-
Notifications
You must be signed in to change notification settings - Fork 12
/
ident.go
258 lines (241 loc) · 5.95 KB
/
ident.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
// Public Domain (-) 2018-present, The Web4 Authors.
// See the Web4 UNLICENSE file for details.
// Package ident provides support for converting identifiers between different
// naming conventions.
package ident
import (
"bytes"
"fmt"
)
// Parts represents the normalized elements of an identifier.
type Parts [][]byte
func (p Parts) String() string {
return string(bytes.Join(p, []byte{','}))
}
// ToCamel converts the identifier into a camelCased string.
func (p Parts) ToCamel() string {
var out []byte
for idx, part := range p {
if idx == 0 {
out = append(out, bytes.ToLower(part)...)
} else {
out = append(out, part...)
}
}
return string(out)
}
// ToKebab converts the identifier into a kebab-cased string.
func (p Parts) ToKebab() string {
var out []byte
for idx, part := range p {
if idx != 0 {
out = append(out, '-')
}
out = append(out, bytes.ToLower(part)...)
}
return string(out)
}
// ToPascal converts the identifier into a PascalCased string.
func (p Parts) ToPascal() string {
var out []byte
for _, part := range p {
out = append(out, part...)
}
return string(out)
}
// ToScreamingSnake converts the identifier into a SCREAMING_SNAKE_CASED string.
func (p Parts) ToScreamingSnake() string {
var out []byte
for idx, part := range p {
if idx != 0 {
out = append(out, '_')
}
out = append(out, bytes.ToUpper(part)...)
}
return string(out)
}
// ToSnake converts the identifier into a snake_cased string.
func (p Parts) ToSnake() string {
var out []byte
for idx, part := range p {
if idx != 0 {
out = append(out, '_')
}
out = append(out, bytes.ToLower(part)...)
}
return string(out)
}
// add appends parts from the given element. It looks for runs of initialisms
// like "HTTPAPIs" and adds them as separate parts, i.e. "HTTP" and "APIs". Once
// all initialisms are detected, the remaining element is added as a single
// part.
func (p Parts) add(elem []byte) Parts {
// Try to match an initialism exactly.
if special, ok := mapping[string(bytes.ToUpper(elem))]; ok {
return append(p, []byte(special))
}
// Try to find the longest initialism matches from the start.
for len(elem) > 0 {
match := ""
pos := -1
for i := 0; i <= len(elem); i++ {
if special, ok := mapping[string(bytes.ToUpper(elem[:i]))]; ok {
match = special
pos = i
}
}
if pos == -1 {
p = append(p, elem)
break
}
p = append(p, []byte(match))
elem = elem[pos:]
}
return p
}
// tryAdd attempts to add parts from the given element. If any initialisms are
// found, they are added in canonical form.
func (p Parts) tryAdd(elem []byte) (Parts, []byte) {
var nelem []byte
// Try to match an initialism exactly.
if special, ok := mapping[string(bytes.ToUpper(elem))]; ok {
return append(p, []byte(special)), nil
}
// Try to match an initialism from the end for the longest identifier with a
// non-uppercase suffix.
last := ""
pos := -1
for i := len(elem) - 1; i >= 0; i-- {
if special, ok := mapping[string(bytes.ToUpper(elem[i:]))]; ok {
last = special
pos = i
}
}
if pos == -1 {
// NOTE(tav): The given elem must be at least 2 characters long. The
// code in FromPascal currently ensures this to be the case.
nelem = elem[len(elem)-2:]
elem = elem[:len(elem)-2]
} else {
elem = elem[:pos]
}
p = p.add(elem)
if len(last) > 0 {
p = append(p, []byte(last))
}
return p, nelem
}
// FromCamel parses the given camelCased identifier into its parts.
func FromCamel(ident string) Parts {
var parts Parts
i := 0
for ; i < len(ident); i++ {
char := ident[i]
if char >= 'A' && char <= 'Z' {
break
}
}
parts = append(parts, normalize([]byte(ident[:i])))
// NOTE(tav): The error must be nil, as ident must be empty or start on an
// uppercase character, per the break clause above.
elems, _ := FromPascal(ident[i:])
return append(parts, elems...)
}
// FromKebab parses the given kebab-cased identifier into its parts.
func FromKebab(ident string) Parts {
var (
elem []byte
parts Parts
)
for i := 0; i < len(ident); i++ {
char := ident[i]
if char == '-' {
if len(elem) == 0 {
continue
}
parts = append(parts, normalize(bytes.ToLower(elem)))
elem = []byte{}
} else {
elem = append(elem, char)
}
}
if len(elem) > 0 {
parts = append(parts, normalize(bytes.ToLower(elem)))
}
return parts
}
// FromPascal parses the given PascalCased identifier into its parts.
func FromPascal(ident string) (Parts, error) {
var (
elem []byte
parts Parts
)
// Ensure the first character is upper case.
if len(ident) > 0 {
char := ident[0]
if char < 'A' || char > 'Z' {
return nil, fmt.Errorf("ident: invalid PascalCased identifier: %q", ident)
}
elem = append(elem, char)
}
caps := true
for i := 1; i < len(ident); i++ {
char := ident[i]
if (char >= 'A' && char <= 'Z') || (char >= '0' && char <= '9') {
if caps {
elem = append(elem, char)
} else {
caps = true
parts = parts.add(elem)
elem = []byte{char}
}
} else if caps {
caps = false
elem = append(elem, char)
parts, elem = parts.tryAdd(elem)
} else {
elem = append(elem, char)
}
}
if len(elem) > 0 {
parts = parts.add(elem)
}
return parts, nil
}
// FromScreamingSnake parses the given SCREAMING_SNAKE_CASED identifier into its
// parts.
func FromScreamingSnake(ident string) Parts {
return FromSnake(ident)
}
// FromSnake parses the given snake_cased identifier into its parts.
func FromSnake(ident string) Parts {
var (
elem []byte
parts Parts
)
for i := 0; i < len(ident); i++ {
char := ident[i]
if char == '_' {
if len(elem) == 0 {
continue
}
parts = append(parts, normalize(bytes.ToLower(elem)))
elem = []byte{}
} else {
elem = append(elem, char)
}
}
if len(elem) > 0 {
parts = append(parts, normalize(bytes.ToLower(elem)))
}
return parts
}
func normalize(elem []byte) []byte {
if special, ok := mapping[string(bytes.ToUpper(elem))]; ok {
return []byte(special)
}
if len(elem) > 0 && 'a' <= elem[0] && elem[0] <= 'z' {
elem[0] -= 32
}
return elem
}