ReentrancyStudy-Data

ReentrancyStudy-Data is a large-scale dataset of reentrancy vulnerability in smart contracts, which is built from 230,548 open-source smart contracts from Etherscan. For more details, please refer to ICSE'23 paper.

Data Annotation

deduplicated_smart_contracts

The deduplicated smart contracts.

reentrant_contracts

The reentrant contracts (code) with labels.

tools_output

The reported reentrancy by analysis tools (oyente, mythril, securify1, securify2, smartian, sailfish).

reentrancy_information.csv

The reentrant labels of deduplicated smart contracts.

In 'true positive' column:

1 means the contract is manually verified to true reentrant contract.
0 means the contract is manually verified to false reentrant contract.
N/A means the contract does not be successfully analyzed by any tool.

In the column of tools (oyente, mythril, securify1, securify2, smartian):

1 means the contract is reported to reentrant contract by the tool.
0 means the contract is not reported to reentrant contract by the tool.
N/A means the contract does not be successfully analyzed by the tool.

contract_information_etherscan.csv

The information of contracts from etherscan.

hash_to_address.json

The mapping from the hash value (for deduplication) to the address.

load_reentrancy_information.py

The python script to load reentrancy_information.csv.

Citing

Zibin Zheng, Neng Zhang, Jianzhong Su, Zhijie Zhong, Mingxi Ye, and Jiachi Chen, “Turn the rudder: A beacon of reentrancy detection for smart contracts on ethereum,” in Proceedings of the 45th IEEE/ACM International Conference on Software Engineering, ser. ICSE ’23, 2023

Contact

If you have any questions about our dataset, please contact [email protected].