parser_test.go 25.6 KB
Newer Older
1
package parser
2
3

import (
Michael Yang's avatar
Michael Yang committed
4
	"bytes"
5
	"crypto/sha256"
6
	"encoding/binary"
7
	"errors"
Michael Yang's avatar
Michael Yang committed
8
9
	"fmt"
	"io"
10
	"maps"
11
	"os"
12
	"path/filepath"
13
14
	"strings"
	"testing"
15
	"unicode/utf16"
16

17
	"github.com/google/go-cmp/cmp"
18
	"github.com/stretchr/testify/assert"
Michael Yang's avatar
lint  
Michael Yang committed
19
	"github.com/stretchr/testify/require"
20
21
	"golang.org/x/text/encoding"
	"golang.org/x/text/encoding/unicode"
22
23

	"github.com/ollama/ollama/api"
24
	"github.com/ollama/ollama/convert"
Michael Yang's avatar
Michael Yang committed
25
	"github.com/ollama/ollama/fs/ggml"
26
27
)

Michael Yang's avatar
Michael Yang committed
28
func TestParseFileFile(t *testing.T) {
29
30
31
32
33
34
	input := `
FROM model1
ADAPTER adapter1
LICENSE MIT
PARAMETER param1 value1
PARAMETER param2 value2
Josh Yan's avatar
Josh Yan committed
35
36
37
38
39
40
41
TEMPLATE """{{ if .System }}<|start_header_id|>system<|end_header_id|>

{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>

{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>

{{ .Response }}<|eot_id|>"""    
42
43
44
45
`

	reader := strings.NewReader(input)

Michael Yang's avatar
Michael Yang committed
46
	modelfile, err := ParseFile(reader)
Michael Yang's avatar
lint  
Michael Yang committed
47
	require.NoError(t, err)
48
49
50
51
52
53
54

	expectedCommands := []Command{
		{Name: "model", Args: "model1"},
		{Name: "adapter", Args: "adapter1"},
		{Name: "license", Args: "MIT"},
		{Name: "param1", Args: "value1"},
		{Name: "param2", Args: "value2"},
Josh Yan's avatar
Josh Yan committed
55
		{Name: "template", Args: "{{ if .System }}<|start_header_id|>system<|end_header_id|>\n\n{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>\n\n{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>\n\n{{ .Response }}<|eot_id|>"},
56
57
	}

Michael Yang's avatar
Michael Yang committed
58
	assert.Equal(t, expectedCommands, modelfile.Commands)
59
60
}

Josh Yan's avatar
Josh Yan committed
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
func TestParseFileTrimSpace(t *testing.T) {
	input := `
FROM "     model 1"
ADAPTER      adapter3
LICENSE "MIT       "
PARAMETER param1        value1
PARAMETER param2    value2
TEMPLATE """   {{ if .System }}<|start_header_id|>system<|end_header_id|>

{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>

{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>

{{ .Response }}<|eot_id|>   """    
`

	reader := strings.NewReader(input)

	modelfile, err := ParseFile(reader)
	require.NoError(t, err)

	expectedCommands := []Command{
		{Name: "model", Args: "     model 1"},
		{Name: "adapter", Args: "adapter3"},
		{Name: "license", Args: "MIT       "},
		{Name: "param1", Args: "value1"},
		{Name: "param2", Args: "value2"},
		{Name: "template", Args: "   {{ if .System }}<|start_header_id|>system<|end_header_id|>\n\n{{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>\n\n{{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>\n\n{{ .Response }}<|eot_id|>   "},
	}

