tools_test.go 30.6 KB
Newer Older
1
2
3
package tools

import (
4
	"strings"
5
	"testing"
6
	"text/template"
7
8
9
10
11

	"github.com/google/go-cmp/cmp"
	"github.com/ollama/ollama/api"
)

12
13
14
15
16
func TestParser(t *testing.T) {
	qwen, err := template.New("qwen").Parse(`{{if .ToolCalls}}<tool_call>{{range .ToolCalls}}{"name": "{{.Function.Name}}", "arguments": {{.Function.Arguments}}}{{end}}</tool_call>{{end}}`)
	if err != nil {
		t.Fatalf("Failed to parse template: %v", err)
	}
17

18
	deepseek, err := template.New("deepseek").Parse("{{if .ToolCalls}}<|tool▁calls▁begin|>{{range .ToolCalls}}<|tool▁call▁begin|>function<|tool▁sep|>get_current_weather\n```json\n{\"location\": \"Tokyo\"}\n```<|tool▁call▁end|>{{end}}<|tool▁calls▁end|><|end▁of▁sentence|>{{end}}")
19
	if err != nil {
20
		t.Fatalf("Failed to parse template: %v", err)
21
22
	}

23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
	json, err := template.New("json").Parse(`{{if .ToolCalls}}{{range .ToolCalls}}{"name": "{{.Function.Name}}", "arguments": {{.Function.Arguments}}}{{end}}{{end}}`)
	if err != nil {
		t.Fatalf("Failed to parse template: %v", err)
	}

	mistral, err := template.New("mistral").Parse(`{{if .ToolCalls}}[TOOL_CALLS] [{{range .ToolCalls}}{"name": "{{.Function.Name}}", "arguments": {{.Function.Arguments}}}{{end}}][/TOOL_CALLS]{{end}}`)
	if err != nil {
		t.Fatalf("Failed to parse template: %v", err)
	}

	list, err := template.New("list").Parse(`{{if .ToolCalls}}[{{range .ToolCalls}}{"name": "{{.Function.Name}}", "arguments": {{.Function.Arguments}}}{{end}}]{{end}}`)
	if err != nil {
		t.Fatalf("Failed to parse template: %v", err)
	}

	tools := []api.Tool{
		{
			Type: "function",
			Function: api.ToolFunction{
				Name:        "get_temperature",
				Description: "Retrieve the temperature for a given location",
44
				Parameters: api.ToolFunctionParameters{
45
46
					Type:     "object",
					Required: []string{"city"},
Devon Rifkin's avatar
Devon Rifkin committed
47
					Properties: map[string]api.ToolProperty{
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
						"format": {
							Type:        api.PropertyType{"string"},
							Description: "The format to return the temperature in",
							Enum:        []any{"fahrenheit", "celsius"},
						},
						"city": {
							Type:        api.PropertyType{"string"},
							Description: "The city to get the temperature for",
						},
					},
				},
			},
		},
		{
			Type: "function",
			Function: api.ToolFunction{
				Name:        "get_conditions",
				Description: "Retrieve the current weather conditions for a given location",
66
				Parameters: api.ToolFunctionParameters{
67
					Type: "object",
Devon Rifkin's avatar
Devon Rifkin committed
68
					Properties: map[string]api.ToolProperty{
69
70
71
72
73
74
75
76
						"location": {
							Type:        api.PropertyType{"string"},
							Description: "The location to get the weather conditions for",
						},
					},
				},
			},
		},
77
78
79
80
81
82
83
		{
			Type: "function",
			Function: api.ToolFunction{
				Name:        "say_hello",
				Description: "Say hello",
			},
		},
84
85
86
87
88
89
90
91
92
93
94
95
		{
			Type: "function",
			Function: api.ToolFunction{
				Name:        "say_hello_world",
				Description: "Say hello world",
			},
		},
		{
			Type: "function",
			Function: api.ToolFunction{
				Name:        "get_address",
				Description: "Get the address of a given location",
96
				Parameters: api.ToolFunctionParameters{
97
					Type: "object",
Devon Rifkin's avatar
Devon Rifkin committed
98
					Properties: map[string]api.ToolProperty{
99
100
101
102
103
104
105
106
107
108
109
110
111
						"location": {
							Type:        api.PropertyType{"string"},
							Description: "The location to get the address for",
						},
					},
				},
			},
		},
		{
			Type: "function",
			Function: api.ToolFunction{
				Name:        "add",
				Description: "Add two numbers",
112
				Parameters: api.ToolFunctionParameters{
113
					Type: "object",
Devon Rifkin's avatar
Devon Rifkin committed
114
					Properties: map[string]api.ToolProperty{
115
116
117
118
119
120
121
122
123
124
125
126
						"a": {
							Type:        api.PropertyType{"string"},
							Description: "The first number to add",
						},
						"b": {
							Type:        api.PropertyType{"string"},
							Description: "The second number to add",
						},
					},
				},
			},
		},
127
	}
128
129

