The DataWorks Copilot Code Programming Assistant uses natural language to help you efficiently generate, optimize, explain, and test SQL and Python code. You can access this feature through the intelligent editor or Copilot Ask. This topic describes the core capabilities and scenarios of the code programming assistant in detail.
Function overview
The DataWorks Copilot Code Programming Assistant is based on a Large Language Model (LLM) and provides intelligent SQL and Python programming support for data developers. It supports two interaction modes: the intelligent code editor (for real-time completion and right-click shortcuts) and Copilot Chat (Ask mode) (for natural language Q&A interaction). Its core capabilities include code generation, refactoring, debugging, optimization, explanation, comment generation, test case design, syntax Q&A, intelligent Notebook cell creation, and cross-engine quick table search. You can add context, such as tables, nodes, data albums, rules, or local files, to improve the accuracy of its responses. It also supports switching between multiple large language models. The assistant is out-of-the-box and requires no coding experience. It significantly improves the efficiency of data modeling, extract, transform, and load (ETL) development, and debugging. This facilitates accessible, high-quality, and efficient data development.
Feature entry points
Intelligent code editor
Scenario 1: Intelligent code completion
Instructions: When developing code-based nodes, Copilot intelligently predicts and recommends code snippets based on the context, such as your code and referenced table schemas. Completion suggestions appear automatically. You can press the Tab key to accept a suggestion.
Scenario 2: Right-click menu shortcuts
Instructions: In the intelligent code editor, you can select the desired code, right-click, and choose Copilot from the shortcut menu.
Copilot Chat (Ask mode)
Ask mode is the default mode for Copilot Chat and is ideal for solving specific coding problems through a question-and-answer format. It helps you complete tasks such as code generation, code refactoring, code debugging, comment generation, code explanation, code optimization, code testing, code Q&A, intelligent Notebook cell generation, and quick table search. While using Copilot Chat in Ask mode, you can select code in the editor to provide context for your requests.
Getting started
This section uses Copilot Chat (Ask mode) as an example to help you get started quickly.
Step 1: Open Copilot Chat (Ask mode)
Log on to the DataWorks console. In the navigation pane on the left, choose . On the page that appears, select the desired workspace and click Enter Data Development (Data Studio).
Click the
icon in the upper-right corner of the top navigation bar on the Data Studio page to open Copilot Chat in Ask mode.
Step 2: Add context (Optional)
To help Copilot better understand your requirements, you can add context. Enter @ in the dialog box or click the @ icon in the lower-right corner of the dialog box to use this feature. This lets you select the type of context to add.
The supported types are:
Table: Reference the metadata of one or more tables.
Node/Code file: Reference the code in a specific node.
Data album: Reference a data album from Data Map.
Rules: Temporarily apply one or more rules to the current conversation.
Local file: Upload a local document as background information.
Step 3: Switch the large language model (Optional)
By default, Copilot uses the default model. To change this, click the 
icon at the bottom of the dialog box and select a different large language model from the menu. The supported models are shown in the following figure.
Step 4: Submit a question and engage in a multi-turn conversation
Enter your requirements in the dialog box. You can use a multi-turn conversation to refine your request by asking follow-up questions or providing additional information. This helps Copilot fully understand your intent and produce the desired result.
Detailed features
Copilot Chat in Ask mode can generate SQL and Python code from natural language. It also provides features such as code continuation, refactoring, optimization, explanation, debugging, and test case generation. The core features include the following:
Capability Type | Description | Example (What you can say to Copilot) |
| Generates code based on your instructions. | "Help me write an SQL query to find the top three products with the highest sales in each city from the |
| Refactors the specified code based on your requirements. | "Rewrite this SQL query that uses |
| Finds and fixes errors in the specified code. | "This SQL query fails with an |
| Generates comments for the specified code. | "Add comments to this complex SQL logic to explain the purpose of each common table expression (CTE)." |
| Explains the specified code. | "Explain what |
| Optimizes the specified code. | "This query is very slow. Help me optimize it and check for any performance bottlenecks." |
| Provides a test plan for the specified code. | "Design some test cases for this SQL query that calculates user retention rate. What edge cases should be considered?" |
| Answers questions about code syntax, functions, and more. | "What is the difference between the |
| Intelligently generates a code cell in a Notebook. | "Create a cell that uses pandas to read the |
| Searches for a table by keyword. | "Find all tables related to 'user'." |
Code generation/SQL generation
Description: Generates code based on your instructions in natural language.
