crawlrs 0.2.0

crawlrs - Enterprise-grade web scraping and crawling platform built with Rust
Documentation
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
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
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
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
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
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
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
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
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
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
846
847
848
849
850
851
852
853
854
855
856
857
858
859
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
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
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
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
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
// Copyright (c) 2025-2026 Kirky.X🌠
// SPDX-License-Identifier: Apache-2.0

use crate::config::settings::Settings;
use crate::domain::models::Crawl;
use crate::domain::models::CreditsTransactionType;
use crate::domain::models::{Task, TaskType};
use crate::domain::repositories::crawl_repository::CrawlRepository;
use crate::domain::repositories::credits_repository::CreditsRepository;
use crate::domain::repositories::credits_repository::CreditsRepositoryError;
use crate::domain::repositories::task_repository::TaskRepository;
use chrono::Utc;
use serde_json::json;
use std::sync::Arc;
use thiserror::Error;

/// Search query parameters (领域层参数对象)
#[derive(Debug, Clone)]
pub struct SearchQuery {
    pub query: String,
    pub limit: Option<u32>,
    pub lang: Option<String>,
    pub country: Option<String>,
    pub engine: Option<String>,
    pub sources: Option<Vec<String>>,
    pub crawl_results: Option<bool>,
    pub crawl_config: Option<SearchCrawlConfig>,
}

/// Search crawl configuration (领域层参数对象)
#[derive(Debug, Clone, serde::Serialize, serde::Deserialize)]
pub struct SearchCrawlConfig {
    pub max_depth: u32,
    pub include_patterns: Option<Vec<String>>,
    pub exclude_patterns: Option<Vec<String>>,
    pub strategy: String,
    pub crawl_delay_ms: Option<u64>,
    pub max_concurrency: u32,
    pub headers: Option<serde_json::Value>,
    pub proxy: Option<String>,
    pub extraction_rules: Option<
        std::collections::HashMap<
            String,
            crate::domain::services::extraction_service::ExtractionRule,
        >,
    >,
    /// 团队域名黑名单(入口预检使用;持久化载体是 `crawl.config`,
    /// worker 执行期与链接过滤从该处读取)。
    ///
    /// 不参与 `CrawlConfigDto` 序列化:该 DTO 带 `deny_unknown_fields`,
    /// 写进派生任务的 config 会让 payload 解析失败。
    #[serde(default, skip_serializing)]
    pub domain_blacklist: Vec<String>,
}

impl Default for SearchCrawlConfig {
    fn default() -> Self {
        Self {
            max_depth: 1,
            include_patterns: None,
            exclude_patterns: None,
            strategy: "bfs".to_string(),
            crawl_delay_ms: None,
            max_concurrency: 10,
            headers: None,
            proxy: None,
            extraction_rules: None,
            domain_blacklist: Vec::new(),
        }
    }
}

/// Search result item (领域层返回对象)
#[derive(Debug, Clone)]
pub struct SearchResult {
    pub title: String,
    pub url: String,
    pub description: Option<String>,
    pub engine: String,
    /// 语义重排相关性得分(0..1)。仅当 rerank 生效时有值;
    /// 未评分条目(重排截断尾部 / 未启用 / fail-open)为 `None`。
    pub score: Option<f64>,
}

/// Search response (领域层返回对象)
#[derive(Debug, Clone)]
pub struct SearchResponse {
    pub query: String,
    pub results: Vec<SearchResult>,
    pub crawl_id: Option<crate::domain::id::Id>,
    pub credits_used: u32,
}

#[derive(Error, Debug)]
pub enum SearchServiceError {
    #[error("Validation failed: {0}")]
    ValidationError(String),
    #[error("Repository error: {0}")]
    Repository(#[from] crate::domain::repositories::task_repository::RepositoryError),
    #[error("Credits repository error: {0}")]
    CreditsRepository(#[from] CreditsRepositoryError),
    #[error("Search engine error: {0}")]
    SearchEngine(String),
    #[error("Insufficient credits: available {available}, required {required}")]
    InsufficientCredits { available: i64, required: i64 },
    /// 未知搜索引擎名(结构化携带引擎名,供 handler 按请求 locale 翻译)
    #[error("unknown search engine: {0}")]
    UnknownEngine(String),
    /// smart 引擎搜索失败(结构化携带底层错误描述,供 handler 翻译)
    #[error("smart search: {0}")]
    SmartSearchFailed(String),
}

// From implementations for SearchServiceError
impl From<String> for SearchServiceError {
    fn from(msg: String) -> Self {
        SearchServiceError::ValidationError(msg)
    }
}

impl From<&str> for SearchServiceError {
    fn from(msg: &str) -> Self {
        SearchServiceError::ValidationError(msg.to_string())
    }
}

impl From<anyhow::Error> for SearchServiceError {
    fn from(err: anyhow::Error) -> Self {
        SearchServiceError::SearchEngine(err.to_string())
    }
}

use crate::domain::services::rerank::SearchReranker;
use crate::search::client::SearchClientTrait;
use crate::search::engine_trait::SearchEngine;

/// Search service trait for trait object support.
#[async_trait::async_trait]
pub trait SearchServiceTrait: Send + Sync {
    /// Perform search operation.
    async fn search(
        &self,
        team_id: crate::domain::id::Id,
        api_key_id: crate::domain::id::Id,
        query: SearchQuery,
    ) -> Result<SearchResponse, SearchServiceError>;
}

/// 单次搜索派生 crawl 的硬上限:结果条数与派生任务数同受此约束,
/// 防止大 `limit`/引擎返回膨胀导致任务表与 crawl 计数失控。
pub const MAX_SEARCH_RESULTS: u32 = 200;

/// 搜索派生 crawl 的最大深度(与 HTTP 创建入口同口径)。
pub const MAX_SEARCH_CRAWL_DEPTH: u32 = 5;

/// Concrete search service implementation.
pub struct SearchService {
    crawl_repo: Arc<dyn CrawlRepository>,
    task_repo: Arc<dyn TaskRepository>,
    credits_repo: Arc<dyn CreditsRepository>,
    search_client: Arc<dyn SearchClientTrait>,
    /// P12:SmartSearchEngine(多引擎 fan-out + RRF 聚合)。
    /// 未注入时 engine=None 的请求回落到单引擎路径(agent-lib / 测试面)。
    smart_engine: Option<Arc<dyn SearchEngine>>,
    max_results: u32,
    reranker: Arc<SearchReranker>,
}

impl SearchService {
    pub fn new(
        crawl_repo: Arc<dyn CrawlRepository>,
        task_repo: Arc<dyn TaskRepository>,
        credits_repo: Arc<dyn CreditsRepository>,
        settings: Arc<Settings>,
        search_client: Arc<dyn SearchClientTrait>,
        reranker: Arc<SearchReranker>,
    ) -> Self {
        Self {
            crawl_repo,
            task_repo,
            credits_repo,
            search_client,
            smart_engine: None,
            max_results: settings.search.max_results,
            reranker,
        }
    }

    /// 注入 SmartSearchEngine(P12 接线:engine=None / "smart" 时走聚合路径)。
    #[must_use]
    pub fn with_smart_engine(mut self, engine: Arc<dyn SearchEngine>) -> Self {
        self.smart_engine = Some(engine);
        self
    }

    pub async fn search(
        &self,
        team_id: crate::domain::id::Id,
        api_key_id: crate::domain::id::Id,
        query: SearchQuery,
    ) -> Result<SearchResponse, SearchServiceError> {
        // 简化验证:检查 query 是否为空
        if query.query.trim().is_empty() {
            return Err(SearchServiceError::ValidationError(
                "Query cannot be empty".to_string(),
            ));
        }

        // Check team credits balance before performing search
        let current_balance = self.credits_repo.get_balance(team_id).await?;
        let search_cost = 1i64; // 1 credit per search as per PRD

        if current_balance < search_cost {
            return Err(SearchServiceError::InsufficientCredits {
                available: current_balance,
                required: search_cost,
            });
        }

        // 1. Perform Search using configured engine
        // Use sources if provided, otherwise use engine
        let engine_param = if let Some(sources) = &query.sources {
            if sources.len() == 1 {
                Some(sources[0].as_str())
            } else {
                // Multiple sources - use aggregator (None means query all engines)
                None
            }
        } else {
            query.engine.as_deref()
        };

        let results = self
            .perform_search(
                &query.query,
                query
                    .limit
                    .unwrap_or(10)
                    .min(self.max_results)
                    .min(MAX_SEARCH_RESULTS),
                query.lang.as_deref(),
                query.country.as_deref(),
                engine_param,
            )
            .await?;

        let mut crawl_id = None;
        let credits_used = search_cost;