	assert.Equal(t, expectedCommands, modelfile.Commands)
}

Michael Yang's avatar
Michael Yang committed
94
func TestParseFileFrom(t *testing.T) {
Michael Yang's avatar
lint  
Michael Yang committed
95
	cases := []struct {
Michael Yang's avatar
tests  
Michael Yang committed
96
97
98
99
		input    string
		expected []Command
		err      error
	}{
Josh Yan's avatar
Josh Yan committed
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
		{
			"FROM \"FOO  BAR  \"",
			[]Command{{Name: "model", Args: "FOO  BAR  "}},
			nil,
		},
		{
			"FROM \"FOO BAR\"\nPARAMETER param1 value1",
			[]Command{{Name: "model", Args: "FOO BAR"}, {Name: "param1", Args: "value1"}},
			nil,
		},
		{
			"FROM     FOOO BAR    ",
			[]Command{{Name: "model", Args: "FOOO BAR"}},
			nil,
		},
		{
			"FROM /what/is/the path ",
			[]Command{{Name: "model", Args: "/what/is/the path"}},
			nil,
		},
Michael Yang's avatar
tests  
Michael Yang committed
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
		{
			"FROM foo",
			[]Command{{Name: "model", Args: "foo"}},
			nil,
		},
		{
			"FROM /path/to/model",
			[]Command{{Name: "model", Args: "/path/to/model"}},
			nil,
		},
		{
			"FROM /path/to/model/fp16.bin",
			[]Command{{Name: "model", Args: "/path/to/model/fp16.bin"}},
			nil,
		},
		{
			"FROM llama3:latest",
			[]Command{{Name: "model", Args: "llama3:latest"}},
			nil,
		},
		{
			"FROM llama3:7b-instruct-q4_K_M",
			[]Command{{Name: "model", Args: "llama3:7b-instruct-q4_K_M"}},
			nil,
		},
		{
			"", nil, errMissingFrom,
		},
		{
			"PARAMETER param1 value1",
			nil,
			errMissingFrom,
		},
		{
			"PARAMETER param1 value1\nFROM foo",
			[]Command{{Name: "param1", Args: "value1"}, {Name: "model", Args: "foo"}},
			nil,
		},
Josh Yan's avatar
Josh Yan committed
158
159
		{
			"PARAMETER what the \nFROM lemons make lemonade ",
Josh Yan's avatar
Josh Yan committed
160
			[]Command{{Name: "what", Args: "the"}, {Name: "model", Args: "lemons make lemonade"}},
Josh Yan's avatar
Josh Yan committed
161
162
			nil,
		},
Michael Yang's avatar
tests  
Michael Yang committed
163
	}
164

Michael Yang's avatar
tests  
Michael Yang committed
165
166
	for _, c := range cases {
		t.Run("", func(t *testing.T) {
Michael Yang's avatar
Michael Yang committed
167
			modelfile, err := ParseFile(strings.NewReader(c.input))
Michael Yang's avatar
lint  
Michael Yang committed
168
			require.ErrorIs(t, err, c.err)
Michael Yang's avatar
Michael Yang committed
169
170
171
			if modelfile != nil {
				assert.Equal(t, c.expected, modelfile.Commands)
			}
Michael Yang's avatar
tests  
Michael Yang committed
172
173
		})
	}
174
175
}

Michael Yang's avatar
Michael Yang committed
176
func TestParseFileParametersMissingValue(t *testing.T) {
177
178
179
180
181
182
183
	input := `
FROM foo
PARAMETER param1
`

	reader := strings.NewReader(input)

Michael Yang's avatar
Michael Yang committed
184
	_, err := ParseFile(reader)
Michael Yang's avatar
lint  
Michael Yang committed
185
	require.ErrorIs(t, err, io.ErrUnexpectedEOF)
186
}
187

Michael Yang's avatar
Michael Yang committed
188
func TestParseFileBadCommand(t *testing.T) {
Michael Yang's avatar
Michael Yang committed
189
190
191
192
	input := `
FROM foo
BADCOMMAND param1 value1
`
193
194
195
196
197
	parserError := &ParserError{
		LineNumber: 3,
		Msg:        errInvalidCommand.Error(),
	}

Michael Yang's avatar
Michael Yang committed
198
	_, err := ParseFile(strings.NewReader(input))
199
200
201
	if !errors.As(err, &parserError) {
		t.Errorf("unexpected error: expected: %s, actual: %s", parserError.Error(), err.Error())
	}
Michael Yang's avatar
Michael Yang committed
202
203
}

Devon Rifkin's avatar
Devon Rifkin committed
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
func TestParseFileRenderer(t *testing.T) {
	input := `
FROM foo
RENDERER renderer1
`

	reader := strings.NewReader(input)

	modelfile, err := ParseFile(reader)
	require.NoError(t, err)

	assert.Equal(t, []Command{{Name: "model", Args: "foo"}, {Name: "renderer", Args: "renderer1"}}, modelfile.Commands)
}

func TestParseFileParser(t *testing.T) {
	input := `
FROM foo
PARSER parser1
`

	reader := strings.NewReader(input)

	modelfile, err := ParseFile(reader)
	require.NoError(t, err)

	assert.Equal(t, []Command{{Name: "model", Args: "foo"}, {Name: "parser", Args: "parser1"}}, modelfile.Commands)
}

