File size: 47,635 Bytes
35e7795 | 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 | """数据集相关的API接口"""
import json
from typing import List, Optional
from fastapi import APIRouter, Depends, File, Form, HTTPException, UploadFile, status
from sqlalchemy.orm import Session
from qa_annotate.api.auth import get_current_active_user, get_current_superuser
from qa_annotate.database.base import get_db
from qa_annotate.database.crud import (
AnnotationResultCRUD,
DatasetAnnotationConfigCRUD,
DatasetCRUD,
ProjectAnnotationConfigCRUD,
ProjectCRUD,
QAPairCRUD,
UserCRUD,
)
from qa_annotate.schema.annotation import AnnotationConfig
from qa_annotate.schema.dataset import Dataset, QAPair
from qa_annotate.schema.task import EvaluationDimension, TaskInfo
from qa_annotate.schema.user import User
router = APIRouter(prefix="/datasets", tags=["datasets"])
async def import_dataset_from_file(
file: UploadFile,
db: Session,
current_user: User,
dataset_name: Optional[str] = None,
dataset_description: Optional[str] = None,
dataset_version: Optional[str] = None,
dataset_category: Optional[str] = None,
dataset_status: Optional[str] = None,
dataset_tags: Optional[str] = None,
dataset_source: Optional[str] = None,
dataset_source_url: Optional[str] = None,
project_id: Optional[int] = None,
annotator_id: Optional[int] = None,
) -> dict:
"""从JSONL文件导入数据集的辅助函数
返回格式:
{
"dataset_id": int,
"dataset_name": str,
"created": bool,
"imported_count": int,
"failed_count": int,
"total_lines": int,
"errors": List[str]
}
"""
# 检查文件类型
if not file.filename or not file.filename.endswith(".jsonl"):
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail="只支持.jsonl格式的文件"
)
# 读取文件内容
try:
content = await file.read()
text_content = content.decode("utf-8")
except Exception as e:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail=f"读取文件失败: {str(e)}"
)
# 解析JSONL文件
lines = text_content.strip().split("\n")
if not lines or not lines[0].strip():
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail="文件为空或格式不正确"
)
dataset_info = {}
qa_start_index = 0
# 如果提供了参数中的元数据,优先使用参数数据
if dataset_name:
dataset_info["name"] = dataset_name
if dataset_description:
dataset_info["description"] = dataset_description
if dataset_version:
dataset_info["version"] = dataset_version
if dataset_category:
dataset_info["category"] = dataset_category
if dataset_status:
dataset_info["status"] = dataset_status
if dataset_tags:
dataset_info["tags"] = [
tag.strip() for tag in dataset_tags.split(",") if tag.strip()
]
if dataset_source:
dataset_info["source"] = dataset_source
if dataset_source_url:
dataset_info["source_url"] = dataset_source_url
else:
# 如果没有提供参数元数据,检查第一行是否是数据集元数据
try:
first_line_data = json.loads(lines[0].strip())
if (
isinstance(first_line_data, dict)
and first_line_data.get("__type__") == "dataset"
):
dataset_info = first_line_data.copy()
# 移除类型标记
dataset_info.pop("__type__", None)
qa_start_index = 1
except (json.JSONDecodeError, AttributeError):
# 第一行不是数据集元数据,所有行都是QA对
pass
# 如果没有数据集信息,使用文件名作为数据集名称
if "name" not in dataset_info or not dataset_info.get("name"):
dataset_name_from_file = file.filename.replace(".jsonl", "").replace(
".json", ""
)
dataset_info["name"] = dataset_name_from_file
if "description" not in dataset_info:
dataset_info["description"] = f"从文件 {file.filename} 导入"
# 验证数据集信息
if "name" not in dataset_info or not dataset_info.get("name"):
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="数据集名称不能为空(请提供数据集名称,或在文件第一行提供数据集元数据)",
)
# 创建新数据集(总是创建新数据集,即使同名已存在)
final_dataset_name = dataset_info["name"]
dataset_data = {
"name": final_dataset_name,
"description": dataset_info.get("description"),
"version": dataset_info.get("version"),
"status": dataset_info.get("status", "active"),
"category": dataset_info.get("category"),
