Jaesung Tae

15 papers A 1B 1Journal 8Unranked 5
YearRankTypeTitle / Venue / Authors
2025 conf
ACL (1)
Jaesung Tae, Hamish Ivison, Sachin Kumar, Arman Cohan
2025 J jnl
CoRR
Jaesung Tae, Hamish Ivison, Sachin Kumar, Arman Cohan
2024 conf
EACL (1)
Rabeeh Karimi Mahabadi, Hamish Ivison, Jaesung Tae, James Henderson, Iz Beltagy, Matthew E. Peters, Arman Cohan
2023 J jnl
CoRR
Linyong Nan, Yilun Zhao, Weijin Zou, Narutatsu Ri, Jaesung Tae, Ellen Zhang, Arman Cohan, Dragomir Radev
2023 conf
EMNLP (Findings)
Linyong Nan, Yilun Zhao, Weijin Zou, Narutatsu Ri, Jaesung Tae, Ellen Zhang, Arman Cohan, Dragomir Radev
2023 J jnl
CoRR
Jaesung Tae
2023 J jnl
CoRR
Rabeeh Karimi Mahabadi, Jaesung Tae, Hamish Ivison, James Henderson, Iz Beltagy, Matthew E. Peters, Arman Cohan
2022 A conf
INTERSPEECH
Jaesung Tae, Hyeongju Kim, Taesu Kim
2022 B conf
LREC
Irene Li, Alexander R. Fabbri, Rina Kawamura, Yixin Liu, Xiangru Tang, Jaesung Tae, Chang Shen, Sally Ma, Tomoe Mizutani, Dragomir Radev
2022 conf
EMNLP (Findings)
Teven Le Scao, Thomas Wang, Daniel Hesslow, Lucile Saulnier, Stas Bekman, M. Saiful Bari, Stella Biderman, Hady Elsahar, Niklas Muennighoff, Jason Phang, Ofir Press, Colin Raffel, Victor Sanh, Sheng Shen, Lintang Sutawika, Jaesung Tae, Zheng Xin Yong, Julien Launay, Iz Beltagy
2022 J jnl
CoRR
Teven Le Scao, Thomas Wang, Daniel Hesslow, Lucile Saulnier, Stas Bekman, M. Saiful Bari, Stella Biderman, Hady Elsahar, Niklas Muennighoff, Jason Phang, Ofir Press, Colin Raffel, Victor Sanh, Sheng Shen, Lintang Sutawika, Jaesung Tae, Zheng Xin Yong, Julien Launay, Iz Beltagy
2021 J jnl
CoRR
Jaesung Tae, Hyeongju Kim, Taesu Kim
2021 conf
MLSP
Jaesung Tae, Hyeongju Kim, Younggun Lee
2021 J jnl
CoRR
Jaesung Tae, Hyeongju Kim, Younggun Lee
2021 J jnl
CoRR
Irene Li, Alexander R. Fabbri, Rina Kawamura, Yixin Liu, Xiangru Tang, Jaesung Tae, Chang Shen, Sally Ma, Tomoe Mizutani, Dragomir R. Radev
redb/extractors/pe_extractors/pe_resources.py
← Index redb/extractors/pe_extractors/pe_resources.py python
from hashlib import sha256
import inspect
from datetime import datetime, timezone
from typing import Any

import magic
from magika import Magika
import pefile
from pefile import UnicodeStringWrapperPostProcessor

from redb.extractors.enum import Tag
from redb.extractors.pe_extractor import PEExtractor
from redb.models.dataclasses import PEResource


class PEResourceExtractor(PEExtractor):

    def __init__(
        self,
        filepath,
        log,
        exporters=None,
        index_prefix=None,
        elastic_index=None,
        known_benign=False,
        known_malicious=False,
        pe=None,
    ):
        super().__init__(
            filepath,
            log,
            exporters,
            index_prefix,
            elastic_index,
            known_benign,
            known_malicious,
            pe,
        )
        self.elastic_index = self.index_prefix + "-pe_resources"
        self.log.debug(inspect.currentframe().f_code.co_name)

