1 // Code generated by running "go run gen.go -core" in golang.org/x/text. DO NOT EDIT.
3 // Copyright 2016 The Go Authors. All rights reserved.
4 // Use of this source code is governed by a BSD-style
5 // license that can be found in the LICENSE file.
7 // Package bidirule implements the Bidi Rule defined by RFC 5893.
9 // This package is under development. The API may change without notice and
10 // without preserving backward compatibility.
17 "golang_org/x/text/transform"
18 "golang_org/x/text/unicode/bidi"
21 // This file contains an implementation of RFC 5893: Right-to-Left Scripts for
22 // Internationalized Domain Names for Applications (IDNA)
24 // A label is an individual component of a domain name. Labels are usually
25 // shown separated by dots; for example, the domain name "www.example.com" is
26 // composed of three labels: "www", "example", and "com".
28 // An RTL label is a label that contains at least one character of class R, AL,
29 // or AN. An LTR label is any label that is not an RTL label.
31 // A "Bidi domain name" is a domain name that contains at least one RTL label.
33 // The following guarantees can be made based on the above:
35 // o In a domain name consisting of only labels that satisfy the rule,
36 // the requirements of Section 3 are satisfied. Note that even LTR
37 // labels and pure ASCII labels have to be tested.
39 // o In a domain name consisting of only LDH labels (as defined in the
40 // Definitions document [RFC5890]) and labels that satisfy the rule,
41 // the requirements of Section 3 are satisfied as long as a label
42 // that starts with an ASCII digit does not come after a
43 // right-to-left label.
45 // No guarantee is given for other combinations.
47 // ErrInvalid indicates a label is invalid according to the Bidi Rule.
48 var ErrInvalid
= errors
.New("bidirule: failed Bidi Rule")
53 ruleInitial ruleState
= iota
61 type ruleTransition
struct {
66 var transitions
= [...][2]ruleTransition
{
67 // [2.1] The first character must be a character with Bidi property L, R, or
68 // AL. If it has the R or AL property, it is an RTL label; if it has the L
69 // property, it is an LTR label.
71 {ruleLTRFinal
, 1 << bidi
.L
},
72 {ruleRTLFinal
, 1<<bidi
.R |
1<<bidi
.AL
},
75 // [2.3] In an RTL label, the end of the label must be a character with
76 // Bidi property R, AL, EN, or AN, followed by zero or more characters
77 // with Bidi property NSM.
78 {ruleRTLFinal
, 1<<bidi
.R |
1<<bidi
.AL |
1<<bidi
.EN |
1<<bidi
.AN
},
80 // [2.2] In an RTL label, only characters with the Bidi properties R,
81 // AL, AN, EN, ES, CS, ET, ON, BN, or NSM are allowed.
82 // We exclude the entries from [2.3]
83 {ruleRTL
, 1<<bidi
.ES |
1<<bidi
.CS |
1<<bidi
.ET |
1<<bidi
.ON |
1<<bidi
.BN |
1<<bidi
.NSM
},
86 // [2.3] In an RTL label, the end of the label must be a character with
87 // Bidi property R, AL, EN, or AN, followed by zero or more characters
88 // with Bidi property NSM.
89 {ruleRTLFinal
, 1<<bidi
.R |
1<<bidi
.AL |
1<<bidi
.EN |
1<<bidi
.AN |
1<<bidi
.NSM
},
91 // [2.2] In an RTL label, only characters with the Bidi properties R,
92 // AL, AN, EN, ES, CS, ET, ON, BN, or NSM are allowed.
93 // We exclude the entries from [2.3] and NSM.
94 {ruleRTL
, 1<<bidi
.ES |
1<<bidi
.CS |
1<<bidi
.ET |
1<<bidi
.ON |
1<<bidi
.BN
},
97 // [2.6] In an LTR label, the end of the label must be a character with
98 // Bidi property L or EN, followed by zero or more characters with Bidi
100 {ruleLTRFinal
, 1<<bidi
.L |
1<<bidi
.EN
},
102 // [2.5] In an LTR label, only characters with the Bidi properties L,
103 // EN, ES, CS, ET, ON, BN, or NSM are allowed.
104 // We exclude the entries from [2.6].
105 {ruleLTR
, 1<<bidi
.ES |
1<<bidi
.CS |
1<<bidi
.ET |
1<<bidi
.ON |
1<<bidi
.BN |
1<<bidi
.NSM
},
108 // [2.6] In an LTR label, the end of the label must be a character with
109 // Bidi property L or EN, followed by zero or more characters with Bidi
111 {ruleLTRFinal
, 1<<bidi
.L |
1<<bidi
.EN |
1<<bidi
.NSM
},
113 // [2.5] In an LTR label, only characters with the Bidi properties L,
114 // EN, ES, CS, ET, ON, BN, or NSM are allowed.
115 // We exclude the entries from [2.6].
116 {ruleLTR
, 1<<bidi
.ES |
1<<bidi
.CS |
1<<bidi
.ET |
1<<bidi
.ON |
1<<bidi
.BN
},
124 // [2.4] In an RTL label, if an EN is present, no AN may be present, and
126 const exclusiveRTL
= uint16(1<<bidi
.EN |
1<<bidi
.AN
)
129 // An RTL label is a label that contains at least one character of type
132 // An LTR label is any label that is not an RTL label.
