# State Management Standards for PostgreSQL
This document outlines the coding standards and best practices for state management within PostgreSQL applications. It aims to guide developers in building maintainable, performant, and secure database-driven applications.
## 1. Introduction to State Management in PostgreSQL
State management is a crucial aspect of application development, referring to how an application maintains and utilizes data between user interactions or different points in time. In the context of PostgreSQL, state management encompasses not only the data stored in tables but also transient data, session information, application-specific states, and data flows controlled within the database itself using SQL and procedural languages. Effective state management is vital for:
* **Maintainability:** Well-defined state helps in understanding the system's behavior and simplifies debugging and modifications.
* **Performance:** Efficient state management avoids unnecessary data reads and writes, thereby improving overall system performance.
* **Consistency:** Reliable state handling prevents data corruption and ensures predictable application behavior.
* **Scalability:** The ability to manage state efficiently is essential for scaling applications to handle increased load.
## 2. Approaches to State Management in PostgreSQL
### 2.1 Data Persistence with Tables
The primary responsibility of a PostgreSQL database is the persistent storage of data within tables. Data modelling is the MOST important aspect of state management in tables.
**Do This:**
* **Normalization:** Adhere to database normalization principles (1NF, 2NF, 3NF, etc.) to reduce redundancy and improve data integrity.
* **Data Types:** Use appropriate data types for each column (e.g., "INTEGER", "TEXT", "DATE", "JSONB", "UUID"). Choosing the correct type significantly impacts storage and performance.
* **Constraints:** Employ constraints ("NOT NULL", "UNIQUE", "CHECK", "FOREIGN KEY") to enforce data integrity and business rules directly within the database.
* **Indexes:** Use indexes strategically on frequently queried columns to speed up data retrieval. Consider composite indexes for complex queries involving multiple columns.
* **Partitioning:** For large tables, consider partitioning to improve query performance and manageability. Partitioning allows you to divide a large table into smaller, more manageable pieces. Use declarative partitioning (introduced in PostgreSQL 10) because it often has better query performance and less complexity (compared to trigger based approaches).
**Don't Do This:**
* **Over-normalization:** Avoid excessive normalization that can lead to complex joins and reduced query performance.
* **Generic Data Types:** Using "TEXT" for everything. Use the most specific applicable type. For example, use "INTEGER" instead of "TEXT" when storing an integer value.
* **Ignoring Contraints:** Omitting constraints can introduce schema-level errors, causing cascading problems down the road.
* **Unnecessary Indexes:** Adding too many indexes can slow down write operations and increase storage costs. Regularly review and remove unused indexes.
* **Ignoring Data Locality:** Consider data access when implementing partitions to further improve performance.
**Example:**
"""sql
-- Creating a table with appropriate data types and constraints
CREATE TABLE users (
user_id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
username VARCHAR(50) UNIQUE NOT NULL,
email VARCHAR(255) UNIQUE NOT NULL,
password_hash VARCHAR(255) NOT NULL,
created_at TIMESTAMP WITH TIME ZONE DEFAULT now()
);
-- Creating an index on the email column
CREATE INDEX idx_users_email ON users (email);
-- Creating a partitioned table (example for user activity)
CREATE TABLE user_activity (
activity_id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID NOT NULL,
activity_type VARCHAR(50) NOT NULL,
activity_time TIMESTAMP WITH TIME ZONE NOT NULL
) PARTITION BY RANGE (activity_time);
-- Creating partitions for different time ranges
CREATE TABLE user_activity_2023_01 PARTITION OF user_activity
FOR VALUES FROM ('2023-01-01') TO ('2023-02-01');
CREATE TABLE user_activity_2023_02 PARTITION OF user_activity
FOR VALUES FROM ('2023-02-01') TO ('2023-03-01');
ALTER TABLE user_activity
ADD CONSTRAINT fk_user_activity_user_id
FOREIGN KEY (user_id) REFERENCES users(user_id);
"""
### 2.2 Using Temporary Tables
Temporary tables are useful for storing intermediate results during complex queries and operations. They exist only for the duration of the session or transaction in which they were created.
**Do This:**
* **Transaction-specific vs. Session-specific:** Choose between "CREATE TEMP TABLE" (available for the entire session) and temporary tables created within a transaction. Transaction specific tables are automatically dropped at the end of the transaction.
* **Unlogged Temporary Tables:** For improved performance with data that doesn't need to be durable, use "CREATE UNLOGGED TEMP TABLE". They are faster because write-ahead logging is skipped. Data in unlogged tables is not crash-safe.
* **Optimize Usage:** Use temporary tables where they significantly simplify complex queries or improve performance by pre-computing intermediate results.
**Don't Do This:**
* **Overuse:** Avoid excessive use of temporary tables, as creating and managing them has overhead. Evaluate if CTEs (Common Table Expressions) can achieve the same result more efficiently.
* **Unnecessary Durability:** Using plain "CREATE TEMP TABLE" if data doesn't require durability at the transaction commit.
* **Ignoring Indexing:** For non-trivial temporary tables use indexes on where clauses.
**Example:**
"""sql
-- Using a temporary table to pre-compute intermediate results
CREATE TEMP TABLE monthly_sales AS
SELECT
EXTRACT(MONTH FROM order_date) AS month,
SUM(order_total) AS total_sales
FROM
orders
WHERE
EXTRACT(YEAR FROM order_date) = 2023
GROUP BY
month;
-- Querying the temporary table
SELECT
month,
total_sales
FROM
monthly_sales
ORDER BY
month;
-- using an UNLOGGED table
CREATE UNLOGGED TEMP TABLE unlogged_example (id INT, val TEXT);
INSERT INTO unlogged_example (id, val) VALUES (1, 'test');
SELECT * FROM unlogged_example;
"""
### 2.3 Using Common Table Expressions (CTEs)
CTEs (Common Table Expressions) allow you to define temporary result sets within a query, improving readability and maintainability.
**Do This:**
* **Recursive CTEs:** Utilize recursive CTEs for hierarchical data structures or iterative computations.
* **Readability:** Use CTEs to break down complex queries into smaller, more understandable parts.
**Don't Do This:**
* **Over-nesting:** Avoid deeply nested CTEs that can become difficult to manage.
* **Performance Misconceptions:** Understand that CTEs are primarily for readability and modularity, not necessarily for performance optimization. CTEs might not always be materialized (optimized by the query planner).
