parser.go 5.08 KB
Newer Older
1
2
3
4
package parser

import (
	"bufio"
5
	"bytes"
6
	"errors"
7
	"fmt"
8
	"io"
Michael Yang's avatar
Michael Yang committed
9
10
	"strconv"
	"strings"
11
12
13
14
)

type Command struct {
	Name string
15
16
17
	Args string
}

Michael Yang's avatar
Michael Yang committed
18
type state int
19

Michael Yang's avatar
Michael Yang committed
20
21
22
23
24
25
26
27
const (
	stateNil state = iota
	stateName
	stateValue
	stateParameter
	stateMessage
	stateComment
)
28

Michael Yang's avatar
tests  
Michael Yang committed
29
30
31
32
var (
	errMissingFrom = errors.New("no FROM line")
	errInvalidRole = errors.New("role must be one of \"system\", \"user\", or \"assistant\"")
)
Michael Yang's avatar
Michael Yang committed
33

Michael Yang's avatar
Michael Yang committed
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
func Format(cmds []Command) string {
	var b bytes.Buffer
	for _, cmd := range cmds {
		name := cmd.Name
		args := cmd.Args

		switch cmd.Name {
		case "model":
			name = "from"
			args = cmd.Args
		case "license", "template", "system", "adapter":
			args = quote(args)
			// pass
		case "message":
			role, message, _ := strings.Cut(cmd.Args, ": ")
			args = role + " " + quote(message)
		default:
			name = "parameter"
			args = cmd.Name + " " + cmd.Args
		}

		fmt.Fprintln(&b, strings.ToUpper(name), args)
	}

	return b.String()
}

Michael Yang's avatar
Michael Yang committed
61
62
63
64
65
66
67
68
69
70
71
72
73
74
func Parse(r io.Reader) (cmds []Command, err error) {
	var cmd Command
	var curr state
	var b bytes.Buffer
	var role string

	br := bufio.NewReader(r)
	for {
		r, _, err := br.ReadRune()
		if errors.Is(err, io.EOF) {
			break
		} else if err != nil {
			return nil, err
		}
75

Michael Yang's avatar
Michael Yang committed
76
77
78
79
80
		next, r, err := parseRuneForState(r, curr)
		if errors.Is(err, io.ErrUnexpectedEOF) {
			return nil, fmt.Errorf("%w: %s", err, b.String())
		} else if err != nil {
			return nil, err
81
82
		}

Michael Yang's avatar
Michael Yang committed
83
		// process the state transition, some transitions need to be intercepted and redirected
Michael Yang's avatar
Michael Yang committed
84
85
86
		if next != curr {
			switch curr {
			case stateName, stateParameter:
Michael Yang's avatar
Michael Yang committed
87
				// next state sometimes depends on the current buffer value
Michael Yang's avatar
Michael Yang committed
88
89
90
91
				switch s := strings.ToLower(b.String()); s {
				case "from":
					cmd.Name = "model"
				case "parameter":
Michael Yang's avatar
Michael Yang committed
92
					// transition to stateParameter which sets command name
Michael Yang's avatar
Michael Yang committed
93
94
					next = stateParameter
				case "message":
Michael Yang's avatar
Michael Yang committed
95
					// transition to stateMessage which validates the message role
Michael Yang's avatar
Michael Yang committed
96
97
98
99
100
101
					next = stateMessage
					fallthrough
				default:
					cmd.Name = s
				}
			case stateMessage:
Michael Yang's avatar
Michael Yang committed
102
				if !isValidMessageRole(b.String()) {
Michael Yang's avatar
Michael Yang committed
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
					return nil, errInvalidRole
				}

				role = b.String()
			case stateComment, stateNil:
				// pass
			case stateValue:
				s, ok := unquote(b.String())
				if !ok || isSpace(r) {
					if _, err := b.WriteRune(r); err != nil {
						return nil, err
					}

					continue
				}

				if role != "" {
					s = role + ": " + s
					role = ""
				}

				cmd.Args = s
				cmds = append(cmds, cmd)
Michael Yang's avatar
Michael Yang committed
126
127
			}

Michael Yang's avatar
Michael Yang committed
128
129
130
131
132
133
134
			b.Reset()
			curr = next
		}

		if strconv.IsPrint(r) {
			if _, err := b.WriteRune(r); err != nil {
				return nil, err
Michael Yang's avatar
Michael Yang committed
135
			}
Michael Yang's avatar
Michael Yang committed
136
137
138
139
140
141
142
143
		}
	}

