Automated Test Data Generation for Enterprise Protobuf Systems: A Metaclass-Enhanced Statistical Approach
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911082163994624 |
|---|---|
| author | Du, Y. |
| author_facet | Du, Y. |
| contents | Large-scale enterprise systems utilizing Protocol Buffers (protobuf) present significant challenges for performance testing, particularly when targeting intermediate business interfaces with complex nested data structures. Traditional test data generation approaches are inadequate for handling the intricate hierarchical and graph-like structures inherent in enterprise protobuf schemas. This paper presents a novel test data generation framework that leverages Python's metaclass system for dynamic type enhancement and statistical analysis of production logs for realistic value domain extraction. Our approach combines automatic schema introspection, statistical value distribution analysis, and recursive descent algorithms for handling deeply nested structures. Experimental evaluation on three real-world enterprise systems demonstrates up to 95\% reduction in test data preparation time and 80\% improvement in test coverage compared to existing approaches. The framework successfully handles protobuf structures with up to 15 levels of nesting and generates comprehensive test suites containing over 100,000 test cases within seconds. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_22070 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Automated Test Data Generation for Enterprise Protobuf Systems: A Metaclass-Enhanced Statistical Approach Du, Y. Software Engineering Computational Engineering, Finance, and Science Programming Languages Large-scale enterprise systems utilizing Protocol Buffers (protobuf) present significant challenges for performance testing, particularly when targeting intermediate business interfaces with complex nested data structures. Traditional test data generation approaches are inadequate for handling the intricate hierarchical and graph-like structures inherent in enterprise protobuf schemas. This paper presents a novel test data generation framework that leverages Python's metaclass system for dynamic type enhancement and statistical analysis of production logs for realistic value domain extraction. Our approach combines automatic schema introspection, statistical value distribution analysis, and recursive descent algorithms for handling deeply nested structures. Experimental evaluation on three real-world enterprise systems demonstrates up to 95\% reduction in test data preparation time and 80\% improvement in test coverage compared to existing approaches. The framework successfully handles protobuf structures with up to 15 levels of nesting and generates comprehensive test suites containing over 100,000 test cases within seconds. |
| title | Automated Test Data Generation for Enterprise Protobuf Systems: A Metaclass-Enhanced Statistical Approach |
| topic | Software Engineering Computational Engineering, Finance, and Science Programming Languages |
| url | https://arxiv.org/abs/2507.22070 |