Overlong Source Code Detection With Segmented Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing software refactoring tools struggle to accurately identify overlong methods in source code due to the use of single line count thresholds, resulting in numerous false positives and non-actionable results, making it difficult to improve code quality and maintainability.
Innovation Solution
A system utilizing machine learning, natural language processing, and static/dynamic code analysis applies tailored code length thresholds and filters to identify and prioritize overlong code segments for refactoring, considering metrics like autogeneration, method type, and cyclomatic complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single line count threshold is used to identify overlong methods, then the identification process is simple and fast, but the result contains numerous false positives and is non-actionable
Solution Approach 1:
The patent segments the identification process into multiple stages: filtering out autogenerated code segments, separating test methods from normal methods, and applying different evaluation criteria to each segment type. This segmentation allows for more precise identification by treating different code segments according to their specific characteristics rather than using a uniform threshold.
Solution Approach 2:
The patent applies local quality by using tailored code length thresholds specific to each code segment type. Test methods receive different threshold criteria compared to normal methods, and autogenerated segments are excluded entirely. This localized approach improves measurement precision by matching the evaluation criteria to the specific properties of each code segment.
2Measurement precision
If multiple filters and tailored thresholds are applied to each code segment, then the identification accuracy improves and false positives are reduced, but the complexity of the identification system increases
Solution Approach 1:
The patent reduces system complexity through segmentation by dividing code segments into distinct categories (autogenerated, test methods, normal methods) that can be processed independently. Each category has its own simplified evaluation rules, which reduces the overall computational complexity compared to applying a single complex evaluation system to all code uniformly.
Solution Approach 2:
The patent manages complexity by dynamically changing evaluation parameters based on the detected type of code segment. The system automatically adjusts which thresholds and criteria are applied based on the segment classification, allowing for high precision without requiring a permanently complex system structure for all cases.
3Reliability
If code segments are filtered based on multiple metrics including autogeneration and method type, then the actionable identification of overlong methods improves, but the time required for analysis increases
Solution Approach 1:
The patent applies preliminary action by performing filtering operations before the main evaluation process. Autogenerated code segments are identified and removed early in the process, and test methods are separated from normal methods before applying the overlong method detection thresholds. This preliminary filtering reduces the amount of code that requires detailed analysis, thereby reducing overall analysis time while maintaining high reliability.
Data Source
AI summary
A method and apparatus are disclosed for finding overlong source code segments (e.g., methods) by evaluating input source code segments for a plurality of predetermined code metric values in order to identify candidate source code segments (e.g., non-autogenerated methods) which do not meet a first code metric value and to assess each candidate source code segment against a second code metric value to identify different sets of candidate source code segments (e.g., test methods and normal methods) so that each set of candidate source code segments may be assessed against a tailored set of code length thresholds to identify any overlong source code segment having a code length which meets or exceeds at least two of the tailored set of code length thresholds.