dwctl 8.38.2

The Doubleword Control Layer - A self-hostable observability and analytics platform for LLM applications
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

//! API request/response models for batch processing.

use serde::{Deserialize, Serialize};
use std::collections::HashMap;
use utoipa::{IntoParams, ToSchema};

use super::dwext::BatchDwExtResponse;

/// Batch-level errors
#[derive(Debug, Serialize, Deserialize, ToSchema, Clone)]
#[schema(example = json!({
    "data": [{
        "code": "invalid_request",
        "line": 5,
        "message": "Invalid JSON on line 5"
    }]
}))]
pub struct BatchErrors {
    /// Array of error details
    pub data: Vec<BatchError>,
}

/// Individual error in a batch
#[derive(Debug, Serialize, Deserialize, ToSchema, Clone)]
#[schema(example = json!({
    "code": "invalid_request",
    "line": 5,
    "message": "Invalid JSON on line 5"
}))]
pub struct BatchError {
    /// An error code identifying the error type
    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = "invalid_request")]
    pub code: Option<String>,

    /// The line number of the input file where the error occurred, if applicable
    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = 5)]
    pub line: Option<i32>,

    /// A human-readable message providing more details about the error
    #[schema(example = "Invalid JSON on line 5")]
    pub message: String,

    /// The name of the parameter that caused the error, if applicable
    #[serde(skip_serializing_if = "Option::is_none")]
    pub param: Option<String>,
}

/// Request body for creating a batch
#[derive(Debug, Serialize, Deserialize, ToSchema)]
#[schema(example = json!({
    "input_file_id": "file-abc123",
    "endpoint": "/v1/chat/completions",
    "completion_window": "24h"
}))]
pub struct CreateBatchRequest {
    /// The ID of an uploaded file that contains requests for the new batch
    #[schema(example = "file-abc123")]
    pub input_file_id: String,

    /// The endpoint to be used for all requests in the batch.
    /// Supported values are configured via `batches.allowed_url_paths`.
    #[schema(example = "/v1/chat/completions")]
    pub endpoint: String,

    /// The time window within which the batch should be processed (e.g., "24h", "1h").
    /// Allowed values are configured per instance.
    #[schema(example = "24h")]
    pub completion_window: String,

    /// Optional metadata (up to 16 key-value pairs)
    #[serde(default)]
    pub metadata: Option<HashMap<String, String>>,
}

/// Request body for retrying specific requests
#[derive(Debug, Deserialize, ToSchema)]
#[schema(example = json!({
    "request_ids": ["req-abc123", "req-def456"]
}))]
pub struct RetryRequestsRequest {
    /// Array of request IDs to retry
    pub request_ids: Vec<String>,
}

/// Batch object response (OpenAI-compatible)
#[derive(Debug, Serialize, Deserialize, ToSchema)]
#[schema(example = json!({
    "id": "batch-abc123",
    "object": "batch",
    "endpoint": "/v1/chat/completions",
    "input_file_id": "file-abc123",
    "completion_window": "24h",
    "status": "completed",
    "output_file_id": "file-xyz789",
    "created_at": 1703187200,
    "completed_at": 1703190800,
    "request_counts": {
        "total": 100,
        "completed": 98,
        "failed": 2
    }
}))]
pub struct BatchResponse {
    #[schema(example = "batch-abc123")]
    pub id: String,

    #[serde(rename = "object")]
    pub object_type: BatchObjectType,

    #[schema(example = "/v1/chat/completions")]
    pub endpoint: String,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub errors: Option<BatchErrors>,

    #[schema(example = "file-abc123")]
    pub input_file_id: String,

    /// The time window within which the batch should be processed (e.g., "24h", "1h").
    #[schema(example = "24h")]
    pub completion_window: String,

    #[schema(example = "completed")]
    pub status: String,

    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = "file-xyz789")]
    pub output_file_id: Option<String>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub error_file_id: Option<String>,

    #[schema(example = 1703187200)]
    pub created_at: i64, // Unix timestamp

    #[serde(skip_serializing_if = "Option::is_none")]
    pub in_progress_at: Option<i64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub expires_at: Option<i64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub finalizing_at: Option<i64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = 1703190800)]
    pub completed_at: Option<i64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub failed_at: Option<i64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub expired_at: Option<i64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub cancelling_at: Option<i64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub cancelled_at: Option<i64>,

    pub request_counts: RequestCounts,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub metadata: Option<HashMap<String, String>>,

    /// Aggregated analytics metrics (only included when requested via `include=analytics`)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub analytics: Option<BatchAnalytics>,

    /// Doubleword-specific extensions (source provenance, etc.)
    #[serde(default, skip_serializing_if = "Option::is_none")]
    pub dwext: Option<BatchDwExtResponse>,
}