	tests := []struct {
130
131
132
133
134
		name    string
		inputs  []string
		tmpl    *template.Template
		content string
		calls   []api.ToolCall
135
136
	}{
		{
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
			name:    "no tool calls - just text",
			inputs:  []string{"Hello, how can I help you today?"},
			content: "Hello, how can I help you today?",
			tmpl:    qwen,
			calls:   nil,
		},
		{
			name:    "empty input",
			inputs:  []string{""},
			content: "",
			tmpl:    qwen,
			calls:   nil,
		},
		{
			name:    "tool call",
			inputs:  []string{`<tool_call>{"name": "get_conditions", "arguments": {"location": "San Francisco"}}</tool_call>`},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
156
157
				{
					Function: api.ToolCallFunction{
158
159
160
161
						Index: 0,
						Name:  "get_conditions",
						Arguments: api.ToolCallFunctionArguments{
							"location": "San Francisco",
162
163
164
165
						},
					},
				},
			},
166
		},
167
		{
168
169
170
171
172
173
174
175
176
177
178
179
180
181
			name:    "empty args",
			inputs:  []string{`<tool_call>{"name": "get_conditions", "arguments": {}}</tool_call>`},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index:     0,
						Name:      "get_conditions",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
			},
		},
182
183
184
185
186
187
		{
			name:    "text before tool call",
			inputs:  []string{`Let me check the weather. <tool_call>{"name": "get_temperature", "arguments": {"city": "New York"}}</tool_call>`},
			content: "Let me check the weather. ",
			tmpl:    qwen,
			calls: []api.ToolCall{
188
189
				{
					Function: api.ToolCallFunction{
190
191
192
193
194
195
196
197
198
						Index: 0,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city": "New York",
						},
					},
				},
			},
		},
199
200
201
202
203
204
205
		{
			name:    "qwen no args with text",
			inputs:  []string{"Let me say hello to the user. I'll use the say_hello tool. "},
			content: "Let me say hello to the user. I'll use the say_hello tool. ",
			tmpl:    qwen,
			calls:   nil,
		},
206
207
208
209
210
211
212
213
214
215
216
217
218
		{
			name:    "two tool calls in a list",
			inputs:  []string{`[TOOL_CALLS] [{"name": "get_temperature", "arguments": {"city": "London", "format": "fahrenheit"}}, {"name": "get_conditions", "arguments": {"location": "Tokyo"}}][/TOOL_CALLS]`},
			content: "",
			tmpl:    mistral,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index: 0,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city":   "London",
							"format": "fahrenheit",
219
220
221
222
223
						},
					},
				},
				{
					Function: api.ToolCallFunction{
224
225
226
227
						Index: 1,
						Name:  "get_conditions",
						Arguments: api.ToolCallFunctionArguments{
							"location": "Tokyo",
228
229
230
231
						},
					},
				},
			},
232
233
		},
		{
234
			name:    "qwen two tool calls",
235
236
237
238
			inputs:  []string{`Okay, let's call both tools! <tool_call>{"name": "get_temperature", "arguments": {"city": "London", "format": "fahrenheit"}}</tool_call><tool_call>{"name": "get_conditions", "arguments": {"location": "Tokyo"}}</tool_call>`},
			content: "Okay, let's call both tools! ",
			tmpl:    qwen,
			calls: []api.ToolCall{
239
240
				{
					Function: api.ToolCallFunction{
241
242
243
244
245
						Index: 0,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city":   "London",
							"format": "fahrenheit",
246
247
248
249
250
						},
					},
				},
				{
					Function: api.ToolCallFunction{
251
252
253
254
						Index: 1,
						Name:  "get_conditions",
						Arguments: api.ToolCallFunctionArguments{
							"location": "Tokyo",
255
256
257
258
						},
					},
				},
			},
259
		},
260
		{
261
262
263
			name:    "empty args followed by args",
			inputs:  []string{`Let me say hello and check the weather. <tool_call>{"name": "say_hello", "arguments": {}}</tool_call><tool_call>{"name": "get_temperature", "arguments": {"city": "London", "format": "fahrenheit"}}</tool_call>`},
			content: "Let me say hello and check the weather. ",
264
265
266
267
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
268
269
270
						Index:     0,
						Name:      "say_hello",
						Arguments: api.ToolCallFunctionArguments{},
271
272
					},
				},
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
				{
					Function: api.ToolCallFunction{
						Index: 1,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city":   "London",
							"format": "fahrenheit",
						},
					},
				},
			},
		},
		{
			name:    "qwen empty followed by args",
			inputs:  []string{`Let me check the weather. <tool_call>{"name": "get_conditions", "arguments": {}}</tool_call><tool_call>{"name": "get_conditions", "arguments": {"location": "Tokyo"}}`},
			content: "Let me check the weather. ",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index:     0,
						Name:      "get_conditions",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