        // 2. If crawl_results is true, create a crawl task
        if query.crawl_results.unwrap_or(false) && !results.is_empty() {
            let cid = crate::domain::id::Id::generate();
            let _now = Utc::now();

            let config = query.crawl_config.unwrap_or_default();
            // 深度上限是创建入口的共同契约:handler 之外再守一层,
            // 防止 SDk/内部调用绕过。
            if config.max_depth > MAX_SEARCH_CRAWL_DEPTH {
                return Err(SearchServiceError::ValidationError(format!(
                    "crawl_config.max_depth must be <= {MAX_SEARCH_CRAWL_DEPTH}"
                )));
            }

            let now = chrono::Utc::now();
            // 先过滤超长 URL 并去重,再一次性写入 total_tasks:
            // 逐任务自增会让 worker 在循环中途看到“部分 total”,
            // 快失败时提前满足 completed+failed==total 并取消剩余任务。
            let mut seen_urls = std::collections::HashSet::new();
            // 黑名单只归一化一次:逐结果调用 `domain_in_denylist` 会反复
            // 解析条目,R 条结果 × N 条规则时退化为 O(R·N²)。
            let normalized_blacklist =
                crate::domain::services::team_service::normalize_denylist_entries(
                    &config.domain_blacklist,
                );
            let mut eligible = Vec::new();
            for result in &results {
                if eligible.len() >= MAX_SEARCH_RESULTS as usize {
                    log::warn!(
                        "search crawl truncated at {} results (hard cap)",
                        MAX_SEARCH_RESULTS
                    );
                    break;
                }
                // tasks.url 的普通 btree 索引有行上限:与 crawl/scrape/extract
                // 同口径过滤超长 URL(字节)。
                if result.url.len() > 2048 {
                    log::warn!(
                        "skip oversized search result url ({} bytes)",
                        result.url.len()
                    );
                    #[cfg(feature = "metrics")]
                    metrics::counter!("crawlrs_search_oversized_url_total").increment(1);
                    continue;
                }
                if let Some(reason) =
                    crate::domain::services::team_service::domain_in_normalized_denylist(
                        &result.url,
                        &normalized_blacklist,
                    )
                {
                    log::warn!("skip blacklisted search result: {}", reason);
                    #[cfg(feature = "metrics")]
                    metrics::counter!("crawlrs_search_blacklisted_url_total").increment(1);
                    continue;
                }
                if seen_urls.insert(result.url.clone()) {
                    eligible.push(result);
                }
            }

            // 无可用结果时不创建 crawl:total=0 的空 crawl 永远无法进入终态。
            if !eligible.is_empty() {
                let mut crawl_config = json!(config);
                if let Some(obj) = crawl_config.as_object_mut() {
                    obj.insert("domain_blacklist".to_string(), json!(normalized_blacklist));
                }
                let crawl = Crawl::with_all_fields(
                    cid,
                    team_id,
                    format!("Search Crawl: {}", query.query),
                    "search://".to_string() + &query.query,
                    "search://".to_string() + &query.query,
                    crate::domain::models::CrawlStatus::Queued,
                    crawl_config,
                    eligible.len() as i32,
                    0,
                    0,
                    now,
                    now,
                    None,
                );

                self.crawl_repo.create(&crawl).await?;

                for result in eligible {
                    let mut task = Task::new(
                        crate::domain::id::Id::generate(),
                        TaskType::Crawl,
                        team_id,
                        api_key_id,
                        result.url.clone(),
                        json!({
                            "crawl_id": cid,
                            "depth": 0,
                            "config": config,
                        }),
                    );
                    task.priority = 100;
                    task.crawl_id = Some(cid);
                    if let Err(e) = self.task_repo.create(&task).await {
                        // 已写 total=N 但任务只落了部分:显式把 crawl 标 Failed,
                        // 避免 completed+failed 永不达 total 的永久 queued。
                        log::error!("Failed to enqueue search crawl task (crawl {}): {}", cid, e);
                        if let Err(mark_err) =
                            self.crawl_repo.finalize_failed_after_enqueue(cid).await
                        {
                            log::error!(
                                "Failed to mark incomplete search crawl {} as Failed: {}",
                                cid,
                                mark_err
                            );
                        }
                        return Err(e.into());
                    }
                }

                crawl_id = Some(cid);
            }
        }

        // Deduct credits for the search operation
        self.credits_repo
            .deduct_credits(
                team_id,
                search_cost,
                CreditsTransactionType::Search,
                format!("Search query: {}", query.query),
                None,
            )
            .await?;

        Ok(SearchResponse {
            query: query.query,
            results,
            crawl_id,
            credits_used: credits_used as u32,
        })
    }

    async fn perform_search(
        &self,
        query: &str,
        limit: u32,
        _lang: Option<&str>,
        _country: Option<&str>,
        engine: Option<&str>,
    ) -> Result<Vec<SearchResult>, SearchServiceError> {
        // P12:路由决策——engine 为 None 或 "smart" 时走 SmartSearchEngine
        // (多引擎 fan-out + RRF 聚合);显式引擎名走 SearchClient 单引擎路径。
        if let Some(results) = self.try_smart_search(query, limit, engine).await? {
            return Ok(results);
        }

        // 显式引擎名必须可解析——此前未知引擎名会静默替换为默认 bing
        if let Some(name) = engine {
            if name != "smart" && crate::search::types::SearchEngineType::from_name(name).is_none()
            {
                return Err(SearchServiceError::UnknownEngine(name.to_string()));
            }
        }

        let mut command = self.search_client.search(query).await;

        command = command.limit(limit);

        if let Some(engine_name) = engine {
            command = command.with_engine(engine_name);
        }

        let response = command
            .execute()
            .await
            .map_err(|e| SearchServiceError::SearchEngine(e.to_string()))?;

        let filtered_results: Vec<SearchResult> = response
            .items
            .into_iter()
            .take(limit as usize)
            .map(into_search_result)
            .collect();

        // 语义重排(精准推荐):RRF/启发式融合之后补 query-doc 相关性。
        // reranker 未启用时直通;provider 故障时 fail-open 保留原序。
        // 输入已 take(limit),重排只改变顺序不改变数量。
        let filtered_results = self.reranker.rerank_results(query, filtered_results).await;

        Ok(filtered_results)
    }

    /// smart 聚合路径(自 [`SearchService::perform_search`] 提取):
    /// 返回 `Some` 表示已由 smart 引擎处理(含重排),`None` 表示需单引擎回退。
    async fn try_smart_search(
        &self,
        query: &str,
        limit: u32,
        engine: Option<&str>,
    ) -> Result<Option<Vec<SearchResult>>, SearchServiceError> {
        let use_smart = engine.is_none() || engine == Some("smart");
        if !use_smart {
            return Ok(None);
        }

        if let Some(smart) = &self.smart_engine {
            let request = crate::search::engine_trait::SearchRequest::new(query).with_limit(limit);
            let response = smart
                .search(&request)
                .await
                .map_err(|e| SearchServiceError::SmartSearchFailed(e.to_string()))?;
            let results: Vec<SearchResult> = response
                .items
                .into_iter()
                .take(limit as usize)
                .map(into_search_result)
                .collect();
            // 与单引擎路径同款语义重排(数量不变,仅调序)
            return Ok(Some(self.reranker.rerank_results(query, results).await));
        }

        // 未注入 smart engine(agent-lib / 测试面):
        // engine=None 真降级到单引擎;engine="smart" 将在下方被
        // from_name 解析后因 SearchClient 未注册 Smart 引擎而显性报错
        if engine.is_none() {
            log::warn!("{}", crate::i18n::tr_log("search-smart-not-injected"));
        }
        Ok(None)
    }
}

/// 搜索响应条目 → 领域搜索结果(smart 与单引擎路径共用)。
fn into_search_result(item: crate::search::response::ResponseItem) -> SearchResult {
    SearchResult {
        title: item.title,
        url: item.url,
        description: Some(item.description),
        engine: item.engine.name().to_string(),
        score: None,
    }
}

/// Implement SearchServiceTrait for SearchService
#[async_trait::async_trait]
impl SearchServiceTrait for SearchService {
    async fn search(
        &self,
        team_id: crate::domain::id::Id,
        api_key_id: crate::domain::id::Id,
        query: SearchQuery,
    ) -> Result<SearchResponse, SearchServiceError> {
        Self::search(self, team_id, api_key_id, query).await
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::domain::models::{CrawlStatus, CreditsTransaction, CreditsTransactionType};
    use crate::domain::repositories::crawl_repository::CrawlRepository;
    use crate::domain::repositories::credits_repository::CreditsRepositoryError;
    use crate::domain::repositories::task_repository::{
        RepositoryError, TaskQueryParams, TaskRepository,
    };
    use crate::search::client::{SearchClient, SearchClientTrait, SearchCommand};
    use crate::search::engine_trait::{SearchEngine, SearchRequest};
    use crate::search::error::SearchError;
    use crate::search::response::{Response, ResponseItem};
    use crate::search::types::SearchEngineType;
    use async_trait::async_trait;
    use std::collections::HashSet;
    use std::sync::Mutex;

    // Mocks

    /// Configurable mock for CreditsRepository.
    struct MockCreditsRepo {
        balance: i64,
        get_balance_fails: bool,
        deduct_fails: bool,
        deduct_calls: Mutex<Vec<(crate::domain::id::Id, i64)>>,
    }

    impl MockCreditsRepo {
        fn with_balance(balance: i64) -> Self {
            Self {
                balance,
                get_balance_fails: false,
                deduct_fails: false,
                deduct_calls: Mutex::new(Vec::new()),
            }
        }

        fn failing() -> Self {
            Self {
                balance: 0,
                get_balance_fails: true,
                deduct_fails: false,
                deduct_calls: Mutex::new(Vec::new()),
            }
        }
    }