"tags": dataset_info.get("tags"),
"source": dataset_info.get("source", "imported"),
"source_url": dataset_info.get("source_url"),
"project_id": project_id,
"metadata": {
k: v
for k, v in dataset_info.items()
if k
not in [
"name",
"description",
"version",
"status",
"category",
"tags",
"source",
"source_url",
]
},
}
dataset = Dataset(**dataset_data)
dataset.creator_id = current_user.id
dataset.creator = current_user.username
# 处理标注者信息
if annotator_id is not None:
dataset.annotator_id = annotator_id
# 从数据库获取标注者用户名
annotator_user = UserCRUD.get_by_id(db, user_id=annotator_id)
if annotator_user:
dataset.annotator_name = annotator_user.username
else:
dataset.annotator_name = None
elif "annotator_id" in dataset_info:
dataset.annotator_id = dataset_info.get("annotator_id")
if dataset.annotator_id:
annotator_user = UserCRUD.get_by_id(db, user_id=dataset.annotator_id)
if annotator_user:
dataset.annotator_name = annotator_user.username
else:
dataset.annotator_name = None
else:
dataset.annotator_name = None
created_dataset = DatasetCRUD.create(db=db, dataset=dataset)
dataset_id = created_dataset.id
# 导入QA对
imported_count = 0
failed_count = 0
errors = []
for line_num, line in enumerate(lines[qa_start_index:], start=qa_start_index + 1):
line = line.strip()
if not line:
continue
try:
# 解析JSON对象
data = json.loads(line)
# 验证必需字段
if "question" not in data or "answer" not in data:
failed_count += 1
errors.append(f"第{line_num}行: 缺少必需字段 'question' 或 'answer'")
continue
# 创建QAPair对象
qa_pair_data = {
"dataset_id": dataset_id,
"question": str(data["question"]),
"answer": str(data["answer"]),
}
# 添加额外字段(如果有)
for key, value in data.items():
if key not in ["id", "dataset_id", "question", "answer"]:
qa_pair_data[key] = value
qa_pair = QAPair(**qa_pair_data)
# 保存到数据库
QAPairCRUD.create(db=db, qa_pair=qa_pair)
imported_count += 1
except json.JSONDecodeError as e:
failed_count += 1
errors.append(f"第{line_num}行: JSON解析错误 - {str(e)}")
except Exception as e:
failed_count += 1
errors.append(f"第{line_num}行: 导入失败 - {str(e)}")
return {
"dataset_id": dataset_id,
"dataset_name": final_dataset_name,
"created": True, # 总是创建新数据集
"imported_count": imported_count,
"failed_count": failed_count,
"total_lines": len([line for line in lines[qa_start_index:] if line.strip()]),
"errors": errors[:10], # 只返回前10个错误,避免响应过大
}
def apply_project_inheritance(
db: Session, dataset: Dataset, include_configs: bool = False
) -> Dataset:
"""应用项目继承逻辑到数据集
- display_extra_fields: 如果数据集未设置,则使用项目的display_extra_fields
- 标注配置: 如果数据集没有配置,则继承项目关联的标注配置;如果数据集已有配置,则不继承
"""
if not dataset.project_id:
return dataset
# 获取项目信息
project = ProjectCRUD.get_by_id(db, project_id=dataset.project_id)
if not project:
return dataset
# 继承display_extra_fields
if not dataset.display_extra_fields and project.display_extra_fields:
dataset.display_extra_fields = project.display_extra_fields
# 如果需要包含配置,则处理标注配置继承
if include_configs:
# 获取数据集自己的配置
dataset_configs = DatasetAnnotationConfigCRUD.get_configs_by_dataset(
db=db, dataset_id=dataset.id
)
# 如果数据集没有配置,则继承项目的配置
if not dataset_configs:
project_configs = ProjectAnnotationConfigCRUD.get_configs_by_project(
db=db, project_id=dataset.project_id
)
# 将继承的配置添加到dataset对象(如果Dataset支持annotation_configs字段)
if hasattr(dataset, "annotation_configs"):
dataset.annotation_configs = project_configs
else:
# 数据集已有配置,不继承项目的配置
if hasattr(dataset, "annotation_configs"):
dataset.annotation_configs = dataset_configs
return dataset
def build_task_info(db: Session, dataset: Dataset) -> TaskInfo:
"""构建任务信息
从数据集和关联的项目中提取完整的任务信息,包括:
- 数据集基本信息
- 目标标注数量(计算得出)
- 项目信息(如果数据集属于项目)
- 评估目的和完成时间(从项目 metadata 获取)
- 评估维度(从项目 annotation_configs 获取)
Args:
db: 数据库会话
dataset: 数据集对象
Returns:
完整的任务信息
"""
