parser.go 5.59 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
var (
Michael Yang's avatar
Michael Yang committed
30
31
32
	errMissingFrom        = errors.New("no FROM line")
	errInvalidMessageRole = errors.New("message role must be one of \"system\", \"user\", or \"assistant\"")
	errInvalidCommand     = errors.New("command must be one of \"from\", \"license\", \"template\", \"system\", \"adapter\", \"parameter\", or \"message\"")
Michael Yang's avatar
tests  
Michael Yang committed
33
)
Michael Yang's avatar
Michael Yang committed
34

Michael Yang's avatar
Michael Yang committed
35
func Format(cmds []Command) string {
36
	var sb strings.Builder
Michael Yang's avatar
Michael Yang committed
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
	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)
		case "message":
			role, message, _ := strings.Cut(cmd.Args, ": ")
			args = role + " " + quote(message)
		default:
			name = "parameter"
52
			args = cmd.Name + " " + quote(cmd.Args)
Michael Yang's avatar
Michael Yang committed
53
54
		}

55
		fmt.Fprintln(&sb, strings.ToUpper(name), args)
Michael Yang's avatar
Michael Yang committed
56
57
	}

58
	return sb.String()
Michael Yang's avatar
Michael Yang committed
59
60
}

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
		if next != curr {
			switch curr {
Michael Yang's avatar
Michael Yang committed
86
87
88
89
90
			case stateName:
				if !isValidCommand(b.String()) {
					return nil, errInvalidCommand
				}

Michael Yang's avatar
Michael Yang committed
91
				// next state sometimes depends on the current buffer value
Michael Yang's avatar
Michael Yang committed
92
93
94
95
				switch s := strings.ToLower(b.String()); s {
				case "from":
					cmd.Name = "model"
				case "parameter":
Michael Yang's avatar
Michael Yang committed
96
					// transition to stateParameter which sets command name
Michael Yang's avatar
Michael Yang committed
97
98
					next = stateParameter
				case "message":
Michael Yang's avatar
Michael Yang committed
99
					// transition to stateMessage which validates the message role
Michael Yang's avatar
Michael Yang committed
100
101
102
103
104
					next = stateMessage
					fallthrough
				default:
					cmd.Name = s
				}
Michael Yang's avatar
Michael Yang committed
105
106
			case stateParameter:
				cmd.Name = b.String()
Michael Yang's avatar
Michael Yang committed
107
			case stateMessage:
Michael Yang's avatar
Michael Yang committed
108
				if !isValidMessageRole(b.String()) {
Michael Yang's avatar
Michael Yang committed
109
					return nil, errInvalidMessageRole
Michael Yang's avatar
Michael Yang committed
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
				}

				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
132
133
			}

Michael Yang's avatar
Michael Yang committed
134
135
136
137
138
139
140
			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
141
			}
Michael Yang's avatar
Michael Yang committed
142
143
144
145
146
147
148
149
		}
	}

	// flush the buffer
	switch curr {
	case stateComment, stateNil:
		// pass; nothing to flush
	case stateValue:
Michael Yang's avatar
Michael Yang committed
150
151
		s, ok := unquote(b.String())
		if !ok {
Michael Yang's avatar
Michael Yang committed
152
			return nil, io.ErrUnexpectedEOF
153
		}
154

Michael Yang's avatar
Michael Yang committed
155
156
157
158
159
		if role != "" {
			s = role + ": " + s
		}

		cmd.Args = s
Michael Yang's avatar
Michael Yang committed
160
161
162
		cmds = append(cmds, cmd)
	default:
		return nil, io.ErrUnexpectedEOF
163
164
	}

Michael Yang's avatar
Michael Yang committed
165
166
167
168
	for _, cmd := range cmds {
		if cmd.Name == "model" {
			return cmds, nil
		}
169
170
	}

Michael Yang's avatar
tests  
Michael Yang committed
171
	return nil, errMissingFrom
172
}
173

Michael Yang's avatar
Michael Yang committed
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
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:
Michael Yang's avatar
Michael Yang committed
192
			return stateNil, 0, errInvalidCommand
Michael Yang's avatar
Michael Yang committed
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
		}
	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("")
230
	}
Michael Yang's avatar
Michael Yang committed
231
}
232

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

239
		return `"` + s + `"`
Michael Yang's avatar
Michael Yang committed
240
241
242
243
244
	}

	return s
}

Michael Yang's avatar
Michael Yang committed
245
246
247
func unquote(s string) (string, bool) {
	if len(s) == 0 {
		return "", false
248
	}
249

Michael Yang's avatar
Michael Yang committed
250
251
252
253
254
255
256
	// 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
257
258
	}

Michael Yang's avatar
Michael Yang committed
259
260
261
262
263
264
	if len(s) >= 1 && s[0] == '"' {
		if len(s) >= 2 && s[len(s)-1] == '"' {
			return s[1 : len(s)-1], true
		}

		return "", false
265
266
	}

Michael Yang's avatar
Michael Yang committed
267
	return s, true
268
269
}

Michael Yang's avatar
Michael Yang committed
270
271
272
func isAlpha(r rune) bool {
	return r >= 'a' && r <= 'z' || r >= 'A' && r <= 'Z'
}
273

Michael Yang's avatar
Michael Yang committed
274
275
276
func isNumber(r rune) bool {
	return r >= '0' && r <= '9'
}
277

Michael Yang's avatar
Michael Yang committed
278
279
280
func isSpace(r rune) bool {
	return r == ' ' || r == '\t'
}
Michael Yang's avatar
Michael Yang committed
281

Michael Yang's avatar
Michael Yang committed
282
283
284
func isNewline(r rune) bool {
	return r == '\r' || r == '\n'
}
285

Michael Yang's avatar
Michael Yang committed
286
func isValidMessageRole(role string) bool {
Michael Yang's avatar
Michael Yang committed
287
	return role == "system" || role == "user" || role == "assistant"
288
}
Michael Yang's avatar
Michael Yang committed
289
290
291
292
293
294
295
296
297

func isValidCommand(cmd string) bool {
	switch strings.ToLower(cmd) {
	case "from", "license", "template", "system", "adapter", "parameter", "message":
		return true
	default:
		return false
	}
}