134 // Direction reports the direction of the given label as defined by RFC 5893.
135 // The Bidi Rule does not have to be applied to labels of the category
137 func Direction(b
[]byte) bidi
.Direction
{
138 for i
:= 0; i
< len(b
); {
139 e
, sz
:= bidi
.Lookup(b
[i
:])
144 if c
== bidi
.R || c
== bidi
.AL || c
== bidi
.AN
{
145 return bidi
.RightToLeft
149 return bidi
.LeftToRight
152 // DirectionString reports the direction of the given label as defined by RFC
153 // 5893. The Bidi Rule does not have to be applied to labels of the category
155 func DirectionString(s
string) bidi
.Direction
{
156 for i
:= 0; i
< len(s
); {
157 e
, sz
:= bidi
.LookupString(s
[i
:])
162 if c
== bidi
.R || c
== bidi
.AL || c
== bidi
.AN
{
163 return bidi
.RightToLeft
167 return bidi
.LeftToRight
170 // Valid reports whether b conforms to the BiDi rule.
171 func Valid(b
[]byte) bool {
173 if n
, ok
:= t
.advance(b
); !ok || n
< len(b
) {
179 // ValidString reports whether s conforms to the BiDi rule.
180 func ValidString(s
string) bool {
182 if n
, ok
:= t
.advanceString(s
); !ok || n
< len(s
) {
188 // New returns a Transformer that verifies that input adheres to the Bidi Rule.
189 func New() *Transformer
{
190 return &Transformer
{}
193 // Transformer implements transform.Transform.
194 type Transformer
struct {
200 // A rule can only be violated for "Bidi Domain names", meaning if one of the
201 // following categories has been observed.
202 func (t
*Transformer
) isRTL() bool {
203 const isRTL
= 1<<bidi
.R |
1<<bidi
.AL |
1<<bidi
.AN
204 return t
.seen
&isRTL
!= 0
207 func (t
*Transformer
) isFinal() bool {
211 return t
.state
== ruleLTRFinal || t
.state
== ruleRTLFinal || t
.state
== ruleInitial
214 // Reset implements transform.Transformer.
215 func (t
*Transformer
) Reset() { *t
= Transformer
{} }
217 // Transform implements transform.Transformer. This Transformer has state and
218 // needs to be reset between uses.
219 func (t
*Transformer
) Transform(dst
, src
[]byte, atEOF
bool) (nDst
, nSrc
int, err error
) {
220 if len(dst
) < len(src
) {
223 err
= transform
.ErrShortDst
225 n
, err1
:= t
.Span(src
, atEOF
)
227 if err
== nil || err1
!= nil && err1
!= transform
.ErrShortSrc
{
233 // Span returns the first n bytes of src that conform to the Bidi rule.
234 func (t
*Transformer
) Span(src
[]byte, atEOF
bool) (n
int, err error
) {
235 if t
.state
== ruleInvalid
&& t
.isRTL() {
238 n
, ok
:= t
.advance(src
)
244 err
= transform
.ErrShortSrc
254 // Precomputing the ASCII values decreases running time for the ASCII fast path
256 var asciiTable
[128]bidi
.Properties
259 for i
:= range asciiTable
{
260 p
, _
:= bidi
.LookupRune(rune(i
))
265 func (t
*Transformer
) advance(s
[]byte) (n
int, ok
bool) {
266 var e bidi
.Properties
269 if s
[n
] < utf8
.RuneSelf
{
270 e
, sz
= asciiTable
[s
[n
]], 1
272 e
, sz
= bidi
.Lookup(s
[n
:])
275 // We always consider invalid UTF-8 to be invalid, even if
276 // the string has not yet been determined to be RTL.
277 // TODO: is this correct?
280 return n
, true // incomplete UTF-8 encoding
283 // TODO: using CompactClass would result in noticeable speedup.
284 // See unicode/bidi/prop.go:Properties.CompactClass.
285 c
:= uint16(1 << e
.Class())
287 if t
.seen
&exclusiveRTL
== exclusiveRTL
{
288 t
.state
= ruleInvalid
291 switch tr
:= transitions
[t
.state
]; {
292 case tr
[0].mask
&c
!= 0:
294 case tr
[1].mask
&c
!= 0:
297 t
.state
= ruleInvalid
307 func (t
*Transformer
) advanceString(s
string) (n
int, ok
bool) {
308 var e bidi
.Properties
311 if s
[n
] < utf8
.RuneSelf
{
312 e
, sz
= asciiTable
[s
[n
]], 1
314 e
, sz
= bidi
.LookupString(s
[n
:])
317 return n
, false // invalid UTF-8
319 return n
, true // incomplete UTF-8 encoding
322 // TODO: using CompactClass results in noticeable speedup.
323 // See unicode/bidi/prop.go:Properties.CompactClass.
324 c
:= uint16(1 << e
.Class())
326 if t
.seen
&exclusiveRTL
== exclusiveRTL
{
327 t
.state
= ruleInvalid
330 switch tr
:= transitions
[t
.state
]; {
331 case tr
[0].mask
&c
!= 0:
333 case tr
[1].mask
&c
!= 0:
336 t
.state
= ruleInvalid