-
-
Notifications
You must be signed in to change notification settings - Fork 0
/
run.go
451 lines (388 loc) · 12.8 KB
/
run.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
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
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
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
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
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
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
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
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
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
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
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
package openai
import (
"context"
"fmt"
"net/http"
"net/url"
)
type Run struct {
ID string `json:"id"`
Object string `json:"object"`
CreatedAt int64 `json:"created_at"`
ThreadID string `json:"thread_id"`
AssistantID string `json:"assistant_id"`
Status RunStatus `json:"status"`
RequiredAction *RunRequiredAction `json:"required_action,omitempty"`
LastError *RunLastError `json:"last_error,omitempty"`
ExpiresAt int64 `json:"expires_at"`
StartedAt *int64 `json:"started_at,omitempty"`
CancelledAt *int64 `json:"cancelled_at,omitempty"`
FailedAt *int64 `json:"failed_at,omitempty"`
CompletedAt *int64 `json:"completed_at,omitempty"`
Model string `json:"model"`
Instructions string `json:"instructions,omitempty"`
Tools []Tool `json:"tools"`
FileIDS []string `json:"file_ids"` //nolint:revive // backwards-compatibility
Metadata map[string]any `json:"metadata"`
Usage Usage `json:"usage,omitempty"`
Temperature *float32 `json:"temperature,omitempty"`
// The maximum number of prompt tokens that may be used over the course of the run.
// If the run exceeds the number of prompt tokens specified, the run will end with status 'incomplete'.
MaxPromptTokens int `json:"max_prompt_tokens,omitempty"`
// The maximum number of completion tokens that may be used over the course of the run.
// If the run exceeds the number of completion tokens specified, the run will end with status 'incomplete'.
MaxCompletionTokens int `json:"max_completion_tokens,omitempty"`
// ThreadTruncationStrategy defines the truncation strategy to use for the thread.
TruncationStrategy *ThreadTruncationStrategy `json:"truncation_strategy,omitempty"`
httpHeader
}
type RunStatus string
const (
RunStatusQueued RunStatus = "queued"
RunStatusInProgress RunStatus = "in_progress"
RunStatusRequiresAction RunStatus = "requires_action"
RunStatusCancelling RunStatus = "cancelling"
RunStatusFailed RunStatus = "failed"
RunStatusCompleted RunStatus = "completed"
RunStatusIncomplete RunStatus = "incomplete"
RunStatusExpired RunStatus = "expired"
RunStatusCancelled RunStatus = "cancelled"
)
type RunRequiredAction struct {
Type RequiredActionType `json:"type"`
SubmitToolOutputs *SubmitToolOutputs `json:"submit_tool_outputs,omitempty"`
}
type RequiredActionType string
const (
RequiredActionTypeSubmitToolOutputs RequiredActionType = "submit_tool_outputs"
)
type SubmitToolOutputs struct {
ToolCalls []ToolCall `json:"tool_calls"`
}
type RunLastError struct {
Code RunError `json:"code"`
Message string `json:"message"`
}
type RunError string
const (
RunErrorServerError RunError = "server_error"
RunErrorRateLimitExceeded RunError = "rate_limit_exceeded"
)
type RunRequest struct {
AssistantID string `json:"assistant_id"`
Model string `json:"model,omitempty"`
Instructions string `json:"instructions,omitempty"`
AdditionalInstructions string `json:"additional_instructions,omitempty"`
Tools []Tool `json:"tools,omitempty"`
Metadata map[string]any `json:"metadata,omitempty"`
// Sampling temperature between 0 and 2. Higher values like 0.8 are more random.
// lower values are more focused and deterministic.
Temperature *float32 `json:"temperature,omitempty"`
TopP *float32 `json:"top_p,omitempty"`
// The maximum number of prompt tokens that may be used over the course of the run.
// If the run exceeds the number of prompt tokens specified, the run will end with status 'incomplete'.
MaxPromptTokens int `json:"max_prompt_tokens,omitempty"`
// The maximum number of completion tokens that may be used over the course of the run.
// If the run exceeds the number of completion tokens specified, the run will end with status 'incomplete'.
MaxCompletionTokens int `json:"max_completion_tokens,omitempty"`
// ThreadTruncationStrategy defines the truncation strategy to use for the thread.
