Hai Zhu

71 papers A* 4A 2B 1C 9Misc 1Journal 41Unranked 13
YearRankTypeTitle / Venue / Authors
2026 J jnl
CoRR
Hai Zhu, Yantao Yu, Lei Shen, Bing Wang, Xiaoyi Zeng
2026 J jnl
CoRR
Chunqi Wang, Bingchao Wu, Taotian Pang, Jiahao Wang, Jie Yang, Jia Liu, Hao Zhang, Hai Zhu, Lei Shen, Shizhun Wang, Bing Wang, Xiaoyi Zeng
2025 J jnl
CoRR
Shun Zhang, Hai Zhu, Zhili Chen, Neal N. Xiong
2025 J jnl
CoRR
Hai Zhu, Chia-Nan Yeh, Miguel A. Morales, Leslie Greengard, Shidong Jiang, Jason Kaye
2025 conf
COLING (Industry)
Hai Zhu, Yuankai Guo, Ronggang Dou, Kai Liu
2025 J jnl
IEEE Trans. Veh. Technol.
Hai Zhu, Baoquan Li, Haolin Yin, Xuebo Zhang
2025 J jnl
CoRR
Haojie Bai, Yang Wang, Cong Guo, Xiongwei Zhao, Hai Zhu
2025 J jnl
SIAM J. Sci. Comput.
Ruowen Liu, Hai Zhu, Hanliang Guo, Marc Bonnet, Shravan K. Veerapaneni
2024 B conf
SMC
Meiying Yang, Hai Zhu, Xingyu Xia, Zhe Liu, Wen Yao
2024 A* conf
AAAI
Hai Zhu, Qingyang Zhao, Weiwei Shang, Yuren Wu, Kai Liu
2024 J jnl
Sensors
Hai Zhu, Baoquan Li, Ruiyang Tong, Haolin Yin, Canlin Zhu
2024 J jnl
CoRR
Shidong Jiang, Hai Zhu
2024 conf
ROBIO
Haojie Bai, Hai Zhu, Xiongwei Zhao, Huafu Li, Yang Wang
2024 Misc conf
ICASSP
Hai Zhu, Xin Wang, Kun Wang, Huayi Zhan
2023 J jnl
J. Comput. Phys.
Thomas G. Anderson, Hai Zhu, Shravan K. Veerapaneni
2023 conf
EITCE
Xin Jing, Hai Zhu
2023 conf
PAKDD (2)
Hai Zhu, Qinyang Zhao, Yuren Wu
2023 J jnl
CoRR
Hai Zhu, Qingyang Zhao, Yuren Wu
2023 J jnl
J. Ind. Inf. Integr.
Hai Zhu, Wenan Tan, Mei Yang, Kai Guo, Jiaojiao Li
2023 A conf
INTERSPEECH
Hai Zhu, Huayi Zhan, Hong Cheng, Ying Wu
2023 J jnl
Inf. Technol. Control.
Haolin Yin, Baoquan Li, Hai Zhu, Lintao Shi
2023 J jnl
CoRR
Hai Zhu, Zhaoqing Yang, Weiwei Shang, Yuren Wu
2023 J jnl
CoRR
Hai Zhu, Yuankai Guo, Ronggang Dou, Kai Liu
2023 J jnl
SIAM J. Sci. Comput.
Marc Bonnet, Ruowen Liu, Shravan K. Veerapaneni, Hai Zhu
2022 J jnl
CoRR
Thomas G. Anderson, Hai Zhu, Shravan K. Veerapaneni
2022 J jnl
Enterp. Inf. Syst.
Wenan Tan, Hai Zhu, Jinjing Tan, Yao Zhao, Li Da Xu, Kai Guo
2022 conf
BDSIC
Su Dai, Hai Zhu, Zhixiang Sun, Huiqin Zhao, Bowen Liu
2022 J jnl
SIAM J. Sci. Comput.
Hai Zhu, Shravan K. Veerapaneni
2022 J jnl
J. Ind. Inf. Integr.
Kai Guo, Shanshan Hu, Hai Zhu, Wenan Tan
2022 J jnl
IEEE Trans. Instrum. Meas.
Yongqiang Ge, Jiawang Chen, Chen Cao, Peng Zhou, Chunying Xu, Hai Zhu, Qixiao Zhou
2022 conf
IALP
Junsong Ruan, Hailin Zhang, Hai Zhu, Ruoyao Ding
2022 A* conf
ACM Multimedia
Zhe Xue, Junping Du, Hai Zhu, Zhongchao Guan, Yunfei Long, Yu Zang, MeiYu Liang
2022 J jnl
Comput. Secur.
Zhen Wang, Yitao Zheng, Hai Zhu, Chang Yang, Tianyi Chen
2021 J jnl
CoRR
Hai Zhu, Shravan K. Veerapaneni
2021 J jnl
J. Ind. Inf. Integr.
Wenan Tan, Li Huang, Mikhail Yu. Kataev, Yong Sun, Lu Zhao, Hai Zhu, Kai Guo, Na Xie
2021 conf
