File size: 18,999 Bytes
4304c6d |
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 |
import uuid
from datetime import datetime, timezone
import pandas as pd
from flask import request
from flask_login import current_user
from flask_restful import Resource, marshal, reqparse
from werkzeug.exceptions import Forbidden, NotFound
import services
from controllers.console import api
from controllers.console.app.error import ProviderNotInitializeError
from controllers.console.datasets.error import InvalidActionError, NoFileUploadedError, TooManyFilesError
from controllers.console.setup import setup_required
from controllers.console.wraps import (
account_initialization_required,
cloud_edition_billing_knowledge_limit_check,
cloud_edition_billing_resource_check,
)
from core.errors.error import LLMBadRequestError, ProviderTokenNotInitError
from core.model_manager import ModelManager
from core.model_runtime.entities.model_entities import ModelType
from extensions.ext_database import db
from extensions.ext_redis import redis_client
from fields.segment_fields import segment_fields
from libs.login import login_required
from models.dataset import DocumentSegment
from services.dataset_service import DatasetService, DocumentService, SegmentService
from tasks.batch_create_segment_to_index_task import batch_create_segment_to_index_task
from tasks.disable_segment_from_index_task import disable_segment_from_index_task
from tasks.enable_segment_to_index_task import enable_segment_to_index_task
class DatasetDocumentSegmentListApi(Resource):
@setup_required
@login_required
@account_initialization_required
def get(self, dataset_id, document_id):
dataset_id = str(dataset_id)
document_id = str(document_id)
dataset = DatasetService.get_dataset(dataset_id)
if not dataset:
raise NotFound('Dataset not found.')
try:
DatasetService.check_dataset_permission(dataset, current_user)
except services.errors.account.NoPermissionError as e:
raise Forbidden(str(e))
document = DocumentService.get_document(dataset_id, document_id)
if not document:
raise NotFound('Document not found.')
parser = reqparse.RequestParser()
parser.add_argument('last_id', type=str, default=None, location='args')
parser.add_argument('limit', type=int, default=20, location='args')
parser.add_argument('status', type=str,
action='append', default=[], location='args')
parser.add_argument('hit_count_gte', type=int,
default=None, location='args')
parser.add_argument('enabled', type=str, default='all', location='args')
parser.add_argument('keyword', type=str, default=None, location='args')
args = parser.parse_args()
last_id = args['last_id']
limit = min(args['limit'], 100)
status_list = args['status']
hit_count_gte = args['hit_count_gte']
keyword = args['keyword']
query = DocumentSegment.query.filter(
DocumentSegment.document_id == str(document_id),
DocumentSegment.tenant_id == current_user.current_tenant_id
)
if last_id is not None:
last_segment = DocumentSegment.query.get(str(last_id))
if last_segment:
query = query.filter(
DocumentSegment.position > last_segment.position)
else:
return {'data': [], 'has_more': False, 'limit': limit}, 200
if status_list:
query = query.filter(DocumentSegment.status.in_(status_list))
if hit_count_gte is not None:
query = query.filter(DocumentSegment.hit_count >= hit_count_gte)
if keyword:
query = query.where(DocumentSegment.content.ilike(f'%{keyword}%'))
if args['enabled'].lower() != 'all':
if args['enabled'].lower() == 'true':
query = query.filter(DocumentSegment.enabled == True)
elif args['enabled'].lower() == 'false':
query = query.filter(DocumentSegment.enabled == False)
total = query.count()
segments = query.order_by(DocumentSegment.position).limit(limit + 1).all()
has_more = False
if len(segments) > limit:
has_more = True
segments = segments[:-1]
return {
'data': marshal(segments, segment_fields),
'doc_form': document.doc_form,
'has_more': has_more,
'limit': limit,
'total': total
}, 200
class DatasetDocumentSegmentApi(Resource):
@setup_required
@login_required
@account_initialization_required
@cloud_edition_billing_resource_check('vector_space')
def patch(self, dataset_id, segment_id, action):
dataset_id = str(dataset_id)
dataset = DatasetService.get_dataset(dataset_id)
if not dataset:
raise NotFound('Dataset not found.')
# check user's model setting
DatasetService.check_dataset_model_setting(dataset)
# The role of the current user in the ta table must be admin or owner
if not current_user.is_admin_or_owner:
raise Forbidden()
try:
DatasetService.check_dataset_permission(dataset, current_user)
except services.errors.account.NoPermissionError as e:
raise Forbidden(str(e))
if dataset.indexing_technique == 'high_quality':
# check embedding model setting
try:
model_manager = ModelManager()
model_manager.get_model_instance(
tenant_id=current_user.current_tenant_id,
provider=dataset.embedding_model_provider,
model_type=ModelType.TEXT_EMBEDDING,
model=dataset.embedding_model
)
except LLMBadRequestError:
raise ProviderNotInitializeError(
"No Embedding Model available. Please configure a valid provider "
"in the Settings -> Model Provider.")