Michael Yang's avatar
Michael Yang committed
232
func TestParseFileMessages(t *testing.T) {
Michael Yang's avatar
lint  
Michael Yang committed
233
	cases := []struct {
Michael Yang's avatar
Michael Yang committed
234
235
236
237
238
239
240
		input    string
		expected []Command
		err      error
	}{
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
241
MESSAGE system You are a file parser. Always parse things.
Michael Yang's avatar
Michael Yang committed
242
243
244
`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
245
				{Name: "message", Args: "system: You are a file parser. Always parse things."},
Michael Yang's avatar
Michael Yang committed
246
247
248
249
250
			},
			nil,
		},
		{
			`
251
FROM foo
Michael Yang's avatar
Michael Yang committed
252
MESSAGE system You are a file parser. Always parse things.`,
Michael Yang's avatar
Michael Yang committed
253
254
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
255
				{Name: "message", Args: "system: You are a file parser. Always parse things."},
Michael Yang's avatar
Michael Yang committed
256
257
258
259
260
261
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
262
MESSAGE system You are a file parser. Always parse things.
263
264
MESSAGE user Hey there!
MESSAGE assistant Hello, I want to parse all the things!
Michael Yang's avatar
Michael Yang committed
265
266
267
`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
268
				{Name: "message", Args: "system: You are a file parser. Always parse things."},
Michael Yang's avatar
Michael Yang committed
269
270
271
272
273
274
275
276
277
				{Name: "message", Args: "user: Hey there!"},
				{Name: "message", Args: "assistant: Hello, I want to parse all the things!"},
			},
			nil,
		},
		{
			`
FROM foo
MESSAGE system """
Michael Yang's avatar
Michael Yang committed
278
You are a multiline file parser. Always parse things.
Michael Yang's avatar
Michael Yang committed
279
280
281
282
"""
			`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
283
				{Name: "message", Args: "system: \nYou are a multiline file parser. Always parse things.\n"},
Michael Yang's avatar
Michael Yang committed
284
285
286
287
288
289
			},
			nil,
		},
		{
			`
FROM foo
290
MESSAGE badguy I'm a bad guy!
Michael Yang's avatar
Michael Yang committed
291
`,
292
			nil,
293
294
295
296
			&ParserError{
				LineNumber: 3,
				Msg:        errInvalidMessageRole.Error(),
			},
Michael Yang's avatar
Michael Yang committed
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
		},
		{
			`
FROM foo
MESSAGE system
`,
			nil,
			io.ErrUnexpectedEOF,
		},
		{
			`
FROM foo
MESSAGE system`,
			nil,
			io.ErrUnexpectedEOF,
		},
	}
314

315
	for _, tt := range cases {
Michael Yang's avatar
Michael Yang committed
316
		t.Run("", func(t *testing.T) {
317
318
			modelfile, err := ParseFile(strings.NewReader(tt.input))

Michael Yang's avatar
Michael Yang committed
319
			if modelfile != nil {
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
				assert.Equal(t, tt.expected, modelfile.Commands)
			}

			if tt.err == nil {
				if err != nil {
					t.Fatalf("expected no error, but got %v", err)
				}
				return
			}

			switch tt.err.(type) {
			case *ParserError:
				var pErr *ParserError
				if errors.As(err, &pErr) {
					// got the correct type of error
					return
				}
			}

			if errors.Is(err, tt.err) {
				return
Michael Yang's avatar
Michael Yang committed
341
			}
342
343

			t.Fatalf("unexpected error: expected: %v, actual: %v", tt.err, err)
Michael Yang's avatar
Michael Yang committed
344
345
346
		})
	}
}
347