**Example:**
"""sql
-- Non-Recursive CTE: Calculating average order value
WITH order_summary AS (
SELECT
customer_id,
COUNT(*) AS order_count,
SUM(order_total) AS total_spent
FROM
orders
GROUP BY
customer_id
)
SELECT
AVG(total_spent / order_count) AS average_order_value
FROM
order_summary;
-- Recursive CTE: Generating a sequence of numbers
WITH RECURSIVE number_series AS (
SELECT 1 AS n
UNION ALL
SELECT n + 1 FROM number_series WHERE n < 10
)
SELECT n FROM number_series;
"""
### 2.4 Sessions and Application Context
Setting Application Context variables through the "pg_catalog.set_config" function to set flags, user IDs, and other context-specific data on a session.
**Do This:**
* **Authentication Propagation:** After user authentication, set the "user_id" or role to be used in subsequent queries.
* **Centralize Access:** Implement functions to set and retrieve application context, creating a uniform approach throughout the application.
**Don't Do This:**
* **Direct Access:** Avoid directly setting context variables in ad-hoc queries, which defeats the purpose of having a consistent approach.
* **Misusing for Configuration Parameters** While "set_config" *can* be used for configuration, it's best left to *application context*. Use "ALTER SYSTEM" or parameters in "postgresql.conf" for global configurations.
**Example:**
"""sql
-- Setting a context value after user authentication
SELECT pg_catalog.set_config('myapp.user_id', '123', false);
-- Function to retrieve the user ID from the application context
CREATE OR REPLACE FUNCTION get_current_user_id()
RETURNS TEXT AS $$
BEGIN
RETURN pg_catalog.current_setting('myapp.user_id', true);
END;
$$ LANGUAGE plpgsql;
-- Using the function in a query
SELECT * FROM orders WHERE user_id = get_current_user_id();
"""
### 2.5 Using JSONB for Flexible State
PostgreSQL's "JSONB" data type allows you to store semi-structured data within a column.
**Do This:**
* **Configuration Options:** Store user preferences, application settings, or dynamic configurations as JSONB objects.
* **Use Indexes:** Index JSONB columns to efficiently query data within the JSON structure using "@>", "?", "?|", "?&". Consider GIN indexes.
**Don't Do This:**
* **Abuse JSONB:** Don't use JSONB as a replacement for proper relational data modeling. Use it only for genuinely semi-structured data with variable schemas.
**Example:**
"""sql
-- Storing user preferences as JSONB
CREATE TABLE user_preferences (
user_id UUID PRIMARY KEY,
preferences JSONB
);
-- Inserting user preferences
INSERT INTO user_preferences (user_id, preferences) VALUES (
'a1b2c3d4-e5f6-7890-1234-567890abcdef',
'{"theme": "dark", "notifications": {"email": true, "sms": false}}'
);
-- Querying for users with email notifications enabled
SELECT user_id
FROM user_preferences
WHERE preferences -> 'notifications' ->> 'email' = 'true';
-- Indexing jsonb column
CREATE INDEX idx_user_preferences_notifications ON user_preferences USING GIN (preferences jsonb_path_ops);
"""
### 2.6 Row-Level Security (RLS)
Row-Level Security (RLS) allows you to define policies that restrict access to rows in a table based on user attributes or application context.
**Do This:**
* **Data Isolation:** Use RLS to enforce access control policies, such as ensuring users can only see their own data.
* **Audit Trails:** Implement RLS policies that log access attempts, which are useful for auditing purposes.
**Don't Do This:**
* **Complex Policies:** Avoid creating overly complex RLS policies that can impact performance. Ensure policies are well-indexed and optimized.
* **Ignoring Performance:** Before adopting RLS in production, performance test with production-scale loads.
**Example:**
"""sql
-- Enable RLS on the orders table
ALTER TABLE orders ENABLE ROW LEVEL SECURITY;
-- Create a policy that allows users to only see their own orders
CREATE POLICY user_orders_policy ON orders
FOR SELECT
USING (user_id = get_current_user_id()); -- Assuming you have implemented get_current_user_id()
-- Create a dummy function
CREATE OR REPLACE FUNCTION get_current_user_id()
RETURNS UUID AS $$
BEGIN
RETURN 'a1b2c3d4-e5f6-7890-1234-567890abcdef'; -- Replace with a mechanism to get the real user ID
END;
$$ LANGUAGE plpgsql;
"""
### 2.7 Publish/Subscribe with "NOTIFY/LISTEN"
PostgreSQL's "NOTIFY/LISTEN" mechanism provides a simple publish/subscribe functionality useful for real-time updates and event-driven architectures.
**Do This:**
* **Real-time Updates:** Use "NOTIFY" to signal clients when data has changed, allowing them to refresh their views.
* **Background Processing:** Use "LISTEN" to trigger background tasks or worker processes when specific events occur.
**Don't Do This:**
* **Reliable Message Queuing:** Don't rely on "NOTIFY/LISTEN" for guaranteed message delivery, as messages can be lost if the client is disconnected. Use a dedicated message queue system (e.g., RabbitMQ, Kafka) for reliable messaging.
* **Overusing Triggers:** Avoid overuse of triggers invoking "NOTIFY", which can lead to performance bottlenecks.
**Example:**
"""sql
-- Trigger function to notify clients when a new order is created
CREATE OR REPLACE FUNCTION notify_new_order()
RETURNS TRIGGER AS $$
BEGIN
PERFORM pg_notify('new_order_channel', row_to_json(NEW)::text);
RETURN NEW;
END;
$$ LANGUAGE plpgsql;
-- Creating a trigger on the orders table
CREATE TRIGGER new_order_trigger
AFTER INSERT ON orders
FOR EACH ROW
EXECUTE FUNCTION notify_new_order();
-- Example usage from the client side (psql)
LISTEN new_order_channel;
-- Then insert a new record to "orders"
--INSERT INTO orders (order_id, user_id, order_total) VALUES (gen_random_uuid(), '849aa4ef-9c92-4599-b50a-47d23a24c85b', 100.00);
"""
### 2.8 Advisory Locks
Advisory locks allow you to implement application-level locking mechanisms to coordinate access to shared resources.
**Do This:**
* **Task Synchronization:** Use advisory locks to prevent concurrent execution of critical tasks or operations.
* **Resource Protection:** Use advisory locks to protect shared resources, such as files or external systems.
* **"pg_advisory_lock" vs. "pg_advisory_lock_shared":** Choose the appropriate lock type depending on whether you need exclusive or shared access to the resource.
**Don't Do This:**
* **Deadlocks:** Be careful to avoid deadlocks when using advisory locks. Acquire locks in a consistent order. Always release locks in a timely manner.
* **Unreleased Locks:** When an application terminates abnormally, locks might not be released. Implement mechanisms to automatically release locks in such cases.
**Example:**
"""sql
-- Acquiring an advisory lock
SELECT pg_advisory_lock(123);
-- Attempting to acquire a lock that's already held
SELECT pg_try_advisory_lock(123); -- Returns false if the lock is not acquired
-- Releasing the advisory lock
SELECT pg_advisory_unlock(123);
"""
## 3. Managing Data Flow and Reactivity
Managing the flow of data through the database involves strategies to ensure that data changes cascade appropriately and that applications react quickly to state transitions.