298
299
300
301
302
303
304
305
306
307
308
				{
					Function: api.ToolCallFunction{
						Index: 1,
						Name:  "get_conditions",
						Arguments: api.ToolCallFunctionArguments{
							"location": "Tokyo",
						},
					},
				},
			},
		},
309
310
311
312
313
314
		{
			name:    "deepseek",
			inputs:  []string{"<think>Wait, I need to call a tool</think><|tool▁calls▁begin|><|tool▁call▁begin|>function<|tool▁sep|>get_temperature\n```json\n{\"city\": \"Tokyo\"}\n```<|tool▁call▁end|><|tool▁calls▁end|><|end▁of▁sentence|>"},
			content: "<think>Wait, I need to call a tool</think>",
			tmpl:    deepseek,
			calls: []api.ToolCall{
315
316
				{
					Function: api.ToolCallFunction{
317
318
319
320
						Index: 0,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city": "Tokyo",
321
322
323
						},
					},
				},
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
			},
		},
		{
			name: "deepseek incremental",
			inputs: []string{
				"<think>Wait",
				", I need",
				" to call",
				" a tool</think><|too",
				"l▁calls▁begin",
				"|>",
				"<|tool▁call▁begin|>function<|tool▁sep|>get_temperature\n",
				"```json\n",
				"{\"city\": \"Tokyo\"}\n",
				"```",
				"<|tool▁c", "all▁end|>",
				"<|tool▁calls▁end|>",
				"<|end▁of▁sentence|>",
			},
			content: "<think>Wait, I need to call a tool</think>",
			tmpl:    deepseek,
			calls: []api.ToolCall{
346
347
				{
					Function: api.ToolCallFunction{
348
349
350
351
						Index: 0,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city": "Tokyo",
352
353
354
355
356
357
						},
					},
				},
			},
		},
		{
358
359
360
361
362
363
364
365
366
367
368
369
			name: "json",
			inputs: []string{
				"{",
				"\"name\": \"get_temperature\",",
				"\"arguments\": {",
				"\"city\": \"Tokyo\"",
				"}",
				"}",
			},
			content: "",
			tmpl:    json,
			calls: []api.ToolCall{
370
371
				{
					Function: api.ToolCallFunction{
372
373
374
375
						Index: 0,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city": "Tokyo",
376
377
378
379
						},
					},
				},
			},
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
		},
		{
			name: "json maybe a tool call",
			inputs: []string{
				"{",
				"\"name\": \"get_temperature\",",
				"\"arguments\": {",
			},
			content: "",
			tmpl:    json,
			calls:   nil,
		},
		{
			name: "json not a tool call",
			inputs: []string{
				"{",
				"\"name\": \"search\", ",
				"\"arguments\": {",
				"\"query\": \"What is the capital of Canada?\"",
				"}",
				"}",
			},
			content: "{\"name\": \"search\", \"arguments\": {\"query\": \"What is the capital of Canada?\"}}",
			tmpl:    json,
			calls:   nil,
		},
		{
			name: "json object followed by tool call",
			inputs: []string{
				"{\"name\": \"jeff\"}",
				"{\"name\": \"get_conditions\", \"arguments\": {\"location\": \"San Francisco\"}}",
			},
			content: "{\"name\": \"jeff\"}{\"name\": \"get_conditions\", \"arguments\": {\"location\": \"San Francisco\"}}",
			tmpl:    json,
		},
		{
			name: "json object followed by tool call split",
			inputs: []string{
				"{\"name\": \"jeff\"} {",
				"\"name\": \"get_conditions\", \"arguments\": {\"location\": \"San Francisco\"}}",
			},
			content: "{\"name\": \"jeff\"} {\"name\": \"get_conditions\", \"arguments\": {\"location\": \"San Francisco\"}}",
			tmpl:    json,
		},
		{
			name: "json code",
			inputs: []string{
				"for { fmt.Println(\"hello\") }",
			},
			content: "for { fmt.Println(\"hello\") }",
			tmpl:    json,
		},
		{
			name: "list multiple",
			inputs: []string{
				"[",
				"{",
				"\"name\": \"get_temperature\", ",
				"\"arguments\": {",
				"\"city\": \"London\"",
				"}",
				"},",
				"{",
				"\"name\": \"get_conditions\", ",
				"\"arguments\": {",
				"\"location\": \"Tokyo\"",
				"}",
				"}]",
			},
			content: "",
			tmpl:    list,
			calls: []api.ToolCall{
452
453
				{
					Function: api.ToolCallFunction{
454
455
456
457
						Index: 0,
						Name:  "get_temperature",
						Arguments: api.ToolCallFunctionArguments{
							"city": "London",
458
459
460
						},
					},
				},
461
462
463
464
465
466
467
468
469
470
471
472
473
				{
					Function: api.ToolCallFunction{
						Index: 1,
						Name:  "get_conditions",
						Arguments: api.ToolCallFunctionArguments{
							"location": "Tokyo",
						},
					},
				},
			},
		},
		{
			name: "list partial",
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
			inputs: []string{
				"[{",
				"\"name\": \"get_conditions\", ",
				"\"arguments\": {",
				"\"location\": \"Tokyo\"",
				"}",
				"}",
			},
			content: "",
			tmpl:    list,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index: 0,
						Name:  "get_conditions",
						Arguments: api.ToolCallFunctionArguments{
							"location": "Tokyo",
						},
					},
				},
			},
		},
		{
			name: "list invalid",
498
499
500
501
502
503
504
505
506
507
508
509
510
			inputs: []string{