Michael Yang's avatar
Michael Yang committed
348
func TestParseFileQuoted(t *testing.T) {
Michael Yang's avatar
lint  
Michael Yang committed
349
	cases := []struct {
Michael Yang's avatar
Michael Yang committed
350
351
352
353
354
355
356
		multiline string
		expected  []Command
		err       error
	}{
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
357
SYSTEM """
Michael Yang's avatar
Michael Yang committed
358
This is a
Michael Yang's avatar
Michael Yang committed
359
multiline system.
Michael Yang's avatar
Michael Yang committed
360
361
362
363
"""
			`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
364
				{Name: "system", Args: "\nThis is a\nmultiline system.\n"},
Michael Yang's avatar
Michael Yang committed
365
366
367
368
369
370
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
371
SYSTEM """
Michael Yang's avatar
Michael Yang committed
372
This is a
Michael Yang's avatar
Michael Yang committed
373
multiline system."""
Michael Yang's avatar
Michael Yang committed
374
375
376
			`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
377
				{Name: "system", Args: "\nThis is a\nmultiline system."},
Michael Yang's avatar
Michael Yang committed
378
379
380
381
382
383
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
384
385
SYSTEM """This is a
multiline system."""
Michael Yang's avatar
Michael Yang committed
386
387
388
			`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
389
				{Name: "system", Args: "This is a\nmultiline system."},
Michael Yang's avatar
Michael Yang committed
390
391
392
393
394
395
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
396
SYSTEM """This is a multiline system."""
Michael Yang's avatar
Michael Yang committed
397
398
399
			`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
400
				{Name: "system", Args: "This is a multiline system."},
Michael Yang's avatar
Michael Yang committed
401
402
403
404
405
406
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
407
SYSTEM """This is a multiline system.""
Michael Yang's avatar
Michael Yang committed
408
409
410
411
412
413
414
			`,
			nil,
			io.ErrUnexpectedEOF,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
415
SYSTEM "
Michael Yang's avatar
Michael Yang committed
416
417
418
419
420
421
422
			`,
			nil,
			io.ErrUnexpectedEOF,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
423
424
SYSTEM """
This is a multiline system with "quotes".
Michael Yang's avatar
Michael Yang committed
425
426
427
428
"""
`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
429
				{Name: "system", Args: "\nThis is a multiline system with \"quotes\".\n"},
Michael Yang's avatar
Michael Yang committed
430
431
432
433
434
435
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
436
SYSTEM """"""
Michael Yang's avatar
Michael Yang committed
437
438
439
`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
440
				{Name: "system", Args: ""},
Michael Yang's avatar
Michael Yang committed
441
442
443
444
445
446
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
447
SYSTEM ""
Michael Yang's avatar
Michael Yang committed
448
449
450
`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
451
				{Name: "system", Args: ""},
Michael Yang's avatar
Michael Yang committed
452
453
454
455
456
457
			},
			nil,
		},
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
458
SYSTEM "'"
Michael Yang's avatar
Michael Yang committed
459
460
461
`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
462
				{Name: "system", Args: "'"},
Michael Yang's avatar
Michael Yang committed
463
464
465
			},
			nil,
		},
Michael Yang's avatar
tests  
Michael Yang committed
466
467
468
		{
			`
FROM foo
Michael Yang's avatar
Michael Yang committed
469
SYSTEM """''"'""'""'"'''''""'""'"""
Michael Yang's avatar
tests  
Michael Yang committed
470
471
472
`,
			[]Command{
				{Name: "model", Args: "foo"},
Michael Yang's avatar
Michael Yang committed
473
474
475
476
477
478
479
480
481
482
483
484
485
				{Name: "system", Args: `''"'""'""'"'''''""'""'`},
			},
			nil,
		},
		{
			`
FROM foo
TEMPLATE """
{{ .Prompt }}
"""`,
			[]Command{
				{Name: "model", Args: "foo"},
				{Name: "template", Args: "\n{{ .Prompt }}\n"},
Michael Yang's avatar
tests  
Michael Yang committed
486
487
488
			},
			nil,
		},
489
490
	}

Michael Yang's avatar
Michael Yang committed
491
492
	for _, c := range cases {
		t.Run("", func(t *testing.T) {
Michael Yang's avatar
Michael Yang committed
493
			modelfile, err := ParseFile(strings.NewReader(c.multiline))
Michael Yang's avatar
lint  
Michael Yang committed
494
			require.ErrorIs(t, err, c.err)
Michael Yang's avatar
Michael Yang committed
495
496
497
			if modelfile != nil {
				assert.Equal(t, c.expected, modelfile.Commands)
			}
Michael Yang's avatar
Michael Yang committed
498
499
		})
	}
500
501
}

Michael Yang's avatar
Michael Yang committed
502
func TestParseFileParameters(t *testing.T) {
Michael Yang's avatar
lint  
Michael Yang committed
503
	cases := map[string]struct {
Michael Yang's avatar
tests  
Michael Yang committed
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
		name, value string
	}{
		"numa true":                    {"numa", "true"},
		"num_ctx 1":                    {"num_ctx", "1"},
		"num_batch 1":                  {"num_batch", "1"},
		"num_gqa 1":                    {"num_gqa", "1"},
		"num_gpu 1":                    {"num_gpu", "1"},
		"main_gpu 1":                   {"main_gpu", "1"},
		"use_mmap true":                {"use_mmap", "true"},
		"num_thread 1":                 {"num_thread", "1"},
		"num_keep 1":                   {"num_keep", "1"},
		"seed 1":                       {"seed", "1"},
		"num_predict 1":                {"num_predict", "1"},
		"top_k 1":                      {"top_k", "1"},
		"top_p 1.0":                    {"top_p", "1.0"},
519
		"min_p 0.05":                   {"min_p", "0.05"},
Michael Yang's avatar
tests  
Michael Yang committed
520
521
522
523
524
525
526
527
		"typical_p 1.0":                {"typical_p", "1.0"},
		"repeat_last_n 1":              {"repeat_last_n", "1"},
		"temperature 1.0":              {"temperature", "1.0"},
		"repeat_penalty 1.0":           {"repeat_penalty", "1.0"},
		"presence_penalty 1.0":         {"presence_penalty", "1.0"},
		"frequency_penalty 1.0":        {"frequency_penalty", "1.0"},
		"penalize_newline true":        {"penalize_newline", "true"},
		"stop ### User:":               {"stop", "### User:"},
Josh Yan's avatar
Josh Yan committed
528
		"stop ### User: ":              {"stop", "### User:"},
Michael Yang's avatar
tests  
Michael Yang committed
529
530
531
532
533
534
535
		"stop \"### User:\"":           {"stop", "### User:"},
		"stop \"### User: \"":          {"stop", "### User: "},
		"stop \"\"\"### User:\"\"\"":   {"stop", "### User:"},
		"stop \"\"\"### User:\n\"\"\"": {"stop", "### User:\n"},
		"stop <|endoftext|>":           {"stop", "<|endoftext|>"},
		"stop <|eot_id|>":              {"stop", "<|eot_id|>"},
		"stop </s>":                    {"stop", "</s>"},
Michael Yang's avatar
Michael Yang committed
536
	}
537