### 3.1 Triggers
Triggers can execute custom functions in response to INSERT, UPDATE, or DELETE operations.
**Do This:**
* **Auditing:** Create audit logs automatically whenever data is modified.
* **Data Validation:** Enforce complex validation rules that cannot be expressed with "CHECK" constraints.
**Don't Do This:**
* **Complex Business Logic:** Avoid placing complex business logic in triggers, as it can make the system harder to understand and debug. Prefer application-layer logic.
* **Cascading Operations:** Limit the scope of triggered actions to avoid complex cascading operations that can lead to performance issues.
**Example:**
"""sql
-- Creating a trigger function to audit changes to the products table
CREATE OR REPLACE FUNCTION audit_products()
RETURNS TRIGGER AS $$
BEGIN
INSERT INTO product_audit (product_id, old_name, new_name, updated_at)
VALUES (OLD.product_id, OLD.name, NEW.name, now());
RETURN NEW;
END;
$$ LANGUAGE plpgsql;
-- Creating a trigger on the products table
CREATE TRIGGER products_audit_trigger
AFTER UPDATE OF name ON products
FOR EACH ROW
EXECUTE FUNCTION audit_products();
"""
### 3.2 Materialized Views
Materialized views store the results of a query as a table, which can be refreshed periodically or on demand.
**Do This:**
* **Pre-computed Aggregations:** Use materialized views to store pre-computed aggregations for frequently accessed data.
* **Complex Joins:** Materialize the results of complex joins for faster retrieval.
**Don't Do This:**
* **Real-time Data:** Don't use materialized views for data that requires real-time updates, as they need to be refreshed explicitly.
* **Ignoring Refresh Costs:** Ensure refreshing the materialized view doesn't become a performance bottleneck. Consider "REFRESH MATERIALIZED VIEW CONCURRENTLY"
**Example:**
"""sql
-- Creating a materialized view of daily sales totals
CREATE MATERIALIZED VIEW daily_sales_summary AS
SELECT
order_date,
SUM(order_total) AS total_sales
FROM
orders
GROUP BY
order_date;
-- Refreshing the materialized view
REFRESH MATERIALIZED VIEW CONCURRENTLY daily_sales_summary;
"""
## 4. Conclusion
These standards provide a foundation for building robust and maintainable applications using PostgreSQL. Adhering to these guidelines will ensure best practices are followed and data management is optimized for performance and scalability. This document should be a living document, updated as new features and best practices emerge within the PostgreSQL ecosystem.
danielsogl
Created Mar 6, 2025
This guide explains how to effectively use .clinerules with Cline, the AI-powered coding assistant.
The .clinerules file is a powerful configuration file that helps Cline understand your project's requirements, coding standards, and constraints. When placed in your project's root directory, it automatically guides Cline's behavior and ensures consistency across your codebase.
Place the .clinerules file in your project's root directory. Cline automatically detects and follows these rules for all files within the project.
# Project Overview project: name: 'Your Project Name' description: 'Brief project description' stack: - technology: 'Framework/Language' version: 'X.Y.Z' - technology: 'Database' version: 'X.Y.Z'
# Code Standards standards: style: - 'Use consistent indentation (2 spaces)' - 'Follow language-specific naming conventions' documentation: - 'Include JSDoc comments for all functions' - 'Maintain up-to-date README files' testing: - 'Write unit tests for all new features' - 'Maintain minimum 80% code coverage'
# Security Guidelines security: authentication: - 'Implement proper token validation' - 'Use environment variables for secrets' dataProtection: - 'Sanitize all user inputs' - 'Implement proper error handling'
Be Specific
Maintain Organization
Regular Updates
# Common Patterns Example patterns: components: - pattern: 'Use functional components by default' - pattern: 'Implement error boundaries for component trees' stateManagement: - pattern: 'Use React Query for server state' - pattern: 'Implement proper loading states'
Commit the Rules
.clinerules in version controlTeam Collaboration
Rules Not Being Applied
Conflicting Rules
Performance Considerations
# Basic .clinerules Example project: name: 'Web Application' type: 'Next.js Frontend' standards: - 'Use TypeScript for all new code' - 'Follow React best practices' - 'Implement proper error handling' testing: unit: - 'Jest for unit tests' - 'React Testing Library for components' e2e: - 'Cypress for end-to-end testing' documentation: required: - 'README.md in each major directory' - 'JSDoc comments for public APIs' - 'Changelog updates for all changes'
# Advanced .clinerules Example project: name: 'Enterprise Application' compliance: - 'GDPR requirements' - 'WCAG 2.1 AA accessibility' architecture: patterns: - 'Clean Architecture principles' - 'Domain-Driven Design concepts' security: requirements: - 'OAuth 2.0 authentication' - 'Rate limiting on all APIs' - 'Input validation with Zod'
# Database: Create RLS policies You're a Supabase Postgres expert in writing row level security policies. Your purpose is to generate a policy with the constraints given by the user. You should first retrieve schema information to write policies for, usually the 'public' schema. The output should use the following instructions: - The generated SQL must be valid SQL. - You can use only CREATE POLICY or ALTER POLICY queries, no other queries are allowed. - Always use double apostrophe in SQL strings (eg. 'Night''s watch') - You can add short explanations to your messages. - The result should be a valid markdown. The SQL code should be wrapped in ``` (including sql language tag). - Always use "auth.uid()" instead of "current_user". - SELECT policies should always have USING but not WITH CHECK - INSERT policies should always have WITH CHECK but not USING - UPDATE policies should always have WITH CHECK and most often have USING - DELETE policies should always have USING but not WITH CHECK - Don't use `FOR ALL`. Instead separate into 4 separate policies for select, insert, update, and delete. - The policy name should be short but detailed text explaining the policy, enclosed in double quotes. - Always put explanations as separate text. Never use inline SQL comments. - If the user asks for something that's not related to SQL policies, explain to the user that you can only help with policies. - Discourage `RESTRICTIVE` policies and encourage `PERMISSIVE` policies, and explain why. The output should look like this: ```sql CREATE POLICY "My descriptive policy." ON books FOR INSERT to authenticated USING ( (select auth.uid()) = author_id ) WITH ( true ); ``` Since you are running in a Supabase environment, take note of these Supabase-specific additions below. ## Authenticated and unauthenticated roles Supabase maps every request to one of the roles: - `anon`: an unauthenticated request (the user is not logged in) - `authenticated`: an authenticated request (the user is logged in) These are actually [Postgres Roles](/docs/guides/database/postgres/roles). You can use these roles within your Policies using the `TO` clause: ```sql create policy "Profiles are viewable by everyone" on profiles for select to authenticated, anon using ( true ); -- OR create policy "Public profiles are viewable only by authenticated users" on profiles for select to authenticated using ( true ); ``` Note that `for ...` must be added after the table but before the roles. `to ...` must be added after `for ...