-
Notifications
You must be signed in to change notification settings - Fork 20
/
lexer.go
268 lines (237 loc) · 6.74 KB
/
lexer.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
259
260
261
262
263
264
265
266
267
268
package hl7
import (
"errors"
"fmt"
"strconv"
"github.com/iNamik/go_lexer"
"github.com/iNamik/go_lexer/rangeutil"
)
const (
tokEOF lexer.TokenType = lexer.T_EOF
tokNIL = lexer.T_EOF + iota
tokError
tokSegmentName
tokSeparators
tokSegmentTerminator
tokFieldSeparator
tokComponentSeparator
tokFieldRepeatSeparator
tokSubComponentSeparator
tokField
//tokEscaped
)
const (
fieldSeparatorPos = iota
componentSeparatorPos
fieldRepeatSeparatorPos
escapeCharacterPos
subComponentSeparatorPos
)
var (
bytesUpperChars = rangeutil.RangeToBytes("A-Z0-9")
bytesSegmentGarbage = []byte{'\r', '\n'}
)
// tokenTypeAsString converts a token type to a string.
func tokenTypeAsString(t lexer.TokenType) string {
var typeString string
switch t {
case tokEOF:
typeString = "tokEOF"
case tokNIL:
typeString = "tokNIL"
case tokError:
typeString = "tokError"
case tokSegmentName:
typeString = "tokSegmentName"
case tokSeparators:
typeString = "tokSeparators"
case tokSegmentTerminator:
typeString = "tokSegmentTerminator"
case tokFieldSeparator:
typeString = "tokFieldSeparator"
case tokComponentSeparator:
typeString = "tokComponentSeparator"
case tokFieldRepeatSeparator:
typeString = "tokFieldRepeatSeparator"
case tokSubComponentSeparator:
typeString = "tokSubComponentSeparator"
case tokField:
typeString = "tokField"
//case tokEscaped:
// typeString = "tokEscaped"
default:
typeString = strconv.Itoa(int(t))
}
return typeString
}
// lexerState represents the state for the lexer.
type lexerState struct {
segmentTerminator rune
fieldSeparator rune
componentSeparator rune
fieldRepeatSeparator rune
escapeCharacter rune
subComponentSeparator rune
err error
lexFieldSeparator func(lexer.Lexer) lexer.StateFn
lexComponentSeparator func(lexer.Lexer) lexer.StateFn
lexFieldRepeatSeparator func(lexer.Lexer) lexer.StateFn
lexSubComponentSeparator func(lexer.Lexer) lexer.StateFn
}
// newLexerState returns the state that helps in lexing
func newLexerState() *lexerState {
s := &lexerState{
segmentTerminator: '\r',
}
// make a function to error if these functions are called
// before we find the message header
f := func(l lexer.Lexer) lexer.StateFn {
s.err = errors.New("did not find message header")
l.EmitToken(tokError)
l.EmitEOF()
return nil
}
s.lexFieldSeparator = f
s.lexComponentSeparator = f
s.lexFieldRepeatSeparator = f
s.lexSubComponentSeparator = f
return s
}
// lexHeader scans for the HL7 message header
func (s *lexerState) lexHeader(l lexer.Lexer) lexer.StateFn {
l.MatchZeroOrMoreRunes([]rune{s.segmentTerminator, '\n'})
// if l.MatchEOF() {
// l.EmitEOF()
// return nil
// }
if matchRunes(l, []rune("MSH")) {
l.EmitTokenWithBytes(tokSegmentName)
} else {
s.err = errors.New("could not find message header")
l.EmitToken(tokError)
return nil
}
return s.lexHeaderSeparators
}
// lexHeaderSeparators scans for the separators used to parse the
// rest of the HL7 message.