				"[",
				"{",
				"\"name\": \"search\", ",
				"\"arguments\": {",
				"\"query\": \"What is the capital of Canada?\"",
				"}",
				"}",
			},
			content: "",
			tmpl:    list,
			calls:   nil,
		},
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
		{
			name: "list trailing ]",
			inputs: []string{
				"[",
				"{",
				"\"name\": \"get_conditions\", ",
				"\"arguments\": {",
				"\"location\": \"Tokyo\"",
				"}",
				"}",
				"]",
				"]",
			},
			content: "",
			tmpl:    list,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index: 0,
						Name:  "get_conditions",
						Arguments: api.ToolCallFunctionArguments{
							"location": "Tokyo",
						},
					},
				},
			},
		},
538
539
540
541
542
543
		{
			name: "list not a tool call",
			inputs: []string{
				"[special",
				" del",
				"ivery]",
544
			},
545
546
547
			content: "[special delivery]",
			tmpl:    list,
			calls:   nil,
548
		},
549
550
551
552
553
554
555
		{
			name: "tool name with collision",
			inputs: []string{
				"<tool_call>",
				"{",
				"\"name\": \"say_hello",
				"_world\",",
556
				"\"arguments\": {}}",
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
				"}",
			},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index:     0,
						Name:      "say_hello_world",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
			},
		},
		{
			name: "tool name with collision multiple",
			inputs: []string{
				"<tool_call>",
				"{",
				"\"name\": \"say_hello",
				"_world\",",
578
				"\"arguments\": {}}",
579
580
581
582
583
				"</tool_call>",
				"<tool_call>",
				"{",
				"\"name\": \"say_hello",
				"\",",
584
				"\"arguments\": {}}",
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
				"</tool_call>",
			},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index:     0,
						Name:      "say_hello_world",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
				{
					Function: api.ToolCallFunction{
						Index:     1,
						Name:      "say_hello",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
			},
		},
		{
			name: "tool name with collision non streaming",
			inputs: []string{
				`<tool_call>{"name": "say_hello`,
			},
			content: "",
			tmpl:    qwen,
			calls:   nil,
		},
		{
			name: "tool name with collision non streaming multiple",
			inputs: []string{
618
				`<tool_call>{"name": "say_hello", "arguments": {}}</tool_call><tool_call>{"name": "say_hello_world", "arguments": {}}`,
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
			},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index:     0,
						Name:      "say_hello",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
				{
					Function: api.ToolCallFunction{
						Index:     1,
						Name:      "say_hello_world",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
			},
		},
		{
			name: "tool name with collision non streaming shorter",
			inputs: []string{
642
				`<tool_call>{"name": "say_hello", "arguments": {}}</tool_call>`,
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
			},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index:     0,
						Name:      "say_hello",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
			},
		},
		{
			name: "tool name with collision non streaming longer",
			inputs: []string{
659
				`<tool_call>{"name": "say_hello_world", "arguments": {}}</tool_call>`,
660
661
662
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
704
705
706
707
708
709
710
711
712
713
714
715
			},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index:     0,
						Name:      "say_hello_world",
						Arguments: api.ToolCallFunctionArguments{},
					},
				},
			},
		},
		{
			name: "tool name with substring of another",
			inputs: []string{
				"{",
				"\"name\": \"get_address\",",
				"\"arguments\": {",
				"\"location\": \"London\"",
				"}",
				"}",
			},
			content: "",
			tmpl:    json,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index: 0,
						Name:  "get_address",
						Arguments: api.ToolCallFunctionArguments{
							"location": "London",
						},
					},
				},
			},
		},
		{
			name: "tool name with substring of another",
			inputs: []string{
				`<tool_call>{"name": "get_address", "arguments": {"location": "London"}}</tool_call>`,
			},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index: 0,
						Name:  "get_address",
						Arguments: api.ToolCallFunctionArguments{
							"location": "London",
						},
					},
				},
			},
		},
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
		{
			name: "args before name",
			inputs: []string{
				`<tool_call>{"arguments": {"a": "5", "b": "10"}, "name": "add"}</tool_call>`,
			},
			content: "",
			tmpl:    qwen,
			calls: []api.ToolCall{
				{
					Function: api.ToolCallFunction{
						Index: 0,
						Name:  "add",
						Arguments: api.ToolCallFunctionArguments{
							"a": "5",
							"b": "10",
						},
					},
				},
			},
		},
736
737
738
739
	}

