Navanath Saharia

19 papers Journal 8Unranked 11
YearRankTypeTitle / Venue / Authors
2023 J jnl
Multim. Tools Appl.
Maibam Debina Devi, Navanath Saharia
2020 conf
MIND (2)
Maibam Debina Devi, Navanath Saharia
2020 J jnl
CoRR
Mirinso Shadang, Navanath Saharia, Thoudam Doren Singh
2018 J jnl
CoRR
Himangshu Sarma, Navanath Saharia
2018 conf
ICACCI
Smriti Kumari, Navanath Saharia
2017 J jnl
ACM Trans. Asian Low Resour. Lang. Inf. Process.
Himangshu Sarma, Navanath Saharia, Utpal Sharma
2016 conf
COMM
Horia-Niolai Teodorescu, Navanath Saharia
2015 conf
SpeD
Horia-Nicolai Teodorescu, Navanath Saharia
2015 conf
SpeD
Navanath Saharia
2014 conf
SIRS
Himangshu Sarma, Navanath Saharia, Utpal Sharma
2014 conf
CICLing (1)
Nayan Jyoti Kalita, Navanath Saharia, Smriti Kumar Sinha
2014 J jnl
CoRR
Nayan Jyoti Kalita, Navanath Saharia, Smriti Kumar Sinha
2014 J jnl
ACM Trans. Asian Lang. Inf. Process.
Navanath Saharia, Utpal Sharma, Jugal K. Kalita
2013 conf
CICLing (1)
Navanath Saharia, Kishori M. Konwar, Utpal Sharma, Jugal K. Kalita
2013 J jnl
CoRR
Himangshu Sarma, Navanath Saharia, Utpal Sharma, Smriti Kumar Sinha, Mancha Jyoti Malakar
2013 J jnl
CoRR
Nayan Jyoti Kalita, Navanath Saharia, Smriti Kumar Sinha
2012 conf
ICACCI
Navanath Saharia, Utpal Sharma, Jugal Kalita
2010 conf
IALP
Navanath Saharia, Utpal Sharma, Jugal K. Kalita
2009 conf
ACL/IJCNLP (2)
Navanath Saharia, Dhrubajyoti Das, Utpal Sharma, Jugal K. Kalita
redb/extractors/decompiler/_archive/DecompileGhidra-old.py
← Index redb/extractors/decompiler/_archive/DecompileGhidra-old.py python
from hashlib import sha256
import inspect
from pathlib import Path
import subprocess
import json
import subprocess
import json
import os
import tempfile
import uuid
import shutil
import time

from dotenv import load_dotenv

from redb.extractors.enum import Tag
from redb.models.dataclasses import Decompiled
from redb.extractors.extractor import Extractor


class DecompileGhidra(Extractor):
    def __init__(
        self,
        filepath,
        log,
        index_prefix=None,
        elastic_index=None,
        known_benign=False,
        known_malicious=False,
    ):
        super().__init__(
            filepath, log, index_prefix, elastic_index, known_benign, known_malicious
        )
        self.log.debug(inspect.currentframe().f_code.co_name)
        self.elastic_index = self.index_prefix + "-ghidra"
        self.ghidra_path = "/opt/ghidra"
        self.java_script_path = (
            self.ghidra_path
            + "/Ghidra/Features/Base/ghidra_scripts/GhidraDecompilerScript.java"
        )
        self.decompiled = None
        load_dotenv()
        self.decompiled_folder = os.getenv("DECOMPILED_FOLDER", "/opt/decompiled")
        self.log.debug(f"Decompiled folder: {self.decompiled_folder}")

    def run_command(self, cmd, env=None):
        try:
            self.log.info(f"Starting command: {' '.join(cmd)}")
            start_time = time.time()
            TIMEOUT = 1200  # 20 minutes in seconds
            process = subprocess.Popen(
                cmd, env=env, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True
            )

