
||||
}
|
||||
}
|
||||
n.value = v
|
||||
}
|
||||
return n.value
|
||||
}
|
||||
|
||||
func printValueBlock(nr int, n *trieNode, offset int) {
|
||||
boff := nr * blockSize
|
||||
fmt.Printf("\n// Block %#x, offset %#x", nr, boff)
|
||||
var printnewline bool
|
||||
for i := 0; i < blockSize; i++ {
|
||||
if i%6 == 0 {
|
||||
printnewline = true
|
||||
}
|
||||
v := 0
|
||||
if nn := n.table[i+offset]; nn != nil {
|
||||
v = nn.value
|
||||
}
|
||||
if v != 0 {
|
||||
if printnewline {
|
||||
fmt.Printf("\n")
|
||||
printnewline = false
|
||||
}
|
||||
fmt.Printf("%#04x:%#04x, ", boff+i, v)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func printSparseBlock(nr int, n *trieNode) {
|
||||
boff := -n.value
|
||||
fmt.Printf("\n// Block %#x, offset %#x", nr, boff)
|
||||
v := 0
|
||||
//stride := f(n)
|
||||
stride := n.mostFrequentStride()
|
||||
c := n.countSparseEntries()
|
||||
fmt.Printf("\n{value:%#04x,lo:%#02x},", stride, uint8(c))
|
||||
for i, nn := range n.table[0x80 : 0x80+blockSize] {
|
||||
nv := 0
|
||||
if nn != nil {
|
||||
nv = nn.value
|
||||
}
|
||||
if nv-v != stride {
|
||||
if v != 0 {
|
||||
fmt.Printf(",hi:%#02x},", 0x80+i-1)
|
||||
}
|
||||
if nv != 0 {
|
||||
fmt.Printf("\n{value:%#04x,lo:%#02x", nv, nn.b)
|
||||
}
|
||||
}
|
||||
v = nv
|
||||
}
|
||||
if v != 0 {
|
||||
fmt.Printf(",hi:%#02x},", 0x80+blockSize-1)
|
||||
}
|
||||
}
|
||||
|
||||
func printLookupBlock(nr int, n *trieNode, offset, cutoff int) {
|
||||
boff := nr * blockSize
|
||||
fmt.Printf("\n// Block %#x, offset %#x", nr, boff)
|
||||
var printnewline bool
|
||||
for i := 0; i < blockSize; i++ {
|
||||
if i%8 == 0 {
|
||||
printnewline = true
|
||||
}
|
||||
v := 0
|
||||
if nn := n.table[i+offset]; nn != nil {
|
||||
v = nn.value
|
||||
}
|
||||
if v != 0 {
|
||||
if v < 0 {
|
||||
v = -v - 1 + cutoff
|
||||
}
|
||||
if printnewline {
|
||||
fmt.Printf("\n")
|
||||
printnewline = false
|
||||
}
|
||||
fmt.Printf("%#03x:%#02x, ", boff+i, v)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// printTables returns the size in bytes of the generated tables.
|
||||
func (t *trieNode) printTables(name string) int {
|
||||
index := newIndex()
|
||||
// Values for 7-bit ASCII are stored in first two block, followed by nil block.
|
||||
index.valueBlocks = append(index.valueBlocks, nil, nil, nil)
|
||||
// First byte of multi-byte UTF-8 codepoints are indexed in 4th block.
|
||||
index.lookupBlocks = append(index.lookupBlocks, nil, nil, nil, nil)
|
||||
// Index starter bytes of multi-byte UTF-8.
|
||||
for i := 0xC0; i < 0x100; i++ {
|
||||
if t.table[i] != nil {
|
||||
computeOffsets(index, t.table[i])
|
||||
}
|
||||
}
|
||||
|
||||
nv := len(index.valueBlocks) * blockSize
|
||||
fmt.Printf("// %sValues: %d entries, %d bytes\n", name, nv, nv*2)
|
||||
fmt.Printf("// Block 2 is the null block.\n")
|
||||
fmt.Printf("var %sValues = [%d]uint16 {", name, nv)
|
||||
printValueBlock(0, t, 0)
|
||||
printValueBlock(1, t, 64)
|
||||
printValueBlock(2, newNode(), 0)
|
||||
for i := 3; i < len(index.valueBlocks); i++ {
|
||||
printValueBlock(i, index.valueBlocks[i], 0x80)
|
||||
}
|
||||
fmt.Print("\n}\n\n")
|
||||
|
||||
ls := len(index.sparseBlocks)
|
||||
fmt.Printf("// %sSparseOffset: %d entries, %d bytes\n", name, ls, ls*2)
|
||||
fmt.Printf("var %sSparseOffset = %#v\n\n", name, index.sparseOffset[1:])
|
||||
|
||||
ns := index.sparseCount
|
||||
fmt.Printf("// %sSparseValues: %d entries, %d bytes\n", name, ns, ns*4)
|
||||
fmt.Printf("var %sSparseValues = [%d]valueRange {", name, ns)
|
||||
for i, n := range index.sparseBlocks {
|
||||
printSparseBlock(i, n)
|
||||
}
|
||||
fmt.Print("\n}\n\n")
|
||||
|
||||
cutoff := len(index.valueBlocks) - blockOffset
|
||||
ni := len(index.lookupBlocks) * blockSize
|
||||
fmt.Printf("// %sLookup: %d bytes\n", name, ni)
|
||||
fmt.Printf("// Block 0 is the null block.\n")
|
||||
fmt.Printf("var %sLookup = [%d]uint8 {", name, ni)
|
||||
printLookupBlock(0, newNode(), 0, cutoff)
|
||||
printLookupBlock(1, newNode(), 0, cutoff)
|
||||
printLookupBlock(2, newNode(), 0, cutoff)
|
||||
printLookupBlock(3, t, 0xC0, cutoff)
|
||||
for i := 4; i < len(index.lookupBlocks); i++ {
|
||||
printLookupBlock(i, index.lookupBlocks[i], 0x80, cutoff)
|
||||
}
|
||||
fmt.Print("\n}\n\n")
|
||||
fmt.Printf("var %sTrie = trie{ %sLookup[:], %sValues[:], %sSparseValues[:], %sSparseOffset[:], %d}\n\n",
|
||||
name, name, name, name, name, cutoff)
|
||||
return nv*2 + ns*4 + ni + ls*2
|
||||
}
|
||||
Generated
Vendored
File renamed without changes.
