選択できるのは25トピックまでです。 トピックは、先頭が英数字で、英数字とダッシュ('-')を使用した35文字以内のものにしてください。

dataset.py 4.1KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596
  1. #
  2. # Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
  3. #
  4. # Licensed under the Apache License, Version 2.0 (the "License");
  5. # you may not use this file except in compliance with the License.
  6. # You may obtain a copy of the License at
  7. #
  8. # http://www.apache.org/licenses/LICENSE-2.0
  9. #
  10. # Unless required by applicable law or agreed to in writing, software
  11. # distributed under the License is distributed on an "AS IS" BASIS,
  12. # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13. # See the License for the specific language governing permissions and
  14. # limitations under the License.
  15. #
  16. from flask import request
  17. from api.db import StatusEnum
  18. from api.db.db_models import APIToken
  19. from api.db.services.knowledgebase_service import KnowledgebaseService
  20. from api.db.services.user_service import TenantService
  21. from api.settings import RetCode
  22. from api.utils import get_uuid
  23. from api.utils.api_utils import get_data_error_result
  24. from api.utils.api_utils import get_json_result
  25. @manager.route('/save', methods=['POST'])
  26. def save():
  27. req = request.json
  28. token = request.headers.get('Authorization').split()[1]
  29. objs = APIToken.query(token=token)
  30. if not objs:
  31. return get_json_result(
  32. data=False, retmsg='Token is not valid!"', retcode=RetCode.AUTHENTICATION_ERROR)
  33. tenant_id = objs[0].tenant_id
  34. e, t = TenantService.get_by_id(tenant_id)
  35. if not e:
  36. return get_data_error_result(retmsg="Tenant not found.")
  37. if "id" not in req:
  38. req['id'] = get_uuid()
  39. req["name"] = req["name"].strip()
  40. if req["name"] == "":
  41. return get_data_error_result(
  42. retmsg="Name is not empty")
  43. if KnowledgebaseService.query(name=req["name"]):
  44. return get_data_error_result(
  45. retmsg="Duplicated knowledgebase name")
  46. req["tenant_id"] = tenant_id
  47. req['created_by'] = tenant_id
  48. req['embd_id'] = t.embd_id
  49. if not KnowledgebaseService.save(**req):
  50. return get_data_error_result(retmsg="Data saving error")
  51. req.pop('created_by')
  52. keys_to_rename = {'embd_id': "embedding_model", 'parser_id': 'parser_method',
  53. 'chunk_num': 'chunk_count', 'doc_num': 'document_count'}
  54. for old_key,new_key in keys_to_rename.items():
  55. if old_key in req:
  56. req[new_key]=req.pop(old_key)
  57. return get_json_result(data=req)
  58. else:
  59. if req["tenant_id"] != tenant_id or req["embd_id"] != t.embd_id:
  60. return get_data_error_result(
  61. retmsg="Can't change tenant_id or embedding_model")
  62. e, kb = KnowledgebaseService.get_by_id(req["id"])
  63. if not e:
  64. return get_data_error_result(
  65. retmsg="Can't find this knowledgebase!")
  66. if not KnowledgebaseService.query(
  67. created_by=tenant_id, id=req["id"]):
  68. return get_json_result(
  69. data=False, retmsg=f'Only owner of knowledgebase authorized for this operation.',
  70. retcode=RetCode.OPERATING_ERROR)
  71. if req["chunk_num"] != kb.chunk_num or req['doc_num'] != kb.doc_num:
  72. return get_data_error_result(
  73. retmsg="Can't change document_count or chunk_count ")
  74. if kb.chunk_num > 0 and req['parser_id'] != kb.parser_id:
  75. return get_data_error_result(
  76. retmsg="if chunk count is not 0, parser method is not changable. ")
  77. if req["name"].lower() != kb.name.lower() \
  78. and len(KnowledgebaseService.query(name=req["name"], tenant_id=req['tenant_id'],
  79. status=StatusEnum.VALID.value)) > 0:
  80. return get_data_error_result(
  81. retmsg="Duplicated knowledgebase name.")
  82. del req["id"]
  83. req['created_by'] = tenant_id
  84. if not KnowledgebaseService.update_by_id(kb.id, req):
  85. return get_data_error_result(retmsg="Data update error ")
  86. return get_json_result(data=True)