ICAIIS
Hailin Zhang, Hai Zhu, Junsong Ruan, Ruoyao Ding
2021 J jnl
CoRR
Shujun Liu, Hai Zhu, Kun Wang, Huajun Wang
2020 J jnl
Neural Comput. Appl.
Kai Guo, Mei Yang, Hai Zhu
2020 J jnl
Neural Comput. Appl.
Mei Yang, Hai Zhu, Kai Guo
2020 J jnl
J. Comput. Phys.
Bowei Wu, Hai Zhu, Alex H. Barnett, Shravan K. Veerapaneni
2019 J jnl
IEEE Robotics Autom. Lett.
Lee-Ling Sharon Ong, Hai Zhu, Debasis Banik, Zhenping Guan, Yinnian Feng, Ellis L. Reinherz, Matthew J. Lang, H. Harry Asada
2019 conf
IST
Hai Zhu, Jiangtao Sun, Lijun Xu, Shijie Sun
2019 A* conf
WWW
Wen Zhang, Bibek Paudel, Liang Wang, Jiaoyan Chen, Hai Zhu, Wei Zhang, Abraham Bernstein, Huajun Chen
2019 J jnl
CoRR
Wen Zhang, Bibek Paudel, Liang Wang, Jiaoyan Chen, Hai Zhu, Wei Zhang, Abraham Bernstein, Huajun Chen
2019 J jnl
IEEE Access
Xudong Zhong, Hao Yin, Yuanzhi He, Hai Zhu
2019 C conf
APSEC
Chen Zhi, Shuiguang Deng, Jianwei Yin, Min Fu, Hai Zhu, Yuanping Li, Tao Xie
2019 J jnl
CoRR
Bowei Wu, Hai Zhu, Alex H. Barnett, Shravan K. Veerapaneni
2018 conf
TrustCom/BigDataSE
Qingqing Tan, Chong Han, Lijuan Sun, Jian Guo, Hai Zhu
2018 J jnl
Inf.
Chong Han, Qingqing Tan, Lijuan Sun, Hai Zhu, Jian Guo
2018 A conf
ICSOC
Min Fu, Chi Man Wong, Hai Zhu, Yanjun Huang, Yuanping Li, Xi Zheng, Jia Wu, Jian Yang, Chi-Man Vong
2018 A* conf
EMNLP
Guanying Wang, Wen Zhang, Ruoxu Wang, Yalin Zhou, Xi Chen, Wei Zhang, Hai Zhu, Huajun Chen
2017 J jnl
J. Electronic Imaging
Lin Feng, Laihang Yu, Hai Zhu
2016 conf
LISS
Xia Luo, Zhi Liu, Hai Zhu
2016 C conf
CIS
Hao Li, Hai Zhu, Guoheng Ren, Hongfeng Wang, Hong Zhang, Liyong Chen
2016 J jnl
Int. J. Pattern Recognit. Artif. Intell.
Hai Zhu, Hongfeng Wang
2016 C conf
CIS
Hai Zhu, Guoheng Ren, Dongxia Qin, Wei Wang, Fei Wei, Ying Cao
2016 conf
LISS
Yanjin Li, Hai Zhu, Yonghong Liu
2016 conf
LISS
Qian Huang, Xia Luo, Hai Zhu
2015 C conf
CIS
Wei Wang, Hai Zhu, Xiaoli Wang
2015 J jnl
Sensors
Huapeng Yu, Hai Zhu, Dayuan Gao, Meng Yu, Wenqi Wu
2015 C conf
CIS
Hai Zhu, Wei Wang, Xiali Wang, Fei Wei
2015 J jnl
计算机科学
Hongfeng Wang, Hai Zhu
2014 C conf
CIS
Hai Zhu, Xiaoli Wang, Hongfeng Wang
2014 C conf
CIS
Hong Zhang, Hai Zhu
2012 J jnl
J. Comput.
Xiaoli Wang, Yuping Wang, Hai Zhu
2011 C conf
CIS
Qiguang Miao, Pengfei Xu, Weisheng Li, Hai Zhu
2011 C conf
CIS
Hai Zhu, Zengfa Dou, Wengang Zhou
2011 J jnl
J. Networks
Hai Zhu, Yuping Wang, Zhanxin Ma, Hecheng Li
2010 J jnl
JOCN
Xian Zhou, Xue Chen, Weiqing Zhou, Yangyang Fan, Hai Zhu, Zhiyu Li
2010 J jnl
J. Comput.
Hai Zhu, Yuping Wang, Lei Fan, Xiaoli Wang
2008 conf
CIS (2)
Hai Zhu, Yuping Wang
redb/extractors/apk_extractors/apk_signature.py
← Index redb/extractors/apk_extractors/apk_signature.py python
import hashlib
import inspect
from datetime import datetime, timezone
from typing import Any, Optional, Tuple