            while True:
                output = process.stdout.readline()
                if output:
                    print(output.strip())
                if process.poll() is not None:
                    break
            try:
                stdout, stderr = process.communicate(timeout=TIMEOUT)
            except subprocess.TimeoutExpired:
                process.kill()
                self.log.error(f"Ghidra timed out after {TIMEOUT} seconds")
                # raise subprocess.TimeoutExpired(process.args, TIMEOUT)
                return None
            end_time = time.time()

            self.log.debug(
                f"Command finished. Execution time: {end_time - start_time:.2f} seconds"
            )
            self.log.debug(f"Return code: {process.returncode}")

            if process.returncode != 0:
                self.log.error(f"Error output:\n{stderr}")
                return None
            return stdout
        except Exception as e:
            self.log.error(f"Error running command {' '.join(cmd)}: {e}")
            return None

    def analyze_binary(self):
        self.log.debug(f"Ghidra path: {self.ghidra_path}")
        self.log.debug(f"Binary path: {self.filepath}")
        self.log.debug(f"Java script path: {self.java_script_path}")

        # Check if Java script exists
        if not os.path.exists(self.java_script_path):
            self.log.error(f"Error: Java script not found at {self.java_script_path}")
            return None

        # Set up environment variables
        env = os.environ.copy()
        java_home = "/usr/lib/jvm/java-17-openjdk-amd64"  # Adjust this path if needed
        env["JAVA_HOME"] = java_home
        env["PATH"] = f"{java_home}/bin:{env['PATH']}"
        env["LD_LIBRARY_PATH"] = f"{java_home}/lib:{env.get('LD_LIBRARY_PATH', '')}"
        env["DECOMPILED_FOLDER"] = self.decompiled_folder

        # Print environment variables for debugging
        self.log.debug(f"JAVA_HOME: {env['JAVA_HOME']}")
        self.log.debug(f"PATH: {env['PATH']}")
        self.log.debug(f"LD_LIBRARY_PATH: {env['LD_LIBRARY_PATH']}")

        # Check Ghidra installation
        analyzeHeadless_path = f"{self.ghidra_path}/support/analyzeHeadless"
        self.log.debug(
            f"analyzeHeadless exists: {os.path.exists(analyzeHeadless_path)}"
        )

        # Create a temporary project directory
        project_path = tempfile.gettempdir() + "/ghidra_" + str(uuid.uuid4())
        os.makedirs(project_path, exist_ok=True)
        self.log.debug(f"Created temporary project path: {project_path}")
        output_file = ""

        try:
            # Run Ghidra's headless analyzer
            analyze_cmd = [
                analyzeHeadless_path,
                project_path,
                "TempProject",
                "-import",
                self.filepath,
                "-postScript",
                self.java_script_path,
                self.sha256,
                "-deleteProject",
            ]

            result = self.run_command(analyze_cmd, env=env)
            if result is None:
                return None

            # Read the output JSON file
            output_file = os.path.join(
                self.decompiled_folder, self.sha256 + "-decompiled.json"
            )
            if os.path.exists(output_file):
                with open(output_file, "r") as f:
                    functions = json.load(f)
                return functions
            else:
                self.log.error(
                    f"Output file {output_file} not found. Ghidra analysis may have failed."
                )
                return None
        finally:
            # Clean up
            if os.path.exists(project_path):
                shutil.rmtree(project_path)
                self.log.debug(f"Deleted temporary project path: {project_path}")

    def extract(self):
        self.log.debug(inspect.currentframe().f_code.co_name)
        try:
            functions = self.analyze_binary()

            if functions:
                self.log.info(f"Extracted functions from {self.filepath}:")
                for func in functions:
                    id = sha256(func["address"].encode()).hexdigest()
                    self.decompiled = Decompiled(
                        _id=id,
                        decompiled_function_name=func["name"],
                        decompiled_function_address=func["address"],
                        decompiled_function=func["decompiled"],
                    )
                    self.export_to_elastic([self.decompiled])
            else:
                self.log.error("No decompiled functions extracted.")
            return True
        except Exception as e:
            self.log.error(f"Error extracting decompiled information: {e}")
            return None

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