	for _, tt := range tests {
		t.Run(tt.name, func(t *testing.T) {
740
			parser := NewParser(tt.tmpl, tools)
741

742
743
744
745
746
747
			var calls []api.ToolCall
			var content string
			for _, input := range tt.inputs {
				tcs, c := parser.Add(input)
				calls = append(calls, tcs...)
				content += c
748
749
			}

750
751
			if content != tt.content {
				t.Errorf("Expected content %q, got %q", tt.content, content)
752
753
			}

754
755
756
757
758
759
760
761
			if len(calls) != len(tt.calls) {
				t.Fatalf("Expected %d tool calls, got %d", len(tt.calls), len(calls))
			}

			for i, want := range tt.calls {
				if diff := cmp.Diff(calls[i], want); diff != "" {
					t.Errorf("Tool call %d mismatch (-got +want):\n%s", i, diff)
				}
762
763
764
765
766
			}
		})
	}
}

767
768
769
770
771
772
func TestDone(t *testing.T) {
	tests := []struct {
		name   string
		tag    string
		buffer []byte
		want   bool
773
774
	}{
		{
775
776
777
778
			name:   "empty",
			tag:    "<tool_call>",
			buffer: []byte{},
			want:   false,
779
780
		},
		{
781
782
783
784
			name:   "empty",
			tag:    "<tool_call>",
			buffer: []byte{},
			want:   false,
785
786
		},
		{
787
788
789
790
			name:   "json open",
			tag:    "{",
			buffer: []byte("{\"name\": \"get_weather\""),
			want:   false,
791
792
		},
		{
793
794
795
796
			name:   "json closed",
			tag:    "{",
			buffer: []byte("{\"name\": \"get_weather\"}"),
			want:   true,
797
798
		},
		{
799
800
801
802
			name:   "json empty",
			tag:    "{",
			buffer: []byte("{}"),
			want:   true,
803
804
		},
		{
805
806
807
808
			name:   "list open",
			tag:    "[",
			buffer: []byte("[{\"name\": \"get_weather\""),
			want:   false,
809
810
		},
		{
811
812
813
814
			name:   "list closed",
			tag:    "[",
			buffer: []byte("[{\"name\": \"get_weather\"}]"),
			want:   true,
815
816
		},
		{
817
818
819
820
			name:   "list empty",
			tag:    "[",
			buffer: []byte("[]"),
			want:   true,
821
		},
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
	}

	for _, tt := range tests {
		t.Run(tt.name, func(t *testing.T) {
			parser := &Parser{
				tag:    tt.tag,
				buffer: tt.buffer,
			}
			got := parser.done()
			if got != tt.want {
				t.Errorf("done() = %t, want %t", got, tt.want)
			}
		})
	}
}

func TestContent(t *testing.T) {
	tests := []struct {
		name    string
		tag     string
		content []byte
		want    string
		n       int
	}{
846
		{
847
848
849
850
851
			name:    "empty",
			content: []byte{},
			tag:     "{",
			want:    "",
			n:       0,
852
853
		},
		{
854
855
856
857
858
			name:    "tag",
			tag:     "<tool_call>",
			content: []byte("<tool_call>{\"name\": \"get_temperature\""),
			want:    "",
			n:       0,
859
860
		},
		{
861
862
863
864
865
			name:    "json object",
			tag:     "{",
			content: []byte("{\"name\": \"get_temperature\"}"),
			want:    "{\"name\": \"get_temperature\"}",
			n:       0,
866
867
		},
		{
868
869
870
871
872
			name:    "json object after called",
			tag:     "{",
			content: []byte("{\"hello\": \"world\"}"),
			want:    "{\"hello\": \"world\"}",
			n:       0,
873
874
		},
		{
875
876
877
878
879
			name:    "json object after called",
			tag:     "{",
			content: []byte("{\"hello\": \"world\"}"),
			want:    "",
			n:       1,
880
881
		},
		{
882
883
884
885
886
			name:    "list",
			tag:     "[",
			content: []byte("[{\"name\": \"get_temperature\"}]"),
			want:    "[{\"name\": \"get_temperature\"}]",
			n:       0,
887
888
		},
		{
889
890
891
892
893
			name:    "code",
			tag:     "{",
			content: []byte("{ fmt.Println(\"hello\")"),
			want:    "{ fmt.Println(\"hello\")",
			n:       0,
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
	}