from redb.extractors.enum import Tag
from redb.extractors.apk_extractor import APKExtractor
from redb.models.dataclasses import APKCertificate, APKCodeSigningInfo


# OID to RFC 4514 short name mapping
_OID_SHORT_NAMES = {
    '2.5.4.3': 'CN', '2.5.4.6': 'C', '2.5.4.7': 'L', '2.5.4.8': 'ST',
    '2.5.4.10': 'O', '2.5.4.11': 'OU', '1.2.840.113549.1.9.1': 'E',
    '2.5.4.5': 'SERIALNUMBER', '2.5.4.12': 'T', '2.5.4.42': 'GN',
    '2.5.4.4': 'SN', '2.5.4.9': 'STREET', '2.5.4.17': 'POSTALCODE',
}


def _asn1_name_to_rfc4514(name):
    """Convert an asn1crypto x509.Name to RFC 4514 string (CN=..., O=...)."""
    parts = []
    for rdn in name.chosen:
        for attr in rdn:
            oid = attr['type'].dotted
            val = attr['value'].native
            short = _OID_SHORT_NAMES.get(oid, oid)
            parts.append(f'{short}={val}')
    return ', '.join(parts)


class APKSignatureExtractor(APKExtractor):

    def __init__(
        self, filepath, log, exporters=None, index_prefix=None,
        known_benign=False, known_malicious=False,
        apk=None,
    ):
        super().__init__(
            filepath, log, exporters, index_prefix,
            known_benign, known_malicious, apk,
        )
        self.signing_info = None
        self.log.debug(inspect.currentframe().f_code.co_name)

    def tag(self):
        return Tag.APK_SIGNATURE.value

    def _detect_signature_schemes(self):
        """Detect which APK signature schemes are present."""
        schemes = []
        try:
            if self.apk.is_signed_v1():
                schemes.append(1)
        except Exception:
            pass
        try:
            if self.apk.is_signed_v2():
                schemes.append(2)
        except Exception:
            pass
        try:
            if self.apk.is_signed_v3():
                schemes.append(3)
        except Exception:
            pass
        return schemes

    def _extract_certificate_info(self, cert, source_scheme=None):
        """Extract certificate details from an androguard certificate object."""
        try:
            # Subject and Issuer — asn1crypto doesn't have rfc4514_string, convert manually
            subject = _asn1_name_to_rfc4514(cert.subject)
            issuer = _asn1_name_to_rfc4514(cert.issuer)

