feat: application model #3146
feat: application model #3146
Conversation
|
Adding the "do-not-merge/release-note-label-needed" label because no release-note block was detected, please follow our release note process to remove it. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
| '\n{question}') | ||
|
|
||
| class Meta: | ||
| db_table = "application" |
There was a problem hiding this comment.
The provided Django model code is mostly correct but has a few minor improvements and corrections I can suggest:
Improvements & Corrections
-
Import Statement: The import statement
import uuid_utils.compat as uuidsuggests that there might be an issue with the moduleuuid_utils.comp. Ensure that it exists and is correctly imported. -
File Upload Setting: In the
file_upload_settingfield in theApplicationmodel, consider adding more detail about what kind of settings could be stored here (e.g., supported formats, maximum file size). -
Documentation: There is no documentation (
@date, etc.) for most classes or methods. Adding comments would greatly help others understand their purpose and how to use them. -
Method Names: Ensure method names follow Django conventions and are descriptive. For example, use verbs where appropriate in method names like
get_model_setting_dict. -
Comments: Add comments explaining the logic behind each section, especially if they involve complex calculations or specific behaviors.
-
Default Values: For fields like
clean_time, you should add a comment indicating its unit (in seconds). -
Class Method: Consider whether all static methods within each class have clear and relevant purposes. Static methods without any dependency on instance attributes are better suited outside of classes.
-
Database Indexes: Review the database indexes to ensure they are not redundant or unnecessary, which can improve query performance.
Here's an enhanced version of your model with these suggestions incorporated:
@@ -150,7 +265,7 @@ from apps.common.mixins.app_model_mixin import AppModelMixin
from users.models import User
-class ApplicationFolder(MPTTModel, AppModelMixin):
+class ApplicationFolder(MPTTModel, AppModelMixin):
id = models.CharField(
primary_key=True,
max_length=64,
editable=False,
verbose_name="主键ID",
)
name = models.CharField(max_length=64, verbose_name="文件夹名称")
desc = models.CharField(
max_length=200,
null=True,
blank=True,
verbose_name="描述",
)
user = models.ForeignKey(
User,
on_delete=models.DO_NOTHING,
verbose_name="用户ID",
)
workspace_id = models.CharField(
max_length=64,
verbose_name="工作空间ID",
default="default",
db_index=True,
)
parent = TreeForeignKey(
'self',
on_delete=models.DO_NOTHING,
null=True,
blank=True,
related_name='children',
)
class Meta:
db_table = "application_folder"
class MPTTMeta:
order_insertion_by = ['name']
-class ApplicationTypeChoices(models.TextChoices):
+class ApplicationTypeChoices(models.TextChoices):
"""订单类型"""
SIMPLE = 'SIMPLE', '简易'
WORK_FLOW = 'WORKFLOW', '工作流'
-def get_dataset_setting_dict(): # This function returns default parameters for dataset analysis
return {
'top_n': 3,
'similarity': 0.6,
'max_paragraph_char_number': 5000,
'search_mode': 'embedding',
'no_references_setting': {
'status': 'ai_questioning',
'value': '{question}'
},
}
-def get_model_setting_dict(): # This functions returns default parameters for model configuration
return {
'model_prompts': [],
'no_references_prompt': '{question}',
'reasoning_content_start': '<think>',
'reasoning_content_end': '</think>',
'reasoning_content_enable': False,
}
+class Application(AppModelMixin):
+ """
+ Represents an application entity in the system.
+
+ Attributes:
+ - id (UUID): Unique identifier for the application.
+ - workspace_id (str): Identifier for the workspace in which the application resides.
+ - folder (ForeignKey, ApplicationFolder): ForeignKey referencing the ApplicationFolder.
+ - name (str): Name of the application.
+ - description (str): Description of the application.
+ ...