`: ### Incorrect ```sql create policy "Public profiles are viewable only by authenticated users" on profiles to authenticated for select using ( true ); ``` ### Correct ```sql create policy "Public profiles are viewable only by authenticated users" on profiles for select to authenticated using ( true ); ``` ## Multiple operations PostgreSQL policies do not support specifying multiple operations in a single FOR clause. You need to create separate policies for each operation. ### Incorrect ```sql create policy "Profiles can be created and deleted by any user" on profiles for insert, delete -- cannot create a policy on multiple operators to authenticated with check ( true ) using ( true ); ``` ### Correct ```sql create policy "Profiles can be created by any user" on profiles for insert to authenticated with check ( true ); create policy "Profiles can be deleted by any user" on profiles for delete to authenticated using ( true ); ``` ## Helper functions Supabase provides some helper functions that make it easier to write Policies. ### `auth.uid()` Returns the ID of the user making the request. ### `auth.jwt()` Returns the JWT of the user making the request. Anything that you store in the user's `raw_app_meta_data` column or the `raw_user_meta_data` column will be accessible using this function. It's important to know the distinction between these two: - `raw_user_meta_data` - can be updated by the authenticated user using the `supabase.auth.update()` function. It is not a good place to store authorization data. - `raw_app_meta_data` - cannot be updated by the user, so it's a good place to store authorization data. The `auth.jwt()` function is extremely versatile. For example, if you store some team data inside `app_metadata`, you can use it to determine whether a particular user belongs to a team. For example, if this was an array of IDs: ```sql create policy "User is in team" on my_table to authenticated using ( team_id in (select auth.jwt() -> 'app_metadata' -> 'teams')); ``` ### MFA The `auth.jwt()` function can be used to check for [Multi-Factor Authentication](/docs/guides/auth/auth-mfa#enforce-rules-for-mfa-logins). For example, you could restrict a user from updating their profile unless they have at least 2 levels of authentication (Assurance Level 2): ```sql create policy "Restrict updates." on profiles as restrictive for update to authenticated using ( (select auth.jwt()->>'aal') = 'aal2' ); ``` ## RLS performance recommendations Every authorization system has an impact on performance. While row level security is powerful, the performance impact is important to keep in mind. This is especially true for queries that scan every row in a table - like many `select` operations, including those using limit, offset, and ordering. Based on a series of [tests](https://github.com/GaryAustin1/RLS-Performance), we have a few recommendations for RLS: ### Add indexes Make sure you've added [indexes](/docs/guides/database/postgres/indexes) on any columns used within the Policies which are not already indexed (or primary keys). For a Policy like this: ```sql create policy "Users can access their own records" on test_table to authenticated using ( (select auth.uid()) = user_id ); ``` You can add an index like: ```sql create index userid on test_table using btree (user_id); ``` ### Call functions with `select` You can use `select` statement to improve policies that use functions. For example, instead of this: ```sql create policy "Users can access their own records" on test_table to authenticated using ( auth.uid() = user_id ); ``` You can do: ```sql create policy "Users can access their own records" on test_table to authenticated using ( (select auth.uid()) = user_id ); ``` This method works well for JWT functions like `auth.uid()` and `auth.jwt()` as well as `security definer` Functions. Wrapping the function causes an `initPlan` to be run by the Postgres optimizer, which allows it to "cache" the results per-statement, rather than calling the function on each row. Caution: You can only use this technique if the results of the query or function do not change based on the row data. ### Minimize joins You can often rewrite your Policies to avoid joins between the source and the target table. Instead, try to organize your policy to fetch all the relevant data from the target table into an array or set, then you can use an `IN` or `ANY` operation in your filter. For example, this is an example of a slow policy which joins the source `test_table` to the target `team_user`: ```sql create policy "Users can access records belonging to their teams" on test_table to authenticated using ( (select auth.uid()) in ( select user_id from team_user where team_user.team_id = team_id -- joins to the source "test_table.team_id" ) ); ``` We can rewrite this to avoid this join, and instead select the filter criteria into a set: ```sql create policy "Users can access records belonging to their teams" on test_table to authenticated using ( team_id in ( select team_id from team_user where user_id = (select auth.uid()) -- no join ) ); ``` ### Specify roles in your policies Always use the Role of inside your policies, specified by the `TO` operator. For example, instead of this query: ```sql create policy "Users can access their own records" on rls_test using ( auth.uid() = user_id ); ``` Use: ```sql create policy "Users can access their own records" on rls_test to authenticated using ( (select auth.uid()) = user_id ); ``` This prevents the policy `( (select auth.uid()) = user_id )` from running for any `anon` users, since the execution stops at the `to authenticated` step.
# Database: Create migration You are a Postgres Expert who loves creating secure database schemas. This project uses the migrations provided by the Supabase CLI. ## Creating a migration file Given the context of the user's message, create a database migration file inside the folder `supabase/migrations/`. The file MUST following this naming convention: The file MUST be named in the format `YYYYMMDDHHmmss_short_description.sql` with proper casing for months, minutes, and seconds in UTC time: 1. `YYYY` - Four digits for the year (e.g., `2024`). 2. `MM` - Two digits for the month (01 to 12). 3. `DD` - Two digits for the day of the month (01 to 31). 4. `HH` - Two digits for the hour in 24-hour format (00 to 23). 5. `mm` - Two digits for the minute (00 to 59). 6. `ss` - Two digits for the second (00 to 59). 7. Add an appropriate description for the migration. For example: ``` 20240906123045_create_profiles.sql ``` ## SQL Guidelines Write Postgres-compatible SQL code for Supabase migration files that: - Includes a header comment with metadata about the migration, such as the purpose, affected tables/columns, and any special considerations. - Includes thorough comments explaining the purpose and expected behavior of each migration step. - Write all SQL in lowercase. - Add copious comments for any destructive SQL commands, including truncating, dropping, or column alterations. - When creating a new table, you MUST enable Row Level Security (RLS) even if the table is intended for public access. - When creating RLS Policies - Ensure the policies cover all relevant access scenarios (e.g. select, insert, update, delete) based on the table's purpose and data sensitivity. - If the table is intended for public access the policy can simply return `true`. - RLS Policies should be granular: one policy for `select`, one for `insert` etc) and for each supabase role (`anon` and `authenticated`). DO NOT combine Policies even if the functionality is the same for both roles. - Include comments explaining the rationale and intended behavior of each security policy The generated SQL code should be production-ready, well-documented, and aligned with Supabase's best practices.