except ProviderTokenNotInitError as ex:
raise ProviderNotInitializeError(ex.description)
segment = DocumentSegment.query.filter(
DocumentSegment.id == str(segment_id),
DocumentSegment.tenant_id == current_user.current_tenant_id
).first()
if not segment:
raise NotFound('Segment not found.')
if segment.status != 'completed':
raise NotFound('Segment is not completed, enable or disable function is not allowed')
document_indexing_cache_key = 'document_{}_indexing'.format(segment.document_id)
cache_result = redis_client.get(document_indexing_cache_key)
if cache_result is not None:
raise InvalidActionError("Document is being indexed, please try again later")
indexing_cache_key = 'segment_{}_indexing'.format(segment.id)
cache_result = redis_client.get(indexing_cache_key)
if cache_result is not None:
raise InvalidActionError("Segment is being indexed, please try again later")
if action == "enable":
if segment.enabled:
raise InvalidActionError("Segment is already enabled.")
segment.enabled = True
segment.disabled_at = None
segment.disabled_by = None
db.session.commit()
# Set cache to prevent indexing the same segment multiple times
redis_client.setex(indexing_cache_key, 600, 1)
enable_segment_to_index_task.delay(segment.id)
return {'result': 'success'}, 200
elif action == "disable":
if not segment.enabled:
raise InvalidActionError("Segment is already disabled.")
segment.enabled = False
segment.disabled_at = datetime.now(timezone.utc).replace(tzinfo=None)
segment.disabled_by = current_user.id
db.session.commit()
# Set cache to prevent indexing the same segment multiple times
redis_client.setex(indexing_cache_key, 600, 1)
disable_segment_from_index_task.delay(segment.id)
return {'result': 'success'}, 200
else:
raise InvalidActionError()
class DatasetDocumentSegmentAddApi(Resource):
@setup_required
@login_required
@account_initialization_required
@cloud_edition_billing_resource_check('vector_space')
@cloud_edition_billing_knowledge_limit_check('add_segment')
def post(self, dataset_id, document_id):
# check dataset
dataset_id = str(dataset_id)
dataset = DatasetService.get_dataset(dataset_id)
if not dataset:
raise NotFound('Dataset not found.')
# check document
document_id = str(document_id)
document = DocumentService.get_document(dataset_id, document_id)
if not document:
raise NotFound('Document not found.')
# The role of the current user in the ta table must be admin or owner
if not current_user.is_admin_or_owner:
raise Forbidden()
# check embedding model setting
if dataset.indexing_technique == 'high_quality':
try:
model_manager = ModelManager()
model_manager.get_model_instance(
tenant_id=current_user.current_tenant_id,
provider=dataset.embedding_model_provider,
model_type=ModelType.TEXT_EMBEDDING,
model=dataset.embedding_model
)
except LLMBadRequestError:
raise ProviderNotInitializeError(
"No Embedding Model available. Please configure a valid provider "
"in the Settings -> Model Provider.")
except ProviderTokenNotInitError as ex:
raise ProviderNotInitializeError(ex.description)
try:
DatasetService.check_dataset_permission(dataset, current_user)
except services.errors.account.NoPermissionError as e:
raise Forbidden(str(e))
# validate args
parser = reqparse.RequestParser()
parser.add_argument('content', type=str, required=True, nullable=False, location='json')
parser.add_argument('answer', type=str, required=False, nullable=True, location='json')
parser.add_argument('keywords', type=list, required=False, nullable=True, location='json')
args = parser.parse_args()
SegmentService.segment_create_args_validate(args, document)
segment = SegmentService.create_segment(args, document, dataset)
return {
'data': marshal(segment, segment_fields),
'doc_form': document.doc_form
}, 200
class DatasetDocumentSegmentUpdateApi(Resource):
@setup_required
@login_required
@account_initialization_required
@cloud_edition_billing_resource_check('vector_space')
def patch(self, dataset_id, document_id, segment_id):
# check dataset
dataset_id = str(dataset_id)
dataset = DatasetService.get_dataset(dataset_id)
if not dataset:
raise NotFound('Dataset not found.')
# check user's model setting
DatasetService.check_dataset_model_setting(dataset)
# check document
document_id = str(document_id)
document = DocumentService.get_document(dataset_id, document_id)
if not document:
raise NotFound('Document not found.')
if dataset.indexing_technique == 'high_quality':
# check embedding model setting
try:
model_manager = ModelManager()
model_manager.get_model_instance(
tenant_id=current_user.current_tenant_id,
provider=dataset.embedding_model_provider,
model_type=ModelType.TEXT_EMBEDDING,
model=dataset.embedding_model
)
except LLMBadRequestError:
raise ProviderNotInitializeError(
"No Embedding Model available. Please configure a valid provider "
"in the Settings -> Model Provider.")
except ProviderTokenNotInitError as ex:
raise ProviderNotInitializeError(ex.description)
# check segment
segment_id = str(segment_id)
segment = DocumentSegment.query.filter(
DocumentSegment.id == str(segment_id),
DocumentSegment.tenant_id == current_user.current_tenant_id
).first()
if not segment:
raise NotFound('Segment not found.')
