Merge branch 'fix/requirements-detection' into tests/final-tests

This commit is contained in:
dqy 2024-06-06 16:21:03 +08:00
commit 430d2b8f8a
2 changed files with 212 additions and 61 deletions

View File

@ -6,6 +6,8 @@ from reportlab.lib.styles import getSampleStyleSheet
from reportlab.platypus import Paragraph, Spacer, SimpleDocTemplate
from detection.pickle_detection import pickleDataDetection
from .requirements_detection import requirement_detection
from .Regexdetection import find_dangerous_functions
from .GPTdetection import detectGPT, GPTdetectFileList
@ -371,7 +373,12 @@ def checkModeAndDetect(mode: str, filePath: str, fileExtension: str, pycdc_addr:
def process_path(
path: str, output_format: str, mode: str, pycdc_addr: str, output_file=None
path: str,
output_format: str,
mode: str,
pycdc_addr: str,
output_file=None,
requirement_path=None,
):
results = {"high": [], "medium": [], "low": [], "none": [], "pickles": []}
if os.path.isdir(path):
@ -381,6 +388,7 @@ def process_path(
for file_path in Path(path).rglob("*")
if file_path.suffix in SUPPORTED_EXTENSIONS
]
<<<<<<< HEAD
if mode == "llm":
results = GPTdetectFileList(all_files)
else:
@ -425,7 +433,8 @@ def process_path(
else:
print("Invalid path.")
sys.exit(1)
if requirement_path is not None:
requirement_detection(requirement_path, output_file)
output_results(results, output_format, output_file)
@ -446,6 +455,18 @@ def main():
help="Path to pycdc.exe to decompile",
default=os.getenv("PATH"),
)
parser.add_argument(
"-P",
"--Pickle",
help="Path to pickle file to analyze",
default=None,
)
parser.add_argument(
"-r",
"--requirement",
help="Path to requirement file to analyze",
default=None,
)
args = parser.parse_args()
output_format = "txt" # Default output format
output_file = None
@ -461,7 +482,9 @@ def main():
)
output_file = args.output.rsplit(".", 1)[0] + ".txt"
# 如果未指定输出文件,则输出到 stdout否则写入文件
process_path(args.path, output_format, args.mode, args.pycdc, output_file)
process_path(
args.path, output_format, args.mode, args.pycdc, output_file, args.requirement
)
if PYCDC_FLAG == False:
print(
"ERROR: Detected Python 3.11 or above .pyc files. You need to install pycdc and compile it yourself to obtain pycdc."