	// flush the buffer
	switch curr {
	case stateComment, stateNil:
		// pass; nothing to flush
	case stateValue:
Michael Yang's avatar
Michael Yang committed
144
145
		s, ok := unquote(b.String())
		if !ok {
Michael Yang's avatar
Michael Yang committed
146
			return nil, io.ErrUnexpectedEOF
147
		}
148

Michael Yang's avatar
Michael Yang committed
149
150
151
152
153
		if role != "" {
			s = role + ": " + s
		}

		cmd.Args = s
Michael Yang's avatar
Michael Yang committed
154
155
156
		cmds = append(cmds, cmd)
	default:
		return nil, io.ErrUnexpectedEOF
157
158
	}

Michael Yang's avatar
Michael Yang committed
159
160
161
162
	for _, cmd := range cmds {
		if cmd.Name == "model" {
			return cmds, nil
		}
163
164
	}

Michael Yang's avatar
tests  
Michael Yang committed
165
	return nil, errMissingFrom
166
}
167

Michael Yang's avatar
Michael Yang committed
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
func parseRuneForState(r rune, cs state) (state, rune, error) {
	switch cs {
	case stateNil:
		switch {
		case r == '#':
			return stateComment, 0, nil
		case isSpace(r), isNewline(r):
			return stateNil, 0, nil
		default:
			return stateName, r, nil
		}
	case stateName:
		switch {
		case isAlpha(r):
			return stateName, r, nil
		case isSpace(r):
			return stateValue, 0, nil
		default:
			return stateNil, 0, errors.New("invalid")
		}
	case stateValue:
		switch {
		case isNewline(r):
			return stateNil, r, nil
		case isSpace(r):
			return stateNil, r, nil
		default:
			return stateValue, r, nil
		}
	case stateParameter:
		switch {
		case isAlpha(r), isNumber(r), r == '_':
			return stateParameter, r, nil
		case isSpace(r):
			return stateValue, 0, nil
		default:
			return stateNil, 0, io.ErrUnexpectedEOF
		}
	case stateMessage:
		switch {
		case isAlpha(r):
			return stateMessage, r, nil
		case isSpace(r):
			return stateValue, 0, nil
		default:
			return stateNil, 0, io.ErrUnexpectedEOF
		}
	case stateComment:
		switch {
		case isNewline(r):
			return stateNil, 0, nil
		default:
			return stateComment, 0, nil
		}
	default:
		return stateNil, 0, errors.New("")
224
	}
Michael Yang's avatar
Michael Yang committed
225
}
226

Michael Yang's avatar
Michael Yang committed
227
228
229
230
231
232
233
234
235
236
237
238
func quote(s string) string {
	if strings.Contains(s, "\n") || strings.HasSuffix(s, " ") {
		if strings.Contains(s, "\"") {
			return `"""` + s + `"""`
		}

		return strconv.Quote(s)
	}

	return s
}

Michael Yang's avatar
Michael Yang committed
239
240
241
func unquote(s string) (string, bool) {
	if len(s) == 0 {
		return "", false
242
	}
243

Michael Yang's avatar
Michael Yang committed
244
245
246
247
248
249
250
	// TODO: single quotes
	if len(s) >= 3 && s[:3] == `"""` {
		if len(s) >= 6 && s[len(s)-3:] == `"""` {
			return s[3 : len(s)-3], true
		}

		return "", false
251
252
	}

Michael Yang's avatar
Michael Yang committed
253
254
255
256
257
258
	if len(s) >= 1 && s[0] == '"' {
		if len(s) >= 2 && s[len(s)-1] == '"' {
			return s[1 : len(s)-1], true
		}

		return "", false
259
260
	}

Michael Yang's avatar
Michael Yang committed
261
	return s, true
262
263
}

Michael Yang's avatar
Michael Yang committed
264
265
266
func isAlpha(r rune) bool {
	return r >= 'a' && r <= 'z' || r >= 'A' && r <= 'Z'
}
267

Michael Yang's avatar
Michael Yang committed
268
269
270
func isNumber(r rune) bool {
	return r >= '0' && r <= '9'
}
271

Michael Yang's avatar
Michael Yang committed
272
273
274
func isSpace(r rune) bool {
	return r == ' ' || r == '\t'
}
Michael Yang's avatar
Michael Yang committed
275

Michael Yang's avatar
Michael Yang committed
276
277
278
func isNewline(r rune) bool {
	return r == '\r' || r == '\n'
}
279

Michael Yang's avatar
Michael Yang committed
280
func isValidMessageRole(role string) bool {
Michael Yang's avatar
Michael Yang committed
281
	return role == "system" || role == "user" || role == "assistant"
282
}