# 计算目标标注数量
target_count = DatasetCRUD.get_items_count(db, dataset.id)
# 初始化任务信息
task_info = TaskInfo(
dataset_id=dataset.id,
dataset_name=dataset.name,
task_description=dataset.description,
category=dataset.category,
target_annotation_count=target_count,
project_id=dataset.project_id,
project_name=None,
evaluation_purpose=None,
deadline=None,
evaluation_dimensions=[],
)
# 如果数据集属于项目,获取项目信息
if dataset.project_id:
project = ProjectCRUD.get_by_id(db, project_id=dataset.project_id)
if project:
task_info.project_name = project.name
# 如果数据集的描述为空,从项目获取
if not task_info.task_description and project.description:
task_info.task_description = project.description
# 如果数据集的分类为空,从项目获取
if not task_info.category and project.category:
task_info.category = project.category
# 从项目 metadata 中获取评估目的和完成时间
if project.metadata:
task_info.evaluation_purpose = project.metadata.get(
"evaluation_purpose"
)
deadline = project.metadata.get("deadline")
if deadline:
# 确保 deadline 是字符串格式(ISO 8601)
if isinstance(deadline, str):
task_info.deadline = deadline
else:
# 如果是 datetime 对象,转换为字符串
task_info.deadline = deadline.isoformat()
# 获取评估维度(从项目的 annotation_configs)
configs = ProjectAnnotationConfigCRUD.get_configs_by_project(
db, project_id=project.id
)
task_info.evaluation_dimensions = [
EvaluationDimension(name=config.name, description=config.description)
for config in configs
]
# 计算标注进度
annotated_count = 0
progress_rate = 0.0
if target_count > 0:
# 获取标注配置(考虑项目继承)
configs = get_dataset_configs_with_inheritance(
db=db, dataset_id=dataset.id, include_inherited=True
)
if configs:
# 获取所有标注结果
all_results = AnnotationResultCRUD.get_all(
db=db, dataset_id=dataset.id, skip=0, limit=100000
)
# 统计已标注的QA对数量(考虑必填配置)
valid_config_ids = set(c.id for c in configs)
required_configs = [c for c in configs if c.required]
if required_configs:
# 有必填配置:所有必填配置都已标注
required_config_ids = set(c.id for c in required_configs)
items_configs = {}
for result in all_results:
if result.annotation_config_id not in valid_config_ids:
continue
item_id = result.dataset_item_id
if item_id not in items_configs:
items_configs[item_id] = set()
items_configs[item_id].add(result.annotation_config_id)
annotated_count = sum(
1
for item_id, config_ids in items_configs.items()
if required_config_ids.issubset(config_ids)
)
else:
# 没有必填配置:至少一个配置有标注结果
items_with_results = set(
r.dataset_item_id
for r in all_results
if r.annotation_config_id in valid_config_ids
)
annotated_count = len(items_with_results)
progress_rate = (
(annotated_count / target_count * 100) if target_count > 0 else 0.0
)
task_info.annotated_count = annotated_count
task_info.progress_rate = round(progress_rate, 2)
return task_info
@router.post("/", response_model=Dataset, status_code=status.HTTP_201_CREATED)
def create_dataset(
dataset: Dataset,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""创建数据集(需要超级用户权限)"""
# 如果提供了ID,检查是否已存在
if dataset.id is not None:
existing = DatasetCRUD.get_by_id(db, dataset_id=dataset.id)
if existing:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"数据集 ID {dataset.id} 已存在",
)
# 设置创建者信息
if not dataset.creator_id:
dataset.creator_id = current_user.id
if not dataset.creator:
dataset.creator = current_user.username
# 处理标注者信息:如果提供了 annotator_id,自动获取用户名
if dataset.annotator_id is not None:
annotator_user = UserCRUD.get_by_id(db, user_id=dataset.annotator_id)
if annotator_user:
dataset.annotator_name = annotator_user.username
else:
dataset.annotator_name = None
elif dataset.annotator_id is None and dataset.annotator_name:
# 如果只提供了 annotator_name 但没有 annotator_id,尝试通过用户名查找
annotator_user = UserCRUD.get_by_username(db, username=dataset.annotator_name)
if annotator_user:
dataset.annotator_id = annotator_user.id
created = DatasetCRUD.create(db=db, dataset=dataset)