# Postgres SQL Style Guide ## General - Use lowercase for SQL reserved words to maintain consistency and readability. - Employ consistent, descriptive identifiers for tables, columns, and other database objects. - Use white space and indentation to enhance the readability of your code. - Store dates in ISO 8601 format (`yyyy-mm-ddThh:mm:ss.sssss`). - Include comments for complex logic, using '/_ ... _/' for block comments and '--' for line comments. ## Naming Conventions - Avoid SQL reserved words and ensure names are unique and under 63 characters. - Use snake_case for tables and columns. - Prefer plurals for table names - Prefer singular names for columns. ## Tables - Avoid prefixes like 'tbl\_' and ensure no table name matches any of its column names. - Always add an `id` column of type `identity generated always` unless otherwise specified. - Create all tables in the `public` schema unless otherwise specified. - Always add the schema to SQL queries for clarity. - Always add a comment to describe what the table does. The comment can be up to 1024 characters. ## Columns - Use singular names and avoid generic names like 'id'. - For references to foreign tables, use the singular of the table name with the `_id` suffix. For example `user_id` to reference the `users` table - Always use lowercase except in cases involving acronyms or when readability would be enhanced by an exception. #### Examples: ```sql create table books ( id bigint generated always as identity primary key, title text not null, author_id bigint references authors (id) ); comment on table books is 'A list of all the books in the library.'; ``` ## Queries - When the query is shorter keep it on just a few lines. As it gets larger start adding newlines for readability - Add spaces for readability. Smaller queries: ```sql select * from employees where end_date is null; update employees set end_date = '2023-12-31' where employee_id = 1001; ``` Larger queries: ```sql select first_name, last_name from employees where start_date between '2021-01-01' and '2021-12-31' and status = 'employed'; ``` ### Joins and Subqueries - Format joins and subqueries for clarity, aligning them with related SQL clauses. - Prefer full table names when referencing tables. This helps for readability. ```sql select employees.employee_name, departments.department_name from employees join departments on employees.department_id = departments.department_id where employees.start_date > '2022-01-01'; ``` ## Aliases - Use meaningful aliases that reflect the data or transformation applied, and always include the 'as' keyword for clarity. ```sql select count(*) as total_employees from employees where end_date is null; ``` ## Complex queries and CTEs - If a query is extremely complex, prefer a CTE. - Make sure the CTE is clear and linear. Prefer readability over performance. - Add comments to each block. ```sql with department_employees as ( -- Get all employees and their departments select employees.department_id, employees.first_name, employees.last_name, departments.department_name from employees join departments on employees.department_id = departments.department_id ), employee_counts as ( -- Count how many employees in each department select department_name, count(*) as num_employees from department_employees group by department_name ) select department_name, num_employees from employee_counts order by department_name; ```
# API Integration Standards for PostgreSQL This document outlines the coding standards for integrating PostgreSQL with external APIs and backend services. These standards promote maintainability, performance, and security when building applications that rely on data and functionality outside of the database itself. It focuses on modern approaches compatible with the latest PostgreSQL version. ## 1. Architectural Considerations for API Integration ### 1.1. Standard: Define Clear API Boundaries **Do This:** * Clearly define the responsibilities of PostgreSQL and external APIs. Use PostgreSQL for data persistence, relational logic, and indexing. Offload complex computations, specialized data processing, and external data access to APIs. * Use clear and consistent naming conventions for database functions/procedures interacting with APIs. Prefix them (e.g., "api_", "ext_") to easily identify external API integration code. * Document the contract (input/output) with each API thoroughly. **Don't Do This:** * Overload PostgreSQL with tasks that APIs are better suited for (e.g., image processing, complex machine learning tasks that are not data-intensive). * Embed undocumented or magic API calls directly within SQL queries. **Why:** Defining clear boundaries ensures modularity, easier maintenance, and optimized performance. It avoids turning the database into a monolithic application component. **Example:** """sql -- Good: Function for fetching user profiles from an external API. CREATE OR REPLACE FUNCTION api_get_user_profile(user_id INT) RETURNS JSONB AS $$ BEGIN -- Call external API to get user profile details. -- Using a hypothetical extension for API calls RETURN http_get('https://api.example.com/users/' || user_id)::jsonb; EXCEPTION WHEN OTHERS THEN RAISE EXCEPTION 'Error fetching user profile from API: %', SQLERRM; END; $$ LANGUAGE plpgsql; -- Bad: Embedding API logic directly within a complex query. -- SELECT * FROM users WHERE ... AND api_call(...) ... ; -- Avoid! """ ### 1.2. Standard: Asynchronous vs. Synchronous API Interactions **Do This:** * Use asynchronous API calls (e.g., message queues, background workers) where possible to prevent long-running database transactions from blocking other operations. Implement retries and error handling for asynchronous tasks. * For synchronous calls, keep the execution time as short as possible to avoid holding database connections for extended periods. **Don't Do This:** * Make blocking API calls directly within critical transaction paths. This will significantly impact database performance and availability. * Assume API calls will always succeed. Implement robust error handling and retries. **Why:** Asynchronous operations improve scalability and responsiveness. Synchronous operations can lead to deadlocks and performance degradation if not managed carefully. **Example (using pg_amqp or similar queue extensions):** """sql -- Asynchronous API call using a message queue. (Hypothetical Example) CREATE OR REPLACE FUNCTION api_process_user_data(user_id INT) RETURNS VOID AS $$ BEGIN -- Send a message to a queue for processing user data via an external API. PERFORM amqp.publish('process_user_data_queue', json_build_object('user_id', user_id)); -- Hypothetical RETURN; END; $$ LANGUAGE plpgsql; -- Example of a background worker (using pg_background) that consumes from the queue to call the external API -- Code for the background worker would be in a separate file and process the queue. """ ### 1.3. Standard: Data Transformation and Mapping **Do This:** * Define clear data mapping between PostgreSQL data types and API request/response formats (e.g., JSON, XML). Use PostgreSQL's JSONB and XML support effectively. * Validate data received from APIs before inserting it into the database using "CHECK" constraints or other validation mechanisms. * Log API requests and responses for debugging and auditing purposes. **Don't Do This:** * Directly insert untrusted data received from APIs into the database without validation. This can lead to SQL injection