	for _, tt := range tests {
		t.Run(tt.name, func(t *testing.T) {
			parser := &Parser{
				tag:    tt.tag,
				buffer: tt.content,
				n:      tt.n,
			}
			got := parser.Content()
			if got != tt.want {
				t.Errorf("Content() = %q, want %q", got, tt.want)
			}
		})
	}
}

func TestFindTag(t *testing.T) {
	cases := []struct {
		name   string
		buffer []byte
		tag    string
		i      int
		found  bool
	}{
920
		{
921
922
923
924
925
			name:   "no overlap",
			buffer: []byte("hello world"),
			tag:    "<tool_call>",
			i:      -1,
			found:  false,
926
927
		},
		{
928
929
930
931
932
			name:   "full overlap",
			buffer: []byte("<tool_call>"),
			tag:    "<tool_call>",
			i:      0,
			found:  true,
933
934
		},
		{
935
936
937
938
939
			name:   "whitespace",
			buffer: []byte("    <tool_call>\n {\"name\": \"bob\"}"),
			tag:    "<tool_call>",
			i:      4,
			found:  true,
940
941
		},
		{
942
943
944
945
946
			name:   "over",
			buffer: []byte("<tool_call>{\"name\""),
			tag:    "<tool_call>",
			i:      0,
			found:  true,
947
948
		},
		{
949
950
951
952
953
			name:   "partial overlap",
			buffer: []byte("text <tool_call>"),
			tag:    "<tool_call>",
			i:      5,
			found:  true,
954
955
		},
		{
956
957
958
959
960
			name:   "overlap with extra",
			buffer: []byte("<tool_calls><tool_call>"),
			tag:    "<tool_calls>",
			i:      0,
			found:  true,
961
962
		},
		{
963
964
965
966
967
			name:   "delimiter longer than string",
			buffer: []byte("<tool>"),
			tag:    "<tool_call>",
			i:      -1,
			found:  false,
968
969
		},
		{
970
971
972
973
974
			name:   "empty string",
			buffer: []byte{},
			tag:    "<tool_call>",
			i:      -1,
			found:  false,
975
976
		},
		{
977
978
979
980
981
			name:   "single char overlap",
			buffer: []byte("test<"),
			tag:    "<tool_call>",
			i:      4,
			found:  false,
982
983
		},
		{
984
985
986
987
988
			name:   "partial tool call",
			buffer: []byte("hello <tool_"),
			tag:    "<tool_call>",
			i:      6,
			found:  false,
989
990
		},
		{
991
992
993
994
995
			name:   "square bracket",
			buffer: []byte("calling tools: ["),
			tag:    "[",
			i:      15,
			found:  true,
996
997
		},
		{
998
999
1000
1001
1002
			name:   "bracket",
			buffer: []byte("{\"name\": \"bob\""),
			tag:    "{",
			i:      0,
			found:  true,
1003
1004
		},
		{
1005
1006
1007
1008
1009
			name:   "bracket with whitespace",
			buffer: []byte("\n\n{\n\"name\": \"bob\""),
			tag:    "{",
			i:      2,
			found:  true,
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
	}

	for _, tt := range cases {
		t.Run(tt.name, func(t *testing.T) {
			parser := &Parser{
				tag:    tt.tag,
				buffer: tt.buffer,
				n:      0,
			}
			i, found := parser.findTag()
			if i != tt.i {
				t.Errorf("findTag(%q, %q) = %d; want %d", tt.buffer, tt.tag, i, tt.i)
			}
			if found != tt.found {
				t.Errorf("findTag(%q, %q) = %t; want %t", tt.buffer, tt.tag, found, tt.found)
			}
		})
	}
}