return apply_project_inheritance(db, created)
@router.get("/", response_model=List[Dataset])
def list_datasets(
skip: int = 0,
limit: int = 100,
name_search: Optional[str] = None,
include_configs: bool = False,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""获取数据集列表(需要超级用户权限)"""
datasets = DatasetCRUD.get_all(
db=db, skip=skip, limit=limit, name_search=name_search
)
return [
apply_project_inheritance(db, dataset, include_configs=include_configs)
for dataset in datasets
]
@router.get("/{dataset_id}", response_model=Dataset)
def get_dataset(
dataset_id: int,
include_configs: bool = False,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""根据ID获取数据集(需要超级用户权限)"""
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
return apply_project_inheritance(db, dataset, include_configs=include_configs)
@router.put("/{dataset_id}", response_model=Dataset)
def update_dataset(
dataset_id: int,
dataset: Dataset,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""更新数据集(需要超级用户权限)"""
# 检查数据集是否存在
existing = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not existing:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 确保ID一致
dataset.id = dataset_id
# 处理标注者信息:如果提供了 annotator_id,自动获取用户名
if dataset.annotator_id is not None:
annotator_user = UserCRUD.get_by_id(db, user_id=dataset.annotator_id)
if annotator_user:
dataset.annotator_name = annotator_user.username
else:
dataset.annotator_name = None
elif dataset.annotator_id is None and dataset.annotator_name:
# 如果只提供了 annotator_name 但没有 annotator_id,尝试通过用户名查找
annotator_user = UserCRUD.get_by_username(db, username=dataset.annotator_name)
if annotator_user:
dataset.annotator_id = annotator_user.id
updated = DatasetCRUD.update(db=db, dataset_id=dataset_id, dataset=dataset)
if not updated:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
return apply_project_inheritance(db, updated)
@router.delete("/{dataset_id}", status_code=status.HTTP_204_NO_CONTENT)
def delete_dataset(
dataset_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""删除数据集(需要超级用户权限)"""
success = DatasetCRUD.delete(db=db, dataset_id=dataset_id)
if not success:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
return None
@router.get("/{dataset_id}/items", response_model=List[QAPair])
def list_dataset_items(
dataset_id: int,
skip: int = 0,
limit: int = 100,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""获取数据集的所有QA对(需要超级用户权限)"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
return QAPairCRUD.get_by_dataset(
db=db, dataset_id=dataset_id, skip=skip, limit=limit
)
@router.post(
"/{dataset_id}/items", response_model=QAPair, status_code=status.HTTP_201_CREATED
)
def create_dataset_item(
dataset_id: int,
qa_pair: QAPair,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""创建数据集的QA对(需要超级用户权限)"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 确保dataset_id一致
qa_pair.dataset_id = dataset_id
return QAPairCRUD.create(db=db, qa_pair=qa_pair)
@router.get("/{dataset_id}/items/{item_id}", response_model=QAPair)
def get_dataset_item(
dataset_id: int,
item_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""获取数据集的单个QA对(需要超级用户权限)"""
qa_pair = QAPairCRUD.get_by_id(db, qa_pair_id=item_id)
if not qa_pair:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail=f"QA对 ID {item_id} 不存在"
)
if qa_pair.dataset_id != dataset_id:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"QA对 ID {item_id} 不属于数据集 ID {dataset_id}",
)
return qa_pair
@router.put("/{dataset_id}/items/{item_id}", response_model=QAPair)
def update_dataset_item(
dataset_id: int,
item_id: int,
qa_pair: QAPair,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""更新数据集的QA对(需要超级用户权限)"""
# 检查QA对是否存在
existing = QAPairCRUD.get_by_id(db, qa_pair_id=item_id)