Michael Yang's avatar
tests  
Michael Yang committed
538
539
	for k, v := range cases {
		t.Run(k, func(t *testing.T) {
Michael Yang's avatar
Michael Yang committed
540
541
			var b bytes.Buffer
			fmt.Fprintln(&b, "FROM foo")
Michael Yang's avatar
tests  
Michael Yang committed
542
			fmt.Fprintln(&b, "PARAMETER", k)
Michael Yang's avatar
Michael Yang committed
543
			modelfile, err := ParseFile(&b)
Michael Yang's avatar
lint  
Michael Yang committed
544
			require.NoError(t, err)
Michael Yang's avatar
tests  
Michael Yang committed
545
546
547
548

			assert.Equal(t, []Command{
				{Name: "model", Args: "foo"},
				{Name: v.name, Args: v.value},
Michael Yang's avatar
Michael Yang committed
549
			}, modelfile.Commands)
Michael Yang's avatar
Michael Yang committed
550
551
552
553
		})
	}
}

Michael Yang's avatar
Michael Yang committed
554
func TestParseFileComments(t *testing.T) {
Michael Yang's avatar
lint  
Michael Yang committed
555
	cases := []struct {
Michael Yang's avatar
Michael Yang committed
556
557
558
559
560
561
		input    string
		expected []Command
	}{
		{
			`
# comment
562
FROM foo
Michael Yang's avatar
Michael Yang committed
563
564
565
566
567
568
	`,
			[]Command{
				{Name: "model", Args: "foo"},
			},
		},
	}
569

Michael Yang's avatar
Michael Yang committed
570
571
	for _, c := range cases {
		t.Run("", func(t *testing.T) {
Michael Yang's avatar
Michael Yang committed
572
			modelfile, err := ParseFile(strings.NewReader(c.input))
Michael Yang's avatar
lint  
Michael Yang committed
573
			require.NoError(t, err)
Michael Yang's avatar
Michael Yang committed
574
			assert.Equal(t, c.expected, modelfile.Commands)
Michael Yang's avatar
Michael Yang committed
575
576
		})
	}
577
}
Michael Yang's avatar
Michael Yang committed
578

Michael Yang's avatar
Michael Yang committed
579
func TestParseFileFormatParseFile(t *testing.T) {
Michael Yang's avatar
lint  
Michael Yang committed
580
	cases := []string{
Michael Yang's avatar
Michael Yang committed
581
582
583
584
585
586
587
		`
FROM foo
ADAPTER adapter1
LICENSE MIT
PARAMETER param1 value1
PARAMETER param2 value2
TEMPLATE template1
Michael Yang's avatar
Michael Yang committed
588
MESSAGE system You are a file parser. Always parse things.
Michael Yang's avatar
Michael Yang committed
589
590
591
592
593
594
595
596
597
598
599
MESSAGE user Hey there!
MESSAGE assistant Hello, I want to parse all the things!
`,
		`
FROM foo
ADAPTER adapter1
LICENSE MIT
PARAMETER param1 value1
PARAMETER param2 value2
TEMPLATE template1
MESSAGE system """
600
You are a store greeter. Always respond with "Hello!".
Michael Yang's avatar
Michael Yang committed
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
"""
MESSAGE user Hey there!
MESSAGE assistant Hello, I want to parse all the things!
`,
		`
FROM foo
ADAPTER adapter1
LICENSE """
Very long and boring legal text.
Blah blah blah.
"Oh look, a quote!"
"""

PARAMETER param1 value1
PARAMETER param2 value2
TEMPLATE template1
MESSAGE system """
618
You are a store greeter. Always respond with "Hello!".
Michael Yang's avatar
Michael Yang committed
619
620
621
"""
MESSAGE user Hey there!
MESSAGE assistant Hello, I want to parse all the things!
622
623
624
625
`,
		`
FROM foo
SYSTEM ""
Michael Yang's avatar
Michael Yang committed
626
627
628
629
630
`,
	}

	for _, c := range cases {
		t.Run("", func(t *testing.T) {
Michael Yang's avatar
Michael Yang committed
631
			modelfile, err := ParseFile(strings.NewReader(c))
Michael Yang's avatar
lint  
Michael Yang committed
632
			require.NoError(t, err)
Michael Yang's avatar
Michael Yang committed
633

