Fine-grained Source Code Changes Research Articles

Comparing fine-grained source code changes and code churn for bug prediction - A retrospective

More than two decades ago, researchers started to mine the data stored in software repositories to help software developers in making informed decisions for developing and testing software systems. Bug prediction was one of the most promising and popular research directions that uses the data stored in software repositories to predict the bug-proneness or number of bugs in source files. On that topic and as part of Emanuel's PhD studies, we submitted a paper with the title Comparing fine-grained source code changes and code churn for bug prediction [8] to the 8th Working Conference on Mining Software Engineering, held 2011 in beautiful Honolulu, Hawaii. Ten years later, it got selected as one of the finalists to receive the MSR 2021 Most Influential Paper Award. In the following, we provide a retrospective on our work, describing the road to publishing this paper, its impact in the field of bug prediction, and the road ahead.

Augmenting commit classification by using fine-grained source code changes and a pre-trained deep neural language model

Analyzing software maintenance activities is very helpful in ensuring cost-effective evolution and development activities. The categorization of commits into maintenance tasks supports practitioners in making decisions about resource allocation and managing technical debt. In this paper, we propose to use a pre-trained language neural model, namely BERT (Bidirectional Encoder Representations from Transformers) for the classification of commits into three categories of maintenance tasks — corrective, perfective and adaptive. The proposed commit classification approach will help the classifier better understand the context of each word in the commit message. We built a balanced dataset of 1793 labeled commits that we collected from publicly available datasets. We used several popular code change distillers to extract fine-grained code changes that we have incorporated into our dataset as additional features to BERT’s word representation features. In our study, a deep neural network (DNN) classifier has been used as an additional layer to fine-tune the BERT model on the task of commit classification. Several models have been evaluated to come up with a deep analysis of the impact of code changes on the classification performance of each commit category. Experimental results have shown that the DNN model trained on BERT’s word representations and Fixminer code changes ([email protected]+Fix_cc) provided the best performance and achieved 79.66% accuracy and a macro-average f1 score of 0.8. Comparison with the state-of-the-art model that combines keywords and code changes ([email protected]+CD_cc) has shown that our model achieved approximately 8% improvement in accuracy. Results have also shown that a DNN model using only BERT’s word representation features achieved an improvement of 5% in accuracy compared to the [email protected]+CD_cc model.

A study on the changes of dynamic feature code when fixing bugs: towards the benefits and costs of Python dynamic features

Dynamic features in programming languages support the modification of the execution status at runtime, which is often considered helpful in rapid development and prototyping. However, it was also reported that some dynamic feature code tends to be change-prone or error-prone. We present the first study that analyzes the changes of dynamic feature code and the roles of dynamic features in bug-fix activities for the Python language. We used an AST-based differencing tool to capture fine-grained source code changes from 17926 bug-fix commits in 17 Python projects. Using this data, we conducted an empirical study on the changes of dynamic feature code when fixing bugs in Python. First, we investigated the characteristics of dynamic feature code changes, by comparing the changes between dynamic feature code and non-dynamic feature code when fixing bugs, and comparing dynamic feature changes between bug-fix and non-bugfix activities. Second, we explored 226 bug-fix commits to investigate the motivation and behaviors of dynamic feature changes when fixing bugs. The study results reveal that (1) the changes of dynamic feature code are significantly related to bug-fix activities rather than non-bugfix activities; (2) compared with non-dynamic feature code, dynamic feature code is inserted or updated more frequently when fixing bugs; (3) developers often insert dynamic feature code as type checks or attribute checks to fix type errors and attribute errors; (4) the misuse of dynamic features introduces bugs in dynamic feature code, and the bugs are often fixed by adding a check or adding an exception handling. As a benefit of this paper, we gain insights into the manner in which developers and researchers handle the changes of dynamic feature code when fixing bugs.