if not existing:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail=f"QA对 ID {item_id} 不存在"
)
if existing.dataset_id != dataset_id:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"QA对 ID {item_id} 不属于数据集 ID {dataset_id}",
)
# 确保ID和dataset_id一致
qa_pair.id = item_id
qa_pair.dataset_id = dataset_id
updated = QAPairCRUD.update(db=db, qa_pair_id=item_id, qa_pair=qa_pair)
if not updated:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail=f"QA对 ID {item_id} 不存在"
)
return updated
@router.delete("/{dataset_id}/items/{item_id}", status_code=status.HTTP_204_NO_CONTENT)
def delete_dataset_item(
dataset_id: int,
item_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""删除数据集的QA对(需要超级用户权限)"""
# 检查QA对是否存在
existing = QAPairCRUD.get_by_id(db, qa_pair_id=item_id)
if not existing:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail=f"QA对 ID {item_id} 不存在"
)
if existing.dataset_id != dataset_id:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"QA对 ID {item_id} 不属于数据集 ID {dataset_id}",
)
success = QAPairCRUD.delete(db=db, qa_pair_id=item_id)
if not success:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail=f"QA对 ID {item_id} 不存在"
)
return None
@router.get("/{dataset_id}/stats")
def get_dataset_stats(
dataset_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""获取数据集的统计信息(需要超级用户权限)"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
items_count = QAPairCRUD.count(db=db, dataset_id=dataset_id)
configs_count = DatasetAnnotationConfigCRUD.count_configs_by_dataset(
db=db, dataset_id=dataset_id
)
return {
"dataset_id": dataset_id,
"items_count": items_count,
"configs_count": configs_count,
}
def get_dataset_configs_with_inheritance(
db: Session, dataset_id: int, include_inherited: bool = True
) -> List:
"""获取数据集关联的所有标注配置(考虑项目继承)
如果include_inherited=True,且数据集没有自己的配置,则返回从项目继承的配置
如果数据集已有配置,则不继承项目的配置
Args:
db: 数据库会话
dataset_id: 数据集ID
include_inherited: 是否包含继承的配置
Returns:
标注配置列表
"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 获取数据集自己的配置
dataset_configs = DatasetAnnotationConfigCRUD.get_configs_by_dataset(
db=db, dataset_id=dataset_id
)
# 如果数据集已有配置,直接返回,不继承项目的配置
if dataset_configs:
return dataset_configs
# 如果数据集没有配置,且需要包含继承的配置,且数据集属于某个项目
if include_inherited and dataset.project_id:
project_configs = ProjectAnnotationConfigCRUD.get_configs_by_project(
db=db, project_id=dataset.project_id
)
return project_configs
return dataset_configs
@router.get("/{dataset_id}/annotation-progress")
def get_dataset_annotation_progress(
dataset_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""获取数据集的标注进展(需要超级用户权限)"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 获取总QA对数量
total_items = QAPairCRUD.count(db=db, dataset_id=dataset_id)
# 获取所有标注配置(考虑项目继承)
configs = get_dataset_configs_with_inheritance(
db=db, dataset_id=dataset_id, include_inherited=True
)
# 获取所有标注结果
all_results = AnnotationResultCRUD.get_all(
db=db, dataset_id=dataset_id, skip=0, limit=100000
)
# 统计每个配置的标注情况
config_progress = []
for config in configs:
# 统计该配置的标注结果数量
config_results = [r for r in all_results if r.annotation_config_id == config.id]
annotated_items = len(set(r.dataset_item_id for r in config_results))
progress_rate = (annotated_items / total_items * 100) if total_items > 0 else 0
config_progress.append(
{
"config_id": config.id,
"config_name": config.name,
"annotated_items": annotated_items,
"total_items": total_items,
"progress_rate": round(progress_rate, 2),
}
)
# 统计总体标注情况
# 如果有必填配置:所有必填配置都已标注的QA对数量
# 如果没有必填配置:至少一个配置有标注结果的QA对数量
# 只统计当前有效配置的标注结果,过滤掉已取消配置的标注结果
valid_config_ids = set(c.id for c in configs)
required_configs = [c for c in configs if c.required]
if required_configs:
# 有必填配置:检查每个QA对是否所有必填配置都有标注结果
required_config_ids = set(c.id for c in required_configs)
# 按QA对分组标注结果,只保留属于当前有效配置的结果
items_configs = {}