Michael Yang's avatar
Michael Yang committed
634
			modelfile2, err := ParseFile(strings.NewReader(modelfile.String()))
Michael Yang's avatar
lint  
Michael Yang committed
635
			require.NoError(t, err)
Michael Yang's avatar
Michael Yang committed
636

Michael Yang's avatar
Michael Yang committed
637
			assert.Equal(t, modelfile, modelfile2)
Michael Yang's avatar
Michael Yang committed
638
639
640
		})
	}
}
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655

func TestParseFileUTF16ParseFile(t *testing.T) {
	data := `FROM bob
PARAMETER param1 1
PARAMETER param2 4096
SYSTEM You are a utf16 file.
`

	expected := []Command{
		{Name: "model", Args: "bob"},
		{Name: "param1", Args: "1"},
		{Name: "param2", Args: "4096"},
		{Name: "system", Args: "You are a utf16 file."},
	}

656
657
658
659
	t.Run("le", func(t *testing.T) {
		var b bytes.Buffer
		require.NoError(t, binary.Write(&b, binary.LittleEndian, []byte{0xff, 0xfe}))
		require.NoError(t, binary.Write(&b, binary.LittleEndian, utf16.Encode([]rune(data))))
660

661
662
		actual, err := ParseFile(&b)
		require.NoError(t, err)
663

664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
		assert.Equal(t, expected, actual.Commands)
	})

	t.Run("be", func(t *testing.T) {
		var b bytes.Buffer
		require.NoError(t, binary.Write(&b, binary.BigEndian, []byte{0xfe, 0xff}))
		require.NoError(t, binary.Write(&b, binary.BigEndian, utf16.Encode([]rune(data))))

		actual, err := ParseFile(&b)
		require.NoError(t, err)
		assert.Equal(t, expected, actual.Commands)
	})
}

func TestParseMultiByte(t *testing.T) {
	input := `FROM test
	SYSTEM 你好👋`

	expect := []Command{
		{Name: "model", Args: "test"},
		{Name: "system", Args: "你好👋"},
	}

	encodings := []encoding.Encoding{
		unicode.UTF8,
		unicode.UTF16(unicode.LittleEndian, unicode.UseBOM),
		unicode.UTF16(unicode.BigEndian, unicode.UseBOM),
	}

	for _, encoding := range encodings {
		t.Run(fmt.Sprintf("%s", encoding), func(t *testing.T) {
			s, err := encoding.NewEncoder().String(input)
			require.NoError(t, err)

			actual, err := ParseFile(strings.NewReader(s))
			require.NoError(t, err)

			assert.Equal(t, expect, actual.Commands)
		})
	}
704
}
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772

func TestCreateRequest(t *testing.T) {
	cases := []struct {
		input    string
		expected *api.CreateRequest
	}{
		{
			`FROM test`,
			&api.CreateRequest{From: "test"},
		},
		{
			`FROM test
TEMPLATE some template
`,
			&api.CreateRequest{
				From:     "test",
				Template: "some template",
			},
		},
		{
			`FROM test
LICENSE single license
PARAMETER temperature 0.5
MESSAGE user Hello
`,
			&api.CreateRequest{
				From:       "test",
				License:    []string{"single license"},
				Parameters: map[string]any{"temperature": float32(0.5)},
				Messages: []api.Message{
					{Role: "user", Content: "Hello"},
				},
			},
		},
		{
			`FROM test
PARAMETER temperature 0.5
PARAMETER top_k 1
SYSTEM You are a bot.
LICENSE license1
LICENSE license2
MESSAGE user Hello there!
MESSAGE assistant Hi! How are you?
`,
			&api.CreateRequest{
				From:       "test",
				License:    []string{"license1", "license2"},
				System:     "You are a bot.",
				Parameters: map[string]any{"temperature": float32(0.5), "top_k": int64(1)},
				Messages: []api.Message{
					{Role: "user", Content: "Hello there!"},
					{Role: "assistant", Content: "Hi! How are you?"},
				},
			},
		},
	}

	for _, c := range cases {
		s, err := unicode.UTF8.NewEncoder().String(c.input)
		if err != nil {
			t.Fatal(err)
		}

		p, err := ParseFile(strings.NewReader(s))
		if err != nil {
			t.Error(err)
		}

773
		actual, err := p.CreateRequest("")
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
		if err != nil {
			t.Error(err)
		}

		if diff := cmp.Diff(actual, c.expected); diff != "" {
			t.Errorf("mismatch (-got +want):\n%s", diff)
		}
	}
}

func getSHA256Digest(t *testing.T, r io.Reader) (string, int64) {
	t.Helper()

	h := sha256.New()
	n, err := io.Copy(h, r)
	if err != nil {
		t.Fatal(err)
	}

	return fmt.Sprintf("sha256:%x", h.Sum(nil)), n
}

796
func createBinFile(t *testing.T, kv map[string]any, ti []*ggml.Tensor) (string, string) {
797
798
799
800
801
802
803
804
	t.Helper()

	f, err := os.CreateTemp(t.TempDir(), "testbin.*.gguf")
	if err != nil {
		t.Fatal(err)
	}
	defer f.Close()