func TestFindArguments(t *testing.T) {
	tests := []struct {
		name   string
		buffer []byte
		want   map[string]any
	}{
1037
		{
1038
1039
1040
			name:   "empty string",
			buffer: []byte{},
			want:   nil,
1041
1042
		},
		{
1043
1044
1045
			name:   "whitespace only",
			buffer: []byte("   \n\t  "),
			want:   nil,
1046
1047
		},
		{
1048
1049
1050
			name:   "unbalanced braces - missing closing",
			buffer: []byte(`{"format": "fahrenheit", "location": "San Francisco"`),
			want:   nil,
1051
1052
		},
		{
1053
1054
1055
1056
1057
			name:   "unbalanced braces - extra closing",
			buffer: []byte(`{"format": "fahrenheit"}}`),
			want: map[string]any{
				"format": "fahrenheit",
			},
1058
1059
		},
		{
1060
1061
1062
			name:   "invalid JSON",
			buffer: []byte(`{format: fahrenheit, location: "San Francisco"}`),
			want:   nil,
1063
1064
		},
		{
1065
1066
1067
1068
1069
1070
			name:   "valid json",
			buffer: []byte(`{"name": "get_temperature", "arguments": {"format": "fahrenheit", "location": "San Francisco, CA"}}`),
			want: map[string]any{
				"format":   "fahrenheit",
				"location": "San Francisco, CA",
			},
1071
1072
		},
		{
1073
1074
1075
1076
1077
1078
			name:   "valid arguments with special tokens",
			buffer: []byte(`[tool]get_temperature[args]{"format": "fahrenheit", "location": "San Francisco, CA"}[end]`),
			want: map[string]any{
				"format":   "fahrenheit",
				"location": "San Francisco, CA",
			},
1079
1080
		},
		{
1081
			name:   "valid arguments in array",
1082
			buffer: []byte(`[{"name": "get_temperature", "arguments": {"format": "fahrenheit", "location": "San Francisco, CA"}}`),
1083
1084
1085
1086
			want: map[string]any{
				"format":   "fahrenheit",
				"location": "San Francisco, CA",
			},
1087
1088
		},
		{
1089
1090
1091
1092
1093
1094
			name:   "nested deep",
			buffer: []byte(`{"function": {"name": "get_temperature", "arguments": {"format": "fahrenheit", "location": "San Francisco, CA"}}}`),
			want: map[string]any{
				"format":   "fahrenheit",
				"location": "San Francisco, CA",
			},
1095
1096
		},
		{
1097
			name:   "one arg",
1098
			buffer: []byte(`get_temperature({"location": "San Francisco, CA"})`),
1099
1100
1101
			want: map[string]any{
				"location": "San Francisco, CA",
			},
1102
1103
		},
		{
1104
			name:   "two args",
1105
			buffer: []byte(`[{"name": "get_temperature", "arguments": {"location": "San Francisco, CA", "format": "fahrenheit"}}, {"name": "get_weather", "arguments": {"location": "San Francisco, CA", "format": "fahrenheit"}}]`),
1106
1107
1108
1109
			want: map[string]any{
				"location": "San Francisco, CA",
				"format":   "fahrenheit",
			},
1110
1111
		},
		{
1112
			name:   "deepseek",
1113
			buffer: []byte("<|tool▁calls▁begin|><|tool▁call▁begin|>function<|tool▁sep|>get_temperature\n```json\n{\"location\": \"Tokyo\"}\n```<|tool▁call▁end|><|tool▁calls▁end|><|end▁of▁sentence|>"),
1114
1115
1116
			want: map[string]any{
				"location": "Tokyo",
			},
1117
		},
1118
1119
		{
			name:   "deepseek",
1120
			buffer: []byte(`"arguments": {"location": "Tokyo"}}</tool_call>`),
1121
1122
1123
1124
			want: map[string]any{
				"location": "Tokyo",
			},
		},
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
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
		{
			name:   "string with braces",
			buffer: []byte(`{"name": "process_code", "arguments": {"code": "if (x > 0) { return true; }"}}`),
			want: map[string]any{
				"code": "if (x > 0) { return true; }",
			},
		},
		{
			name:   "string with nested json",
			buffer: []byte(`{"name": "send_data", "arguments": {"payload": "{\"nested\": {\"key\": \"value\"}}"}}`),
			want: map[string]any{
				"payload": `{"nested": {"key": "value"}}`,
			},
		},
		{
			name:   "string with escaped quotes and braces",
			buffer: []byte(`{"name": "analyze", "arguments": {"text": "The JSON is: {\"key\": \"val{ue}\"}"}}`),
			want: map[string]any{
				"text": `The JSON is: {"key": "val{ue}"}`,
			},
		},
		{
			name:   "multiple objects with string containing braces",
			buffer: []byte(`{"name": "test", "arguments": {"query": "find } in text"}} {"name": "other"}`),
			want: map[string]any{
				"query": "find } in text",
			},
		},
		{
