feat: add initial model structure for knowledge management

apps/common/utils/tool_code.py

@@ -3,7 +3,7 @@
 import os
 import subprocess
 import sys
-import uuid
+import uuid_utils as uuid
 from textwrap import dedent
 
 from diskcache import Cache
@@ -18,7 +18,7 @@ class ToolExecutor:
     def __init__(self, sandbox=False):
         self.sandbox = sandbox
         if sandbox:
-            self.sandbox_path = '/opt/maxkb/app/sandbox'
+            self.sandbox_path = '/opt/maxkb-app/sandbox'
             self.user = 'sandbox'
         else:
             self.sandbox_path = os.path.join(PROJECT_DIR, 'data', 'sandbox')
@@ -35,7 +35,7 @@ class ToolExecutor:
             os.umask(old_mask)
 
     def exec_code(self, code_str, keywords):
-        _id = str(uuid.uuid1())
+        _id = str(uuid.uuid7())
         success = '{"code":200,"msg":"成功","data":exec_result}'
         err = '{"code":500,"msg":str(e),"data":None}'
         path = r'' + self.sandbox_path + ''

apps/knowledge/admin.py

@@ -0,0 +1,3 @@
+from django.contrib import admin
+
+# Register your models here.

apps/knowledge/apps.py

@@ -0,0 +1,6 @@
+from django.apps import AppConfig
+
+
+class KnowledgeConfig(AppConfig):
+    default_auto_field = 'django.db.models.BigAutoField'
+    name = 'knowledge'

apps/knowledge/models/knowledge.py

@@ -0,0 +1,59 @@
+import uuid_utils.compat as uuid
+from django.db import models
+from django.db.models.signals import pre_delete
+from django.dispatch import receiver
+
+from common.db.sql_execute import select_one
+from common.mixins.app_model_mixin import AppModelMixin
+from models_provider.models import Model
+from users.models import User
+
+
+class KnowledgeType(models.TextChoices):
+    base = 0, '通用类型'
+    web = 1, 'web站点类型'
+    lark = 2, '飞书类型'
+    yuque = 3, '语雀类型'
+
+
+def default_model():
+    # todo : 这里需要从数据库中获取默认的模型
+    return uuid.UUID('42f63a3d-427e-11ef-b3ec-a8a1595801ab')
+
+
+class DataSet(AppModelMixin):
+    id = models.UUIDField(primary_key=True, max_length=128, default=uuid.uuid7, editable=False, verbose_name="主键id")
+    name = models.CharField(max_length=150, verbose_name="知识库名称")
+    desc = models.CharField(max_length=256, verbose_name="描述")
+    user = models.ForeignKey(User, on_delete=models.DO_NOTHING, verbose_name="所属用户")
+    type = models.IntegerField(verbose_name='类型', choices=KnowledgeType.choices, default=KnowledgeType.base)
+    embedding_mode = models.ForeignKey(Model, on_delete=models.DO_NOTHING, verbose_name="向量模型",
+                                       default=default_model)
+    meta = models.JSONField(verbose_name="元数据", default=dict)
+
+    class Meta:
+        db_table = "knowledge"
+
+
+class File(AppModelMixin):
+    id = models.UUIDField(primary_key=True, max_length=128, default=uuid.uuid7, editable=False, verbose_name="主键id")
+    file_name = models.CharField(max_length=256, verbose_name="文件名称", default="")
+    loid = models.IntegerField(verbose_name="loid")
+    meta = models.JSONField(verbose_name="文件关联数据", default=dict)
+
+    class Meta:
+        db_table = "file"
+
+    def save(self, bytea=None, force_insert=False, force_update=False, using=None, update_fields=None):
+        result = select_one("SELECT lo_from_bytea(%s, %s::bytea) as loid", [0, bytea])
+        self.loid = result['loid']
+        super().save()
+
+    def get_bytes(self):
+        result = select_one(f'SELECT lo_get({self.loid}) as "data"', [])
+        return result['data']
+
+
+@receiver(pre_delete, sender=File)
+def on_delete_file(sender, instance, **kwargs):
+    select_one(f'SELECT lo_unlink({instance.loid})', [])

apps/knowledge/tests.py

@@ -0,0 +1,3 @@
+from django.test import TestCase
+
+# Create your tests here.

apps/knowledge/urls.py

@@ -0,0 +1,3 @@
+from django.shortcuts import render
+
+# Create your views here.

apps/models_provider/serializers/model_serializer.py

@@ -293,7 +293,7 @@ class ModelSerializer(serializers.Serializer):
 
             credential = self.data.get('credential')
             model_data = {
-                'id': uuid.uuid1(),
+                'id': uuid.uuid7(),
                 'status': status,
                 'user_id': self.data.get('user_id'),
                 'name': self.data.get('name'),