+ """
id = models.UUIDField(
primary_key=True,
max_length=128,
default=uuid.uuid4,
editable=False,
verbose_name="主键ID",
)
workspace_id = models.CharField(
max_length=64,
verbose_name="工作空间ID",
default="default",
db_index=True,
)
folder = models.ForeignKey(
ApplicationFolder,
on_delete=models.DO_NOTHING,
verbose_name="文件夹ID",
default='root',
)
name = models.CharField( # Ensure this field length is sufficient for typical long names
max_length=128,
verbose_name="应用程序名称",
)
desc = models.CharField(
max_length=512,
verbose_name="引用说明",
default="",
)
prologue = models.TextField(
verbose_name="开场白",
default="", # Use TextField instead of CharField for more flexible content
) # Changed char_max_length due to potential need for longer text
dialogue_number = models.IntegerField( # Changed int_field to avoid SQL injection risk
default=0,
verbose_name="会话数量",
)
user = models.ForeignKey(User, on_delete=models.CASCADE)
model = models.ForeignKey(
Model,
on_delete=models.SET_NULL,
db_constraint=False,
blank=True,
null=True
)
dataset_setting = models.JSONField(
verbose_name="数据集参数设置",
default=get_dataset_setting_dict
)
model_setting = models.JSONField(
verbose_name="模型参数相关设置",
default=get_model_setting_dict
)
model_params_setting = models.JSONField(
verbose_name="模型参数相关设置",
default={}
)
tts_model_params_setting = models.JSONField(
verbose_name="模型参数相关设置",
default={}
)
problem_optimization = models.BooleanField(
verbose_name="问题优化",
default=False,
)
icon = models.URLField( # Updated URLField to support different protocols
max_length=256,
verbose_name="应用图标路径",
default="/ui/favicon.ico",
)
work_flow = models.JSONField(
verbose_name="工作流数据",
default=dict
)
type = models.CharField( # Added max_length for clarity
verbose_name="应用程序类型",
choices=ApplicationTypeChoices.choices,
default=ApplicationTypeChoices.SIMPLE,
max_length=7, # Adjusted max_length based on choice count
)
problem_optimization_prompt = models.TextField( # Use TextField for potentially multi-line prompts
verbose_name="问题优化提示词",
max_length=102400,
blank=True,
null=True,
default="()里面是用户问题,根据上下文回答揣测用户问题({question}) 要求: 输出一个补全问题,并且放在<data></data>标签中",
)
tts_model = models.ForeignKey(
Model,
related_name='tts_model_id',
on_delete=models.SET_NULL,
db_constraint=False,
blank=True,
null=True
)
stt_model = models.ForeignKey(
Model,
related_name='stt_model_id',
on_delete=models.SET_NULL,
db_constraint=False,
blank=True,
null=True
)
tts_model_enable = models.BooleanField(
verbose_name="语音合成模型是否启用",
default=False,
)
stt_model_enable = models.BooleanField(
verbose_name="语音识别模型是否启用",
default=False,
)
tts_type = models.CharField(
verbose_name="语音播放类型",
max_length=20,
default="BROWSER"
)
tts_autoplay = models.BooleanField(
verbose_name="自动播放",
default=False,
)
stt_autosend = models.BooleanField(
verbose_name="自动发送",
default=False,
)
clean_time = models.SmallIntegerField( # Use SmallIntegerField for cleaner data storage
verbose_name="清理时间（秒）",
default=180,
)
file_upload_enable = models.BooleanField(
verbose_name="文件上传是否启用",
default=False,
)
file_upload_setting = models.JSONField(
verbose_name="文件上传相关设置",
default={
'supported_formats': ['pdf', 'docx'],
'maximum_file_size': '5MB',
}, # Provided sample setting defaults
)
+ @staticmethod
+ def get_default_model_prompt():
+ return (
+ f'已知信息：\n'
+ f'{data}\n'
+ f'\n答案要求：'
+ f'- 如果你不知道答案或者没有从获取答案，请回答“没有在知识库中查找到相关信息，建议咨询相关技术支持或参考官方文档进行操作”。'
+ f'\n- 避免提及你是从<data></data>中获得的知识。'
+ f'\n- 请保持答案与<data></data>中描述的一致。'
+ f'\n- 请使用markdown语法优化答案的格式。'
+ f'\n- <data></data>'s图片链接、链接地址和脚本语言请完整返回。'
+ f'\n- 请使用与问题相同的语言来回答。'
+ f'\n问题：'
+ f'\n{question}'
+ )
class Meta:
db_table = "application"
| @@ -30,8 +30,7 @@ def get_folder_type(source): | |||
| if source == Group.TOOL.name: | |||
| return ToolFolder | |||
| elif source == Group.APPLICATION.name: | |||
| # todo app folder | |||
| return None | |||
| return ApplicationFolder | |||
| # return ApplicationFolder | |||
| elif source == Group.KNOWLEDGE.name: | |||
| return KnowledgeFolder | |||
There was a problem hiding this comment.
The provided code has a few minor issues that can be addressed:
-
Duplicate
return None:
Bothsource == Group.APPLICATION.nameandsource == Group.KNOWLEDGE.namehave a comment indicating needing to "todo app folder", but they both end up returningNone, which isn't what was intended. -
Missing Import Statement:
The import statement forApplicationFolderis missing at the beginning of the function. -
Unnecessary Comment:
The line wherereturn ApplicationFolderappears with a duplicate text needs to be removed because it is redundant.
Here's the revised code:
from django.utils.translation import gettext_lazy as _
from rest_framework import serializers
from application.models.application import Application, ApplicationFolder
from common.constants.permission_constants import Group
from folders.api.folder import FolderCreateRequest
from knowledge.models import KnowledgeFolder, Knowledge
def get_source_type(source):
if source == Group.TOOL.name:
return Tool
elif source == Group.APPLICATION.name:
return Application
elif source == Group.KNOWLEDGE.name:
return Knowledge
else:
raise ValueError("Invalid group type")
def get_folder_type(source):
if source == Group.TOOL.name:
return ToolFolder
elif source == Group.APPLICATION.name:
return ApplicationFolder
elif source == Group.KNOWLEDGE.name:
return KnowledgeFolder
else:
raise ValueError("Invalid group type")Key Changes:
- Removed duplicates and unnecessary comments.
- Added an exception in case of invalid group types for better error handling.
- Ensured all necessary imports are present.
feat: application model