and other security vulnerabilities. * Rely on implicit type conversions between PostgreSQL and API data formats. Be explicit. **Why:** Proper data transformation and validation prevent data corruption and security breaches. Logging helps troubleshoot issues and track API usage. **Example:** """sql -- Validating and inserting JSON data from an API. CREATE TABLE api_user_profiles ( user_id INT PRIMARY KEY, profile_data JSONB -- CHECK constraint is appropriate here to require the JSON object ALWAYS conform to a schema ); CREATE OR REPLACE FUNCTION api_import_user_profile(user_id INT, profile_json JSONB) RETURNS VOID AS $$ DECLARE -- Validate JSON data against a schema (hypothetical function). is_valid BOOLEAN; BEGIN -- Validate that the JSON is valid against a schema is_valid := jsonb_matches_schema('{"type": "object", "properties": {"name": {"type": "string"},"email": {"type": "string", "format": "email"} }}', profile_json); IF NOT is_valid THEN RAISE EXCEPTION 'Invalid profile data format.'; END IF; INSERT INTO api_user_profiles (user_id, profile_data) VALUES (user_id, profile_json); RETURN; EXCEPTION WHEN OTHERS THEN RAISE EXCEPTION 'Error importing user profile: %', SQLERRM; END; $$ LANGUAGE plpgsql; """ ## 2. Implementation Details ### 2.1. Standard: Choosing the Right API Interaction Method **Do This:** * Evaluate these methods: * **HTTP Requests (using extensions like "http" or "curl"):** Suitable for RESTful APIs. * **Message Queues (using extensions like "pg_amqp" or "pg_kafka"):** Ideal for asynchronous communication. * **Foreign Data Wrappers (FDWs):** For integrating with other databases or data stores directly. * Choose the method that best fits the API's protocol, data format, and communication pattern. **Don't Do This:** * Force a specific integration method because it's familiar. Consider alternatives based on the API's characteristics. * Build custom, ad-hoc solutions when standard extensions and FDWs provide the necessary functionality. **Why:** Selecting the right method simplifies integration, improves performance, and reduces development effort. **Example (using "http" extension for a REST API):** """sql -- Example using the http extension to call a REST API CREATE EXTENSION IF NOT EXISTS http; CREATE OR REPLACE FUNCTION api_get_weather(city TEXT) RETURNS JSONB AS $$ DECLARE api_url TEXT := 'https://api.weatherapi.com/v1/current.json?key=YOUR_API_KEY&q=' || city; response HTTPResponse; BEGIN response := http_get(api_url); IF response.status_code = 200 THEN RETURN response.content::jsonb; ELSE RAISE EXCEPTION 'Weather API error: %', response.content; END IF; EXCEPTION WHEN OTHERS THEN RAISE EXCEPTION 'Error fetching weather data: %', SQLERRM; END; $$ LANGUAGE plpgsql; -- SELECT api_get_weather('London'); """ ### 2.2. Standard: Error Handling and Retries **Do This:** * Implement robust error handling for API calls. Catch exceptions, log errors, and implement retry mechanisms with exponential backoff. * Distinguish between transient and permanent errors. Retry transient errors (e.g., network timeouts), and log permanent errors (e.g., invalid API key) for investigation. * Set appropriate timeouts for API calls to prevent indefinite blocking. * Consider using "TRY...CATCH" blocks for error handling within PL/pgSQL functions. **Don't Do This:** * Ignore errors from API calls. At a minimum, log the error so it can be investigated later. * Retry indefinitely without a limit or backoff strategy. This can overload the API or the database. **Why:** Robust error handling ensures resilience and prevents cascading failures. It provides valuable insights into API issues. **Example:** """sql CREATE OR REPLACE FUNCTION api_get_data_with_retry(url TEXT, max_retries INT DEFAULT 3) RETURNS JSONB AS $$ DECLARE response HTTPResponse; retries INT := 0; delay INTERVAL := '1 second'; BEGIN LOOP BEGIN response := http_get(url); IF response.status_code = 200 THEN RETURN response.content::jsonb; ELSE RAISE WARNING 'API call failed with status code: %', response.status_code; -- Check for non-retryable errors here! -- IF response.status_code = 400 THEN RETURN NULL; -- Bad Request (do not retry) END IF; EXCEPTION WHEN OTHERS THEN RAISE WARNING 'API call error: %', SQLERRM; END; retries := retries + 1; IF retries >= max_retries THEN RAISE EXCEPTION 'Max retries exceeded for API call.'; END IF; RAISE NOTICE 'Retrying in %', delay; PERFORM pg_sleep(extract(epoch from delay)); delay := delay * 2; -- Exponential backoff END LOOP; EXCEPTION WHEN OTHERS THEN RAISE EXCEPTION 'Failed to get data after multiple retries: %', SQLERRM; END; $$ LANGUAGE plpgsql; """ ### 2.3. Standard: Security Considerations **Do This:** * Store API keys and secrets securely using PostgreSQL's configuration parameters or a dedicated secrets management solution. NEVER hardcode API keys in SQL code. * Use HTTPS for all API calls to encrypt data in transit. * Validate API responses to prevent data injection (e.g., JSON injection). * Implement rate limiting to prevent abuse. * Use least privilege principle when granting permissions to API interaction functions. **Don't Do This:** * Hardcode API keys or secrets in SQL code or store them in plain text in the database. * Trust API responses implicitly. Always validate the data. * Expose your PostgreSQL database directly to the internet without proper firewall and security measures. **Why:** Security is paramount. Protecting API keys, encrypting data, and rate limiting prevent unauthorized access and malicious attacks. **Example:** """sql -- Storing API key securely using postgresql.conf -- In postgresql.conf: -- api.weather_api_key = 'YOUR_API_KEY' -- SQL to retrieve the API key CREATE OR REPLACE FUNCTION api_get_weather_secure(city TEXT) RETURNS JSONB AS $$ DECLARE api_url TEXT := 'https://api.weatherapi.com/v1/current.json?key=' || current_setting('api.weather_api_key') || '&q=' || city; response HTTPResponse; BEGIN response := http_get(api_url); IF response.status_code = 200 THEN RETURN response.content::jsonb; ELSE RAISE EXCEPTION 'Weather API error: %', response.content; END IF; EXCEPTION WHEN OTHERS THEN RAISE EXCEPTION 'Error fetching weather data: %', SQLERRM; END; $$ LANGUAGE plpgsql SECURITY DEFINER; -- SECURITY DEFINER crucial for accessing external configurations -- Revoke execute permission from public REVOKE EXECUTE ON FUNCTION api_get_weather_secure(TEXT) FROM PUBLIC; -- Grant access to specific roles GRANT EXECUTE ON FUNCTION api_get_weather_secure(TEXT) TO your_application_role; """ ### 2.4. Standard: Performance Optimization **Do This:** * Cache API responses to reduce the number of API calls, especially for frequently accessed data. Use "MATERIALIZED VIEW" or a custom cache table. * Use connection pooling to minimize the overhead of establishing new connections to APIs. Some HTTP extensions do this internally. * Optimize data transfer by requesting only the necessary fields from the API. Use appropriate query parameters. **Don't Do This:** * Make redundant API calls. Identify opportunities for caching or batching. * Retrieve large amounts of data from APIs when only a small subset is needed. **Why:** Performance optimization improves application responsiveness and reduces API usage costs. **Example (using a materialized view for caching):** """sql CREATE MATERIALIZED VIEW weather_cache AS SELECT city, api_get_weather(city) AS weather_data, NOW() AS last_updated FROM (VALUES ('London'), ('New York'), ('Tokyo')) AS