Datasets for Paper "Towards Understanding Bugs in An Open Source Cloud Management Stack: An Empirical Study of OpenStack Software Bugs"

1.Folder 1-48255 original data contains the original XML files obtained from OpenStack bug stack, including 48255 bug reports which have Critical, High and Medium severity level and Complete and Fixed status. We uploaded bug607068 as a sample and uploaded other remaining data as two zip files for it has more than 40000 files, while GitHub only allows uploading up to 100 files at a time. All these data were obtained using a data crawl tool. Specific steps are as follows: 1)Create a new task. 2)Add an opened web page and enter the web address. 3)Set the page of the OpenStack bug page to loop so it could flip automatically after each page is obtained. 4)Create a group of elements so that the data could be crawled one by one. 5)Open each element in the group and crawl the required data. 6) Loop Step 3-4-5 until the task has been completed. 2.Folder 2-32547 bugs is XML files that have been preprogressed using basic text parsing and discourse matching method. This file includes 32547 satisfactory bug reports. We also uploaded them as two zip files and a sample file. 3.Folder 3-Preprogress contains the preprogressing code. 4.Folder 4-800 bugs for analysis contains 800 bug reports which have randomly selected for analysis. The folder has a XML file as a sample and a zip file including other needed data. 5.Folder 5-Benchmarks and data analysis results contains 3 Excel files. Benchmarks including the analyzed bugs in different dimension and each sheet is correspond to a statistical approach, 5 sheets in total. Deployment of components contains the information of deployments of components obtained from the OpenStack bug repository. Fixing information of duration and comments includes the information of each bug, each sheet is corresponding to a severity, all captured from the original data using a data crawling tool.1. 文件夹1-48255原始数据：包含从OpenStack缺陷栈获取的原始XML文件，内含48255份缺陷报告，其严重等级涵盖Critical（致命）、High（高）与Medium（中），状态为Complete（已完成）与Fixed（已修复）。我们上传了bug607068作为示例样本；由于文件总数超过40000个，且GitHub单次仅允许上传最多100个文件，因此将剩余数据打包为两个ZIP压缩文件上传。所有数据均通过数据爬取工具获取，具体爬取步骤如下：① 创建新任务；② 添加已打开的网页并输入目标网址；③ 设置OpenStack缺陷页面循环加载，使每页数据抓取完成后自动翻页；④ 创建元素组以实现逐条抓取数据；⑤ 打开元素组中的每个条目并提取所需数据；⑥ 循环执行步骤③至⑤直至任务完成。 2. 文件夹2-32547缺陷数据：内含通过基础文本解析与语篇匹配方法完成预处理的XML文件，共包含32547份合格缺陷报告。我们同样将其打包为两个ZIP压缩文件与一个示例文件上传。 3. 文件夹3-预处理代码：存放预处理相关代码。 4. 文件夹4-待分析800份缺陷数据：内含随机选取的800份缺陷报告用于分析任务。该文件夹包含一份XML格式示例文件，其余所需数据打包为一个ZIP压缩文件上传。 5. 文件夹5-基准测试集与数据分析结果：内含3个Excel文件。基准测试集涵盖多维度分析的缺陷数据，每个工作表对应一种统计分析方法，共包含5个工作表。其中“组件部署信息”板块收录从OpenStack缺陷仓库获取的组件部署相关数据；“修复时长与注释信息”板块收录每份缺陷的对应信息，每个工作表对应一种缺陷严重等级，所有数据均通过数据爬取工具从原始数据中提取得到。