Generated
Vendored
+1
-1
@@ -8,11 +8,11 @@ package bcrypt
|
||||
|
||||
// The code is a port of Provos and Mazières's C implementation.
|
||||
import (
|
||||
"code.google.com/p/go.crypto/blowfish"
|
||||
"crypto/rand"
|
||||
"crypto/subtle"
|
||||
"errors"
|
||||
"fmt"
|
||||
"golang.org/x/crypto/blowfish"
|
||||
"io"
|
||||
"strconv"
|
||||
)
|
||||
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
Generated
Vendored
+2
-2
@@ -8,8 +8,8 @@ import (
|
||||
"fmt"
|
||||
"unicode"
|
||||
|
||||
"code.google.com/p/go.text/transform"
|
||||
"code.google.com/p/go.text/unicode/norm"
|
||||
"golang.org/x/text/transform"
|
||||
"golang.org/x/text/unicode/norm"
|
||||
)
|
||||
|
||||
func ExampleRemoveFunc() {
|
||||
Generated
Vendored
+33
-9
@@ -55,10 +55,17 @@ type Transformer interface {
|
||||
// either error may be returned. Other than the error conditions listed
|
||||
// here, implementations are free to report other errors that arise.
|
||||
Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error)
|
||||
|
||||
// Reset resets the state and allows a Transformer to be reused.
|
||||
Reset()
|
||||
}
|
||||
|
||||
// TODO: Do we require that a Transformer be reusable if it returns a nil error
|
||||
// or do we always require a reset after use? Is Reset mandatory or optional?
|
||||
// NopResetter can be embedded by implementations of Transformer to add a nop
|
||||
// Reset method.
|
||||
type NopResetter struct{}
|
||||
|
||||
// Reset implements the Reset method of the Transformer interface.
|
||||
func (NopResetter) Reset() {}
|
||||
|
||||
// Reader wraps another io.Reader by transforming the bytes read.
|
||||
type Reader struct {
|
||||
@@ -84,8 +91,9 @@ type Reader struct {
|
||||
const defaultBufSize = 4096
|
||||
|
||||
// NewReader returns a new Reader that wraps r by transforming the bytes read
|
||||
// via t.
|
||||
// via t. It calls Reset on t.
|
||||
func NewReader(r io.Reader, t Transformer) *Reader {
|
||||
t.Reset()
|
||||
return &Reader{
|
||||
r: r,
|
||||
t: t,
|
||||
@@ -170,8 +178,9 @@ type Writer struct {
|
||||
}
|
||||
|
||||
// NewWriter returns a new Writer that wraps w by transforming the bytes written
|
||||
// via t.
|
||||
// via t. It calls Reset on t.
|
||||
func NewWriter(w io.Writer, t Transformer) *Writer {
|
||||
t.Reset()
|
||||
return &Writer{
|
||||
w: w,
|
||||
t: t,
|
||||
@@ -247,7 +256,7 @@ func (w *Writer) Close() error {
|
||||
return nil
|
||||
}
|
||||
|
||||
type nop struct{}
|
||||
type nop struct{ NopResetter }
|
||||
|
||||
func (nop) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
n := copy(dst, src)
|
||||
@@ -257,7 +266,7 @@ func (nop) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
return n, n, err
|
||||
}
|
||||
|
||||
type discard struct{}
|
||||
type discard struct{ NopResetter }
|
||||
|
||||
func (discard) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
return 0, len(src), nil
|
||||
@@ -327,6 +336,16 @@ func Chain(t ...Transformer) Transformer {
|
||||
return c
|
||||
}
|
||||
|
||||
// Reset resets the state of Chain. It calls Reset on all the Transformers.