			name:   "unmatched closing brace in string",
			buffer: []byte(`{"name": "search", "arguments": {"pattern": "regex: }"}}`),
			want: map[string]any{
				"pattern": "regex: }",
			},
		},
		{
			name:   "complex nested with mixed braces",
			buffer: []byte(`{"name": "analyze", "arguments": {"data": "{\"items\": [{\"value\": \"}\"}, {\"code\": \"if (x) { return y; }\"}]}"}}`),
			want: map[string]any{
				"data": `{"items": [{"value": "}"}, {"code": "if (x) { return y; }"}]}`,
			},
		},
		{
			name:   "string with newline and braces",
			buffer: []byte(`{"name": "format", "arguments": {"template": "{\n  \"key\": \"value\"\n}"}}`),
			want: map[string]any{
				"template": "{\n  \"key\": \"value\"\n}",
			},
		},
		{
			name:   "string with unicode escape",
			buffer: []byte(`{"name": "test", "arguments": {"text": "Unicode: \u007B and \u007D"}}`),
			want: map[string]any{
				"text": "Unicode: { and }",
			},
		},
		{
			name:   "array arguments",
			buffer: []byte(`{"name": "batch", "arguments": ["item1", "item2", "{\"nested\": true}"]}`),
			want:   nil, // This should return nil because arguments is not a map
		},
		{
			name:   "escaped backslash before quote",
			buffer: []byte(`{"name": "path", "arguments": {"dir": "C:\\Program Files\\{App}\\"}}`),
			want: map[string]any{
				"dir": `C:\Program Files\{App}\`,
			},
		},
		{
			name:   "single quotes not treated as string delimiters",
			buffer: []byte(`{"name": "query", "arguments": {"sql": "SELECT * FROM users WHERE name = '{admin}'"}}`),
			want: map[string]any{
				"sql": "SELECT * FROM users WHERE name = '{admin}'",
			},
		},
		{
			name:   "incomplete json at buffer end",
			buffer: []byte(`{"name": "test", "arguments": {"data": "some {"`),
			want:   nil,
		},
		{
			name:   "multiple escaped quotes",
			buffer: []byte(`{"name": "echo", "arguments": {"msg": "He said \"Hello {World}\" loudly"}}`),
			want: map[string]any{
				"msg": `He said "Hello {World}" loudly`,
			},
		},
		{
			name:   "json with comments style string",
			buffer: []byte(`{"name": "code", "arguments": {"snippet": "// This is a comment with { and }"}}`),
			want: map[string]any{
				"snippet": "// This is a comment with { and }",
			},
		},
		{
			name:   "consecutive escaped backslashes",
			buffer: []byte(`{"name": "test", "arguments": {"path": "C:\\\\{folder}\\\\"}}`),
			want: map[string]any{
				"path": `C:\\{folder}\\`,
			},
		},
		{
			name:   "empty string with braces after",
			buffer: []byte(`{"name": "test", "arguments": {"a": "", "b": "{value}"}}`),
			want: map[string]any{
				"a": "",
				"b": "{value}",
			},
		},
		{
			name:   "unicode in key names",
			buffer: []byte(`{"name": "test", "arguments": {"key{": "value", "key}": "value2"}}`),
			want: map[string]any{
				"key{": "value",
				"key}": "value2",
			},
		},
		{
			name:   "very long string with braces",
			buffer: []byte(`{"name": "test", "arguments": {"data": "` + strings.Repeat("a{b}c", 100) + `"}}`),
			want: map[string]any{
				"data": strings.Repeat("a{b}c", 100),
			},
		},
		{
			name:   "tab characters and braces",
			buffer: []byte(`{"name": "test", "arguments": {"code": "\tif (true) {\n\t\treturn;\n\t}"}}`),
			want: map[string]any{
				"code": "\tif (true) {\n\t\treturn;\n\t}",
			},
		},
		{
			name:   "null byte in string",
			buffer: []byte(`{"name": "test", "arguments": {"data": "before\u0000{after}"}}`),
			want: map[string]any{
				"data": "before\x00{after}",
			},
		},
		{
			name:   "escaped quote at end of string",
			buffer: []byte(`{"name": "test", "arguments": {"data": "text with quote at end\\\""}}`),
			want: map[string]any{
				"data": `text with quote at end\"`,
			},
		},
		{
			name:   "mixed array and object in arguments",
			buffer: []byte(`{"name": "test", "arguments": {"items": ["{", "}", {"key": "value"}]}}`),
			want: map[string]any{
				"items": []any{"{", "}", map[string]any{"key": "value"}},
			},
		},
1277
1278
	}

1279
	for _, tt := range tests {
1280
		t.Run(tt.name, func(t *testing.T) {
1281
			got, _ := findArguments(tt.buffer)
1282

1283
1284
1285
			if diff := cmp.Diff(got, tt.want); diff != "" {
				t.Errorf("scanArguments() args mismatch (-got +want):\n%s", diff)
			}
1286
1287
1288
		})
	}
}