# The role of the current user in the ta table must be admin or owner
if not current_user.is_admin_or_owner:
raise Forbidden()
try:
DatasetService.check_dataset_permission(dataset, current_user)
except services.errors.account.NoPermissionError as e:
raise Forbidden(str(e))
# validate args
parser = reqparse.RequestParser()
parser.add_argument('content', type=str, required=True, nullable=False, location='json')
parser.add_argument('answer', type=str, required=False, nullable=True, location='json')
parser.add_argument('keywords', type=list, required=False, nullable=True, location='json')
args = parser.parse_args()
SegmentService.segment_create_args_validate(args, document)
segment = SegmentService.update_segment(args, segment, document, dataset)
return {
'data': marshal(segment, segment_fields),
'doc_form': document.doc_form
}, 200
@setup_required
@login_required
@account_initialization_required
def delete(self, dataset_id, document_id, segment_id):
# check dataset
dataset_id = str(dataset_id)
dataset = DatasetService.get_dataset(dataset_id)
if not dataset:
raise NotFound('Dataset not found.')
# check user's model setting
DatasetService.check_dataset_model_setting(dataset)
# check document
document_id = str(document_id)
document = DocumentService.get_document(dataset_id, document_id)
if not document:
raise NotFound('Document not found.')
# check segment
segment_id = str(segment_id)
segment = DocumentSegment.query.filter(
DocumentSegment.id == str(segment_id),
DocumentSegment.tenant_id == current_user.current_tenant_id
).first()
if not segment:
raise NotFound('Segment not found.')
# The role of the current user in the ta table must be admin or owner
if not current_user.is_admin_or_owner:
raise Forbidden()
try:
DatasetService.check_dataset_permission(dataset, current_user)
except services.errors.account.NoPermissionError as e:
raise Forbidden(str(e))
SegmentService.delete_segment(segment, document, dataset)
return {'result': 'success'}, 200
class DatasetDocumentSegmentBatchImportApi(Resource):
@setup_required
@login_required
@account_initialization_required
@cloud_edition_billing_resource_check('vector_space')
@cloud_edition_billing_knowledge_limit_check('add_segment')
def post(self, dataset_id, document_id):
# check dataset
dataset_id = str(dataset_id)
dataset = DatasetService.get_dataset(dataset_id)
if not dataset:
raise NotFound('Dataset not found.')
# check document
document_id = str(document_id)
document = DocumentService.get_document(dataset_id, document_id)
if not document:
raise NotFound('Document not found.')
# get file from request
file = request.files['file']
# check file
if 'file' not in request.files:
raise NoFileUploadedError()
if len(request.files) > 1:
raise TooManyFilesError()
# check file type
if not file.filename.endswith('.csv'):
raise ValueError("Invalid file type. Only CSV files are allowed")
try:
# Skip the first row
df = pd.read_csv(file)
result = []
for index, row in df.iterrows():
if document.doc_form == 'qa_model':
data = {'content': row[0], 'answer': row[1]}
else:
data = {'content': row[0]}
result.append(data)
if len(result) == 0:
raise ValueError("The CSV file is empty.")
# async job
job_id = str(uuid.uuid4())
indexing_cache_key = 'segment_batch_import_{}'.format(str(job_id))
# send batch add segments task
redis_client.setnx(indexing_cache_key, 'waiting')
batch_create_segment_to_index_task.delay(str(job_id), result, dataset_id, document_id,
current_user.current_tenant_id, current_user.id)
except Exception as e:
return {'error': str(e)}, 500
return {
'job_id': job_id,
'job_status': 'waiting'
}, 200
@setup_required
@login_required
@account_initialization_required
def get(self, job_id):
job_id = str(job_id)
indexing_cache_key = 'segment_batch_import_{}'.format(job_id)
cache_result = redis_client.get(indexing_cache_key)
if cache_result is None:
raise ValueError("The job is not exist.")
return {
'job_id': job_id,
'job_status': cache_result.decode()
}, 200
api.add_resource(DatasetDocumentSegmentListApi,
'/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/segments')
api.add_resource(DatasetDocumentSegmentApi,
'/datasets/<uuid:dataset_id>/segments/<uuid:segment_id>/<string:action>')
api.add_resource(DatasetDocumentSegmentAddApi,
'/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/segment')
api.add_resource(DatasetDocumentSegmentUpdateApi,
'/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/segments/<uuid:segment_id>')
api.add_resource(DatasetDocumentSegmentBatchImportApi,
'/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/segments/batch_import',
'/datasets/batch_import_status/<uuid:job_id>')
|