|
||||
func (c *chain) Reset() {
|
||||
for i, l := range c.link {
|
||||
if l.t != nil {
|
||||
l.t.Reset()
|
||||
}
|
||||
c.link[i].p, c.link[i].n = 0, 0
|
||||
}
|
||||
}
|
||||
|
||||
// Transform applies the transformers of c in sequence.
|
||||
func (c *chain) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
// Set up src and dst in the chain.
|
||||
@@ -425,6 +444,8 @@ func RemoveFunc(f func(r rune) bool) Transformer {
|
||||
|
||||
type removeF func(r rune) bool
|
||||
|
||||
func (removeF) Reset() {}
|
||||
|
||||
// Transform implements the Transformer interface.
|
||||
func (t removeF) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
for r, sz := rune(0), 0; len(src) > 0; src = src[sz:] {
|
||||
@@ -436,7 +457,7 @@ func (t removeF) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err err
|
||||
|
||||
if sz == 1 {
|
||||
// Invalid rune.
|
||||
if !atEOF && !utf8.FullRune(src[nSrc:]) {
|
||||
if !atEOF && !utf8.FullRune(src) {
|
||||
err = ErrShortSrc
|
||||
break
|
||||
}
|
||||
@@ -485,12 +506,14 @@ func grow(b []byte, n int) []byte {
|
||||
const initialBufSize = 128
|
||||
|
||||
// String returns a string with the result of converting s[:n] using t, where
|
||||
// n <= len(s). If err == nil, n will be len(s).
|
||||
// n <= len(s). If err == nil, n will be len(s). It calls Reset on t.
|
||||
func String(t Transformer, s string) (result string, n int, err error) {
|
||||
if s == "" {
|
||||
return "", 0, nil
|
||||
}
|
||||
|
||||
t.Reset()
|
||||
|
||||
// Allocate only once. Note that both dst and src escape when passed to
|
||||
// Transform.
|
||||
buf := [2 * initialBufSize]byte{}
|
||||
@@ -571,8 +594,9 @@ func String(t Transformer, s string) (result string, n int, err error) {
|
||||
}
|
||||
|
||||
// Bytes returns a new byte slice with the result of converting b[:n] using t,
|
||||
// where n <= len(b). If err == nil, n will be len(b).
|
||||
// where n <= len(b). If err == nil, n will be len(b). It calls Reset on t.
|
||||
func Bytes(t Transformer, b []byte) (result []byte, n int, err error) {
|
||||
t.Reset()
|
||||
dst := make([]byte, len(b))
|
||||
pDst, pSrc := 0, 0
|
||||
for {
|
||||
Generated
Vendored
+22
-15
@@ -16,7 +16,7 @@ import (
|
||||
"unicode/utf8"
|
||||
)
|
||||
|
||||
type lowerCaseASCII struct{}
|
||||
type lowerCaseASCII struct{ transform.NopResetter }
|
||||
|
||||
func (lowerCaseASCII) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
n := len(src)
|
||||
@@ -34,7 +34,7 @@ func (lowerCaseASCII) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, er
|
||||
|
||||
var errYouMentionedX = errors.New("you mentioned X")
|
||||
|
||||
type dontMentionX struct{}
|
||||
type dontMentionX struct{ transform.NopResetter }
|
||||
|
||||
func (dontMentionX) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
n := len(src)
|
||||
@@ -52,7 +52,7 @@ func (dontMentionX) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err
|
||||
|
||||
// doublerAtEOF is a strange Transformer that transforms "this" to "tthhiiss",
|
||||
// but only if atEOF is true.
|
||||
type doublerAtEOF struct{}
|
||||
type doublerAtEOF struct{ transform.NopResetter }
|
||||
|
||||
func (doublerAtEOF) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
if !atEOF {
|
||||
@@ -71,7 +71,7 @@ func (doublerAtEOF) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err
|
||||
// rleDecode and rleEncode implement a toy run-length encoding: "aabbbbbbbbbb"
|
||||
// is encoded as "2a10b". The decoding is assumed to not contain any numbers.
|
||||
|
||||
type rleDecode struct{}
|
||||
type rleDecode struct{ transform.NopResetter }
|
||||
|
||||
func (rleDecode) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
loop:
|
||||
@@ -104,6 +104,8 @@ loop:
|
||||
}
|
||||
|
||||
type rleEncode struct {
|
||||
transform.NopResetter
|
||||
|
||||
// allowStutter means that "xxxxxxxx" can be encoded as "5x3x"
|
||||
// instead of always as "8x".
|
||||
allowStutter bool
|
||||
@@ -136,6 +138,10 @@ func (e rleEncode) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err e
|
||||
// trickler consumes all input bytes, but writes a single byte at a time to dst.
|
||||
type trickler []byte
|
||||
|
||||
func (t *trickler) Reset() {
|
||||
*t = nil
|
||||
}
|
||||
|
||||
func (t *trickler) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
*t = append(*t, src...)