    #[async_trait]
    impl CreditsRepository for MockCreditsRepo {
        async fn get_balance(
            &self,
            _team_id: crate::domain::id::Id,
        ) -> Result<i64, CreditsRepositoryError> {
            if self.get_balance_fails {
                return Err(CreditsRepositoryError::DatabaseError(
                    "mock get_balance failure".to_string(),
                ));
            }
            Ok(self.balance)
        }

        async fn deduct_credits(
            &self,
            team_id: crate::domain::id::Id,
            amount: i64,
            _transaction_type: CreditsTransactionType,
            _description: String,
            _reference_id: Option<crate::domain::id::Id>,
        ) -> Result<(), CreditsRepositoryError> {
            if self.deduct_fails {
                return Err(CreditsRepositoryError::DatabaseError(
                    "mock deduct failure".to_string(),
                ));
            }
            self.deduct_calls.lock().unwrap().push((team_id, amount));
            Ok(())
        }

        async fn add_credits(
            &self,
            _team_id: crate::domain::id::Id,
            _amount: i64,
            _transaction_type: CreditsTransactionType,
            _description: String,
            _reference_id: Option<crate::domain::id::Id>,
        ) -> Result<i64, CreditsRepositoryError> {
            Ok(0)
        }

        async fn get_transaction_history(
            &self,
            _team_id: crate::domain::id::Id,
            _limit: Option<u32>,
        ) -> Result<Vec<CreditsTransaction>, CreditsRepositoryError> {
            Ok(vec![])
        }

        async fn initialize_team_credits(
            &self,
            _team_id: crate::domain::id::Id,
            _initial_balance: i64,
        ) -> Result<i64, CreditsRepositoryError> {
            Ok(0)
        }
    }

    /// Minimal mock for TaskRepository (only create is used in search_service).
    struct MockTaskRepo {
        create_fails: bool,
        created: Mutex<Vec<Task>>,
    }

    impl MockTaskRepo {
        fn new() -> Self {
            Self {
                create_fails: false,
                created: Mutex::new(Vec::new()),
            }
        }

        fn failing() -> Self {
            Self {
                create_fails: true,
                created: Mutex::new(Vec::new()),
            }
        }

        fn created_count(&self) -> usize {
            self.created.lock().unwrap().len()
        }

        fn created_urls(&self) -> Vec<String> {
            self.created
                .lock()
                .unwrap()
                .iter()
                .map(|task| task.url.clone())
                .collect()
        }
    }

    #[async_trait]
    impl TaskRepository for MockTaskRepo {
        async fn create(&self, task: &Task) -> Result<Task, RepositoryError> {
            if self.create_fails {
                return Err(RepositoryError::Database(anyhow::anyhow!(
                    "mock task create failure"
                )));
            }
            self.created.lock().unwrap().push(task.clone());
            Ok(task.clone())
        }

        async fn find_by_id(
            &self,
            _id: crate::domain::id::Id,
        ) -> Result<Option<Task>, RepositoryError> {
            Ok(None)
        }

        async fn update(&self, task: &Task) -> Result<Task, RepositoryError> {
            Ok(task.clone())
        }

        async fn acquire_next(
            &self,
            _worker_id: crate::domain::id::Id,
        ) -> Result<Option<Task>, RepositoryError> {
            Ok(None)
        }

        async fn mark_completed(
            &self,
            _id: crate::domain::id::Id,
            _lock_token: Option<crate::domain::id::Id>,
        ) -> Result<u64, RepositoryError> {
            Ok(1)
        }

        async fn mark_failed(
            &self,
            _id: crate::domain::id::Id,
            _lock_token: Option<crate::domain::id::Id>,
        ) -> Result<u64, RepositoryError> {
            Ok(1)
        }

        async fn mark_cancelled(&self, _id: crate::domain::id::Id) -> Result<u64, RepositoryError> {
            Ok(1)
        }

        async fn exists_by_url(&self, _url: &str) -> Result<bool, RepositoryError> {
            Ok(false)
        }

        async fn find_existing_urls(
            &self,
            _urls: &[String],
        ) -> Result<HashSet<String>, RepositoryError> {
            Ok(HashSet::new())
        }

        async fn reset_stuck_tasks(
            &self,
            _timeout: chrono::Duration,
        ) -> Result<u64, RepositoryError> {
            Ok(0)
        }

        async fn cancel_tasks_by_crawl_id(
            &self,
            _crawl_id: crate::domain::id::Id,
        ) -> Result<u64, RepositoryError> {
            Ok(0)
        }

        async fn expire_tasks(&self) -> Result<u64, RepositoryError> {
            Ok(0)
        }

        async fn find_by_crawl_id(
            &self,
            _crawl_id: crate::domain::id::Id,
        ) -> Result<Vec<Task>, RepositoryError> {
            Ok(vec![])
        }

        async fn query_tasks(
            &self,
            _params: TaskQueryParams,
        ) -> Result<(Vec<Task>, u64), RepositoryError> {
            Ok((vec![], 0))
        }

        async fn batch_cancel(
            &self,
            _task_ids: Vec<crate::domain::id::Id>,
            _team_id: crate::domain::id::Id,
            _force: bool,
        ) -> Result<
            (
                Vec<crate::domain::id::Id>,
                Vec<(crate::domain::id::Id, String)>,
            ),
            RepositoryError,
        > {
            Ok((vec![], vec![]))
        }

        async fn renew_lock(
            &self,
            _task_id: crate::domain::id::Id,
            _worker_id: crate::domain::id::Id,
            _extend_seconds: i64,
        ) -> Result<bool, RepositoryError> {
            Ok(true)
        }
    }

    /// Minimal mock for CrawlRepository (only create is used in search_service).
    struct MockCrawlRepo {
        create_fails: bool,
        created: Mutex<Vec<Crawl>>,
        /// 入队失败补偿调用次数(验证失败路径真正落终态)
        compensations: Mutex<u32>,
    }

    impl MockCrawlRepo {
        fn new() -> Self {
            Self {
                create_fails: false,
                created: Mutex::new(Vec::new()),
                compensations: Mutex::new(0),
            }
        }

        fn failing() -> Self {
            Self {
                create_fails: true,
                created: Mutex::new(Vec::new()),
                compensations: Mutex::new(0),
            }
        }

        fn created_count(&self) -> usize {
            self.created.lock().unwrap().len()
        }

        fn compensation_count(&self) -> u32 {
            *self.compensations.lock().unwrap()
        }

        fn last_created_config(&self) -> Option<serde_json::Value> {
            self.created
                .lock()
                .unwrap()
                .last()
                .map(|c| c.config().clone())
        }
    }

    #[async_trait]
    impl CrawlRepository for MockCrawlRepo {
        async fn create(&self, crawl: &Crawl) -> Result<Crawl, RepositoryError> {
            if self.create_fails {
                return Err(RepositoryError::Database(anyhow::anyhow!(
                    "mock crawl create failure"
                )));
            }
            self.created.lock().unwrap().push(crawl.clone());
            Ok(crawl.clone())
        }

        async fn find_by_id(
            &self,
            _id: crate::domain::id::Id,
        ) -> Result<Option<Crawl>, RepositoryError> {
            Ok(None)
        }

        async fn finalize_failed_after_enqueue(
            &self,
            _id: crate::domain::id::Id,
        ) -> Result<bool, RepositoryError> {
            *self.compensations.lock().unwrap() += 1;
            Ok(true)
        }

        async fn update(&self, crawl: &Crawl) -> Result<Crawl, RepositoryError> {
            Ok(crawl.clone())
        }

        async fn increment_completed_tasks(
            &self,
            _id: crate::domain::id::Id,
        ) -> Result<(), RepositoryError> {
            Ok(())
        }

        async fn increment_failed_tasks(
            &self,
            _id: crate::domain::id::Id,
        ) -> Result<(), RepositoryError> {
            Ok(())
        }

        async fn update_status(
            &self,
            _id: crate::domain::id::Id,
            _status: CrawlStatus,
        ) -> Result<(), RepositoryError> {
            Ok(())
        }

        async fn increment_total_tasks(
            &self,
            _id: crate::domain::id::Id,
        ) -> Result<(), RepositoryError> {
            Ok(())
        }

        async fn find_by_team_id_paginated(
            &self,
            _team_id: crate::domain::id::Id,
            _limit: u32,
            _offset: u32,
        ) -> Result<Vec<Crawl>, RepositoryError> {
            Ok(vec![])
        }

        async fn count_by_team_id(
            &self,
            _team_id: crate::domain::id::Id,
        ) -> Result<u64, RepositoryError> {
            Ok(0)
        }
    }

    /// Mock SearchEngine that returns a preconfigured Response or error.
    ///
    /// Enables testing `perform_search` success paths by injecting this engine
    /// into a `SearchClient` via `SearchClient::new_with_engines`. The engine
    /// records the last request it saw so tests can assert forwarding behavior.
    struct MockSearchEngine {
        engine_type: SearchEngineType,
        items: Vec<ResponseItem>,
        fail: bool,
        last_request: Mutex<Option<SearchRequest>>,
    }

    impl MockSearchEngine {
        /// Build an engine that returns a successful response with the given items.
        fn success_with_items(engine_type: SearchEngineType, items: Vec<ResponseItem>) -> Self {
            Self {
                engine_type,
                items,
                fail: false,
                last_request: Mutex::new(None),
            }
        }

        /// Build an engine that always fails with `NoEngineAvailable`.
        fn failing(engine_type: SearchEngineType) -> Self {
            Self {
                engine_type,
                items: Vec::new(),
                fail: true,
                last_request: Mutex::new(None),
            }
        }

        /// Return the last SearchRequest seen by `search`, if any.
        #[allow(dead_code)]
        fn last_request(&self) -> Option<SearchRequest> {
            self.last_request.lock().unwrap().clone()
        }
    }