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

    def _extract_resources(self):
        """
        Returns:
        resources: a list of dictionaries, one per each resources type found.
                    each dictionary the key represents the name of the content,
                    which is the value itself.
                    Empty list if no resources present.
        """
        self.log.debug(inspect.currentframe().f_code.co_name)
        resources_list = []
        try:
            if hasattr(self.pe, "DIRECTORY_ENTRY_RESOURCE"):
                for resource_type in self.pe.DIRECTORY_ENTRY_RESOURCE.entries:
                    # if resource_type.name is not None:
                    #     name = resource_type.name
                    # else:
                    #     name = pefile.RESOURCE_TYPE.get(resource_type.struct.Id)
                    # if not name:
                    #     name = resource_type.struct.Id
                    name = (
                        resource_type.name
                        if resource_type.name is not None
                        else pefile.RESOURCE_TYPE.get(resource_type.struct.Id)
                    )
                    if isinstance(name, UnicodeStringWrapperPostProcessor):
                        name = name.decode()
                    try:
                        if hasattr(resource_type, "directory"):
                            for resource_id in resource_type.directory.entries:
                                if hasattr(resource_id, "directory"):
                                    for resource_lang in resource_id.directory.entries:
                                        rsrc_data = self.pe.get_data(
                                            resource_lang.data.struct.OffsetToData,
                                            resource_lang.data.struct.Size,
                                        )
                                        file_type = magic.from_buffer(rsrc_data)
                                        magik = Magika().identify_bytes(rsrc_data).output.label

                                        rsrc_entropy = (
                                            "%.2f"
                                            % pefile.SectionStructure.entropy_H(
                                                self.pe, rsrc_data
                                            )
                                        )
                                        rsrc_sha256 = sha256(rsrc_data).hexdigest()
                                        lang = pefile.LANG.get(
                                            resource_lang.data.lang, "*unknown*"
                                        )
                                        sublang = pefile.get_sublang_name_for_lang(
                                            resource_lang.data.lang,
                                            resource_lang.data.sublang,
                                        )
                                        pe_resource = PEResource(
                                            _id=rsrc_sha256,
                                            resource_type=name,
                                            resource_entropy=rsrc_entropy,
                                            resource_sha256=rsrc_sha256,
                                            resource_filetype=file_type,
                                            resource_magika=magik,
                                            resource_language=lang,
                                            resource_rva=resource_lang.data.struct.OffsetToData,
                                            resource_size=resource_lang.data.struct.Size,
                                            resource_sub_lang=sublang,
                                        )
                                        resources_list.append(pe_resource)
                    except Exception as e:
                        self.log.warning(
                            f"Continue after Error in {self.hash.sha256}: {resource_type.name} "
                            f"Exception: {e}",
                            stack_info=True,
                        )
                        # resources_list.append({f"{e} - {resource_type.name}"})
                        continue
        except Exception as e:
            self.log.exception(
                f"Extract exports error {self.hash.sha256} Exception: {e}"
            )
        self.log.debug(f"Resource list {resources_list}")
        return resources_list

    def extract(self):
        try:
            self.log.debug(inspect.currentframe().f_code.co_name)
            resources = self._extract_resources()
            # self.export_to_elastic(resources)  # Let the exporters handle this
            return resources
        except Exception as e:
            self.log.error(f"Extract resources error {self.hash.sha256} Exception: {e}")
            return None

    def prepare_export_data(self, exporter_type: str) -> Any:
        if exporter_type == "ElasticsearchExporter":
            return self.extract()
        elif exporter_type == "ClickHouseExporter":
            resources = self.extract()
            if resources is None:
                return None
            
            data = []
            current_time = datetime.now(timezone.utc)
            
            for resource in resources:
                data.append([
                    self.sha256,                    # sha256
                    self.md5,                       # md5
                    self.sha1,                      # sha1
                    resource.resource_type,         # resource_type
                    resource.resource_entropy,      # resource_entropy
                    resource.resource_sha256,       # resource_sha256
                    resource.resource_filetype,     # resource_filetype
                    resource.resource_magika,       # resource_magika
                    resource.resource_language,     # resource_language
                    resource.resource_sub_lang,     # resource_sub_lang
                    resource.resource_size,         # resource_size
                    resource.resource_rva,          # resource_rva
                    current_time                    # analysis_date
                ])
            
            column_names = [
                'sha256', 'md5', 'sha1', 'resource_type', 'resource_entropy',
                'resource_sha256', 'resource_filetype', 'resource_magika',
                'resource_language', 'resource_sub_lang', 'resource_size',
                'resource_rva', 'analysis_date'
            ]
            
            if not data:
                return None

            column_type_names = [
                'FixedString(64)', 'FixedString(32)', 'FixedString(40)',
                'LowCardinality(Nullable(String))', 'Float64',
                'FixedString(64)', 'LowCardinality(Nullable(String))', 'LowCardinality(Nullable(String))',
                'LowCardinality(Nullable(String))', 'LowCardinality(Nullable(String))', 'UInt64',
                'UInt64', 'DateTime64(3, \'UTC\')'
            ]

            return (data, column_names, column_type_names)

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