            # Serial number
            serial = str(cert.serial_number)

            # Validity
            valid_from = cert.not_valid_before
            valid_to = cert.not_valid_after
            if isinstance(valid_from, datetime):
                valid_from = valid_from.strftime("%Y-%m-%d %H:%M:%S")
            else:
                valid_from = str(valid_from)
            if isinstance(valid_to, datetime):
                valid_to = valid_to.strftime("%Y-%m-%d %H:%M:%S")
            else:
                valid_to = str(valid_to)

            # Thumbprints
            cert_der = cert.dump()
            thumbprint_sha1 = hashlib.sha1(cert_der).hexdigest()
            thumbprint_sha256 = hashlib.sha256(cert_der).hexdigest()

            # Algorithm
            algorithm = None
            try:
                algorithm = cert.hash_algo
            except Exception:
                try:
                    algorithm = cert.signature_algo
                except Exception:
                    pass

            # Key size
            key_size = None
            try:
                pub_key = cert.public_key
                if hasattr(pub_key, 'bit_size'):
                    key_size = pub_key.bit_size
                elif hasattr(pub_key, 'key_size'):
                    key_size = pub_key.key_size
            except Exception:
                pass

            # Self-signed check
            is_self_signed = cert.subject == cert.issuer

            return APKCertificate(
                subject=subject,
                issuer=issuer,
                serial_number=serial,
                valid_from=valid_from,
                valid_to=valid_to,
                thumbprint_sha1=thumbprint_sha1,
                thumbprint_sha256=thumbprint_sha256,
                algorithm=algorithm,
                key_size=key_size,
                is_self_signed=is_self_signed,
                certificate_source_scheme=source_scheme,
            )
        except Exception as e:
            self.log.error(f"Error extracting certificate info: {e}")
            return None

    def _get_signer_certificate(self, schemes) -> Tuple[Optional[APKCertificate], Optional[str]]:
        """Identify the actual signer certificate using V3>V2>V1 priority.

        V2/V3: certificates[0] is the signer per spec.
        V1: androguard's get_certificate() does SignerInfo matching internally.

        Returns (signer_cert_info, source_scheme) or (None, None).
        """
        # V3 takes priority (Android 9+)
        if 3 in schemes:
            try:
                certs = self.apk.get_certificates_v3()
                if certs:
                    cert_info = self._extract_certificate_info(certs[0], source_scheme='v3')
                    if cert_info:
                        return cert_info, 'v3'
            except Exception as e:
                self.log.debug(f"V3 signer extraction failed: {e}")

        # V2 next (Android 7.0+)
        if 2 in schemes:
            try:
                certs = self.apk.get_certificates_v2()
                if certs:
                    cert_info = self._extract_certificate_info(certs[0], source_scheme='v2')
                    if cert_info:
                        return cert_info, 'v2'
            except Exception as e:
                self.log.debug(f"V2 signer extraction failed: {e}")

        # V1 fallback — androguard does SignerInfo.issuerAndSerialNumber matching
        if 1 in schemes:
            try:
                names = self.apk.get_signature_names()
                if names:
                    cert = self.apk.get_certificate(names[0])
                    if cert:
                        cert_info = self._extract_certificate_info(cert, source_scheme='v1')
                        if cert_info:
                            return cert_info, 'v1'
            except Exception as e:
                self.log.debug(f"V1 signer extraction failed: {e}")

        return None, None

    def _extract_all_certificates(self, schemes):
        """Extract all certificates with their source scheme tracked.

        Uses scheme-specific methods to preserve source information,
        deduplicating by thumbprint_sha1 and merging source schemes.
        """
        # cert thumbprint -> (APKCertificate, set of source schemes)
        seen = {}

        for scheme_num, method_name in [
            (3, 'get_certificates_v3'),
            (2, 'get_certificates_v2'),
            (1, 'get_certificates_v1'),
        ]:
            if scheme_num not in schemes:
                continue
            try:
                method = getattr(self.apk, method_name, None)
                if method is None:
                    continue
                certs = method()
                if not certs:
                    continue
                for cert in certs:
                    cert_info = self._extract_certificate_info(cert, source_scheme=f'v{scheme_num}')
                    if cert_info:
                        key = cert_info.thumbprint_sha1
                        if key in seen:
                            # Merge source schemes
                            existing_cert, existing_schemes = seen[key]
                            existing_schemes.add(f'v{scheme_num}')
                        else:
                            seen[key] = (cert_info, {f'v{scheme_num}'})
            except Exception as e:
                self.log.warning(f"Error extracting v{scheme_num} certificates: {e}")