    #[async_trait]
    impl SearchEngine for MockSearchEngine {
        fn name(&self) -> &'static str {
            match self.engine_type {
                SearchEngineType::Bing => "MockBing",
                SearchEngineType::Baidu => "MockBaidu",
                SearchEngineType::Sogou => "MockSogou",
                _ => "MockEngine",
            }
        }

        fn engine_type(&self) -> SearchEngineType {
            self.engine_type
        }

        fn health(&self) -> crate::search::types::EngineHealth {
            crate::search::types::EngineHealth::Healthy
        }

        async fn search(
            &self,
            request: &SearchRequest,
        ) -> Result<Response<ResponseItem>, SearchError> {
            *self.last_request.lock().unwrap() = Some(request.clone());
            if self.fail {
                return Err(SearchError::NoEngineAvailable);
            }
            let total = self.items.len() as u64;
            Ok(Response {
                items: self.items.clone(),
                total_results: Some(total),
                engine: self.engine_type,
            })
        }
    }

    /// Build a sample ResponseItem for tests.
    fn make_response_item(title: &str, url: &str, engine: SearchEngineType) -> ResponseItem {
        ResponseItem {
            title: title.to_string(),
            url: url.to_string(),
            description: format!("desc for {}", title),
            engine,
        }
    }

    /// Mock for SearchClientTrait backed by a `SearchClient` whose engines are
    /// `MockSearchEngine` instances. This lets `perform_search` succeed or fail
    /// deterministically without touching the network.
    struct MockSearchClient {
        inner: SearchClient,
    }

    impl MockSearchClient {
        /// Build a client whose default (Bing) engine returns `items`.
        fn new_with_items(items: Vec<ResponseItem>) -> Self {
            let engine: Arc<dyn SearchEngine> = Arc::new(MockSearchEngine::success_with_items(
                SearchEngineType::Bing,
                items,
            ));
            Self {
                inner: SearchClient::new_with_engines(vec![engine], SearchEngineType::Bing),
            }
        }

        /// Build a client whose default engine always fails.
        fn failing() -> Self {
            let engine: Arc<dyn SearchEngine> =
                Arc::new(MockSearchEngine::failing(SearchEngineType::Bing));
            Self {
                inner: SearchClient::new_with_engines(vec![engine], SearchEngineType::Bing),
            }
        }

        /// Default success client with two generic results (kept for
        /// existing `make_service` callers that don't assert result shape).
        fn new() -> Self {
            Self::new_with_items(vec![
                make_response_item("Result 1", "https://example.com/1", SearchEngineType::Bing),
                make_response_item("Result 2", "https://example.com/2", SearchEngineType::Bing),
            ])
        }
    }

    #[async_trait]
    impl SearchClientTrait for MockSearchClient {
        async fn search(&self, query: &str) -> SearchCommand {
            self.inner.search(query)
        }

        async fn search_with_engine(
            &self,
            _query: &str,
            _engine: SearchEngineType,
        ) -> Result<Response<ResponseItem>, SearchError> {
            Err(SearchError::NoEngineAvailable)
        }

        fn default_engine(&self) -> SearchEngineType {
            SearchEngineType::Bing
        }
    }

    // P12:smart 引擎路由

    fn make_service_with_smart(
        search_client: MockSearchClient,
        smart: Arc<dyn SearchEngine>,
    ) -> SearchService {
        SearchService {
            crawl_repo: Arc::new(MockCrawlRepo::new()),
            task_repo: Arc::new(MockTaskRepo::new()),
            credits_repo: Arc::new(MockCreditsRepo::with_balance(100)),
            search_client: Arc::new(search_client),
            smart_engine: Some(smart),
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(None, 25)),
        }
    }

    /// engine=None 时必须路由到 SmartSearchEngine(fan-out + RRF),
    /// 而非单引擎路径
    #[tokio::test]
    async fn test_engine_none_routes_to_smart_engine() {
        let smart: Arc<dyn SearchEngine> = Arc::new(MockSearchEngine::success_with_items(
            SearchEngineType::Smart,
            vec![make_response_item(
                "smart consensus",
                "https://smart.example/1",
                SearchEngineType::Smart,
            )],
        ));
        let client = MockSearchClient::new_with_items(vec![make_response_item(
            "bing single",
            "https://bing.example/1",
            SearchEngineType::Bing,
        )]);
        let service = make_service_with_smart(client, smart);

        let results = service
            .perform_search("rust async", 10, None, None, None)
            .await
            .expect("perform_search");
        assert!(
            results.iter().all(|r| r.title == "smart consensus"),
            "engine=None must route to smart engine, got: {:?}",
            results.iter().map(|r| r.title.clone()).collect::<Vec<_>>()
        );
    }

    /// 显式引擎名走单引擎路径,不得被 smart 拦截
    #[tokio::test]
    async fn test_explicit_engine_routes_to_single_engine() {
        let smart: Arc<dyn SearchEngine> = Arc::new(MockSearchEngine::success_with_items(
            SearchEngineType::Smart,
            vec![make_response_item(
                "smart consensus",
                "https://smart.example/1",
                SearchEngineType::Smart,
            )],
        ));
        let client = MockSearchClient::new_with_items(vec![make_response_item(
            "bing single",
            "https://bing.example/1",
            SearchEngineType::Bing,
        )]);
        let service = make_service_with_smart(client, smart);

        let results = service
            .perform_search("rust async", 10, None, None, Some("bing"))
            .await
            .expect("perform_search");
        assert!(
            results.iter().all(|r| r.title == "bing single"),
            "explicit engine must bypass smart aggregation, got: {:?}",
            results.iter().map(|r| r.title.clone()).collect::<Vec<_>>()
        );
    }

    /// 未知引擎名必须显性报错——修复前会静默替换为默认 bing
    #[tokio::test]
    async fn test_unknown_engine_name_returns_validation_error() {
        let service = make_service_with_smart(
            MockSearchClient::new(),
            Arc::new(MockSearchEngine::success_with_items(
                SearchEngineType::Smart,
                vec![],
            )),
        );
        let result = service
            .perform_search("rust async", 10, None, None, Some("google"))
            .await;
        match result {
            Err(SearchServiceError::UnknownEngine(name)) => {
                assert_eq!(
                    name, "google",
                    "error should carry the unknown engine name: {name}"
                );
            }
            other => panic!("expected UnknownEngine, got {other:?}"),
        }
    }

    // Helpers

    fn make_service(credits: Arc<dyn CreditsRepository>) -> SearchService {
        SearchService {
            crawl_repo: Arc::new(MockCrawlRepo::new()),
            task_repo: Arc::new(MockTaskRepo::new()),
            credits_repo: credits,
            search_client: Arc::new(MockSearchClient::new()),
            smart_engine: None,
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(None, 25)),
        }
    }

    fn make_query(text: &str) -> SearchQuery {
        SearchQuery {
            query: text.to_string(),
            limit: None,
            lang: None,
            country: None,
            engine: None,
            sources: None,
            crawl_results: None,
            crawl_config: None,
        }
    }

    /// Build a service with recording mock repos, returning the mock handles
    /// so tests can inspect side effects (task/crawl creation, credit calls).
    fn make_service_with_recording(
        credits: Arc<dyn CreditsRepository>,
        search_client: Arc<dyn SearchClientTrait>,
    ) -> (SearchService, Arc<MockCrawlRepo>, Arc<MockTaskRepo>) {
        let crawl_repo = Arc::new(MockCrawlRepo::new());
        let task_repo = Arc::new(MockTaskRepo::new());
        let service = SearchService {
            crawl_repo: crawl_repo.clone(),
            task_repo: task_repo.clone(),
            credits_repo: credits,
            search_client,
            smart_engine: None,
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(None, 25)),
        };
        (service, crawl_repo, task_repo)
    }

    /// Build a service whose search client always fails.
    fn make_failing_search_service(credits: Arc<dyn CreditsRepository>) -> SearchService {
        SearchService {
            crawl_repo: Arc::new(MockCrawlRepo::new()),
            task_repo: Arc::new(MockTaskRepo::new()),
            credits_repo: credits,
            search_client: Arc::new(MockSearchClient::failing()),
            smart_engine: None,
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(None, 25)),
        }
    }

    /// 固定打分重排 mock:"boost-" 前缀文档相关性最高,验证 perform_search 全链路
    struct PrefixScoreReranker {
        fail: bool,
    }

    #[async_trait::async_trait]
    impl crate::domain::services::rerank::RerankProvider for PrefixScoreReranker {
        async fn rerank(
            &self,
            _query: &str,
            documents: &[String],
            _top_k: Option<usize>,
        ) -> Result<
            Vec<crate::domain::services::rerank::RerankScore>,
            crate::domain::services::rerank::RerankError,
        > {
            if self.fail {
                return Err(crate::domain::services::rerank::RerankError::Provider(
                    "down".to_string(),
                ));
            }
            Ok(documents
                .iter()
                .enumerate()
                .map(|(index, doc)| {
                    let score = if doc.starts_with("boost") { 0.99 } else { 0.01 };
                    crate::domain::services::rerank::RerankScore { index, score }
                })
                .collect())
        }

        fn name(&self) -> &str {
            "prefix-mock"
        }
    }

    #[tokio::test]
    async fn test_perform_search_applies_rerank_and_scores() {
        let items = vec![
            make_response_item("plain one", "https://example.com/1", SearchEngineType::Bing),
            make_response_item("boost me", "https://example.com/2", SearchEngineType::Bing),
        ];
        let service = SearchService {
            crawl_repo: Arc::new(MockCrawlRepo::new()),
            task_repo: Arc::new(MockTaskRepo::new()),
            credits_repo: Arc::new(MockCreditsRepo::with_balance(100)),
            search_client: Arc::new(MockSearchClient::new_with_items(items)),
            smart_engine: None,
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(
                Some(Arc::new(PrefixScoreReranker { fail: false })),
                25,
            )),
        };

        let results = service
            .perform_search("anything", 10, None, None, None)
            .await
            .expect("perform_search");