|
||||
if len(*t) == 0 {
|
||||
@@ -157,6 +163,9 @@ func (t *trickler) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err e
|
||||
// to have some tolerance as long as progress can be detected.
|
||||
type delayedTrickler []byte
|
||||
|
||||
func (t *delayedTrickler) Reset() {
|
||||
*t = nil
|
||||
}
|
||||
func (t *delayedTrickler) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
if len(*t) > 0 && len(dst) > 0 {
|
||||
dst[0] = (*t)[0]
|
||||
@@ -447,7 +456,6 @@ var testCases = []testCase{
|
||||
|
||||
func TestReader(t *testing.T) {
|
||||
for _, tc := range testCases {
|
||||
reset(tc.t)
|
||||
r := NewReader(strings.NewReader(tc.src), tc.t)
|
||||
// Differently sized dst and src buffers are not part of the
|
||||
// exported API. We override them manually.
|
||||
@@ -461,13 +469,6 @@ func TestReader(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func reset(t Transformer) {
|
||||
var dst [128]byte
|
||||
for err := ErrShortDst; err != nil; {
|
||||
_, _, err = t.Transform(dst[:], nil, true)
|
||||
}
|
||||
}
|
||||
|
||||
func TestWriter(t *testing.T) {
|
||||
tests := append(testCases, chainTests()...)
|
||||
for _, tc := range tests {
|
||||
@@ -477,7 +478,6 @@ func TestWriter(t *testing.T) {
|
||||
}
|
||||
for _, sz := range sizes {
|
||||
bb := &bytes.Buffer{}
|
||||
reset(tc.t)
|
||||
w := NewWriter(bb, tc.t)
|
||||
// Differently sized dst and src buffers are not part of the
|
||||
// exported API. We override them manually.
|
||||
@@ -735,7 +735,7 @@ func chainTests() []testCase {
|
||||
}
|
||||
|
||||
func doTransform(tc testCase) (res string, iter int, err error) {
|
||||
reset(tc.t)
|
||||
tc.t.Reset()
|
||||
dst := make([]byte, tc.dstSize)
|
||||
out, in := make([]byte, 0, 2*len(tc.src)), []byte(tc.src)
|
||||
for {
|
||||
@@ -884,6 +884,14 @@ func TestRemoveFunc(t *testing.T) {
|
||||
wantIter: 2,
|
||||
},
|
||||
|
||||
{
|
||||
// Test a long buffer greater than the internal buffer size
|
||||
src: "hello\xcc\xcc\xccworld",
|
||||
srcSize: 13,
|
||||
wantStr: "hello\uFFFD\uFFFD\uFFFDworld",
|
||||
wantIter: 1,
|
||||
},
|
||||
|
||||
{
|
||||
src: "\u2345",
|
||||
dstSize: 2,
|
||||
@@ -947,7 +955,6 @@ func testString(t *testing.T, f func(Transformer, string) (string, int, error))
|
||||
// that depend on a specific buffer size being set.
|
||||
continue
|
||||
}
|
||||
reset(tt.t)
|
||||
if tt.wantErr == ErrShortDst || tt.wantErr == ErrShortSrc {
|
||||
// The result string will be different.
|
||||
continue
|
||||
Generated
Vendored
-7
@@ -5,9 +5,6 @@
|
||||
maketables: maketables.go triegen.go
|
||||
go build $^
|
||||
|
||||
maketesttables: maketesttables.go triegen.go
|
||||
go build $^
|
||||
|
||||
normregtest: normregtest.go
|
||||
go build $^
|
||||
|
||||
@@ -15,10 +12,6 @@ tables: maketables
|
||||
./maketables > tables.go
|
||||
gofmt -w tables.go
|
||||
|
||||
trietesttables: maketesttables
|
||||
./maketesttables > triedata_test.go
|
||||
gofmt -w triedata_test.go
|
||||
|
||||
# Downloads from www.unicode.org, so not part
|
||||
# of standard test scripts.
|
||||
test: testtables regtest
|
||||
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
Generated
Vendored
+1
-1
@@ -9,7 +9,7 @@ import (
|
||||
"fmt"
|
||||
"unicode/utf8"
|
||||
|
||||
"code.google.com/p/go.text/unicode/norm"
|
||||
"golang.org/x/text/unicode/norm"
|
||||
)
|
||||
|
||||
// EqualSimple uses a norm.Iter to compare two non-normalized
|
||||
Generated
Vendored
+4
-4
@@ -201,17 +201,17 @@ func lookupInfoNFKC(b input, i int) Properties {
|
||||
// Properties returns properties for the first rune in s.
|
||||
func (f Form) Properties(s []byte) Properties {
|
||||
if f == NFC || f == NFD {
|
||||
return compInfo(nfcTrie.lookup(s))
|
||||
return compInfo(nfcData.lookup(s))
|
||||
}
|
||||
return compInfo(nfkcTrie.lookup(s))
|
||||
return compInfo(nfkcData.lookup(s))
|
||||
}
|
||||
|
||||
// PropertiesString returns properties for the first rune in s.
|
||||
func (f Form) PropertiesString(s string) Properties {
|
||||
if f == NFC || f == NFD {
|
||||
return compInfo(nfcTrie.lookupString(s))
|
||||
return compInfo(nfcData.lookupString(s))
|
||||
}
|
||||
return compInfo(nfkcTrie.lookupString(s))
|
||||
return compInfo(nfkcData.lookupString(s))
|
||||
}
|
||||
|
||||
// compInfo converts the information contained in v and sz
|
||||
Generated
Vendored
File renamed without changes.