func (s *lexerState) lexHeaderSeparators(l lexer.Lexer) lexer.StateFn {
for i := 0; i <= subComponentSeparatorPos-fieldSeparatorPos; i++ {
r := l.NextRune()
if r == lexer.RuneEOF {
s.err = errors.New("found eof while reading message header")
l.EmitToken(tokError)
return nil
} else if i != fieldSeparatorPos && r == s.fieldSeparator {
s.err = fmt.Errorf("missing %d separators", 5-i)
l.EmitToken(tokError)
return nil
}
switch i {
case fieldSeparatorPos:
s.fieldSeparator = r
l.EmitTokenWithBytes(tokField)
case componentSeparatorPos:
s.componentSeparator = r
case fieldRepeatSeparatorPos:
s.fieldRepeatSeparator = r
case escapeCharacterPos:
s.escapeCharacter = r
case subComponentSeparatorPos:
s.subComponentSeparator = r
}
}
s.lexFieldSeparator = s.getLexSeparator("field", s.fieldSeparator, tokFieldSeparator)
s.lexComponentSeparator = s.getLexSeparator("component", s.componentSeparator, tokComponentSeparator)
s.lexFieldRepeatSeparator = s.getLexSeparator("field repeat", s.fieldRepeatSeparator, tokFieldRepeatSeparator)
s.lexSubComponentSeparator = s.getLexSeparator("sub component", s.subComponentSeparator, tokSubComponentSeparator)
l.EmitTokenWithBytes(tokSeparators)
return s.lexFieldSeparator
}
// lexSegment scans for a HL7 segment.
func (s *lexerState) lexSegment(l lexer.Lexer) lexer.StateFn {
if l.MatchEOF() {
l.EmitEOF()
return nil // We're done here
}
if !l.MatchMinMaxBytes(bytesUpperChars, 3, 3) {
s.err = fmt.Errorf("incorrect segment name, found: %c%c%c", l.PeekRune(0), l.PeekRune(1), l.PeekRune(2))
l.EmitToken(tokError)
return nil
}
l.EmitTokenWithBytes(tokSegmentName)
return s.lexFieldSeparator
}
// getLexSeparator is a function generator that returns a function
// that scans for a particular separator. This is used to prevent
// a lot of code duplication for all the separators/terminators.
func (s *lexerState) getLexSeparator(typ string, sep rune, tok lexer.TokenType) lexer.StateFn {
return func(l lexer.Lexer) lexer.StateFn {
r := l.NextRune()
if r != sep {
s.err = fmt.Errorf("expected a %s separator '%c'. got: '%c'", typ, sep, r)
l.EmitToken(tokError)
return nil
}
l.EmitTokenWithBytes(tok)
return s.lexField
}
}
// lexField scans for a HL7 field.
func (s *lexerState) lexField(l lexer.Lexer) lexer.StateFn {
for {
switch r := l.NextRune(); {
case r == s.fieldSeparator:
l.BackupRune()
l.EmitTokenWithBytes(tokField)
return s.lexFieldSeparator
case r == s.componentSeparator:
l.BackupRune()
l.EmitTokenWithBytes(tokField)
return s.lexComponentSeparator
case r == s.fieldRepeatSeparator:
l.BackupRune()
l.EmitTokenWithBytes(tokField)
return s.lexFieldRepeatSeparator
// case r == s.escapeCharacter:
// l.NonMatchOneOrMoreRunes([]rune{s.escapeCharacter})
// l.NextRune()
// l.EmitTokenWithBytes(tokEscaped)
// return s.lexField
case r == s.subComponentSeparator:
l.BackupRune()
l.EmitTokenWithBytes(tokField)
return s.lexSubComponentSeparator
case r == s.segmentTerminator || r == '\n':
l.BackupRune()
l.EmitTokenWithBytes(tokField)
l.NewLine()
l.NextRune()
l.EmitTokenWithBytes(tokSegmentTerminator)
// ignore any multiple \r and \n
l.MatchOneOrMoreBytes(bytesSegmentGarbage)
l.IgnoreToken()
return s.lexSegment
case r == lexer.RuneEOF:
//l.BackupRune()
l.EmitTokenWithBytes(tokField)
l.EmitToken(tokSegmentTerminator)
l.EmitEOF()
return nil
default:
// absorb
}
}
l.EmitEOF()
return nil
}
// matchRunes consumes a specified run of matching runes. Every rune
// must match sequentially.
func matchRunes(l lexer.Lexer, match []rune) bool {
i := 0
return l.MatchMinMaxFunc(func(r rune) bool {
truth := match[i] == r
i++
return truth
}, len(match), len(match))
}