        // "boost me"(0.99)被重排到第一位,且带 score
        assert_eq!(results[0].title, "boost me");
        assert!((results[0].score.unwrap() - 0.99).abs() < 1e-6);
        assert_eq!(results[1].title, "plain one");
        assert!((results[1].score.unwrap() - 0.01).abs() < 1e-6);
    }

    #[tokio::test]
    async fn test_perform_search_rerank_failure_fails_open() {
        let items = vec![
            make_response_item("Result 1", "https://example.com/1", SearchEngineType::Bing),
            make_response_item("Result 2", "https://example.com/2", SearchEngineType::Bing),
        ];
        let service = SearchService {
            crawl_repo: Arc::new(MockCrawlRepo::new()),
            task_repo: Arc::new(MockTaskRepo::new()),
            credits_repo: Arc::new(MockCreditsRepo::with_balance(100)),
            search_client: Arc::new(MockSearchClient::new_with_items(items)),
            smart_engine: None,
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(
                Some(Arc::new(PrefixScoreReranker { fail: true })),
                25,
            )),
        };

        let results = service
            .perform_search("anything", 10, None, None, None)
            .await
            .expect("perform_search");

        // fail-open:原序保留,无 score
        assert_eq!(results[0].title, "Result 1");
        assert_eq!(results[1].title, "Result 2");
        assert!(results.iter().all(|r| r.score.is_none()));
    }

    // SearchQuery tests

    #[test]
    fn test_search_query_construction_with_all_fields() {
        let query = SearchQuery {
            query: "rust web scraping".to_string(),
            limit: Some(20),
            lang: Some("en".to_string()),
            country: Some("us".to_string()),
            engine: Some("google".to_string()),
            sources: Some(vec!["google".to_string(), "bing".to_string()]),
            crawl_results: Some(true),
            crawl_config: Some(SearchCrawlConfig::default()),
        };
        assert_eq!(query.query, "rust web scraping");
        assert_eq!(query.limit, Some(20));
        assert_eq!(query.lang.as_deref(), Some("en"));
        assert_eq!(query.country.as_deref(), Some("us"));
        assert_eq!(query.engine.as_deref(), Some("google"));
        assert_eq!(query.sources.as_ref().unwrap().len(), 2);
        assert_eq!(query.crawl_results, Some(true));
        assert!(query.crawl_config.is_some());
    }

    #[test]
    fn test_search_query_with_minimal_fields() {
        let query = make_query("test");
        assert_eq!(query.query, "test");
        assert!(query.limit.is_none());
        assert!(query.lang.is_none());
        assert!(query.crawl_results.is_none());
    }

    #[test]
    fn test_search_query_empty_string() {
        let query = make_query("");
        assert!(query.query.is_empty());
    }

    // SearchCrawlConfig tests

    #[test]
    fn test_search_crawl_config_default_values() {
        let config = SearchCrawlConfig::default();
        assert_eq!(config.max_depth, 1, "default max_depth should be 1");
        assert_eq!(config.strategy, "bfs", "default strategy should be bfs");
        assert_eq!(
            config.max_concurrency, 10,
            "default max_concurrency should be 10"
        );
        assert!(config.include_patterns.is_none());
        assert!(config.exclude_patterns.is_none());
        assert!(config.crawl_delay_ms.is_none());
        assert!(config.headers.is_none());
        assert!(config.proxy.is_none());
        assert!(config.extraction_rules.is_none());
    }

    #[test]
    fn test_search_crawl_config_serde_roundtrip() {
        let config = SearchCrawlConfig {
            max_depth: 3,
            include_patterns: Some(vec!["*/blog/*".to_string()]),
            exclude_patterns: Some(vec!["*/admin/*".to_string()]),
            strategy: "bfs".to_string(),
            crawl_delay_ms: Some(500),
            max_concurrency: 10,
            headers: Some(serde_json::json!({"User-Agent": "test"})),
            proxy: Some("http://proxy:8080".to_string()),
            extraction_rules: None,
            domain_blacklist: Vec::new(),
        };
        let json = serde_json::to_string(&config).expect("should serialize");
        let back: SearchCrawlConfig = serde_json::from_str(&json).expect("should deserialize");
        assert_eq!(back.max_depth, 3);
        assert_eq!(back.strategy, "bfs");
        assert_eq!(back.max_concurrency, 10);
        assert_eq!(back.crawl_delay_ms, Some(500));
        assert_eq!(
            back.include_patterns.as_ref().unwrap(),
            &vec!["*/blog/*".to_string()]
        );
    }

    #[test]
    fn test_search_crawl_config_serde_minimal() {
        let config = SearchCrawlConfig::default();
        let json = serde_json::to_string(&config).expect("should serialize");
        let back: SearchCrawlConfig = serde_json::from_str(&json).expect("should deserialize");
        assert_eq!(config.max_depth, back.max_depth);
        assert_eq!(config.strategy, back.strategy);
    }

    // SearchResult tests

    #[test]
    fn test_search_result_construction() {
        let result = SearchResult {
            title: "Rust Programming".to_string(),
            url: "https://www.rust-lang.org".to_string(),
            description: Some(
                "A language empowering everyone to build reliable software.".to_string(),
            ),
            engine: "Google".to_string(),
            score: None,
        };
        assert_eq!(result.title, "Rust Programming");
        assert_eq!(result.url, "https://www.rust-lang.org");
        assert!(result.description.is_some());
        assert_eq!(result.engine, "Google");
    }

    #[test]
    fn test_search_result_with_none_description() {
        let result = SearchResult {
            title: "Test".to_string(),
            url: "https://example.com".to_string(),
            description: None,
            engine: "Bing".to_string(),
            score: None,
        };
        assert!(result.description.is_none());
    }

    // SearchResponse tests

    #[test]
    fn test_search_response_construction_with_crawl() {
        let response = SearchResponse {
            query: "rust scraping".to_string(),
            results: vec![SearchResult {
                title: "Result 1".to_string(),
                url: "https://example.com".to_string(),
                description: None,
                engine: "Google".to_string(),
                score: None,
            }],
            crawl_id: Some(crate::domain::id::Id::generate()),
            credits_used: 1,
        };
        assert_eq!(response.query, "rust scraping");
        assert_eq!(response.results.len(), 1);
        assert!(response.crawl_id.is_some());
        assert_eq!(response.credits_used, 1);
    }

    #[test]
    fn test_search_response_construction_without_crawl() {
        let response = SearchResponse {
            query: "test".to_string(),
            results: vec![],
            crawl_id: None,
            credits_used: 1,
        };
        assert!(response.results.is_empty());
        assert!(response.crawl_id.is_none());
    }

    // SearchServiceError From conversions

    #[test]
    fn test_search_service_error_from_string() {
        let err: SearchServiceError = "bad query".to_string().into();
        match err {
            SearchServiceError::ValidationError(msg) => {
                assert_eq!(msg, "bad query");
            }
            other => panic!("expected ValidationError, got {:?}", other),
        }
    }

    #[test]
    fn test_search_service_error_from_str_ref() {
        let err: SearchServiceError = "invalid input".into();
        match err {
            SearchServiceError::ValidationError(msg) => {
                assert_eq!(msg, "invalid input");
            }
            other => panic!("expected ValidationError, got {:?}", other),
        }
    }

    #[test]
    fn test_search_service_error_from_anyhow_error() {
        let err: SearchServiceError = anyhow::anyhow!("engine crashed").into();
        match err {
            SearchServiceError::SearchEngine(msg) => {
                assert!(msg.contains("engine crashed"), "msg: {}", msg);
            }
            other => panic!("expected SearchEngine, got {:?}", other),
        }
    }

    #[test]
    fn test_search_service_error_display_validation() {
        let err = SearchServiceError::ValidationError("missing field".to_string());
        let msg = err.to_string();
        assert!(msg.contains("Validation failed"));
        assert!(msg.contains("missing field"));
    }

    #[test]
    fn test_search_service_error_display_insufficient_credits() {
        let err = SearchServiceError::InsufficientCredits {
            available: 0,
            required: 1,
        };
        let msg = err.to_string();
        assert!(msg.contains("Insufficient credits"));
        assert!(msg.contains("available 0"));
        assert!(msg.contains("required 1"));
    }

    #[test]
    fn test_search_service_error_display_search_engine() {
        let err = SearchServiceError::SearchEngine("timeout".to_string());
        let msg = err.to_string();
        assert!(msg.contains("Search engine error"));
        assert!(msg.contains("timeout"));
    }

    // search pre-validation tests

    /// 回归:搜索派生 crawl 的 payload 必须能被 worker 的 `parse_crawl_payload`
    /// 解析(config 不得含 deny_unknown_fields 之外的字段),且黑名单**只**
    /// 存 crawl.config(payload 不再逐任务复制顶层黑名单)。
    #[test]
    fn search_crawl_payload_is_worker_compatible() {
        let config = SearchCrawlConfig {
            max_depth: 1,
            include_patterns: None,
            exclude_patterns: None,
            strategy: "bfs".to_string(),
            crawl_delay_ms: None,
            max_concurrency: 10,
            headers: None,
            proxy: None,
            extraction_rules: None,
            domain_blacklist: vec!["blocked.com".to_string()],
        };
        let config_json = serde_json::to_value(&config).expect("serialize config");
        assert!(
            config_json.get("domain_blacklist").is_none(),
            "domain_blacklist must not be serialized into CrawlConfigDto"
        );
        let crawl_id = crate::domain::id::Id::generate();
        let mut task = Task::new(
            crate::domain::id::Id::generate(),
            TaskType::Crawl,
            crate::domain::id::Id::generate(),
            crate::domain::id::Id::generate(),
            "https://example.com/".to_string(),
            json!({
                "crawl_id": crawl_id.to_string(),
                "depth": 0,
                "config": config_json,
            }),
        );
        task.crawl_id = Some(crawl_id);
        let parsed = crate::workers::scrape_response_builder::parse_crawl_payload(&task)
            .expect("search crawl payload must parse as CrawlConfigDto");
        assert_eq!(parsed.1, 0);
        assert_eq!(parsed.2.max_depth, 1);
        assert!(
            task.payload.get("domain_blacklist").is_none(),
            "domain_blacklist must not be copied into task payload"
        );
    }