Generated
Vendored
+4
-4
@@ -77,16 +77,16 @@ func (in *input) copySlice(buf []byte, b, e int) int {
|
||||
|
||||
func (in *input) charinfoNFC(p int) (uint16, int) {
|
||||
if in.bytes == nil {
|
||||
return nfcTrie.lookupString(in.str[p:])
|
||||
return nfcData.lookupString(in.str[p:])
|
||||
}
|
||||
return nfcTrie.lookup(in.bytes[p:])
|
||||
return nfcData.lookup(in.bytes[p:])
|
||||
}
|
||||
|
||||
func (in *input) charinfoNFKC(p int) (uint16, int) {
|
||||
if in.bytes == nil {
|
||||
return nfkcTrie.lookupString(in.str[p:])
|
||||
return nfkcData.lookupString(in.str[p:])
|
||||
}
|
||||
return nfkcTrie.lookup(in.bytes[p:])
|
||||
return nfkcData.lookup(in.bytes[p:])
|
||||
}
|
||||
|
||||
func (in *input) hangul(p int) (r rune) {
|
||||
Generated
Vendored
+2
@@ -9,6 +9,8 @@ import (
|
||||
"unicode/utf8"
|
||||
)
|
||||
|
||||
// MaxSegmentSize is the maximum size of a byte buffer needed to consider any
|
||||
// sequence of starter and non-starter runes for the purpose of normalization.
|
||||
const MaxSegmentSize = maxByteBufferSize
|
||||
|
||||
// An Iter iterates over a string or byte slice, while normalizing it
|
||||
Generated
Vendored
File renamed without changes.
Generated
Vendored
+12
-6
@@ -24,7 +24,8 @@ import (
|
||||
"strings"
|
||||
"unicode"
|
||||
|
||||
"code.google.com/p/go.text/internal/ucd"
|
||||
"golang.org/x/text/internal/triegen"
|
||||
"golang.org/x/text/internal/ucd"
|
||||
)
|
||||
|
||||
func main() {
|
||||
@@ -714,13 +715,14 @@ func printCharInfoTables() int {
|
||||
|
||||
varnames := []string{"nfc", "nfkc"}
|
||||
for i := 0; i < FNumberOfFormTypes; i++ {
|
||||
trie := newNode()
|
||||
trie := triegen.NewTrie(varnames[i])
|
||||
|
||||
for r, c := range chars {
|
||||
f := c.forms[i]
|
||||
d := f.expandedDecomp
|
||||
if len(d) != 0 {
|
||||
_, key := mkstr(c.codePoint, &f)
|
||||
trie.insert(rune(r), positionMap[key])
|
||||
trie.Insert(rune(r), uint64(positionMap[key]))
|
||||
if c.ccc != ccc(d[0]) {
|
||||
// We assume the lead ccc of a decomposition !=0 in this case.
|
||||
if ccc(d[0]) == 0 {
|
||||
@@ -730,12 +732,16 @@ func printCharInfoTables() int {
|
||||
} else if c.nLeadingNonStarters > 0 && len(f.expandedDecomp) == 0 && c.ccc == 0 && !f.combinesBackward {
|
||||
// Handle cases where it can't be detected that the nLead should be equal
|
||||
// to nTrail.
|
||||
trie.insert(c.codePoint, positionMap[nLeadStr])
|
||||
trie.Insert(c.codePoint, uint64(positionMap[nLeadStr]))
|
||||
} else if v := makeEntry(&f, &c)<<8 | uint16(c.ccc); v != 0 {
|
||||
trie.insert(c.codePoint, 0x8000|v)
|
||||
trie.Insert(c.codePoint, uint64(0x8000|v))
|
||||
}
|
||||
}
|
||||
size += trie.printTables(varnames[i])
|
||||
sz, err := trie.Gen(os.Stdout, triegen.Compact(&normCompacter{name: varnames[i]}))
|
||||
if err != nil {
|
||||
logger.Fatal(err)
|
||||
}
|
||||
size += sz
|
||||
}
|
||||
return size
|
||||
}
|
||||
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
Generated
Vendored
+1
-1
@@ -23,7 +23,7 @@ import (
|
||||
"unicode"
|
||||
"unicode/utf8"
|
||||
|
||||
"code.google.com/p/go.text/unicode/norm"
|
||||
"golang.org/x/text/unicode/norm"
|
||||
)
|
||||
|
||||
func main() {
|
||||
Generated
Vendored
File renamed without changes.
