dataset_info:
- config_name: bugs
features:
- name: fixCommitSHA1
dtype: large_string
- name: fixCommitParentSHA1
dtype: large_string
- name: bugFilePath
dtype: large_string
- name: fixPatch
dtype: large_string
- name: projectName
dtype: large_string
- name: bugLineNum
dtype: int64
- name: bugNodeStartChar
dtype: int64
- name: bugNodeLength
dtype: int64
- name: fixLineNum
dtype: int64
- name: fixNodeStartChar
dtype: int64
- name: fixNodeLength
dtype: int64
- name: sourceBeforeFix
dtype: large_string
- name: sourceAfterFix
dtype: large_string
splits:
- name: train
num_bytes: 69913130
num_examples: 25539
download_size: 13039503
dataset_size: 69913130
- config_name: bugsLarge
features:
- name: fixCommitSHA1
dtype: large_string
- name: fixCommitParentSHA1
dtype: large_string
- name: bugFilePath
dtype: large_string
- name: fixPatch
dtype: large_string
- name: projectName
dtype: large_string
- name: bugLineNum
dtype: int64
- name: bugNodeStartChar
dtype: int64
- name: bugNodeLength
dtype: int64
- name: fixLineNum
dtype: int64
- name: fixNodeStartChar
dtype: int64
- name: fixNodeLength
dtype: int64
- name: sourceBeforeFix
dtype: large_string
- name: sourceAfterFix
dtype: large_string
splits:
- name: train
num_bytes: 432646122
num_examples: 153652
download_size: 76500373
dataset_size: 432646122
- config_name: sstub_input_no_comments
features:
- name: identifier
dtype: int64
- name: condition
dtype: large_string
- name: bug
dtype: large_string
- name: fix
dtype: large_string
- name: post_conditional
dtype: large_string
- name: bugType
dtype: large_string
- name: buggy_commit
dtype: large_string
- name: repository_url
dtype: large_string
- name: bug_timestamp
dtype: int64
- name: num_commits_til_fix
dtype: int64
- name: originating_commit
dtype: large_string
splits:
- name: train
num_bytes: 601223201
num_examples: 16899
download_size: 158694988
dataset_size: 601223201
- config_name: sstubs
features:
- name: bugType
dtype: large_string
- name: fixCommitSHA1
dtype: large_string
- name: fixCommitParentSHA1
dtype: large_string
- name: bugFilePath
dtype: large_string
- name: fixPatch
dtype: large_string
- name: projectName
dtype: large_string
- name: bugLineNum
dtype: int64
- name: bugNodeStartChar
dtype: int64
- name: bugNodeLength
dtype: int64
- name: fixLineNum
dtype: int64
- name: fixNodeStartChar
dtype: int64
- name: fixNodeLength
dtype: int64
- name: sourceBeforeFix
dtype: large_string
- name: sourceAfterFix
dtype: large_string
splits:
- name: train
num_bytes: 22474181
num_examples: 10231
download_size: 3355994
dataset_size: 22474181
- config_name: sstubsLarge
features:
- name: bugType
dtype: large_string
- name: fixCommitSHA1
dtype: large_string
- name: fixCommitParentSHA1
dtype: large_string
- name: bugFilePath
dtype: large_string
- name: fixPatch
dtype: large_string
- name: projectName
dtype: large_string
- name: bugLineNum
dtype: int64
- name: bugNodeStartChar
dtype: int64
- name: bugNodeLength
dtype: int64
- name: fixLineNum
dtype: int64
- name: fixNodeStartChar
dtype: int64
- name: fixNodeLength
dtype: int64
- name: sourceBeforeFix
dtype: large_string
- name: sourceAfterFix
dtype: large_string
splits:
- name: train
num_bytes: 148605032
num_examples: 63923
download_size: 21951751
dataset_size: 148605032
- config_name: topJavaMavenProjects
features:
- name: repository_url
dtype: large_string
- name: forks
dtype: int64
- name: watchers
dtype: int64
- name: normForks+normWatchers
dtype: float64
splits:
- name: train
num_bytes: 7028
num_examples: 100
download_size: 6282
dataset_size: 7028
- config_name: topProjects
features:
- name: repository_url
dtype: large_string
- name: forks
dtype: int64
- name: watchers
dtype: int64
- name: normForks+normWatchers
dtype: float64
splits:
- name: train
num_bytes: 264573273
num_examples: 3114516
download_size: 74470636
dataset_size: 264573273
configs:
- config_name: bugs
data_files:
- split: train
path: bugs/train-*
- config_name: bugsLarge
data_files:
- split: train
path: bugsLarge/train-*
- config_name: sstub_input_no_comments
data_files:
- split: train
path: sstub_input_no_comments/train-*
- config_name: sstubs
data_files:
- split: train
path: sstubs/train-*
- config_name: sstubsLarge
data_files:
- split: train
path: sstubsLarge/train-*
- config_name: topJavaMavenProjects
data_files:
- split: train
path: topJavaMavenProjects/train-*
- config_name: topProjects
data_files:
- split: train
path: topProjects/train-*
ManySStuBs4J
About
The ManySStuBs4J corpus is a collection of simple fixes to Java bugs, designed for evaluating program repair techniques. We collect all bug-fixing changes using the SZZ heuristic, and then filter these to obtain a data set of small bug fix changes. These are single statement fixes, classified where possible into one of 16 syntactic templates which we call SStuBs. The dataset contains simple statement bugs mined from open-source Java projects hosted in GitHub. There are two variants of the dataset. One mined from the 100 Java Maven Projects and one mined from the top 1000 Java Projects. A project's popularity is determined by computing the sum of z-scores of its forks and watchers. We kept only bug commits that contain only single statement changes and ignore stylistic differences such as spaces or empty as well as differences in comments. Some single statement changes can be caused by refactorings, like changing a variable name rather than bug fixes. We attempted to detect and exclude refactorings such as variable, function, and class renamings, function argument renamings or changing the number of arguments in a function. The commits are classified as bug fixes or not by checking if the commit message contains any of a set of predetermined keywords such as bug, fix, fault etc. We evaluated the accuracy of this method on a random sample of 100 commits that contained SStuBs from the smaller version of the dataset and found it to achieve a satisfactory 94% accuracy. This method has also been used before to extract bug datasets (Ray et al., 2015; Tufano et al., 2018) where it achieved an accuracy of 96% and 97.6% respectively.
