File size: 10,187 Bytes
b9fe2b4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
#
#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
#
#  Licensed under the Apache License, Version 2.0 (the "License");
#  you may not use this file except in compliance with the License.
#  You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
#  Unless required by applicable law or agreed to in writing, software
#  distributed under the License is distributed on an "AS IS" BASIS,
#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#  See the License for the specific language governing permissions and
#  limitations under the License.
#
from api.db import StatusEnum, TenantPermission
from api.db.db_models import Knowledgebase, DB, Tenant, User, UserTenant,Document
from api.db.services.common_service import CommonService
from peewee import fn


class KnowledgebaseService(CommonService):
    model = Knowledgebase

    @classmethod
    @DB.connection_context()
    def is_parsed_done(cls, kb_id):
        """
        Check if all documents in the knowledge base have completed parsing
        
        Args:
            kb_id: Knowledge base ID
            
        Returns:
            If all documents are parsed successfully, returns (True, None)
            If any document is not fully parsed, returns (False, error_message)
        """
        from api.db import TaskStatus
        from api.db.services.document_service import DocumentService
        
        # Get knowledge base information
        kbs = cls.query(id=kb_id)
        if not kbs:
            return False, "Knowledge base not found"
        kb = kbs[0]
        
        # Get all documents in the knowledge base
        docs, _ = DocumentService.get_by_kb_id(kb_id, 1, 1000, "create_time", True, "")
        
        # Check parsing status of each document
        for doc in docs:
            # If document is being parsed, don't allow chat creation
            if doc['run'] == TaskStatus.RUNNING.value or doc['run'] == TaskStatus.CANCEL.value or doc['run'] == TaskStatus.FAIL.value:
                return False, f"Document '{doc['name']}' in dataset '{kb.name}' is still being parsed. Please wait until all documents are parsed before starting a chat."
            # If document is not yet parsed and has no chunks, don't allow chat creation
            if doc['run'] == TaskStatus.UNSTART.value and doc['chunk_num'] == 0:
                return False, f"Document '{doc['name']}' in dataset '{kb.name}' has not been parsed yet. Please parse all documents before starting a chat."
        
        return True, None

    @classmethod
    @DB.connection_context()
    def list_documents_by_ids(cls,kb_ids):
        doc_ids=cls.model.select(Document.id.alias("document_id")).join(Document,on=(cls.model.id == Document.kb_id)).where(
            cls.model.id.in_(kb_ids)
        )
        doc_ids =list(doc_ids.dicts())
        doc_ids = [doc["document_id"] for doc in doc_ids]
        return doc_ids

    @classmethod
    @DB.connection_context()
    def get_by_tenant_ids(cls, joined_tenant_ids, user_id,
                          page_number, items_per_page,
                          orderby, desc, keywords,
                          parser_id=None
                          ):
        fields = [
            cls.model.id,
            cls.model.avatar,
            cls.model.name,
            cls.model.language,
            cls.model.description,
            cls.model.permission,
            cls.model.doc_num,
            cls.model.token_num,
            cls.model.chunk_num,
            cls.model.parser_id,
            cls.model.embd_id,
            User.nickname,
            User.avatar.alias('tenant_avatar'),
            cls.model.update_time
        ]
        if keywords:
            kbs = cls.model.select(*fields).join(User, on=(cls.model.tenant_id == User.id)).where(
                ((cls.model.tenant_id.in_(joined_tenant_ids) & (cls.model.permission ==
                                                                TenantPermission.TEAM.value)) | (
                    cls.model.tenant_id == user_id))
                & (cls.model.status == StatusEnum.VALID.value),
                (fn.LOWER(cls.model.name).contains(keywords.lower()))
            )
        else:
            kbs = cls.model.select(*fields).join(User, on=(cls.model.tenant_id == User.id)).where(
                ((cls.model.tenant_id.in_(joined_tenant_ids) & (cls.model.permission ==
                                                                TenantPermission.TEAM.value)) | (
                    cls.model.tenant_id == user_id))
                & (cls.model.status == StatusEnum.VALID.value)
            )
        if parser_id:
            kbs = kbs.where(cls.model.parser_id == parser_id)
        if desc:
            kbs = kbs.order_by(cls.model.getter_by(orderby).desc())
        else:
            kbs = kbs.order_by(cls.model.getter_by(orderby).asc())

        count = kbs.count()

        kbs = kbs.paginate(page_number, items_per_page)

        return list(kbs.dicts()), count

    @classmethod
    @DB.connection_context()
    def get_kb_ids(cls, tenant_id):
        fields = [
            cls.model.id,
        ]
        kbs = cls.model.select(*fields).where(cls.model.tenant_id == tenant_id)
        kb_ids = [kb.id for kb in kbs]
        return kb_ids