Generated
Vendored
File renamed without changes.
+7549
File diff suppressed because it is too large.
Load diff
Generated
Vendored
+8
-5
@@ -7,13 +7,16 @@ package norm
|
||||
import (
|
||||
"unicode/utf8"
|
||||
|
||||
"code.google.com/p/go.text/transform"
|
||||
"golang.org/x/text/transform"
|
||||
)
|
||||
|
||||
// Transform implements the transform.Transformer interface. It may need to
|
||||
// write segments of up to MaxSegmentSize at once. Users should either catch
|
||||
// ErrShortDst and allow dst to grow or have dst be at least of size
|
||||
// MaxTransformChunkSize to be guaranteed of progress.
|
||||
// Reset implements the Reset method of the transform.Transformer interface.
|
||||
func (Form) Reset() {}
|
||||
|
||||
// Transform implements the Transform method of the transform.Transformer
|
||||
// interface. It may need to write segments of up to MaxSegmentSize at once.
|
||||
// Users should either catch ErrShortDst and allow dst to grow or have dst be at
|
||||
// least of size MaxTransformChunkSize to be guaranteed of progress.
|
||||
func (f Form) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) {
|
||||
n := 0
|
||||
// Cap the maximum number of src bytes to check.
|
||||
Generated
Vendored
+1
-1
@@ -8,7 +8,7 @@ import (
|
||||
"fmt"
|
||||
"testing"
|
||||
|
||||
"code.google.com/p/go.text/transform"
|
||||
"golang.org/x/text/transform"
|
||||
)
|
||||
|
||||
func TestTransform(t *testing.T) {
|
||||
+54
@@ -0,0 +1,54 @@
|
||||
// Copyright 2011 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
package norm
|
||||
|
||||
type valueRange struct {
|
||||
value uint16 // header: value:stride
|
||||
lo, hi byte // header: lo:n
|
||||
}
|
||||
|
||||
type sparseBlocks struct {
|
||||
values []valueRange
|
||||
offset []uint16
|
||||
}
|
||||
|
||||
var nfcSparse = sparseBlocks{
|
||||
values: nfcSparseValues[:],
|
||||
offset: nfcSparseOffset[:],
|
||||
}
|
||||
|
||||
var nfkcSparse = sparseBlocks{
|
||||
values: nfkcSparseValues[:],
|
||||
offset: nfkcSparseOffset[:],
|
||||
}
|
||||
|
||||
var (
|
||||
nfcData = newNfcTrie(0)
|
||||
nfkcData = newNfkcTrie(0)
|
||||
)
|
||||
|
||||
// lookupValue determines the type of block n and looks up the value for b.
|
||||
// For n < t.cutoff, the block is a simple lookup table. Otherwise, the block
|
||||
// is a list of ranges with an accompanying value. Given a matching range r,
|
||||
// the value for b is by r.value + (b - r.lo) * stride.
|
||||
func (t *sparseBlocks) lookup(n uint32, b byte) uint16 {
|
||||
offset := t.offset[n]
|
||||
header := t.values[offset]
|
||||
lo := offset + 1
|
||||
hi := lo + uint16(header.lo)
|
||||
for lo < hi {
|
||||
m := lo + (hi-lo)/2
|
||||
r := t.values[m]
|
||||
if r.lo <= b && b <= r.hi {
|
||||
return r.value + uint16(b-r.lo)*header.value
|
||||
}
|
||||
if b < r.lo {
|
||||
hi = m
|
||||
} else {
|
||||
lo = m + 1
|
||||
}
|
||||
}
|
||||
return 0
|
||||
}
|
||||
+117
@@ -0,0 +1,117 @@
|
||||
// Copyright 2011 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// +build ignore
|
||||
|
||||
// Trie table generator.
|
||||
// Used by make*tables tools to generate a go file with trie data structures
|
||||
// for mapping UTF-8 to a 16-bit value. All but the last byte in a UTF-8 byte
|
||||
// sequence are used to lookup offsets in the index table to be used for the
|
||||
// next byte. The last byte is used to index into a table with 16-bit values.