    /// 黑名单只在 crawl.config 存一份(归一化后),结果过滤与任务派生同源。
    #[tokio::test]
    async fn search_crawl_stores_normalized_blacklist_in_crawl_config() {
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let client = MockSearchClient::new_with_items(vec![
            make_response_item(
                "Blocked",
                "https://Blocked.COM/page",
                SearchEngineType::Bing,
            ),
            make_response_item(
                "Allowed",
                "https://allowed.example/page",
                SearchEngineType::Bing,
            ),
        ]);
        let (service, crawl_repo, task_repo) =
            make_service_with_recording(credits, Arc::new(client));
        let config = SearchCrawlConfig {
            domain_blacklist: vec![" Blocked.com. ".to_string()],
            ..SearchCrawlConfig::default()
        };
        let query = SearchQuery {
            crawl_results: Some(true),
            crawl_config: Some(config),
            ..make_query("blacklist storage")
        };
        let resp = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await
            .expect("search should succeed");
        // 搜索结果本身不回删(响应契约不变);黑名单只作用于派生 crawl 的任务。
        assert_eq!(resp.results.len(), 2, "search response keeps all results");
        let created = crawl_repo.created.lock().unwrap();
        let crawl = created.last().expect("crawl record should exist");
        assert_eq!(crawl.total_tasks(), 1, "total counts eligible results only");
        let stored = crawl.config().clone();
        assert_eq!(
            stored["domain_blacklist"],
            json!(["blocked.com"]),
            "crawl.config must carry the normalized blacklist"
        );
        assert_eq!(task_repo.created_count(), 1, "one task for allowed result");
        assert_eq!(
            task_repo.created_urls(),
            vec!["https://allowed.example/page".to_string()],
            "blacklisted URL must not be enqueued"
        );
    }

    #[tokio::test]
    async fn test_search_crawl_config_depth_above_limit_is_rejected() {
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let (service, crawl_repo, task_repo) =
            make_service_with_recording(credits, Arc::new(MockSearchClient::new()));
        let query = SearchQuery {
            crawl_results: Some(true),
            crawl_config: Some(SearchCrawlConfig {
                max_depth: MAX_SEARCH_CRAWL_DEPTH + 1,
                ..SearchCrawlConfig::default()
            }),
            ..make_query("too deep")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(matches!(
            result,
            Err(SearchServiceError::ValidationError(_))
        ));
        assert_eq!(crawl_repo.created_count(), 0, "no crawl on invalid depth");
        assert_eq!(task_repo.created_count(), 0, "no tasks on invalid depth");
    }

    #[tokio::test]
    async fn test_search_empty_query_returns_validation_error() {
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(100)));
        let query = make_query("");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::ValidationError(msg) => {
                assert!(msg.contains("empty"), "error should mention empty: {}", msg);
            }
            other => panic!("expected ValidationError, got {:?}", other),
        }
    }

    #[tokio::test]
    async fn test_search_whitespace_only_query_returns_validation_error() {
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(100)));
        let query = make_query("   ");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err(), "whitespace-only query should fail");
        match result.unwrap_err() {
            SearchServiceError::ValidationError(_) => {}
            other => panic!("expected ValidationError, got {:?}", other),
        }
    }

    #[tokio::test]
    async fn test_search_insufficient_credits_returns_error() {
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(0)));
        let query = make_query("rust scraping");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::InsufficientCredits {
                available,
                required,
            } => {
                assert_eq!(available, 0);
                assert_eq!(required, 1);
            }
            other => panic!("expected InsufficientCredits, got {:?}", other),
        }
    }

    #[tokio::test]
    async fn test_search_exact_balance_succeeds_validation() {
        // Balance exactly equals search_cost (1) → passes credit check.
        // perform_search succeeds with the mock engine, so we expect Ok with
        // credits_used == 1. The key assertion is that InsufficientCredits is
        // NOT returned.
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(1)));
        let query = make_query("rust scraping");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(
            result.is_ok(),
            "should pass credit check and perform_search, got {:?}",
            result.err()
        );
        let resp = result.unwrap();
        assert_eq!(resp.credits_used, 1, "search cost is 1 credit");
    }

    // engine_param branch tests
    // These tests exercise the sources/engine parameter handling logic in search.
    // perform_search succeeds with the injected MockSearchEngine (default Bing),
    // so each test asserts Ok, proving the code reached perform_search and that the
    // engine_param branches did not short-circuit into ValidationError/InsufficientCredits.

    #[tokio::test]
    async fn test_search_with_single_source_passes_source_as_engine() {
        // sources.len == 1 → engine_param = Some(sources[0]) = Some("bing")
        // MockSearchEngine (Bing) handles the request successfully.
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(10)));
        let query = SearchQuery {
            sources: Some(vec!["bing".to_string()]),
            ..make_query("rust scraping")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(
            result.is_ok(),
            "should reach perform_search and succeed, got {:?}",
            result.err()
        );
    }

    #[tokio::test]
    async fn test_search_with_multiple_sources_uses_aggregator() {
        // sources.len > 1 → engine_param = None (aggregator)
        // perform_search falls back to the default (Bing) mock engine and succeeds.
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(10)));
        let query = SearchQuery {
            sources: Some(vec!["bing".to_string(), "baidu".to_string()]),
            ..make_query("rust scraping")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(
            result.is_ok(),
            "aggregator path should reach perform_search and succeed, got {:?}",
            result.err()
        );
    }

    #[tokio::test]
    async fn test_search_with_engine_specified_passes_engine() {
        // sources = None, engine = Some → engine_param = Some(engine)
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(10)));
        let query = SearchQuery {
            engine: Some("exa".to_string()),
            ..make_query("rust scraping")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        // Exa engine is not registered in mock, so it should fail with NoEngineAvailable
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::ValidationError(_) => {
                panic!("should pass validation");
            }
            SearchServiceError::InsufficientCredits { .. } => {
                panic!("should pass credit check");
            }
            _ => { /* reached perform_search — expected */ }
        }
    }

    #[tokio::test]
    async fn test_search_with_limit_specified_passes_to_perform_search() {
        // Exercise the query.limit.unwrap_or(10) branch with explicit limit.
        // MockSearchClient::new returns 2 items, so limit=5 yields 2 results.
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(10)));
        let query = SearchQuery {
            limit: Some(5),
            lang: Some("en".to_string()),
            country: Some("us".to_string()),
            ..make_query("rust scraping")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(
            result.is_ok(),
            "should reach perform_search and succeed, got {:?}",
            result.err()
        );
        let resp = result.unwrap();
        assert_eq!(
            resp.results.len(),
            2,
            "mock returns 2 items, limit=5 keeps both"
        );
    }

    #[tokio::test]
    async fn test_search_get_balance_failure_returns_credits_error() {
        let service = make_service(Arc::new(MockCreditsRepo::failing()));
        let query = make_query("rust scraping");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::CreditsRepository(err) => match err {
                CreditsRepositoryError::DatabaseError(msg) => {
                    assert!(msg.contains("mock get_balance failure"));
                }
                other => panic!("expected DatabaseError, got {:?}", other),
            },
            other => panic!("expected CreditsRepository, got {:?}", other),
        }
    }

    #[tokio::test]
    async fn test_search_negative_balance_returns_insufficient_credits() {
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(-5)));
        let query = make_query("rust scraping");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::InsufficientCredits {
                available,
                required,
            } => {
                assert_eq!(available, -5, "should report negative balance");
                assert_eq!(required, 1);
            }
            other => panic!("expected InsufficientCredits, got {:?}", other),
        }
    }

    // SearchServiceTrait forwarding test

    #[tokio::test]
    async fn test_search_service_trait_forwards_empty_query_error() {
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(100)));
        let query = make_query("");
        // Use the trait method
        let result: Result<SearchResponse, SearchServiceError> = SearchServiceTrait::search(
            &service,
            crate::domain::id::Id::generate(),
            crate::domain::id::Id::generate(),
            query,
        )
        .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::ValidationError(_) => {}
            other => panic!("expected ValidationError, got {:?}", other),
        }
    }

    // perform_search success path tests
    //
    // These tests rely on MockSearchEngine (injected via SearchClient::new_with_engines)
    // to exercise the success branch of `perform_search` and the downstream crawl/credit
    // flows that were previously unreachable with a real SearchClient.