for result in all_results:
# 过滤掉已取消配置的标注结果
if result.annotation_config_id not in valid_config_ids:
continue
item_id = result.dataset_item_id
if item_id not in items_configs:
items_configs[item_id] = set()
items_configs[item_id].add(result.annotation_config_id)
# 统计所有必填配置都有标注结果的QA对
annotated_items_count = sum(
1
for item_id, config_ids in items_configs.items()
if required_config_ids.issubset(config_ids)
)
else:
# 没有必填配置:至少一个配置有标注结果(只统计有效配置)
items_with_results = set(
r.dataset_item_id
for r in all_results
if r.annotation_config_id in valid_config_ids
)
annotated_items_count = len(items_with_results)
overall_progress_rate = (
(annotated_items_count / total_items * 100) if total_items > 0 else 0
)
return {
"dataset_id": dataset_id,
"total_items": total_items,
"annotated_items": annotated_items_count,
"overall_progress_rate": round(overall_progress_rate, 2),
"config_progress": config_progress,
}
@router.get("/{dataset_id}/configs", response_model=List[AnnotationConfig])
def get_dataset_configs(
dataset_id: int,
include_inherited: bool = True,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""获取数据集关联的所有标注配置(需要超级用户权限)
如果include_inherited=True,且数据集没有自己的配置,则返回从项目继承的配置
如果数据集已有配置,则不继承项目的配置
"""
# 使用统一的辅助函数获取配置(考虑项目继承)
return get_dataset_configs_with_inheritance(
db=db, dataset_id=dataset_id, include_inherited=include_inherited
)
@router.post("/{dataset_id}/import", status_code=status.HTTP_200_OK)
async def import_dataset_from_jsonl(
dataset_id: int,
file: UploadFile = File(..., description="JSONL文件"),
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""从JSONL文件导入QA对到数据集(需要超级用户权限)"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查文件类型
if not file.filename.endswith(".jsonl"):
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail="只支持.jsonl格式的文件"
)
# 读取文件内容
try:
content = await file.read()
text_content = content.decode("utf-8")
except Exception as e:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST, detail=f"读取文件失败: {str(e)}"
)
# 解析JSONL文件
lines = text_content.strip().split("\n")
imported_count = 0
failed_count = 0
errors = []
for line_num, line in enumerate(lines, start=1):
line = line.strip()
if not line:
continue
try:
# 解析JSON对象
data = json.loads(line)
# 验证必需字段
if "question" not in data or "answer" not in data:
failed_count += 1
errors.append(f"第{line_num}行: 缺少必需字段 'question' 或 'answer'")
continue
# 创建QAPair对象
qa_pair_data = {
"dataset_id": dataset_id,
"question": str(data["question"]),
"answer": str(data["answer"]),
}
# 添加额外字段(如果有)
for key, value in data.items():
if key not in ["id", "dataset_id", "question", "answer"]:
qa_pair_data[key] = value
qa_pair = QAPair(**qa_pair_data)
# 保存到数据库
QAPairCRUD.create(db=db, qa_pair=qa_pair)
imported_count += 1
except json.JSONDecodeError as e:
failed_count += 1
errors.append(f"第{line_num}行: JSON解析错误 - {str(e)}")
except Exception as e:
failed_count += 1
errors.append(f"第{line_num}行: 导入失败 - {str(e)}")
return {
"dataset_id": dataset_id,
"imported_count": imported_count,
"failed_count": failed_count,
"total_lines": len([line for line in lines if line.strip()]),
"errors": errors[:10], # 只返回前10个错误,避免响应过大
}
@router.post("/import", status_code=status.HTTP_201_CREATED)
async def import_dataset(
file: UploadFile = File(..., description="JSONL文件"),
name: Optional[str] = Form(
None, description="数据集名称(如果提供,将覆盖文件中的元数据)"
),
description: Optional[str] = Form(None, description="数据集描述"),
version: Optional[str] = Form(None, description="数据集版本"),
category: Optional[str] = Form(None, description="数据集分类"),
status: Optional[str] = Form(None, description="数据集状态"),
tags: Optional[str] = Form(None, description="数据集标签(逗号分隔)"),
source: Optional[str] = Form(None, description="数据来源"),