805
	var base convert.KV = map[string]any{"general.architecture": "test"}
806
807
808
	maps.Copy(base, kv)

	if err := ggml.WriteGGUF(f, base, ti); err != nil {
809
810
811
812
813
814
815
816
817
818
819
820
821
		t.Fatal(err)
	}
	// Calculate sha256 of file
	if _, err := f.Seek(0, 0); err != nil {
		t.Fatal(err)
	}

	digest, _ := getSHA256Digest(t, f)

	return f.Name(), digest
}

func TestCreateRequestFiles(t *testing.T) {
822
823
	n1, d1 := createBinFile(t, nil, nil)
	n2, d2 := createBinFile(t, map[string]any{"foo": "bar"}, nil)
824
825
826
827
828
829

	cases := []struct {
		input    string
		expected *api.CreateRequest
	}{
		{
830
831
832
833
834
835
			fmt.Sprintf("FROM %s", n1),
			&api.CreateRequest{Files: map[string]string{n1: d1}},
		},
		{
			fmt.Sprintf("FROM %s\nFROM %s", n1, n2),
			&api.CreateRequest{Files: map[string]string{n1: d1, n2: d2}},
836
837
838
839
840
841
842
843
844
845
846
847
848
849
		},
	}

	for _, c := range cases {
		s, err := unicode.UTF8.NewEncoder().String(c.input)
		if err != nil {
			t.Fatal(err)
		}

		p, err := ParseFile(strings.NewReader(s))
		if err != nil {
			t.Error(err)
		}

850
		actual, err := p.CreateRequest("")
851
852
853
854
855
856
857
858
859
		if err != nil {
			t.Error(err)
		}

		if diff := cmp.Diff(actual, c.expected); diff != "" {
			t.Errorf("mismatch (-got +want):\n%s", diff)
		}
	}
}
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891