The bugs are stored in a JSON file (each version of the dataset has each own instance of this file). Any bugs that fit one of 16 patterns are also annotated by which pattern(s) they fit in a separate JSON file (each version of the dataset has each own instance of this file). We refer to bugs that fit any of the 16 patterns as simple stupid bugs (SStuBs).
For more information on extracting the dataset and a detailed documentation of the software visit the GitHub repo: https://github.com/mast-group/SStuBs-mining
Files
100 Java Maven Project Bugs bugs (bugs.json) 1000 Java Project Bugs bugsLarge (bugsLarge.json) 100 Java Maven Project SStuBs sstubs (sstubs.json) 1000 Java Project SStuBs sstubsLarge (sstubsLarge.json) Top 100 Ranked Java Maven Projects topJavaMavenProjects.csv Ranked Java Projects topProjects.csv
Due to a bug zenodo returns an error when uploading json files. The .json suffix can be restored by simply renaming the files (e.g. bugs -> bugs.json).
Corpus Statistics
| Projects | Bug Commits | Buggy Statements | Bug Statements per Commit | Pattern Fitting Bugs |
|---|---|---|---|---|
| 100 Java Maven | 12,598 | 25,539 | 2.03 | 7,824 |
| 1000 Java | 86,771 | 153,652 | 1.77 | 51,537 |
The table below provides information about the number of mined single statement bugs that fit each of the SStuB patterns.
| Pattern Name | Instances (100 Projects) | Instances Large (1000 Projects) |
|---|---|---|
| Change Identifier Used | 3,265 | 22,668 |
| Change Numeric Literal | 1,137 | 5,447 |
| Change Boolean Literal | 169 | 1,842 |
| Change Modifier | 1,852 | 5,011 |
| Wrong Function Name | 1,486 | 10,179 |
| Same Function More Args | 758 | 5,100 |
| Same Function Less Args | 179 | 1,588 |
| Same Function Change Caller | 187 | 1,504 |
| Same Function Swap Args | 127 | 612 |
| Change Binary Operator | 275 | 2,241 |
| Change Unary Operator | 170 | 1,016 |
| Change Operand | 120 | 807 |
| Less Specific If | 215 | 2,813 |
| More Specific If | 175 | 2,381 |
| Missing Throws Exception | 68 | 206 |
| Delete Throws Exception | 48 | 508 |
Use
The ManySStuBs4J Corpus is an automatically mined collection of Java bugs at large scale. We note that the automatic extraction could have inserted some noise. However, the amount of inserted noise is deemed to be almost negligible (see about). We also note that the code of the Java projects is not ours but is open-source. Please respect the license of each project.
If you use the data set in a research publication, please cite:
@inproceedings{ManySStuBs4JCorpus2020,
author={Karampatsis, Rafael-Michael and Sutton, Charles},
title={{How Often Do Single-Statement Bugs Occur?\\ The ManySStuBs4J Dataset}},
booktitle={},
year={2020},
pages={},
organization={}
}
Features
The files sstubs.json and sstubsLarge.json contain the following fields:
bugType: The bug type (16 possible values).commitSHA1: The hash of the commit fixing the bug.fixCommitParentSHA1: The hash of the last commit containing the bug.commitFile: Path of the fixed file.patch: The diff of the buggy and fixed file containing all the changes applied by the fix commit. Helps humans to understand the change.projectName: The concatenated repo owner and repo name separated by a..bugLineNum: The line in which the bug exists in the buggy version of the file.bugNodeStartChar: The character index (i.e., the number of characters in the java file that must be read before encountering the first one of the AST node) at which the affected ASTNode starts in the buggy version of the file.bugNodeLength: The length of the affected ASTNode in the buggy version of the file.fixLineNum: The line in which the bug was fixed in the fixed version of the file.fixNodeStartChar: The character index at which the affected ASTNode starts in the fixed version of the file.fixNodeLength: The length of the affected ASTNode in the fixed version of the file.before: The affected AST's tree (sometimes subtree e.g., Change Numeric Literal) text before the fix.after: The affected AST's tree (sometimes subtree e.g., Change Numeric Literal) text after the fix.
The "before", "after", "patch" fields help humans to understand the change. The "bugLineNum", "bugNodeStartChar", "bugNodeLength", "fixLineNum", "fixNodeStartChar", and "fixNodeLength" allow pinpointing of the AST nodes and lines that contained the bug and their equivalent ones in the fixed version of the file.
Similarly the bugs in bugs.json contain the above fields except bugType. All bugs appearing in sstubs.json have also an entry in bugs.json.