parser.go 4.08 KB
Newer Older
1
2
3
4
package parser

import (
	"bufio"
5
	"bytes"
6
	"errors"
7
	"fmt"
8
	"io"
9
	"slices"
Michael Yang's avatar
Michael Yang committed
10
11
	"strconv"
	"strings"
12
13
14
15
)

type Command struct {
	Name string
16
17
18
	Args string
}

Michael Yang's avatar
Michael Yang committed
19
type state int
20

Michael Yang's avatar
Michael Yang committed
21
22
23
24
25
26
27
28
const (
	stateNil state = iota
	stateName
	stateValue
	stateParameter
	stateMessage
	stateComment
)
29

Michael Yang's avatar
Michael Yang committed
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
var errInvalidRole = errors.New("role must be one of \"system\", \"user\", or \"assistant\"")

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
		}
46

Michael Yang's avatar
Michael Yang committed
47
48
49
50
51
		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
52
53
		}

Michael Yang's avatar
Michael Yang committed
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
		if next != curr {
			switch curr {
			case stateName, stateParameter:
				switch s := strings.ToLower(b.String()); s {
				case "from":
					cmd.Name = "model"
				case "parameter":
					next = stateParameter
				case "message":
					next = stateMessage
					fallthrough
				default:
					cmd.Name = s
				}
			case stateMessage:
				if !slices.Contains([]string{"system", "user", "assistant"}, b.String()) {
					return nil, errInvalidRole
				}

				role = b.String()
			case stateComment, stateNil:
				// pass
			case stateValue:
				s := b.String()

				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
95
96
			}

Michael Yang's avatar
Michael Yang committed
97
98
99
100
101
102
103
			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
104
			}
Michael Yang's avatar
Michael Yang committed
105
106
107
108
109
110
111
112
113
114
		}
	}

	// flush the buffer
	switch curr {
	case stateComment, stateNil:
		// pass; nothing to flush
	case stateValue:
		if _, ok := unquote(b.String()); !ok {
			return nil, io.ErrUnexpectedEOF
115
		}
116

Michael Yang's avatar
Michael Yang committed
117
118
119
120
		cmd.Args = b.String()
		cmds = append(cmds, cmd)
	default:
		return nil, io.ErrUnexpectedEOF
121
122
	}

Michael Yang's avatar
Michael Yang committed
123
124
125
126
	for _, cmd := range cmds {
		if cmd.Name == "model" {
			return cmds, nil
		}
127
128
	}

Michael Yang's avatar
Michael Yang committed
129
	return nil, errors.New("no FROM line")
130
}
131

Michael Yang's avatar
Michael Yang committed
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
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("")
188
	}
Michael Yang's avatar
Michael Yang committed
189
}
190

Michael Yang's avatar
Michael Yang committed
191
192
193
func unquote(s string) (string, bool) {
	if len(s) == 0 {
		return "", false
194
	}
195

Michael Yang's avatar
Michael Yang committed
196
197
198
199
200
201
202
	// 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
203
204
	}

Michael Yang's avatar
Michael Yang committed
205
206
207
208
209
210
	if len(s) >= 1 && s[0] == '"' {
		if len(s) >= 2 && s[len(s)-1] == '"' {
			return s[1 : len(s)-1], true
		}

		return "", false
211
212
	}

Michael Yang's avatar
Michael Yang committed
213
	return s, true
214
215
}

Michael Yang's avatar
Michael Yang committed
216
217
218
func isAlpha(r rune) bool {
	return r >= 'a' && r <= 'z' || r >= 'A' && r <= 'Z'
}
219

Michael Yang's avatar
Michael Yang committed
220
221
222
func isNumber(r rune) bool {
	return r >= '0' && r <= '9'
}
223

Michael Yang's avatar
Michael Yang committed
224
225
226
func isSpace(r rune) bool {
	return r == ' ' || r == '\t'
}
Michael Yang's avatar
Michael Yang committed
227

Michael Yang's avatar
Michael Yang committed
228
229
230
func isNewline(r rune) bool {
	return r == '\r' || r == '\n'
}
231

Michael Yang's avatar
Michael Yang committed
232
233
func isValidRole(role string) bool {
	return role == "system" || role == "user" || role == "assistant"
234
}