Workflow Code Generation from Function Description Text
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low-code development tools heavily depend on manual design, lacking intelligent generation and assistant capabilities, which affects development efficiency and fails to meet complex service requirements.
Innovation Solution
A code development platform that intelligently generates workflow code by mining and analyzing function description text to identify logical relationships between call statements, reducing developer dependency and automating trigger/action parameter filling, while supporting complex service logic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If low-code development tools are used to build applications, then development complexity is reduced, but development efficiency deteriorates due to heavy reliance on manual design
Solution Approach 1:
The system enables self-service through automated code generation from natural language descriptions. The code generation module automatically creates workflow code without requiring manual programming, allowing the system to serve itself in generating executable code from user-friendly descriptions.
Solution Approach 2:
Manual mechanical design work is replaced by an intelligent system that uses natural language processing and code generation technologies. The system substitutes the manual mechanical process of coding with an automated computational process that converts natural language descriptions into executable workflow code.
2Adaptability or versatility
If manual design is used for workflow development, then flexibility is maintained, but time consumption increases significantly
Solution Approach 1:
The system performs preliminary action by pre-defining workflow templates and structures that can be automatically instantiated. Common workflow patterns are prepared in advance, allowing the system to quickly generate code by filling in specific parameters rather than creating everything from scratch each time.
Solution Approach 2:
The system utilizes parameter changes to adapt workflows to different scenarios. By modifying parameters in pre-defined templates rather than rewriting entire workflows, the system maintains flexibility while significantly reducing the time required to develop new workflows.
3Productivity
If intelligent generation capabilities are added to low-code tools, then development efficiency improves, but system complexity increases
Solution Approach 1:
The system introduces an intermediary layer between natural language input and code output. This intermediary includes natural language processing modules that translate user descriptions into structured representations, which then guide the code generation process, simplifying the overall complexity by breaking it into manageable stages.
Solution Approach 2:
The intelligent generation system is segmented into distinct functional modules: natural language processing, logical relationship analysis, code generation, and validation. Each module handles a specific aspect of the workflow, reducing overall system complexity by dividing the complex task into smaller, more manageable components.
4Extent of automation
If automated code generation is implemented, then dependency on developers is reduced, but accuracy of generated code may deteriorate
Solution Approach 1:
The system implements feedback mechanisms where generated code is validated against the original natural language description and predefined workflow templates. If inconsistencies are detected, the system can request clarification from the user or automatically adjust the generated code to better match the intended functionality, thereby maintaining accuracy while preserving high automation levels.
Data Source
AI summary
A code development method, applied to a code development platform, includes receiving function description information that is of a target workflow and that is input by a user; obtaining, based on the function description information, a function description text including call statements, where the call statements include a trigger call statement used to start the target workflow and an action interface call statement used to execute the target workflow; and then mining and analyzing the function description text to obtain a logical relationship between the call statements. The code development method further includes generating code of the target workflow based on the logical relationship between the call statements, and rendering a view of the target workflow based on the code of the target workflow.