source_url: Optional[str] = Form(None, description="数据来源URL"),
annotator_id: Optional[int] = Form(None, description="标注者ID"),
db: Session = Depends(get_db),
current_user: User = Depends(get_current_superuser),
):
"""从JSONL文件导入完整数据集(包括数据集信息和QA对,需要超级用户权限)
JSONL文件格式:
1. 第一行(可选):数据集元数据,格式为 {"__type__": "dataset", "name": "...", "description": "...", ...}
2. 后续行:QA对,每行一个JSON对象,必须包含question和answer字段
如果提供了表单中的元数据字段,将优先使用表单中的元数据,而不是文件中的元数据。
如果第一行不是数据集元数据且未提供表单元数据,则使用文件名作为数据集名称。
"""
return await import_dataset_from_file(
file=file,
db=db,
current_user=current_user,
dataset_name=name,
dataset_description=description,
dataset_version=version,
dataset_category=category,
dataset_status=status,
dataset_tags=tags,
dataset_source=source,
dataset_source_url=source_url,
annotator_id=annotator_id,
)
# ==================== 普通用户标注接口 ====================
def check_dataset_access_permission(dataset: Dataset, current_user: User) -> None:
"""检查用户是否有权限访问数据集
规则:
- 管理员(is_superuser=True)可以访问所有数据集
- 普通用户只能访问分配给自己的数据集(annotator_id == current_user.id)
如果没有权限,抛出 HTTPException
"""
# 管理员可以访问所有数据集
if current_user.is_superuser:
return
# 普通用户只能访问分配给自己的数据集
if dataset.annotator_id is None:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail="该数据集尚未分配给任何用户,无权访问",
)
if dataset.annotator_id != current_user.id:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail="无权访问此数据集,该数据集已分配给其他用户",
)
@router.get("/annotation/{dataset_id}/info", response_model=Dataset)
def get_dataset_info_for_annotation(
dataset_id: int,
include_configs: bool = True, # 标注时默认包含配置
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""获取数据集信息(普通用户可访问,用于标注)"""
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查权限
check_dataset_access_permission(dataset, current_user)
return apply_project_inheritance(db, dataset, include_configs=include_configs)
@router.get("/annotation/{dataset_id}/items", response_model=List[QAPair])
def list_dataset_items_for_annotation(
dataset_id: int,
skip: int = 0,
limit: int = 100,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""获取数据集的QA对列表(普通用户可访问,用于标注)"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查权限
check_dataset_access_permission(dataset, current_user)
return QAPairCRUD.get_by_dataset(
db=db, dataset_id=dataset_id, skip=skip, limit=limit
)
@router.get("/annotation/{dataset_id}/items/{item_id}", response_model=QAPair)
def get_dataset_item_for_annotation(
dataset_id: int,
item_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""获取单个QA对(普通用户可访问,用于标注)"""
qa_pair = QAPairCRUD.get_by_id(db, qa_pair_id=item_id)
if not qa_pair:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND, detail=f"QA对 ID {item_id} 不存在"
)
if qa_pair.dataset_id != dataset_id:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"QA对 ID {item_id} 不属于数据集 ID {dataset_id}",
)
# 检查数据集权限
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查权限
check_dataset_access_permission(dataset, current_user)
return qa_pair
@router.get("/annotation/{dataset_id}/configs", response_model=List[AnnotationConfig])
def get_dataset_configs_for_annotation(
dataset_id: int,
include_inherited: bool = True,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""获取数据集关联的所有标注配置(普通用户可访问,用于标注)
如果include_inherited=True,且数据集没有自己的配置,则返回从项目继承的配置
如果数据集已有配置,则不继承项目的配置
"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查权限
check_dataset_access_permission(dataset, current_user)
# 使用统一的辅助函数获取配置(考虑项目继承)
return get_dataset_configs_with_inheritance(
db=db, dataset_id=dataset_id, include_inherited=include_inherited
)
@router.get("/annotation/{dataset_id}/stats")
def get_dataset_stats_for_annotation(
dataset_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""获取数据集的统计信息(普通用户可访问,用于标注)"""
# 检查数据集是否存在
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查权限