cities(city); CREATE UNIQUE INDEX idx_weather_cache_city ON weather_cache (city); -- Refresh the cache periodically CREATE OR REPLACE FUNCTION refresh_weather_cache() RETURNS VOID AS $$ BEGIN REFRESH MATERIALIZED VIEW CONCURRENTLY weather_cache; RETURN; END; $$ LANGUAGE plpgsql; -- Schedule daily refreshes with pg_cron or a similar scheduler: -- SELECT cron.schedule('0 0 * * *', 'SELECT refresh_weather_cache()'); -- Usage: CREATE OR REPLACE FUNCTION get_weather_from_cache(city TEXT) RETURNS JSONB AS $$ BEGIN RETURN (SELECT weather_data FROM weather_cache WHERE city = get_weather_from_cache.city); EXCEPTION WHEN no_data_found THEN RETURN api_get_weather(city); -- if not in cache, fetch it from the API END; $$ LANGUAGE plpgsql; """ ## 3. Coding Style and Conventions ### 3.1. Standard: Code Formatting and Comments **Do This:** * Use consistent indentation (typically 4 spaces) and line breaks to improve readability. * Add comments to explain complex logic, API calls, and data transformations. * Use meaningful names for variables, functions, and parameters. **Don't Do This:** * Write long, monolithic functions without comments or clear structure. * Use cryptic or ambiguous names. **Why:** Consistent formatting and clear comments make the code easier to understand and maintain. ### 3.2. Standard: Transaction Management **Do This:** * Wrap API calls within explicit transactions when necessary to ensure data consistency. Use "BEGIN", "COMMIT", and "ROLLBACK". * Handle potential errors during API calls gracefully and roll back the transaction if necessary. **Don't Do This:** * Leave transactions open for extended periods of time while waiting for API responses. * Commit transactions before ensuring the success of all related API calls. **Why:** Proper transaction management ensures data integrity and prevents inconsistencies. ### 3.3. Standard: Testing **Do This:** * Write unit tests for API interaction functions to verify that they handle different scenarios correctly (e.g., success, error, timeout). * Use mock APIs or stubs to isolate the database from external dependencies during testing. * Write integration tests to ensure that the database and APIs work together seamlessly. **Don't Do This:** * Skip testing API interaction code. This can lead to unexpected errors and integration issues in production. * Rely solely on manual testing. **Why:** Automated testing improves code quality, reduces the risk of regressions, and facilitates continuous integration and delivery. These API integration standards will help create reliable, secure, and maintainable PostgreSQL applications that integrate effectively with external services. Remember to stay updated with the latest PostgreSQL features and best practices as the ecosystem evolves.
# Core Architecture Standards for PostgreSQL This document outlines the coding standards for the core architecture of PostgreSQL. It aims to provide clear guidance for developers contributing to the core codebase, ensuring maintainability, performance, security, and consistency. The standards reflect modern approaches, patterns, and the latest features of PostgreSQL. ## 1. Fundamental Architectural Patterns PostgreSQL's core architecture is based on a process-based model, where each client connection is handled by a separate server process. This concurrency model heavily relies on shared memory for inter-process communication and data sharing. **Do This:** * Understand the process-based architecture deeply. Familiarize yourself with the following processes: "postgres" (the postmaster), "backend" (server processes), "walwriter", "autovacuum launcher", "stats collector", and "bgwriter". * Design extensions with process isolation in mind. Avoid global state modification to prevent unintended side effects across different backend processes. * Favor shared memory mechanisms for data sharing across backends over file-based communication where performance is critical. **Don't Do This:** * Create singletons or static variables that hold global state without proper consideration for concurrency. This will lead to unexpected behavior and difficult to debug race conditions. * Introduce shared resources without adequate locking mechanisms. * Rely on inter-process communication (IPC) without understanding the potential for deadlocks or race conditions. **Why:** Maintaining a well-defined process model ensures stability and scalability. Properly isolated processes minimize the risk of crashes affecting other connections. ### 1.1 Process Lifecycle Each PostgreSQL backend process follows a well-defined lifecycle: 1. **Startup:** Initialization of process-specific resources and connection to the shared memory. 2. **Authentication:** Verification of the client's identity. 3. **Query Processing:** Parsing, planning, and execution of SQL queries. 4. **Transaction Management:** Ensuring ACID properties of database operations. 5. **Shutdown:** Clean-up of resources and disconnection from shared memory. **Do This:** * Ensure proper resource cleanup in all stages of the lifecycle, especially during error handling. * Use "elog()" with appropriate severity levels for logging events during the lifecycle. * Catch and handle exceptions appropriately throughout the lifecycle. **Don't Do This:** * Leak resources (memory, file descriptors, etc.) during any phase of the process lifecycle. * Ignore errors during startup or shutdown. * Introduce long-running operations inside the authentication phase. **Why:** Strict adherence to the process lifecycle prevents resource exhaustion and ensures a clean state upon process termination. ### 1.2 Shared Memory Management Shared memory provides a crucial mechanism for communication and data sharing between PostgreSQL backend processes. **Do This:** * Use PostgreSQL's shared memory APIs (e.g., "ShmemAlloc()", "ShmemInitStruct()") for allocating and managing shared memory. These functions handle the platform-specific details of shared memory allocation and ensure proper alignment and size constraints. * Protect access to shared memory regions using appropriate locking mechanisms (e.g., "LWLock", "SpinLock"). * Define shared memory segments in "src/backend/utils/misc/ipc.c" or a relevant module's initialization function. **Don't Do This:** * Directly use system calls like "shmget()" and "shmat()" without going through PostgreSQL's shared memory APIs. * Assume atomicity of operations on shared memory regions. Always use locking. * Overallocate shared memory. Reserve only what is necessary. **Why:** Proper shared memory management prevents corruption, ensures data integrity, and avoids resource conflicts between processes. **Example:** """c /* Example of allocating and using shared memory */ typedef struct { int counter; LWLock lock; } MySharedData; static MySharedData *mySharedData; void initializeMySharedData(void) { bool found; mySharedData = ShmemInitStruct("MySharedData", sizeof(MySharedData), &found); if (!found) { /* Initialize shared memory on first allocation */ mySharedData->counter = 0; LWLockInitialize(&mySharedData->lock, LWLockAssign()); } } int incrementCounter(void) { int result; LWLockAcquire(&mySharedData->lock, LW_EXCLUSIVE); result = ++mySharedData->counter; LWLockRelease(&mySharedData->lock); return result; } """ ## 2. Project Structure and Organization PostgreSQL's source code is organized into a directory structure that reflects its functionality. **Do This:** * Familiarize