func TestFilesForModel(t *testing.T) {
	tests := []struct {
		name          string
		setup         func(string) error
		wantFiles     []string
		wantErr       bool
		expectErrType error
	}{
		{
			name: "safetensors model files",
			setup: func(dir string) error {
				files := []string{
					"model-00001-of-00002.safetensors",
					"model-00002-of-00002.safetensors",
					"config.json",
					"tokenizer.json",
				}
				for _, file := range files {
					if err := os.WriteFile(filepath.Join(dir, file), []byte("test content"), 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantFiles: []string{
				"model-00001-of-00002.safetensors",
				"model-00002-of-00002.safetensors",
				"config.json",
				"tokenizer.json",
			},
		},
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
		{
			name: "safetensors with both tokenizer.json and tokenizer.model",
			setup: func(dir string) error {
				// Create binary content for tokenizer.model (application/octet-stream)
				binaryContent := make([]byte, 512)
				for i := range binaryContent {
					binaryContent[i] = byte(i % 256)
				}
				files := []string{
					"model-00001-of-00001.safetensors",
					"config.json",
					"tokenizer.json",
				}
				for _, file := range files {
					if err := os.WriteFile(filepath.Join(dir, file), []byte("test content"), 0o644); err != nil {
						return err
					}
				}
				// Write tokenizer.model as binary
				if err := os.WriteFile(filepath.Join(dir, "tokenizer.model"), binaryContent, 0o644); err != nil {
					return err
				}
				return nil
			},
			wantFiles: []string{
				"model-00001-of-00001.safetensors",
				"config.json",
				"tokenizer.json",
				"tokenizer.model",
			},
		},
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
		{
			name: "safetensors with consolidated files - prefers model files",
			setup: func(dir string) error {
				files := []string{
					"model-00001-of-00001.safetensors",
					"consolidated.safetensors",
					"config.json",
				}
				for _, file := range files {
					if err := os.WriteFile(filepath.Join(dir, file), []byte("test content"), 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantFiles: []string{
				"model-00001-of-00001.safetensors", // consolidated files should be excluded
				"config.json",
			},
		},
		{
			name: "safetensors without model-.safetensors files - uses consolidated",
			setup: func(dir string) error {
				files := []string{
					"consolidated.safetensors",
					"config.json",
				}
				for _, file := range files {
					if err := os.WriteFile(filepath.Join(dir, file), []byte("test content"), 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantFiles: []string{
				"consolidated.safetensors",
				"config.json",
			},
		},
		{
			name: "pytorch model files",
			setup: func(dir string) error {
				// Create a file that will be detected as application/zip
				zipHeader := []byte{0x50, 0x4B, 0x03, 0x04} // PK zip header
				files := []string{
					"pytorch_model-00001-of-00002.bin",
					"pytorch_model-00002-of-00002.bin",
					"config.json",
				}
				for _, file := range files {
					content := zipHeader
					if file == "config.json" {
						content = []byte(`{"config": true}`)
					}
					if err := os.WriteFile(filepath.Join(dir, file), content, 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantFiles: []string{
				"pytorch_model-00001-of-00002.bin",
				"pytorch_model-00002-of-00002.bin",
				"config.json",
			},
		},
		{
			name: "consolidated pth files",
			setup: func(dir string) error {
				zipHeader := []byte{0x50, 0x4B, 0x03, 0x04}
				files := []string{
					"consolidated.00.pth",
					"consolidated.01.pth",
					"config.json",
				}
				for _, file := range files {
					content := zipHeader
					if file == "config.json" {
						content = []byte(`{"config": true}`)
					}
					if err := os.WriteFile(filepath.Join(dir, file), content, 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantFiles: []string{
				"consolidated.00.pth",
				"consolidated.01.pth",
				"config.json",
			},
		},
		{
			name: "gguf files",
			setup: func(dir string) error {
				// Create binary content that will be detected as application/octet-stream
				binaryContent := make([]byte, 512)
				for i := range binaryContent {
					binaryContent[i] = byte(i % 256)
				}
				files := []string{
					"model.gguf",
					"config.json",
				}
				for _, file := range files {
					content := binaryContent
					if file == "config.json" {
						content = []byte(`{"config": true}`)
					}
					if err := os.WriteFile(filepath.Join(dir, file), content, 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantFiles: []string{
				"model.gguf",
				"config.json",
			},
		},
		{
			name: "bin files as gguf",
			setup: func(dir string) error {
				binaryContent := make([]byte, 512)
				for i := range binaryContent {
					binaryContent[i] = byte(i % 256)
				}
				files := []string{
					"model.bin",
					"config.json",
				}
				for _, file := range files {
					content := binaryContent
					if file == "config.json" {
						content = []byte(`{"config": true}`)
					}
					if err := os.WriteFile(filepath.Join(dir, file), content, 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantFiles: []string{
				"model.bin",
				"config.json",
			},
		},
		{
			name: "no model files found",
			setup: func(dir string) error {
				// Only create non-model files
				files := []string{"README.md", "config.json"}
				for _, file := range files {
					if err := os.WriteFile(filepath.Join(dir, file), []byte("content"), 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantErr:       true,
			expectErrType: ErrModelNotFound,
		},
		{
			name: "invalid content type for pytorch model",
			setup: func(dir string) error {
				// Create pytorch model file with wrong content type (text instead of zip)
				files := []string{
					"pytorch_model.bin",
					"config.json",
				}
				for _, file := range files {
					content := []byte("plain text content")
					if err := os.WriteFile(filepath.Join(dir, file), content, 0o644); err != nil {
						return err
					}
				}
				return nil
			},
			wantErr: true,
		},
	}

	tmpDir := t.TempDir()

	for _, tt := range tests {
		t.Run(tt.name, func(t *testing.T) {
			testDir := filepath.Join(tmpDir, tt.name)
			if err := os.MkdirAll(testDir, 0o755); err != nil {
				t.Fatalf("Failed to create test directory: %v", err)
			}

			if err := tt.setup(testDir); err != nil {
				t.Fatalf("Setup failed: %v", err)
			}

			files, err := filesForModel(testDir)

			if tt.wantErr {
				if err == nil {
					t.Error("Expected error, but got none")
				}
				if tt.expectErrType != nil && err != tt.expectErrType {
					t.Errorf("Expected error type %v, got %v", tt.expectErrType, err)
				}
				return
			}

			if err != nil {
				t.Errorf("Unexpected error: %v", err)
				return
			}

			var relativeFiles []string
			for _, file := range files {
				rel, err := filepath.Rel(testDir, file)
				if err != nil {
					t.Fatalf("Failed to get relative path: %v", err)
				}
				relativeFiles = append(relativeFiles, rel)
			}

			if len(relativeFiles) != len(tt.wantFiles) {
				t.Errorf("Expected %d files, got %d: %v", len(tt.wantFiles), len(relativeFiles), relativeFiles)
			}

			fileSet := make(map[string]bool)
			for _, file := range relativeFiles {
				fileSet[file] = true
			}

			for _, wantFile := range tt.wantFiles {
				if !fileSet[wantFile] {
					t.Errorf("Missing expected file: %s", wantFile)
				}
			}
		})
	}
}