


How to distinguish between business logic and storage logic in back-end development?
Apr 19, 2025 pm 09:18 PMBack-end three-layer architecture: the boundary between business logic and data access logic
In back-end development, the common three-layer architectures of controller, service and dao are relatively clear in the controller and service layers. They are mainly implemented by separating business logic and presentation logic, such as decoupling message queue (MQ), HTTP, RPC, etc. from business logic. However, the boundary between the service layer and the dao layer, especially after the introduction of the manager layer, often confuses developers.
In Python back-end development, business logic is sometimes mixed in the model layer, such as business query methods such as usermodel.is_super()
, or native database operations such as usermodel.objects.all()
, and even cross-table operations such as usermodel.**
.
Analysis of business logic and non-business logic
The key to business logic and non-business logic lies in whether it directly relates to customer needs. Logic that customers cannot perceive is often considered non-business logic, including:
-
Database structure and association relationship: For example,
usermanager.delete()
anddepartmentmanager.delete()
methods can handle the deletion of the association table (such asuserdeptmodel
) at the same time, without calling the dao layer method twice at the service layer. Even without a manager layer, the dao layer can perform such association or cross-table operations as long as these operations are independent of business logic.class UserManager: def delete(self): userdao.delete() userdeptdao.delete() class DepartmentManager: def delete(self): departmentdao.delete() userdeptdao.delete()
-
Password salt: Users only need to know that the password is not stored in plain text, and the salt addition operation can be processed in the dao or manager layer.
class UserDao: def make_password(self, passwd): return salt(passwd) # add salt function def save(self): self.passwd = self.make_password(self.passwd) super().save()
Naming and definition of dao layer method: Naming dao layer method, for example, whether semantic names such as
get_super_user
are suitable depends on whether they are related to business logic. Ifsuper
is not business-related, it is acceptable to useget_super_user
.HTTP request encapsulation: Backend dependencies (such as services provided by other teams) can be encapsulated into dao-layer methods, rather than service-layer methods.
Implement functions similar to Django filter in Django/Flask
When implementing Django filter-like functions in Django and Flask, you often encounter layer-by-layer penetration problems because the dao layer needs to pass in request parameters. In the absence of a dependency injection framework like Spring, you can consider:
- In Java, frameworks such as MyBatis or JPA are usually used to manage data access and filtering logic through annotations and configuration files.
The relationship between data entities and three-layer architecture
Data entities represent data objects in the system. In a three-layer architecture, controller, service and dao layers do not strictly correspond to one by one:
- The dao layer may contain multiple methods to process different data entities, such as
userdao
anddepartmentdao
. - The service layer may need to combine multiple dao layer methods to implement complete business logic.
In short, the dao layer is only responsible for data storage interaction and does not include business logic; the service layer is responsible for executing business logic. For example, when creating a user, the service layer checks whether the user name is duplicated, and then calls the dao layer method to save the user. This architecture is designed to divide the system by responsibility and improve the maintainability and scalability of the code.
The above is the detailed content of How to distinguish between business logic and storage logic in back-end development?. For more information, please follow other related articles on the PHP Chinese website!

Hot AI Tools

Undress AI Tool
Undress images for free

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Clothoff.io
AI clothes remover

Video Face Swap
Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Article

Hot Tools

Notepad++7.3.1
Easy-to-use and free code editor

SublimeText3 Chinese version
Chinese version, very easy to use

Zend Studio 13.0.1
Powerful PHP integrated development environment

Dreamweaver CS6
Visual web development tools

SublimeText3 Mac version
God-level code editing software (SublimeText3)

Hot Topics

In Python, there are two main ways to call the __init__ method of the parent class. 1. Use the super() function, which is a modern and recommended method that makes the code clearer and automatically follows the method parsing order (MRO), such as super().__init__(name). 2. Directly call the __init__ method of the parent class, such as Parent.__init__(self,name), which is useful when you need to have full control or process old code, but will not automatically follow MRO. In multiple inheritance cases, super() should always be used consistently to ensure the correct initialization order and behavior.

ToconnecttoadatabaseinPython,usetheappropriatelibraryforthedatabasetype.1.ForSQLite,usesqlite3withconnect()andmanagewithcursorandcommit.2.ForMySQL,installmysql-connector-pythonandprovidecredentialsinconnect().3.ForPostgreSQL,installpsycopg2andconfigu

def is suitable for complex functions, supports multiple lines, document strings and nesting; lambda is suitable for simple anonymous functions and is often used in scenarios where functions are passed by parameters. The situation of selecting def: ① The function body has multiple lines; ② Document description is required; ③ Called multiple places. When choosing a lambda: ① One-time use; ② No name or document required; ③ Simple logic. Note that lambda delay binding variables may throw errors and do not support default parameters, generators, or asynchronous. In actual applications, flexibly choose according to needs and give priority to clarity.

Yes, you can parse HTML tables using Python and Pandas. First, use the pandas.read_html() function to extract the table, which can parse HTML elements in a web page or string into a DataFrame list; then, if the table has no clear column title, it can be fixed by specifying the header parameters or manually setting the .columns attribute; for complex pages, you can combine the requests library to obtain HTML content or use BeautifulSoup to locate specific tables; pay attention to common pitfalls such as JavaScript rendering, encoding problems, and multi-table recognition.

The way to access nested JSON objects in Python is to first clarify the structure and then index layer by layer. First, confirm the hierarchical relationship of JSON, such as a dictionary nested dictionary or list; then use dictionary keys and list index to access layer by layer, such as data "details"["zip"] to obtain zip encoding, data "details"[0] to obtain the first hobby; to avoid KeyError and IndexError, the default value can be set by the .get() method, or the encapsulation function safe_get can be used to achieve secure access; for complex structures, recursively search or use third-party libraries such as jmespath to handle.

The key to dealing with API authentication is to understand and use the authentication method correctly. 1. APIKey is the simplest authentication method, usually placed in the request header or URL parameters; 2. BasicAuth uses username and password for Base64 encoding transmission, which is suitable for internal systems; 3. OAuth2 needs to obtain the token first through client_id and client_secret, and then bring the BearerToken in the request header; 4. In order to deal with the token expiration, the token management class can be encapsulated and automatically refreshed the token; in short, selecting the appropriate method according to the document and safely storing the key information is the key.

In Python's for loop, use the continue statement to skip some operations in the current loop and enter the next loop. When the program executes to continue, the current loop will be immediately ended, the subsequent code will be skipped, and the next loop will be started. For example, scenarios such as excluding specific values ??when traversing the numeric range, skipping invalid entries when data cleaning, and skipping situations that do not meet the conditions in advance to make the main logic clearer. 1. Skip specific values: For example, exclude items that do not need to be processed when traversing the list; 2. Data cleaning: Skip exceptions or invalid data when reading external data; 3. Conditional judgment pre-order: filter non-target data in advance to improve code readability. Notes include: continue only affects the current loop layer and will not

ToscrapeawebsitethatrequiresloginusingPython,simulatetheloginprocessandmaintainthesession.First,understandhowtheloginworksbyinspectingtheloginflowinyourbrowser'sDeveloperTools,notingtheloginURL,requiredparameters,andanytokensorredirectsinvolved.Secon