check_dataset_access_permission(dataset, current_user)
items_count = QAPairCRUD.count(db=db, dataset_id=dataset_id)
configs_count = DatasetAnnotationConfigCRUD.count_configs_by_dataset(
db=db, dataset_id=dataset_id
)
return {
"dataset_id": dataset_id,
"items_count": items_count,
"configs_count": configs_count,
}
# ==================== 任务相关 API ====================
@router.get("/tasks/available", response_model=List[TaskInfo])
def get_available_tasks(
skip: int = 0,
limit: int = 100,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""获取当前用户可领取的任务列表
规则:
1. 只显示 annotator_id 为空的数据集
2. 如果用户有 species:
- 显示 category 匹配用户 species 的数据集
- 或没有 category 的数据集
3. 如果用户没有 species:
- 只显示没有 category 的数据集
- 排除有 category 的数据集,因为用户无法领取它们
Returns:
可领取的任务列表,包含完整的项目信息
"""
# 获取可领取的数据集
datasets = DatasetCRUD.get_available_datasets(
db=db, skip=skip, limit=limit, user_species=current_user.species
)
# 为每个数据集构建完整的任务信息
tasks = [build_task_info(db, dataset) for dataset in datasets]
return tasks
@router.post("/tasks/{dataset_id}/claim", response_model=Dataset)
def claim_task(
dataset_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""领取任务(将数据集分配给当前用户)
规则:
1. 数据集必须没有指定标注者(annotator_id 为空)
2. 如果数据集有 category,必须匹配用户的 species
3. 如果数据集没有 category,所有用户都可以领取
Returns:
更新后的数据集对象
"""
# 获取数据集
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查分类匹配(在原子更新前进行业务逻辑检查)
if dataset.category:
if not current_user.species:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail="该任务需要匹配的物种标签,您的账户未设置物种标签",
)
if dataset.category != current_user.species:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail=f"该任务需要物种标签 '{dataset.category}',您的物种标签是 '{current_user.species}'",
)
# 使用原子更新领取任务(并发安全)
# 只有当 annotator_id 为 None 时才会更新,确保不会覆盖其他用户的领取
updated = DatasetCRUD.claim_dataset(
db=db,
dataset_id=dataset_id,
annotator_id=current_user.id,
annotator_name=current_user.username,
)
if not updated:
# 原子更新失败,可能是已被其他用户领取或数据集不存在
# 重新查询以获取最新状态
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
if dataset.annotator_id is not None:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="该任务已被其他用户领取",
)
# 如果仍然为 None,可能是其他原因导致的更新失败
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail="领取任务失败",
)
return updated
@router.get("/tasks/my", response_model=List[TaskInfo])
def get_my_tasks(
skip: int = 0,
limit: int = 100,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""获取分配给当前用户的任务列表
Returns:
当前用户已领取的任务列表,包含完整的项目信息
"""
# 查询分配给当前用户的数据集
datasets = DatasetCRUD.get_by_annotator(
db=db, annotator_id=current_user.id, skip=skip, limit=limit
)
# 为每个数据集构建完整的任务信息
tasks = [build_task_info(db, dataset) for dataset in datasets]
return tasks
@router.post("/tasks/{dataset_id}/release", response_model=Dataset)
def release_task(
dataset_id: int,
db: Session = Depends(get_db),
current_user: User = Depends(get_current_active_user),
):
"""退回任务(将数据集从当前用户释放,使其重新变为可用)
规则:
1. 数据集必须属于当前用户(annotator_id == current_user.id)
2. 退回后,数据集将重新出现在可用任务列表中
Returns:
更新后的数据集对象
"""
# 获取数据集
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
# 检查任务是否属于当前用户
if dataset.annotator_id != current_user.id:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail="该任务不属于您,无法退回",
)
# 使用原子更新退回任务(并发安全)
# 只有当 annotator_id 匹配当前用户时才会更新
updated = DatasetCRUD.release_dataset(
db=db,
dataset_id=dataset_id,
annotator_id=current_user.id,
)
if not updated:
# 原子更新失败,可能是已被其他用户领取或数据集不存在
# 重新查询以获取最新状态
dataset = DatasetCRUD.get_by_id(db, dataset_id=dataset_id)
if not dataset:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"数据集 ID {dataset_id} 不存在",
)
if dataset.annotator_id != current_user.id:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail="该任务不属于您,无法退回",
)
# 如果仍然失败,可能是其他原因导致的更新失败
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail="退回任务失败",
)
return updated
|