    @classmethod
    @DB.connection_context()
    def get_detail(cls, kb_id):
        fields = [
            cls.model.id,
            # Tenant.embd_id,
            cls.model.embd_id,
            cls.model.avatar,
            cls.model.name,
            cls.model.language,
            cls.model.description,
            cls.model.permission,
            cls.model.doc_num,
            cls.model.token_num,
            cls.model.chunk_num,
            cls.model.parser_id,
            cls.model.parser_config,
            cls.model.pagerank]
        kbs = cls.model.select(*fields).join(Tenant, on=(
                (Tenant.id == cls.model.tenant_id) & (Tenant.status == StatusEnum.VALID.value))).where(
            (cls.model.id == kb_id),
            (cls.model.status == StatusEnum.VALID.value)
        )
        if not kbs:
            return
        d = kbs[0].to_dict()
        # d["embd_id"] = kbs[0].tenant.embd_id
        return d

    @classmethod
    @DB.connection_context()
    def update_parser_config(cls, id, config):
        e, m = cls.get_by_id(id)
        if not e:
            raise LookupError(f"knowledgebase({id}) not found.")

        def dfs_update(old, new):
            for k, v in new.items():
                if k not in old:
                    old[k] = v
                    continue
                if isinstance(v, dict):
                    assert isinstance(old[k], dict)
                    dfs_update(old[k], v)
                elif isinstance(v, list):
                    assert isinstance(old[k], list)
                    old[k] = list(set(old[k] + v))
                else:
                    old[k] = v

        dfs_update(m.parser_config, config)
        cls.update_by_id(id, {"parser_config": m.parser_config})

    @classmethod
    @DB.connection_context()
    def get_field_map(cls, ids):
        conf = {}
        for k in cls.get_by_ids(ids):
            if k.parser_config and "field_map" in k.parser_config:
                conf.update(k.parser_config["field_map"])
        return conf

    @classmethod
    @DB.connection_context()
    def get_by_name(cls, kb_name, tenant_id):
        kb = cls.model.select().where(
            (cls.model.name == kb_name)
            & (cls.model.tenant_id == tenant_id)
            & (cls.model.status == StatusEnum.VALID.value)
        )
        if kb:
            return True, kb[0]
        return False, None

    @classmethod
    @DB.connection_context()
    def get_all_ids(cls):
        return [m["id"] for m in cls.model.select(cls.model.id).dicts()]

    @classmethod
    @DB.connection_context()
    def get_list(cls, joined_tenant_ids, user_id,
                 page_number, items_per_page, orderby, desc, id, name):
        kbs = cls.model.select()
        if id:
            kbs = kbs.where(cls.model.id == id)
        if name:
            kbs = kbs.where(cls.model.name == name)
        kbs = kbs.where(
            ((cls.model.tenant_id.in_(joined_tenant_ids) & (cls.model.permission ==
                                                            TenantPermission.TEAM.value)) | (
                     cls.model.tenant_id == user_id))
            & (cls.model.status == StatusEnum.VALID.value)
        )
        if desc:
            kbs = kbs.order_by(cls.model.getter_by(orderby).desc())
        else:
            kbs = kbs.order_by(cls.model.getter_by(orderby).asc())

        kbs = kbs.paginate(page_number, items_per_page)

        return list(kbs.dicts())

    @classmethod
    @DB.connection_context()
    def accessible(cls, kb_id, user_id):
        docs = cls.model.select(
            cls.model.id).join(UserTenant, on=(UserTenant.tenant_id == Knowledgebase.tenant_id)
            ).where(cls.model.id == kb_id, UserTenant.user_id == user_id).paginate(0, 1)
        docs = docs.dicts()
        if not docs:
            return False
        return True

    @classmethod
    @DB.connection_context()
    def get_kb_by_id(cls, kb_id, user_id):
        kbs = cls.model.select().join(UserTenant, on=(UserTenant.tenant_id == Knowledgebase.tenant_id)
            ).where(cls.model.id == kb_id, UserTenant.user_id == user_id).paginate(0, 1)
        kbs = kbs.dicts()
        return list(kbs)

    @classmethod
    @DB.connection_context()
    def get_kb_by_name(cls, kb_name, user_id):
        kbs = cls.model.select().join(UserTenant, on=(UserTenant.tenant_id == Knowledgebase.tenant_id)
            ).where(cls.model.name == kb_name, UserTenant.user_id == user_id).paginate(0, 1)
        kbs = kbs.dicts()
        return list(kbs)

    @classmethod
    @DB.connection_context()
    def accessible4deletion(cls, kb_id, user_id):
        docs = cls.model.select(
            cls.model.id).where(cls.model.id == kb_id, cls.model.created_by == user_id).paginate(0, 1)
        docs = docs.dicts()
        if not docs:
            return False
        return True