PDLogger: Automated Logging Framework for Practical Software Development

2025-07-26

AI Summary

Existing log generation approaches focus on isolated, single-point predictions (e.g., individual log locations, levels, or messages), neglecting intra-method multi-log coordination and cross-method semantic dependencies, while offering limited variable coverage—leading to incomplete logs and poor practical adaptability. This paper proposes PDLogger, the first end-to-end log generation framework designed for realistic multi-log scenarios. It jointly models log location, level, message, and variables via block-type-aware prompting, reverse program slicing, and generalized variable expression expansion. To enhance consistency, it incorporates hierarchical correction and context-sensitive deduplication. Evaluated on 3,113 real-world logs, PDLogger outperforms state-of-the-art methods by +139.0% in location precision, +69.2% in F1-score, +82.3% in level accuracy, +131.8% in variable precision, and +65.7% in message BERTScore.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Knowledge Representation and Reasoning: Logic Programming
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Logging is indispensable for maintaining the reliability and diagnosability of modern software, yet developers still struggle to decide where and how to log effectively. Existing automated logging techniques focus on isolated sub-tasks - predicting a single log position, level, or message - and therefore cannot produce complete, high-quality log statements that reflect real-world practice in which multiple logs often appear inside one method. They also neglect deeper semantic dependencies among methods and consider only a narrow set of candidate variables, leading to superficial or incomplete logs. In this paper, we present PDLogger, the first end-to-end log generation technique expressly designed for practical, multi-log scenarios. PDLogger operates in three phases. (1) Log position prediction: block-type-aware structured prompts guide a large language model (LLM) to suggest candidate positions across all control-flow blocks of a method. (2) Log generation: backward program slicing supplies precise inter-procedural control and data-dependency context, while an expanded variable extractor captures both member and external function expressions; the enriched prompt enables the LLM to emit a full log statement (position, level, message, variables). (3) Log refinement: level correction and context-sensitive deduplication prune false positives and redundant logs. We evaluate PDLogger on 3,113 log statements drawn from two widely used Java projects. Compared with the strongest prior systems, PDLogger improves log-position precision by 139.0 percent, F1 by 69.2 percent, level accuracy by 82.3 percent, variable precision by 131.8 percent, and message quality (BERTScore) by 65.7 percent. The framework consistently performs well with different mainstream LLMs, demonstrating robustness and generality. PDLogger's implementation is available as open source to foster future research and adoption.

Problem

• Automating log position prediction in multi-log scenarios
• Generating complete log statements with semantic dependencies
• Improving log quality and reducing redundancy in logging

Innovation

• Block-type-aware structured prompts for log positions
• Backward program slicing for dependency context
• Level correction and deduplication for refinement