|
||||
|
||||
package main
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"io"
|
||||
)
|
||||
|
||||
const maxSparseEntries = 16
|
||||
|
||||
type normCompacter struct {
|
||||
sparseBlocks [][]uint64
|
||||
sparseOffset []uint16
|
||||
sparseCount int
|
||||
name string
|
||||
}
|
||||
|
||||
func mostFrequentStride(a []uint64) int {
|
||||
counts := make(map[int]int)
|
||||
var v int
|
||||
for _, x := range a {
|
||||
if stride := int(x) - v; v != 0 && stride >= 0 {
|
||||
counts[stride]++
|
||||
}
|
||||
v = int(x)
|
||||
}
|
||||
var maxs, maxc int
|
||||
for stride, cnt := range counts {
|
||||
if cnt > maxc || (cnt == maxc && stride < maxs) {
|
||||
maxs, maxc = stride, cnt
|
||||
}
|
||||
}
|
||||
return maxs
|
||||
}
|
||||
|
||||
func countSparseEntries(a []uint64) int {
|
||||
stride := mostFrequentStride(a)
|
||||
var v, count int
|
||||
for _, tv := range a {
|
||||
if int(tv)-v != stride {
|
||||
if tv != 0 {
|
||||
count++
|
||||
}
|
||||
}
|
||||
v = int(tv)
|
||||
}
|
||||
return count
|
||||
}
|
||||
|
||||
func (c *normCompacter) Size(v []uint64) (sz int, ok bool) {
|
||||
if n := countSparseEntries(v); n <= maxSparseEntries {
|
||||
return (n+1)*4 + 2, true
|
||||
}
|
||||
return 0, false
|
||||
}
|
||||
|
||||
func (c *normCompacter) Store(v []uint64) uint32 {
|
||||
h := uint32(len(c.sparseOffset))
|
||||
c.sparseBlocks = append(c.sparseBlocks, v)
|
||||
c.sparseOffset = append(c.sparseOffset, uint16(c.sparseCount))
|
||||
c.sparseCount += countSparseEntries(v) + 1
|
||||
return h
|
||||
}
|
||||
|
||||
func (c *normCompacter) Handler() string {
|
||||
return c.name + "Sparse.lookup"
|
||||
}
|
||||
|
||||
func (c *normCompacter) Print(w io.Writer) (retErr error) {
|
||||
p := func(f string, x ...interface{}) {
|
||||
if _, err := fmt.Fprintf(w, f, x...); retErr == nil && err != nil {
|
||||
retErr = err
|
||||
}
|
||||
}
|
||||
|
||||
ls := len(c.sparseBlocks)
|
||||
p("// %sSparseOffset: %d entries, %d bytes\n", c.name, ls, ls*2)
|
||||
p("var %sSparseOffset = %#v\n\n", c.name, c.sparseOffset)
|
||||
|
||||
ns := c.sparseCount
|
||||
p("// %sSparseValues: %d entries, %d bytes\n", c.name, ns, ns*4)
|
||||
p("var %sSparseValues = [%d]valueRange {", c.name, ns)
|
||||
for i, b := range c.sparseBlocks {
|
||||
p("\n// Block %#x, offset %#x", i, c.sparseOffset[i])
|
||||
var v int
|
||||
stride := mostFrequentStride(b)
|
||||
n := countSparseEntries(b)
|
||||
p("\n{value:%#04x,lo:%#02x},", stride, uint8(n))
|
||||
for i, nv := range b {
|
||||
if int(nv)-v != stride {
|
||||
if v != 0 {
|
||||
p(",hi:%#02x},", 0x80+i-1)
|
||||
}
|
||||
if nv != 0 {
|
||||
p("\n{value:%#04x,lo:%#02x", nv, 0x80+i)
|
||||
}
|
||||
}
|
||||
v = int(nv)
|
||||
}
|
||||
if v != 0 {
|
||||
p(",hi:%#02x},", 0x80+len(b)-1)
|
||||
}
|
||||
}
|
||||
p("\n}\n\n")
|
||||
return
|
||||
}
|
||||
@@ -167,9 +167,9 @@ func checkRequiredGoVersion() {
|
||||
}
|
||||
|
||||
func setup() {
|
||||
runPrint("go", "get", "-v", "code.google.com/p/go.tools/cmd/cover")
|
||||
runPrint("go", "get", "-v", "code.google.com/p/go.tools/cmd/vet")
|
||||
runPrint("go", "get", "-v", "code.google.com/p/go.net/html")
|
||||
runPrint("go", "get", "-v", "golang.org/x/tools/cmd/cover")
|
||||
runPrint("go", "get", "-v", "golang.org/x/tools/cmd/vet")
|
||||
runPrint("go", "get", "-v", "golang.org/x/net/html")
|
||||
runPrint("go", "get", "-v", "github.com/tools/godep")
|
||||
}
|
||||
|
||||
|
||||
@@ -31,7 +31,6 @@ import (
|
||||
"sync"
|
||||
"time"
|
||||
|
||||
"code.google.com/p/go.crypto/bcrypt"
|
||||
"github.com/calmh/logger"
|
||||
"github.com/syncthing/syncthing/internal/auto"
|
||||
"github.com/syncthing/syncthing/internal/config"
|
||||
@@ -42,6 +41,7 @@ import (
|
||||
"github.com/syncthing/syncthing/internal/protocol"
|
||||
"github.com/syncthing/syncthing/internal/upgrade"
|
||||
"github.com/vitrun/qart/qr"