    #[tokio::test]
    async fn test_search_success_returns_results() {
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let (service, _, _) =
            make_service_with_recording(credits, Arc::new(MockSearchClient::new()));
        let query = make_query("rust web scraping");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_ok(), "expected ok, got {:?}", result.err());
        let resp = result.unwrap();
        assert_eq!(resp.query, "rust web scraping");
        assert_eq!(
            resp.results.len(),
            2,
            "MockSearchClient::new returns 2 items"
        );
        assert_eq!(resp.credits_used, 1, "search cost is 1 credit");
        assert!(resp.crawl_id.is_none(), "crawl_results not set");
    }

    #[tokio::test]
    async fn test_search_success_deducts_credits() {
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let credits_handle: Arc<MockCreditsRepo> = credits.clone();
        let (service, _, _) =
            make_service_with_recording(credits, Arc::new(MockSearchClient::new()));
        let query = make_query("rust");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_ok());
        let calls = credits_handle.deduct_calls.lock().unwrap().clone();
        assert_eq!(calls.len(), 1, "deduct_credits should be called once");
        assert_eq!(calls[0].1, 1, "deducted amount should be 1");
    }

    #[tokio::test]
    async fn test_search_success_with_crawl_creates_tasks_and_crawl() {
        let (service, crawl_repo, task_repo) = make_service_with_recording(
            Arc::new(MockCreditsRepo::with_balance(100)),
            Arc::new(MockSearchClient::new()),
        );
        let query = SearchQuery {
            crawl_results: Some(true),
            ..make_query("rust scraping")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_ok());
        let resp = result.unwrap();
        assert!(resp.crawl_id.is_some(), "crawl_id should be set");
        assert_eq!(
            crawl_repo.created_count(),
            1,
            "exactly one crawl record should be created"
        );
        assert_eq!(
            task_repo.created_count(),
            resp.results.len(),
            "one task per search result should be created"
        );
    }

    #[tokio::test]
    async fn test_search_success_with_crawl_skips_when_no_results() {
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let (service, crawl_repo, task_repo) = make_service_with_recording(
            credits,
            Arc::new(MockSearchClient::new_with_items(vec![])),
        );
        let query = SearchQuery {
            crawl_results: Some(true),
            ..make_query("empty")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_ok());
        let resp = result.unwrap();
        assert!(resp.crawl_id.is_none(), "no crawl when results empty");
        assert_eq!(crawl_repo.created_count(), 0);
        assert_eq!(task_repo.created_count(), 0);
    }

    #[tokio::test]
    async fn test_search_respects_limit_truncates_results() {
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let (service, _, _) = make_service_with_recording(
            credits,
            Arc::new(MockSearchClient::new_with_items(vec![
                make_response_item("a", "https://a.example", SearchEngineType::Bing),
                make_response_item("b", "https://b.example", SearchEngineType::Bing),
                make_response_item("c", "https://c.example", SearchEngineType::Bing),
                make_response_item("d", "https://d.example", SearchEngineType::Bing),
            ])),
        );
        let query = SearchQuery {
            limit: Some(2),
            ..make_query("limited")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_ok());
        let resp = result.unwrap();
        assert_eq!(
            resp.results.len(),
            2,
            "limit=2 should truncate to 2 results"
        );
    }

    #[tokio::test]
    async fn test_search_perform_search_failure_returns_search_engine_error() {
        let service = make_failing_search_service(Arc::new(MockCreditsRepo::with_balance(100)));
        let query = make_query("will fail");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::SearchEngine(msg) => {
                assert!(!msg.is_empty(), "error message should be non-empty");
            }
            other => panic!("expected SearchEngine error, got {:?}", other),
        }
    }

    #[tokio::test]
    async fn test_search_crawl_repo_failure_propagates() {
        let credits: Arc<dyn CreditsRepository> = Arc::new(MockCreditsRepo::with_balance(100));
        let service = SearchService {
            crawl_repo: Arc::new(MockCrawlRepo::failing()),
            task_repo: Arc::new(MockTaskRepo::new()),
            credits_repo: credits,
            search_client: Arc::new(MockSearchClient::new()),
            smart_engine: None,
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(None, 25)),
        };
        let query = SearchQuery {
            crawl_results: Some(true),
            ..make_query("crawl fails")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::Repository(_) => { /* expected */ }
            other => panic!("expected Repository error, got {:?}", other),
        }
    }

    #[tokio::test]
    async fn test_search_task_repo_failure_propagates_and_compensates() {
        let credits: Arc<dyn CreditsRepository> = Arc::new(MockCreditsRepo::with_balance(100));
        let crawl_repo = Arc::new(MockCrawlRepo::new());
        let service = SearchService {
            crawl_repo: crawl_repo.clone(),
            task_repo: Arc::new(MockTaskRepo::failing()),
            credits_repo: credits,
            search_client: Arc::new(MockSearchClient::new()),
            smart_engine: None,
            max_results: 50,
            reranker: Arc::new(SearchReranker::new(None, 25)),
        };
        let query = SearchQuery {
            crawl_results: Some(true),
            ..make_query("task fails")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::Repository(_) => { /* expected */ }
            other => panic!("expected Repository error, got {:?}", other),
        }
        assert_eq!(
            crawl_repo.compensation_count(),
            1,
            "enqueue failure must mark the crawl terminal (not leave it queued)"
        );
    }

    #[tokio::test]
    async fn test_search_deduct_credits_failure_propagates() {
        let credits = Arc::new(MockCreditsRepo {
            balance: 100,
            get_balance_fails: false,
            deduct_fails: true,
            deduct_calls: Mutex::new(Vec::new()),
        });
        let (service, _, _) =
            make_service_with_recording(credits, Arc::new(MockSearchClient::new()));
        let query = make_query("deduct fails");
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_err());
        match result.unwrap_err() {
            SearchServiceError::CreditsRepository(_) => { /* expected */ }
            other => panic!("expected CreditsRepository error, got {:?}", other),
        }
    }

    #[tokio::test]
    async fn test_search_trait_delegates_to_success_path() {
        let service = make_service(Arc::new(MockCreditsRepo::with_balance(100)));
        let query = make_query("trait delegation");
        let result: Result<SearchResponse, SearchServiceError> = SearchServiceTrait::search(
            &service,
            crate::domain::id::Id::generate(),
            crate::domain::id::Id::generate(),
            query,
        )
        .await;
        assert!(result.is_ok(), "trait should forward to success path");
        let resp = result.unwrap();
        assert_eq!(resp.results.len(), 2);
    }

    // Mock stub contract tests
    //
    // search only exercises `create` on the task/crawl repos and
    // `get_balance`/`deduct_credits` on the credits repo. The remaining methods
    // are stubs that return safe defaults. Calling them here documents the stub
    // contract and keeps coverage accounting honest for the mock implementations.

    #[tokio::test]
    async fn test_mock_credits_repo_unused_methods_return_expected_defaults() {
        let credits = MockCreditsRepo::with_balance(10);
        let team_id = crate::domain::id::Id::generate();
        // add_credits returns 0 (stub does not track new balance)
        assert_eq!(
            credits
                .add_credits(
                    team_id,
                    5,
                    CreditsTransactionType::Search,
                    "add".to_string(),
                    None
                )
                .await
                .unwrap(),
            0
        );
        // get_transaction_history returns empty
        assert!(credits
            .get_transaction_history(team_id, None)
            .await
            .unwrap()
            .is_empty());
        // initialize_team_credits returns 0
        assert_eq!(
            credits.initialize_team_credits(team_id, 100).await.unwrap(),
            0
        );
    }

    #[tokio::test]
    async fn test_mock_task_repo_unused_methods_return_expected_defaults() {
        let task_repo = MockTaskRepo::new();
        let task_id = crate::domain::id::Id::generate();
        // Read/query stubs return "not found" / empty
        assert!(task_repo.find_by_id(task_id).await.unwrap().is_none());
        assert!(task_repo
            .acquire_next(crate::domain::id::Id::generate())
            .await
            .unwrap()
            .is_none());
        assert!(!task_repo
            .exists_by_url("https://example.com")
            .await
            .unwrap());
        // State-transition stubs are no-ops returning Ok
        task_repo.mark_completed(task_id, None).await.unwrap();
        task_repo.mark_failed(task_id, None).await.unwrap();
        task_repo.mark_cancelled(task_id).await.unwrap();
        // Maintenance stubs return zero counts
        assert_eq!(task_repo.expire_tasks().await.unwrap(), 0);
        assert_eq!(
            task_repo
                .cancel_tasks_by_crawl_id(crate::domain::id::Id::generate())
                .await
                .unwrap(),
            0
        );
    }

    #[tokio::test]
    async fn test_mock_crawl_repo_unused_methods_return_expected_defaults() {
        let crawl_repo = MockCrawlRepo::new();
        let crawl_id = crate::domain::id::Id::generate();
        // find_by_id returns None (stub)
        assert!(crawl_repo.find_by_id(crawl_id).await.unwrap().is_none());
        // Counter/status stubs are no-ops returning Ok
        crawl_repo
            .increment_completed_tasks(crawl_id)
            .await
            .unwrap();
        crawl_repo.increment_failed_tasks(crawl_id).await.unwrap();
        crawl_repo.increment_total_tasks(crawl_id).await.unwrap();
        crawl_repo
            .update_status(crawl_id, CrawlStatus::Completed)
            .await
            .unwrap();
        // Query stubs return empty / zero
        assert!(crawl_repo
            .find_by_team_id_paginated(crawl_id, 10, 0)
            .await
            .unwrap()
            .is_empty());
        assert_eq!(crawl_repo.count_by_team_id(crawl_id).await.unwrap(), 0);
    }