TruncationStrategy *ThreadTruncationStrategy `json:"truncation_strategy,omitempty"`
// This can be either a string or a ToolChoice object.
ToolChoice any `json:"tool_choice,omitempty"`
// This can be either a string or a ResponseFormat object.
ResponseFormat any `json:"response_format,omitempty"`
}
// ThreadTruncationStrategy defines the truncation strategy to use for the thread.
// https://platform.openai.com/docs/assistants/how-it-works/truncation-strategy.
type ThreadTruncationStrategy struct {
// default 'auto'.
Type TruncationStrategy `json:"type,omitempty"`
// this field should be set if the truncation strategy is set to LastMessages.
LastMessages *int `json:"last_messages,omitempty"`
}
// TruncationStrategy defines the existing truncation strategies existing for thread management in an assistant.
type TruncationStrategy string
const (
// TruncationStrategyAuto messages in the middle of the thread will be dropped to fit the context length of the model.
TruncationStrategyAuto = TruncationStrategy("auto")
// TruncationStrategyLastMessages the thread will be truncated to the n most recent messages in the thread.
TruncationStrategyLastMessages = TruncationStrategy("last_messages")
)
// ReponseFormat specifies the format the model must output.
// https://platform.openai.com/docs/api-reference/runs/createRun#runs-createrun-response_format.
// Type can either be text or json_object.
type ReponseFormat struct {
Type string `json:"type"`
}
type RunModifyRequest struct {
Metadata map[string]any `json:"metadata,omitempty"`
}
// RunList is a list of runs.
type RunList struct {
Runs []Run `json:"data"`
httpHeader
}
type SubmitToolOutputsRequest struct {
ToolOutputs []ToolOutput `json:"tool_outputs"`
}
type ToolOutput struct {
ToolCallID string `json:"tool_call_id"`
Output any `json:"output"`
}
type CreateThreadAndRunRequest struct {
RunRequest
Thread ThreadRequest `json:"thread"`
}
type RunStep struct {
ID string `json:"id"`
Object string `json:"object"`
CreatedAt int64 `json:"created_at"`
AssistantID string `json:"assistant_id"`
ThreadID string `json:"thread_id"`
RunID string `json:"run_id"`
Type RunStepType `json:"type"`
Status RunStepStatus `json:"status"`
StepDetails StepDetails `json:"step_details"`
LastError *RunLastError `json:"last_error,omitempty"`
ExpiredAt *int64 `json:"expired_at,omitempty"`
CancelledAt *int64 `json:"cancelled_at,omitempty"`
FailedAt *int64 `json:"failed_at,omitempty"`
CompletedAt *int64 `json:"completed_at,omitempty"`
Metadata map[string]any `json:"metadata"`
httpHeader
}
type RunStepStatus string
const (
RunStepStatusInProgress RunStepStatus = "in_progress"
RunStepStatusCancelling RunStepStatus = "cancelled"
RunStepStatusFailed RunStepStatus = "failed"
RunStepStatusCompleted RunStepStatus = "completed"
RunStepStatusExpired RunStepStatus = "expired"
)
type RunStepType string
const (
RunStepTypeMessageCreation RunStepType = "message_creation"
RunStepTypeToolCalls RunStepType = "tool_calls"
)
type StepDetails struct {
Type RunStepType `json:"type"`
MessageCreation *StepDetailsMessageCreation `json:"message_creation,omitempty"`
ToolCalls []ToolCall `json:"tool_calls,omitempty"`
}
type StepDetailsMessageCreation struct {
MessageID string `json:"message_id"`
}
// RunStepList is a list of steps.
type RunStepList struct {
RunSteps []RunStep `json:"data"`
FirstID string `json:"first_id"`
LastID string `json:"last_id"`
HasMore bool `json:"has_more"`
httpHeader
}
type Pagination struct {
Limit *int
Order *string
After *string
Before *string
}
// CreateRun creates a new run.
func (c *Client) CreateRun(
ctx context.Context,
threadID string,
request RunRequest,
) (response Run, err error) {
urlSuffix := fmt.Sprintf("/threads/%s/runs", threadID)
req, err := c.newRequest(
ctx,
http.MethodPost,
c.fullURL(urlSuffix),
withBody(request),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// RetrieveRun retrieves a run.