/// Aggregated analytics metrics for batch requests
#[derive(Debug, Clone, Serialize, Deserialize, ToSchema)]
#[schema(example = json!({
    "total_requests": 100,
    "total_prompt_tokens": 50000,
    "total_completion_tokens": 25000,
    "total_tokens": 75000,
    "avg_duration_ms": 1250.5,
    "avg_ttfb_ms": 150.2,
    "total_cost": "0.75"
}))]
pub struct BatchAnalytics {
    /// Total number of requests with analytics data
    #[schema(example = 100)]
    pub total_requests: i64,

    /// Total prompt tokens across all requests
    #[schema(example = 50000)]
    pub total_prompt_tokens: i64,

    /// Total completion tokens across all requests
    #[schema(example = 25000)]
    pub total_completion_tokens: i64,

    /// Total reasoning tokens across all requests
    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = 12000)]
    pub total_reasoning_tokens: Option<i64>,

    /// Total tokens (prompt + completion) across all requests
    #[schema(example = 75000)]
    pub total_tokens: i64,

    /// Average request duration in milliseconds
    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = 1250.5)]
    pub avg_duration_ms: Option<f64>,

    /// Average time to first byte in milliseconds
    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = 150.2)]
    pub avg_ttfb_ms: Option<f64>,

    /// Total cost in credits (if pricing is available)
    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = "0.75")]
    pub total_cost: Option<String>,
}

/// Object type - always "batch"
#[derive(Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "lowercase")]
pub enum BatchObjectType {
    Batch,
}

/// Request counts for a batch
#[derive(Debug, Serialize, Deserialize, ToSchema)]
#[schema(example = json!({
    "total": 100,
    "completed": 98,
    "failed": 2
}))]
pub struct RequestCounts {
    #[schema(example = 100)]
    pub total: i64,
    #[schema(example = 98)]
    pub completed: i64,
    #[schema(example = 2)]
    pub failed: i64,
}

/// Response for batch list
#[derive(Debug, Serialize, Deserialize, ToSchema)]
#[schema(example = json!({
    "object": "list",
    "data": [{
        "id": "batch-abc123",
        "object": "batch",
        "endpoint": "/v1/chat/completions",
        "input_file_id": "file-abc123",
        "completion_window": "24h",
        "status": "completed",
        "created_at": 1703187200,
        "request_counts": {"total": 100, "completed": 98, "failed": 2}
    }],
    "first_id": "batch-abc123",
    "last_id": "batch-abc123",
    "has_more": false
}))]
pub struct BatchListResponse {
    #[serde(rename = "object")]
    pub object_type: ListObjectType,

    pub data: Vec<BatchResponse>,

    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = "batch-abc123")]
    pub first_id: Option<String>,

    #[serde(skip_serializing_if = "Option::is_none")]
    #[schema(example = "batch-abc123")]
    pub last_id: Option<String>,

    #[schema(example = false)]
    pub has_more: bool,
}

/// Object type for lists - always "list"
#[derive(Debug, Serialize, Deserialize, ToSchema)]
#[serde(rename_all = "lowercase")]
pub enum ListObjectType {
    List,
}

use super::pagination::CursorPagination;
use crate::api::models::Pagination;

/// Query parameters for listing batches
#[derive(Debug, Deserialize, IntoParams, ToSchema)]
pub struct ListBatchesQuery {
    /// Pagination parameters
    #[serde(flatten)]
    #[param(inline)]
    pub pagination: CursorPagination,

    /// Search query to filter batches by metadata, input filename, or batch ID (case-insensitive substring match)
    pub search: Option<String>,

    /// Comma-separated list of related resources to include. Supported: "analytics"
    #[param(example = "analytics")]
    pub include: Option<String>,

    /// Filter by member user ID (resolves to api_key_id). Available in org context for any member, or in personal context for platform managers.
    pub member_id: Option<uuid::Uuid>,

    /// Filter by batch status. Supported: "in_progress", "completed", "failed", "cancelled", "expired".
    /// "in_progress" includes validating and finalizing sub-states. "cancelled" includes cancelling.
    /// "expired" matches batches with SLA issues (overdue or finished past deadline).
    pub status: Option<String>,

    /// Only return batches created after this ISO 8601 timestamp
    #[param(example = "2025-01-01T00:00:00Z")]
    pub created_after: Option<chrono::DateTime<chrono::Utc>>,

    /// Only return batches created before this ISO 8601 timestamp
    #[param(example = "2025-12-31T23:59:59Z")]
    pub created_before: Option<chrono::DateTime<chrono::Utc>>,

    /// When true, sort active (non-terminal) batches before completed/failed/cancelled ones.
    /// Each group is sorted by creation time (newest first). Default: false.
    #[serde(default)]
    pub active_first: bool,
}

/// Query parameters for batch results
#[derive(Debug, Deserialize, IntoParams, ToSchema)]
pub struct BatchResultsQuery {
    /// Pagination parameters (limit and skip)
    #[serde(flatten)]
    #[param(inline)]
    pub pagination: Pagination,

    /// Search query to filter by custom_id (case-insensitive substring match)
    pub search: Option<String>,

    /// Filter by request status (completed, failed, pending, in_progress)
    pub status: Option<String>,
}