parser.go 4.37 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
34
35
36
37
38
39
40
41
42
43
44
45
46
47

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
		}
48

Michael Yang's avatar
Michael Yang committed
49
50
51
52
53
		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
54
55
		}

Michael Yang's avatar
Michael Yang committed
56
		// process the state transition, some transitions need to be intercepted and redirected
Michael Yang's avatar
Michael Yang committed
57
58
59
		if next != curr {
			switch curr {
			case stateName, stateParameter:
Michael Yang's avatar
Michael Yang committed
60
				// next state sometimes depends on the current buffer value
Michael Yang's avatar
Michael Yang committed
61
62
63
64
				switch s := strings.ToLower(b.String()); s {
				case "from":
					cmd.Name = "model"
				case "parameter":
Michael Yang's avatar
Michael Yang committed
65
					// transition to stateParameter which sets command name
Michael Yang's avatar
Michael Yang committed
66
67
					next = stateParameter
				case "message":
Michael Yang's avatar
Michael Yang committed
68
					// transition to stateMessage which validates the message role
Michael Yang's avatar
Michael Yang committed
69
70
71
72
73
74
					next = stateMessage
					fallthrough
				default:
					cmd.Name = s
				}
			case stateMessage:
Michael Yang's avatar
Michael Yang committed
75
				if !isValidMessageRole(b.String()) {
Michael Yang's avatar
Michael Yang committed
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
					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
99
100
			}

Michael Yang's avatar
Michael Yang committed
101
102
103
104
105
106
107
			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
108
			}
Michael Yang's avatar
Michael Yang committed
109
110
111
112
113
114
115
116
		}
	}

	// flush the buffer
	switch curr {
	case stateComment, stateNil:
		// pass; nothing to flush
	case stateValue:
Michael Yang's avatar
Michael Yang committed
117
118
		s, ok := unquote(b.String())
		if !ok {
Michael Yang's avatar
Michael Yang committed
119
			return nil, io.ErrUnexpectedEOF
120
		}
121

Michael Yang's avatar
Michael Yang committed
122
123
124
125
126
		if role != "" {
			s = role + ": " + s
		}

		cmd.Args = s
Michael Yang's avatar
Michael Yang committed
127
128
129
		cmds = append(cmds, cmd)
	default:
		return nil, io.ErrUnexpectedEOF
130
131
	}

Michael Yang's avatar
Michael Yang committed
132
133
134
135
	for _, cmd := range cmds {
		if cmd.Name == "model" {
			return cmds, nil
		}
136
137
	}

Michael Yang's avatar
tests  
Michael Yang committed
138
	return nil, errMissingFrom
139
}
140

Michael Yang's avatar
Michael Yang committed
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
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("")
197
	}
Michael Yang's avatar
Michael Yang committed
198
}
199

Michael Yang's avatar
Michael Yang committed
200
201
202
func unquote(s string) (string, bool) {
	if len(s) == 0 {
		return "", false
203
	}
204

Michael Yang's avatar
Michael Yang committed
205
206
207
208
209
210
211
	// 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
212
213
	}

Michael Yang's avatar
Michael Yang committed
214
215
216
217
218
219
	if len(s) >= 1 && s[0] == '"' {
		if len(s) >= 2 && s[len(s)-1] == '"' {
			return s[1 : len(s)-1], true
		}

		return "", false
220
221
	}

Michael Yang's avatar
Michael Yang committed
222
	return s, true
223
224
}

Michael Yang's avatar
Michael Yang committed
225
226
227
func isAlpha(r rune) bool {
	return r >= 'a' && r <= 'z' || r >= 'A' && r <= 'Z'
}
228

Michael Yang's avatar
Michael Yang committed
229
230
231
func isNumber(r rune) bool {
	return r >= '0' && r <= '9'
}
232

Michael Yang's avatar
Michael Yang committed
233
234
235
func isSpace(r rune) bool {
	return r == ' ' || r == '\t'
}
Michael Yang's avatar
Michael Yang committed
236

Michael Yang's avatar
Michael Yang committed
237
238
239
func isNewline(r rune) bool {
	return r == '\r' || r == '\n'
}
240

Michael Yang's avatar
Michael Yang committed
241
func isValidMessageRole(role string) bool {
Michael Yang's avatar
Michael Yang committed
242
	return role == "system" || role == "user" || role == "assistant"
243
}