func (c *Client) RetrieveRun(
ctx context.Context,
threadID string,
runID string,
) (response Run, err error) {
urlSuffix := fmt.Sprintf("/threads/%s/runs/%s", threadID, runID)
req, err := c.newRequest(
ctx,
http.MethodGet,
c.fullURL(urlSuffix),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// ModifyRun modifies a run.
func (c *Client) ModifyRun(
ctx context.Context,
threadID string,
runID string,
request RunModifyRequest,
) (response Run, err error) {
urlSuffix := fmt.Sprintf("/threads/%s/runs/%s", threadID, runID)
req, err := c.newRequest(
ctx,
http.MethodPost,
c.fullURL(urlSuffix),
withBody(request),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// ListRuns lists runs.
func (c *Client) ListRuns(
ctx context.Context,
threadID string,
pagination Pagination,
) (response RunList, err error) {
urlValues := url.Values{}
if pagination.Limit != nil {
urlValues.Add("limit", fmt.Sprintf("%d", *pagination.Limit))
}
if pagination.Order != nil {
urlValues.Add("order", *pagination.Order)
}
if pagination.After != nil {
urlValues.Add("after", *pagination.After)
}
if pagination.Before != nil {
urlValues.Add("before", *pagination.Before)
}
encodedValues := ""
if len(urlValues) > 0 {
encodedValues = "?" + urlValues.Encode()
}
urlSuffix := fmt.Sprintf("/threads/%s/runs%s", threadID, encodedValues)
req, err := c.newRequest(
ctx,
http.MethodGet,
c.fullURL(urlSuffix),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// SubmitToolOutputs submits tool outputs.
func (c *Client) SubmitToolOutputs(
ctx context.Context,
threadID string,
runID string,
request SubmitToolOutputsRequest) (response Run, err error) {
urlSuffix := fmt.Sprintf("/threads/%s/runs/%s/submit_tool_outputs", threadID, runID)
req, err := c.newRequest(
ctx,
http.MethodPost,
c.fullURL(urlSuffix),
withBody(request),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// CancelRun cancels a run.
func (c *Client) CancelRun(
ctx context.Context,
threadID string,
runID string) (response Run, err error) {
urlSuffix := fmt.Sprintf("/threads/%s/runs/%s/cancel", threadID, runID)
req, err := c.newRequest(
ctx,
http.MethodPost,
c.fullURL(urlSuffix),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// CreateThreadAndRun submits tool outputs.
func (c *Client) CreateThreadAndRun(
ctx context.Context,
request CreateThreadAndRunRequest) (response Run, err error) {
urlSuffix := "/threads/runs"
req, err := c.newRequest(
ctx,
http.MethodPost,
c.fullURL(urlSuffix),
withBody(request),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// RetrieveRunStep retrieves a run step.
func (c *Client) RetrieveRunStep(
ctx context.Context,
threadID string,
runID string,
stepID string,
) (response RunStep, err error) {
urlSuffix := fmt.Sprintf("/threads/%s/runs/%s/steps/%s", threadID, runID, stepID)
req, err := c.newRequest(
ctx,
http.MethodGet,
c.fullURL(urlSuffix),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}
// ListRunSteps lists run steps.
func (c *Client) ListRunSteps(
ctx context.Context,
threadID string,
runID string,
pagination Pagination,
) (response RunStepList, err error) {
urlValues := url.Values{}
if pagination.Limit != nil {
urlValues.Add("limit", fmt.Sprintf("%d", *pagination.Limit))
}
if pagination.Order != nil {
urlValues.Add("order", *pagination.Order)
}
if pagination.After != nil {
urlValues.Add("after", *pagination.After)
}
if pagination.Before != nil {
urlValues.Add("before", *pagination.Before)
}
encodedValues := ""
if len(urlValues) > 0 {
encodedValues = "?" + urlValues.Encode()
}
urlSuffix := fmt.Sprintf("/threads/%s/runs/%s/steps%s", threadID, runID, encodedValues)
req, err := c.newRequest(
ctx,
http.MethodGet,
c.fullURL(urlSuffix),
withBetaAssistantVersion(c.config.AssistantVersion))
if err != nil {
return
}
err = c.sendRequest(req, &response)
return
}