Instructions: Use one of the following methods:
In the code editor, you can right-click a blank area and choose to open the Copilot interface. Then, enter your request in natural language to have the large language model generate the code.
In the code editor, click the
icon in the upper-right corner of the top navigation bar on the DataStudio page to open Copilot Chat in Ask mode. In the chat input box, enter /, select Code Generation, and then enter your request in natural language. The Large Language Model (LLM) will then generate the required code.
Code refactoring/SQL refactoring
Description: Refactors the specified code based on your instructions in natural language.
Instructions: Use one of the following methods:
In the code editor, you can select the code you want to refactor, right-click, and choose . This opens the Copilot interface where you can enter your requirements.
In the code editor, select the code you want to refactor and click the
icon in the upper-right corner of the DataStudio top navigation bar to open Copilot Chat in Ask mode. In the chat input box, enter /, select Rewrite Code, enter your requirements, and click Send. Copilot will then return the result.
Code debugging/SQL debugging
Description: DataWorks Copilot can find and fix errors in a code snippet.
Instructions: Use one of the following methods:
In the code editor, select the target code, right-click the selection, and then select to open the Copilot interface.
In the code editor, select the code you want to debug and click the
icon in the upper-right corner of the Data Studio page's top navigation bar to open Copilot Chat in Ask mode. In the dialog box, enter /, select Code Error correction, and click Send. Copilot will then return the result.
Comment generation
Description: DataWorks Copilot can generate comments for SQL code to improve its readability.
Instructions: Use one of the following methods:
In the code editor, you can select the code, right-click, and choose to open the Copilot interface.
In the code editor, select the code and click the
icon in the upper-right corner of the Data Studio top navigation bar to open Copilot Chat in Ask mode. In the chat input box, enter /, select Generate Comments, and click Send. Copilot will then return the result.
Code explanation
Description: DataWorks Copilot can explain SQL code to improve its readability and help you understand its logic.
Instructions: In the code editor, select the code you want explained and click the icon in the upper-right corner of the Data Studio top navigation bar to open Copilot Chat in Ask mode. In the chat input box, enter /, select Code Explanation, and click Send. Copilot will then return the result.
Code optimization
Description: DataWorks Copilot can optimize SQL code to simplify its logic, improve execution efficiency, and reduce the database load.
Instructions: In the code editor, select the code you want to optimize and click the icon in the upper-right corner of the Data Studio navigation bar to open Copilot Chat in Ask mode. In the chat input box, enter /, select Code Optimization, and click Send. Copilot will then return the results.
Code testing
Description: DataWorks Copilot can provide a test plan for your SQL code. This helps you generate test cases to verify that each part of the task works as expected.
Instructions: Click the icon in the upper-right corner of the DataStudio page's top navigation bar to open Copilot Chat in Ask mode. In the chat input box, enter / and select Code Test. Then, select the code you want to test in the editor, click Send, and wait for Copilot to return the results.
Code Q&A
Description: DataWorks Copilot can answer questions about SQL syntax or MaxCompute functions by providing explanations and usage examples.
Instructions: Click the icon in the upper-right corner of the Data Studio page's top navigation bar to open Copilot in Ask mode. In the dialog box, enter your question, click Send, and wait for Copilot to return the result.
Intelligent Notebook cell generation
Description: You can enter a keyword to generate a Notebook cell.
Instructions: Click the icon in the upper-right corner of the DataStudio page's top navigation bar to open Copilot in Ask mode. In the dialog box, enter / and select Generate Notebook Cell. Then, enter keywords in the input field and click Send. Copilot will generate the corresponding Notebook node.
Quick table search
Description: You can enter a keyword to search for tables across all compute engines and data sources.
Instructions: Click the icon in the upper-right corner of the DataStudio page's top navigation bar to open Copilot in Ask mode. In the chat input box, enter / and select Quick Table Search. Then, enter your keywords in the input field and click Send. Copilot will return a list of tables from all attached compute engines and data sources that match your keywords.
