Supplementary Material - Dataset for "Automating Quantum Software Maintenance: Flakiness Detection and Root Cause Analysis"

README

The dataset consists of the following components: - `prompts.txt`: This file contains the prompts used for large language models. - `Dataset` directory: includes general information about the dataset. Specifically, the `dataset.xlsx` file lists flaky and non-flaky tests, along with their root causes and fix types. - `Full` directory contains two subdirectories: `Flaky` and `Non-flaky`. Each of these directories is organized by individual GitHub organization projects, with each project having its list of repository subdirectories. These subdirectories are further divided into “issues”  and “pull requests” (PRs).

- `Method` level subdirectory has a similar structure but contains extracted code snippets at the method level instead of full code listings. The `code.diff` file is copied over and left unaltered.  

- Issue Directories (IRs): Named with an `issueID` template, each issue directory contains a `log.issue` file that includes the extracted description, comments, and metadata. - PR Directories (PRs): Named using the `prID` template, each PR directory contains the text, comments, and metadata in the `pr.log` file. The text of the associated issue is stored in the `log.issue` file. Code listings are stored in a file with the `.bug` suffix, while the corresponding fixed version is in a `.fix` file. The `code.diff` file contains the patch that transforms the `.bug` version into the `.fix` version.

Additional notes:Issues with associated pull requests in `dataset.xlsx` are combined into the pull request directory template. If two pull requests are listed for a row, a PR directory is created for each. Due to updates in the extended dataset, some repositories have been renamed or archived, meaning the current repository directory names in `Dataset` will include both the previous and new names if it has been changed (e.g., a repository previously saved as Qiskit/qiskit-terra may now be saved as Qiskit/qiskit following the renaming from qiskit-terra to qiskit).

Directory Structure:

├── prompts.txt├── Dataset/  └── dataset.xlsx├── Full/   ├── Flaky/      └── //...   ├── Non-Flaky/      └── //...├── Method/   ├── Flaky/     └── //...   ├── Non-flaky/       └── //...