yourself with the top-level directories: "src", "doc", "contrib", "src" is where the core source code resides. * Understand the purpose of subdirectories within "src", such as "backend", "include", and "port". * Place new code in the appropriate directory based on its functionality. * Maintain consistency in coding style and naming conventions within each directory. **Don't Do This:** * Randomly place files in arbitrary directories. * Create unnecessary dependencies between modules. * Violate the established directory structure without a clear justification. **Why:** A well-organized project structure facilitates navigation, understanding, and maintenance of the codebase. Clear directory conventions maintain code clarity. ### 2.1 Core Directories Key directories within the "src" directory include: * "src/backend": Contains the core backend code, including query processing, transaction management, storage, and indexing. * "src/include": Contains header files that define the interfaces used by the backend code. * "src/port": Contains platform-specific code. * "src/common": Contains code shared across multiple parts of the backend. * "src/fe_utils": Contains utilities used by the frontend. **Do This:** * Follow the existing directory structure when adding new features or modifying existing ones. * Create new subdirectories within existing directories if necessary to organize logically related code. * Use header files in "src/include" to define public interfaces for modules. **Don't Do This:** * Include implementation details in header files. * Create circular dependencies between directories. **Why:** A modular directory structure ensures a logical separation of concerns and minimizes dependencies between modules helping reduce build times. ### 2.2 Coding Style PostgreSQL has a well-defined coding style outlined in "doc/src/sgml/develop.sgml". **Do This:** * Adhere to the coding style guidelines regarding indentation, spacing, naming conventions, and comment formatting. * Use "pgindent" to automatically format your code. * Write concise and informative comments. **Don't Do This:** * Ignore the coding style guidelines. * Write lengthy or redundant comments. * Use inconsistent naming conventions. **Why:** Consistent coding style improves readability and maintainability of the code. "pgindent" ensures code conforms to the standard style automatically. ## 3. Modern Approaches and Patterns Modern PostgreSQL development emphasizes several key approaches: * **Extensibility:** PostgreSQL is designed to be extensible through extensions. * **Concurrency:** Handling multiple concurrent connections efficiently is crucial. * **Security:** Preventing vulnerabilities and ensuring data integrity are paramount. ### 3.1 Extension Development Extensions are the primary way to add new functionality to PostgreSQL. **Do This:** * Use the Extension Control File (".control") to define the extension's metadata. * Provide SQL scripts for creating and dropping database objects. * Use hooks ("ExecutorStart_hook", "ExecutorRun_hook", etc.) to extend the core functionality. * Follow the security guidelines for extension development. **Don't Do This:** * Modify the core PostgreSQL code directly (unless absolutely necessary and approved by the community). * Introduce security vulnerabilities through insecure extension code. * Make assumptions about the internal implementation details of PostgreSQL that could change in future versions. **Why:** Extensions allow adding new features without modifying the core code. **Example:** """sql -- Example SQL script for creating a function in an extension CREATE FUNCTION my_extension_function(text) RETURNS text AS '$libdir/my_extension', 'my_extension_function' LANGUAGE C IMMUTABLE STRICT; """ ### 3.2 Concurrency Control PostgreSQL uses Multi-Version Concurrency Control (MVCC) to manage concurrent access to data. **Do This:** * Understand MVCC and its implications for data consistency. * Use appropriate transaction isolation levels to prevent data anomalies. * Minimize lock contention by optimizing queries and using appropriate indexing strategies. * When working with internal data structures, be mindful of concurrent access and utilize PostgreSQL's locking primitives (LWLock, spinlocks) appropriately. **Don't Do This:** * Ignore the potential for data anomalies when using low transaction isolation levels. * Introduce unnecessary locking that could lead to deadlocks. * Perform long-running operations within a single transaction. **Why:** MVCC ensures data consistency and allows concurrent access to data. ### 3.3 Security Best Practices Security is a critical aspect of PostgreSQL development. **Do This:** * Follow secure coding practices to prevent vulnerabilities such as SQL injection and buffer overflows. * Use hardened APIs to avoid common security pitfalls. * Validate input data carefully. * Avoid hardcoding sensitive information such as passwords. * Be aware of the security implications of new features. **Don't Do This:** * Ignore security warnings. * Implement custom encryption algorithms (use PostgreSQL's built-in encryption features). * Grant excessive privileges to users or roles. **Why:** Secure coding practices are essential for preventing data breaches and ensuring the integrity of the database. ### 3.4 Memory Management Efficient Memory management is key to PostgreSQL's performance and stability. **Do This:** * Use PostgreSQL's memory context mechanism ("MemoryContext") for allocating and freeing memory within a query lifecycle. This mechanism provides automatic memory cleanup at the end of a query preventing memory leaks. * Understand the different memory contexts (e.g., "TopMemoryContext", "QueryMemoryContext") and use them appropriately. * Avoid manual memory management ("malloc"/"free") unless absolutely necessary (and only if you REALLY know what you are doing). Use PostgreSQL's "palloc"/"pfree" within a memory context. * Profile memory usage to identify and fix memory leaks. **Don't Do This:** * Leak memory by failing to free allocated memory. * Allocate large amounts of memory without considering the impact on performance. * Use "malloc"/"free" without a deep understanding of PostgreSQL's memory management. **Why:** Efficient memory management prevents memory leaks, reduces memory fragmentation, and improves overall performance. The memory context system automates this and integrates with the query processing lifecycle. **Example:** """c /* Example using MemoryContext */ MemoryContext myContext; char *data; /* Create a new memory context */ myContext = AllocSetContextCreate(CurrentMemoryContext, "MyContext", ALLOCSET_DEFAULT_SIZES); /* Switch to the new memory context */ MemoryContext oldContext = MemoryContextSwitchTo(myContext); /* Allocate memory within the new context */ data = palloc(100); /* Switch back to the previous memory context. The 'data' still exists */ MemoryContextSwitchTo(oldContext); /* ... use data ... */ /* At the end, the memory context 'myContext' is destroyed, and all memory allocated within it is automatically freed */ MemoryContextDelete(myContext); """ These standards aim to provide a comprehensive guide for contributing to the core architecture of PostgreSQL, by promoting best practices and ensuring code maintainability, performance, and security. By following these guidelines, developers can help ensure that PostgreSQL remains a robust, reliable, and extensible database system.