        # Build final list with merged source schemes
        result = []
        for cert_info, source_schemes in seen.values():
            cert_info.certificate_source_scheme = ','.join(sorted(source_schemes))
            result.append(cert_info)

        return result

    def extract(self):
        if not self._is_valid_apk():
            self.log.error(f"Invalid APK for {self.hash.sha256}")
            return None

        schemes = self._detect_signature_schemes()
        is_signed = len(schemes) > 0

        # Extract all certificates with source scheme tracking
        certificates = self._extract_all_certificates(schemes)

        # Identify the actual signer using V3>V2>V1 priority
        signer_cert, signer_scheme = self._get_signer_certificate(schemes)

        apk_signer_subject = signer_cert.subject if signer_cert else None
        apk_signer_issuer = signer_cert.issuer if signer_cert else None

        self.signing_info = APKCodeSigningInfo(
            _id=self.sha256,
            is_signed=is_signed,
            signature_scheme_versions=schemes,
            number_of_certificates=len(certificates),
            x509_certificates=certificates if certificates else None,
            apk_signer_subject=apk_signer_subject,
            apk_signer_issuer=apk_signer_issuer,
        )
        return self.signing_info

    def prepare_export_data(self, exporter_type: str) -> Any:
        if exporter_type == "ClickHouseExporter":
            if not self.signing_info:
                return None

            current_time = datetime.now(timezone.utc)
            sig = self.signing_info

            # Signature summary table
            signature_data = [[
                self.sha256,
                int(sig.is_signed),
                sig.signature_scheme_versions,
                sig.number_of_certificates,
                sig.apk_signer_subject,
                sig.apk_signer_issuer,
                current_time,
            ]]

            # Certificates table (one row per certificate)
            certificates_data = []
            if sig.x509_certificates:
                for cert in sig.x509_certificates:
                    certificates_data.append([
                        self.sha256,
                        cert.subject,
                        cert.issuer,
                        cert.serial_number,
                        datetime.strptime(cert.valid_from, "%Y-%m-%d %H:%M:%S") if cert.valid_from else None,
                        datetime.strptime(cert.valid_to, "%Y-%m-%d %H:%M:%S") if cert.valid_to else None,
                        cert.thumbprint_sha1,
                        cert.algorithm,
                        cert.key_size,
                        int(cert.is_self_signed),
                        cert.certificate_source_scheme,
                        current_time,
                    ])

            return {
                'multi_table': True,
                'signature': {
                    'table': 'redb_apk_signature',
                    'data': signature_data,
                    'column_names': [
                        'sha256',
                        'is_signed', 'signature_scheme_versions',
                        'number_of_certificates',
                        'apk_signer_subject', 'apk_signer_issuer',
                        'analysis_date',
                    ],
                    'column_type_names': [
                        'FixedString(64)',
                        'UInt8', 'Array(UInt8)',
                        'UInt8',
                        'Nullable(String)', 'Nullable(String)',
                        "DateTime64(3, 'UTC')",
                    ],
                },
                'certificates': {
                    'table': 'redb_apk_certificates',
                    'data': certificates_data,
                    'column_names': [
                        'sha256', 'certificate_subject', 'certificate_issuer',
                        'certificate_serial_number',
                        'certificate_valid_from', 'certificate_valid_to',
                        'certificate_thumbprint',
                        'certificate_algorithm', 'key_size', 'is_self_signed',
                        'certificate_source_scheme',
                        'analysis_date',
                    ],
                    'column_type_names': [
                        'FixedString(64)', 'String', 'String', 'String',
                        "DateTime64(3, 'UTC')", "DateTime64(3, 'UTC')",
                        'FixedString(40)',
                        'LowCardinality(String)', 'Nullable(UInt16)', 'UInt8',
                        'LowCardinality(String)',
                        "DateTime64(3, 'UTC')",
                    ],
                },
            }

    def get_clickhouse_table(self) -> str:
        return "redb_apk_signature"