|
||||
"golang.org/x/crypto/bcrypt"
|
||||
)
|
||||
|
||||
type guiError struct {
|
||||
|
||||
@@ -24,8 +24,8 @@ import (
|
||||
"sync"
|
||||
"time"
|
||||
|
||||
"code.google.com/p/go.crypto/bcrypt"
|
||||
"github.com/syncthing/syncthing/internal/config"
|
||||
"golang.org/x/crypto/bcrypt"
|
||||
)
|
||||
|
||||
var (
|
||||
|
||||
@@ -36,7 +36,6 @@ import (
|
||||
"strings"
|
||||
"time"
|
||||
|
||||
"code.google.com/p/go.crypto/bcrypt"
|
||||
"github.com/calmh/logger"
|
||||
"github.com/juju/ratelimit"
|
||||
"github.com/syncthing/syncthing/internal/config"
|
||||
@@ -50,6 +49,7 @@ import (
|
||||
"github.com/syncthing/syncthing/internal/upnp"
|
||||
"github.com/syndtr/goleveldb/leveldb"
|
||||
"github.com/syndtr/goleveldb/leveldb/opt"
|
||||
"golang.org/x/crypto/bcrypt"
|
||||
)
|
||||
|
||||
var (
|
||||
|
||||
@@ -24,7 +24,7 @@ import (
|
||||
"regexp"
|
||||
"strings"
|
||||
|
||||
"code.google.com/p/go.net/html"
|
||||
"golang.org/x/net/html"
|
||||
)
|
||||
|
||||
var trans = make(map[string]string)
|
||||
|
||||
@@ -42,6 +42,7 @@
|
||||
"Folder ID": "Folder ID",
|
||||
"Folder Master": "Folder Master",
|
||||
"Folder Path": "Folder Path",
|
||||
"Folders": "Folders",
|
||||
"GUI Authentication Password": "GUI Authentication Password",
|
||||
"GUI Authentication User": "GUI Authentication User",
|
||||
"GUI Listen Addresses": "GUI Listen Addresses",
|
||||
@@ -131,6 +132,7 @@
|
||||
"The rescan interval must be at least 5 seconds.": "The rescan interval must be at least 5 seconds.",
|
||||
"Unknown": "Unknown",
|
||||
"Unshared": "Unshared",
|
||||
"Unused": "Unused",
|
||||
"Up to Date": "Up to Date",
|
||||
"Upgrade To {%version%}": "Upgrade To {{version}}",
|
||||
"Upgrading": "Upgrading",
|
||||
|
||||
@@ -26,10 +26,10 @@ import (
|
||||
"sort"
|
||||
"strconv"
|
||||
|
||||
"code.google.com/p/go.crypto/bcrypt"
|
||||
"github.com/calmh/logger"
|
||||
"github.com/syncthing/syncthing/internal/osutil"
|
||||
"github.com/syncthing/syncthing/internal/protocol"
|
||||
"golang.org/x/crypto/bcrypt"
|
||||
)
|
||||
|
||||
var l = logger.DefaultLogger
|
||||
|
||||
@@ -15,7 +15,7 @@
|
||||
|
||||
package osutil
|
||||
|
||||
import "code.google.com/p/go.text/unicode/norm"
|
||||
import "golang.org/x/text/unicode/norm"
|
||||
|
||||
func NormalizedFilename(s string) string {
|
||||
return norm.NFC.String(s)
|
||||
|
||||
@@ -17,7 +17,7 @@
|
||||
|
||||
package osutil
|
||||
|
||||
import "code.google.com/p/go.text/unicode/norm"
|
||||
import "golang.org/x/text/unicode/norm"
|
||||
|
||||
func NormalizedFilename(s string) string {
|
||||
return norm.NFC.String(s)
|
||||
|
||||
@@ -18,7 +18,7 @@ package osutil
|
||||
import (
|
||||
"path/filepath"
|
||||
|
||||
"code.google.com/p/go.text/unicode/norm"
|
||||
"golang.org/x/text/unicode/norm"
|
||||
)
|
||||
|
||||
func NormalizedFilename(s string) string {
|
||||
|
||||
@@ -19,7 +19,7 @@ package protocol
|
||||
|
||||
// Darwin uses NFD normalization
|
||||
|
||||
import "code.google.com/p/go.text/unicode/norm"
|
||||
import "golang.org/x/text/unicode/norm"
|
||||
|
||||
type nativeModel struct {
|
||||
next Model
|
||||
|
||||
@@ -18,7 +18,7 @@ package protocol
|
||||
import (
|
||||
"path/filepath"
|
||||
|
||||
"code.google.com/p/go.text/unicode/norm"
|
||||
"golang.org/x/text/unicode/norm"
|
||||
)
|
||||
|
||||
type wireFormatConnection struct {
|
||||
|
||||
@@ -22,12 +22,11 @@ import (
|
||||
"runtime"
|
||||
"strings"
|
||||
|
||||
"code.google.com/p/go.text/unicode/norm"
|
||||
|
||||
"github.com/syncthing/syncthing/internal/ignore"
|
||||
"github.com/syncthing/syncthing/internal/lamport"
|
||||
"github.com/syncthing/syncthing/internal/protocol"
|
||||
"github.com/syncthing/syncthing/internal/symlinks"
|
||||
"golang.org/x/text/unicode/norm"
|
||||
)
|
||||
|
||||
type Walker struct {
|
||||
|
||||
Reference in new issue
Block a user