    #[tokio::test]
    async fn test_mock_task_repo_remaining_methods_return_defaults() {
        let task_repo = MockTaskRepo::new();
        let task = Task::new(
            crate::domain::id::Id::generate(),
            TaskType::Scrape,
            crate::domain::id::Id::generate(),
            crate::domain::id::Id::generate(),
            "https://example.com".to_string(),
            json!({}),
        );
        // update echoes the task back
        let updated = task_repo.update(&task).await.unwrap();
        assert_eq!(updated.id, task.id);
        // find_existing_urls returns empty set
        let existing = task_repo
            .find_existing_urls(&["https://a.com".to_string()])
            .await
            .unwrap();
        assert!(existing.is_empty());
        // reset_stuck_tasks returns 0
        assert_eq!(
            task_repo
                .reset_stuck_tasks(chrono::Duration::minutes(5))
                .await
                .unwrap(),
            0
        );
        // find_by_crawl_id returns empty vec
        assert!(task_repo
            .find_by_crawl_id(crate::domain::id::Id::generate())
            .await
            .unwrap()
            .is_empty());
        // query_tasks returns empty vec + 0 count
        let (tasks, count) = task_repo
            .query_tasks(crate::domain::repositories::task_repository::TaskQueryParams::default())
            .await
            .unwrap();
        assert!(tasks.is_empty());
        assert_eq!(count, 0);
        // batch_cancel returns empty tuples
        let (cancelled, failed) = task_repo
            .batch_cancel(
                vec![crate::domain::id::Id::generate()],
                crate::domain::id::Id::generate(),
                false,
            )
            .await
            .unwrap();
        assert!(cancelled.is_empty());
        assert!(failed.is_empty());
    }

    #[tokio::test]
    async fn test_mock_crawl_repo_update_echoes_back() {
        let crawl_repo = MockCrawlRepo::new();
        let crawl = Crawl::new(
            crate::domain::id::Id::generate(),
            crate::domain::id::Id::generate(),
            "Test".to_string(),
            "https://example.com".to_string(),
            "https://example.com".to_string(),
            json!({}),
        );
        let result = crawl_repo.update(&crawl).await.unwrap();
        assert_eq!(result.id, crawl.id);
    }

    #[test]
    fn test_mock_search_engine_name_engine_type_health_all_variants() {
        // Named branches in name: Google, Bing, Baidu, Sogou
        for engine_type in [
            SearchEngineType::Bing,
            SearchEngineType::Bing,
            SearchEngineType::Baidu,
            SearchEngineType::Sogou,
        ] {
            let engine = MockSearchEngine::success_with_items(engine_type, vec![]);
            assert_eq!(engine.engine_type(), engine_type);
            let _ = engine.name();
            assert!(matches!(
                engine.health(),
                crate::search::types::EngineHealth::Healthy
            ));
        }
        // Catch-all branch in name: Smart, Exa, Tavily
        for engine_type in [
            SearchEngineType::Smart,
            SearchEngineType::Exa,
            SearchEngineType::Tavily,
        ] {
            let engine = MockSearchEngine::success_with_items(engine_type, vec![]);
            assert_eq!(engine.engine_type(), engine_type);
            assert_eq!(engine.name(), "MockEngine");
            assert!(matches!(
                engine.health(),
                crate::search::types::EngineHealth::Healthy
            ));
        }
    }

    #[test]
    fn test_mock_search_engine_last_request_is_none_initially() {
        let engine = MockSearchEngine::success_with_items(SearchEngineType::Bing, vec![]);
        assert!(engine.last_request().is_none());
    }

    fn make_test_settings_for_search() -> Settings {
        use crate::config::settings::*;
        Settings {
            server: ServerSettings::default(),
            robots: RobotsSettings::default(),
            database: DatabaseSettings::default(),
            cors: CorsSettings::default(),
            rate_limiting: RateLimitingSettings::default(),
            concurrency: ConcurrencySettings::default(),
            retention: RetentionSettings::default(),
            webhook: WebhookSettings::default(),
            bing_search: BingSearchSettings::default(),
            search: SearchSettings::default(),
            llm: LLMSettings::default(),
            rag: RagSettings::default(),
            fluxweir: FluxweirSettings::default(),
            proxy: ProxySettings::default(),
            engines: EngineSettings::default(),
            markdown: Default::default(),
            logging: LoggingSettings::default(),
            workers: WorkerSettings::default(),
            sitemap: SitemapSettings::default(),
            crawl: CrawlSettings::default(),
            timeouts: TimeoutSettings::default(),
            cache: CacheSettings::default(),
            websocket: WebSocketSettings::default(),
            trusted_proxies: TrustedProxySettings::default(),
            observability: ObservabilitySettings::default(),
            auth: AuthSettings::default(),
            i18n: I18nSettings::default(),
            politeness: PolitenessSettings::default(),
            session: SessionSettings::default(),
            content_extraction: ContentExtractionSettings::default(),
            id: IdSettings::default(),
        }
    }

    #[test]
    fn test_search_service_new_constructor_assigns_dependencies() {
        let settings = make_test_settings_for_search();
        let service = SearchService::new(
            Arc::new(MockCrawlRepo::new()),
            Arc::new(MockTaskRepo::new()),
            Arc::new(MockCreditsRepo::with_balance(10)),
            Arc::new(settings),
            Arc::new(MockSearchClient::new()),
            Arc::new(SearchReranker::new(None, 25)),
        );
        // Verify the service is usable by performing a search that exercises
        // the injected mocks (empty query → ValidationError before touching repos)
        let rt = tokio::runtime::Runtime::new().unwrap();
        let err = rt
            .block_on(service.search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                make_query(""),
            ))
            .unwrap_err();
        assert!(matches!(
            err,
            SearchServiceError::ValidationError(ref msg) if msg.contains("empty")
        ));
    }

    // crawl_config custom path tests

    #[tokio::test]
    async fn test_search_with_crawl_results_and_custom_crawl_config() {
        // Exercises the `query.crawl_config.unwrap_or(default)` branch where
        // crawl_config is Some(custom). Verifies the custom config values are
        // serialized into the created Crawl record (not the defaults).
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let (service, crawl_repo, task_repo) =
            make_service_with_recording(credits, Arc::new(MockSearchClient::new()));
        let custom_config = SearchCrawlConfig {
            max_depth: 5,
            include_patterns: Some(vec!["*/blog/*".to_string()]),
            exclude_patterns: Some(vec!["*/admin/*".to_string()]),
            strategy: "dfs".to_string(),
            crawl_delay_ms: Some(1000),
            max_concurrency: 20,
            headers: Some(json!({"User-Agent": "CrawlrsBot"})),
            proxy: Some("http://proxy:8080".to_string()),
            extraction_rules: None,
            domain_blacklist: Vec::new(),
        };
        let query = SearchQuery {
            crawl_results: Some(true),
            crawl_config: Some(custom_config),
            ..make_query("custom crawl config")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_ok(), "expected ok, got {:?}", result.err());
        let resp = result.unwrap();
        assert!(resp.crawl_id.is_some(), "crawl_id should be set");
        assert_eq!(crawl_repo.created_count(), 1, "one crawl record created");
        assert_eq!(
            task_repo.created_count(),
            resp.results.len(),
            "one task per result"
        );
        // Verify the custom config (not defaults) was serialized into the crawl
        let config = crawl_repo
            .last_created_config()
            .expect("crawl record should exist");
        assert_eq!(config["max_depth"], 5, "custom max_depth should be 5");
        assert_eq!(config["strategy"], "dfs", "custom strategy should be dfs");
        assert_eq!(
            config["max_concurrency"], 20,
            "custom max_concurrency should be 20"
        );
        assert_eq!(
            config["crawl_delay_ms"], 1000,
            "custom crawl_delay_ms should be 1000"
        );
        assert_eq!(
            config["proxy"], "http://proxy:8080",
            "custom proxy should be set"
        );
        assert_eq!(
            config["include_patterns"][0], "*/blog/*",
            "custom include_patterns should be set"
        );
        assert_eq!(
            config["exclude_patterns"][0], "*/admin/*",
            "custom exclude_patterns should be set"
        );
    }

    #[tokio::test]
    async fn test_search_with_crawl_results_default_config_uses_bfs_strategy() {
        // Exercises the `query.crawl_config.unwrap_or(default)` branch where
        // crawl_config is None. Verifies the default config uses strategy="bfs"
        // and max_depth=1, confirming the default branch is distinct from the
        // custom config branch.
        let credits = Arc::new(MockCreditsRepo::with_balance(100));
        let (service, crawl_repo, _) =
            make_service_with_recording(credits, Arc::new(MockSearchClient::new()));
        let query = SearchQuery {
            crawl_results: Some(true),
            crawl_config: None,
            ..make_query("default crawl config")
        };
        let result = service
            .search(
                crate::domain::id::Id::generate(),
                crate::domain::id::Id::generate(),
                query,
            )
            .await;
        assert!(result.is_ok());
        let resp = result.unwrap();
        assert!(resp.crawl_id.is_some());
        let config = crawl_repo
            .last_created_config()
            .expect("crawl record should exist");
        assert_eq!(config["max_depth"], 1, "default max_depth should be 1");
        assert_eq!(config["strategy"], "bfs", "default strategy should be bfs");
        assert_eq!(
            config["max_concurrency"], 10,
            "default max_concurrency should be 10"
        );
    }

    // SearchServiceError display tests for remaining variants

    #[test]
    fn test_search_service_error_display_repository() {
        let inner = RepositoryError::Database(anyhow::anyhow!("db connection lost"));
        let err = SearchServiceError::Repository(inner);
        let msg = err.to_string();
        assert!(msg.contains("Repository error"));
        assert!(msg.contains("db connection lost"));
    }

    #[test]
    fn test_search_service_error_display_credits_repository() {
        let inner = CreditsRepositoryError::DatabaseError("db timeout".to_string());
        let err = SearchServiceError::CreditsRepository(inner);
        let msg = err.to_string();
        assert!(msg.contains("Credits repository error"));
        assert!(msg.contains("db timeout"));
    }
}