View File

@ -3,6 +3,16 @@ import requests
from bs4 import BeautifulSoup
from packaging.version import Version, InvalidVersion
import sys
from reportlab.lib.pagesizes import letter
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer
from colorama import Fore, Style, init
from tqdm import tqdm
import html
import os
init(autoreset=True) # 初始化colorama并在每次打印后自动重置颜色
def fetch_html(url: str) -> str:
@ -55,7 +65,6 @@ def version_in_range(version, range_str: str) -> bool:
except InvalidVersion:
return False
else:
# 如果没有给版本号,默认使用最新版本
if range_str[-2] == ",":
return True
@ -77,64 +86,183 @@ def version_in_range(version, range_str: str) -> bool:
return True
def check_vulnerabilities(requirements: list, base_url: str, output_file: str):
with open(output_file, "w") as out_file:
for req in requirements:
version = ""
# 如果有版本
if "==" in req:
package_name, version = req.split("==")
# 没有版本
else:
package_name, version = req, None
# 拼接URL
url = f"{base_url}{package_name}"
print(f"Fetching data for {package_name} from {url}")
html_content = fetch_html(url)
if html_content:
# 解析hmtl
extracted_data = parse_html(html_content)
if extracted_data:
relevant_vulns = []
for vuln in extracted_data:
if version_in_range(version, vuln["chip"]):
relevant_vulns.append(vuln)
if relevant_vulns:
out_file.write(f"Vulnerabilities found for {package_name}:\n")
for vuln in relevant_vulns:
out_file.write(f" - {vuln['link']}\n")
out_file.write("\n")
else:
print(f"No relevant data found for {package_name}.")
else:
print(f"Failed to fetch data for {package_name}.")
def check_vulnerabilities(requirements: list, base_url: str) -> str:
results = []
for req in tqdm(requirements, desc="Checking vulnerabilities", unit="dependency"):
version = ""
if "==" in req:
package_name, version = req.split("==")
else:
package_name, version = req, None
url = f"{base_url}{package_name}"
# print(f"\nFetching data for {package_name} from {url}")
html_content = fetch_html(url)
if html_content:
extracted_data = parse_html(html_content)
if extracted_data:
relevant_vulns = []
for vuln in extracted_data:
if version_in_range(version, vuln["chip"]):
relevant_vulns.append(vuln)
if relevant_vulns:
result = f"Vulnerabilities found for {package_name}:\n"
for vuln in relevant_vulns:
result += f" - {vuln['link']}\n"
results.append(result)
return "\n".join(results)
def main():
parser = argparse.ArgumentParser(
description="Check project dependencies for vulnerabilities."
)
parser.add_argument(
"-r",
"--requirement",
help="Path to the requirements file of the project",
required=True,
)
parser.add_argument(
"-o",
"--output",
help="Output file path with extension, e.g., './output/report.txt'",
required=True,
)
args = parser.parse_args()
def save_to_file(output_path: str, data: str):
if output_path.endswith(".html"):
save_as_html(output_path, data)
elif output_path.endswith(".pdf"):
save_as_pdf(output_path, data)
elif output_path.endswith(".md"):
save_as_markdown(output_path, data)
else:
save_as_txt(output_path, data)
def save_as_html(output_path: str, data: str):
escaped_data = html.escape(data)
html_content = f"""
<html>
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="icon" href="https://s2.loli.net/2024/05/30/WDc6MekjbuCU9Qo.png">
<title>Vulnerability Report</title>
<style>
body {{
font-family: Arial, sans-serif;
background-image: url('https://s2.loli.net/2024/05/30/85Mv7leB2IRWNp6.jpg');
background-size: cover;
color: #333;
margin: 0;
padding: 0;
display: flex;
justify-content: center;
align-items: center;
height: 100vh;
}}
.container {{
background: rgba(255, 255, 255, 0.8);
border-radius: 10px;
padding: 20px;
box-shadow: 0 0 10px rgba(0, 0, 0, 0.1);
max-width: 800px;
width: 100%;
margin: 20px;
overflow-y: auto;
max-height: 90vh;
}}
.title {{
font-size: 24px;
font-weight: bold;
text-align: center;
margin-bottom: 20px;
}}
pre {{
white-space: pre-wrap;
word-wrap: break-word;
font-size: 14px;
line-height: 1.5;
color: #333;
background: #f4f4f4;
padding: 10px;
border-radius: 5px;
border: 1px solid #ddd;
overflow: auto;
font-weight: bold;
}}
</style>
</head>
<body>
<div class="container">
<div class="title">Vulnerability Report</div>
<pre>{escaped_data}</pre>
</div>
</body>
</html>
"""
with open(output_path, "w", encoding="utf-8") as file:
file.write(html_content)
def save_as_pdf(output_path: str, data: str):
doc = SimpleDocTemplate(output_path, pagesize=letter)
story = []
styles = getSampleStyleSheet()
# Add the title centered
title_style = ParagraphStyle(
"Title",
parent=styles["Title"],
alignment=1, # Center alignment
fontSize=24,
leading=28,
spaceAfter=20,
fontName="Helvetica-Bold",
)
title = Paragraph("Vulnerability Report", title_style)
story.append(title)
# Normal body text style
normal_style = ParagraphStyle(
"BodyText", parent=styles["BodyText"], fontSize=12, leading=15, spaceAfter=12
)
# Add the vulnerability details
for line in data.split("\n"):
if line.strip(): # Skip empty lines
story.append(Paragraph(line, normal_style))
doc.build(story)
def save_as_markdown(output_path: str, data: str):
with open(output_path, "w") as file:
file.write("## Vulnerability Report: \n\n")
file.write(data)
def save_as_txt(output_path: str, data: str):
with open(output_path, "w") as file:
file.write("Vulnerability Report: \n\n")
file.write(data)
def print_separator(title, char="-", length=50, padding=2):
print(f"{title:^{length + 4*padding}}") # 居中打印标题两侧各有padding个空格
print(char * (length + 2 * padding)) # 打印分割线两侧各有padding个字符的空格
def modify_file_name(file_path: str) -> str:
"""
Modify the file name by adding '-re' before the file extension.
Args:
file_path (str): The original file path.
Returns:
str: The modified file path.
"""
directory, file_name = os.path.split(file_path)
name, ext = os.path.splitext(file_name)
new_file_name = f"{name}-re{ext}"
new_file_path = os.path.join(directory, new_file_name)
return new_file_path
def requirement_detection(requirement_path, output_path=None):
base_url = "https://security.snyk.io/package/pip/"
# 分析项目依赖,包括名称和版本(如果有的话)
requirements = load_requirements(args.requirement)
# 传入依赖信息url前缀扫描结果输出位置
check_vulnerabilities(requirements, base_url, args.output)
print("Vulnerability scan complete. Results saved to", args.output)
if __name__ == "__main__":
main()
requirements = load_requirements(requirement_path)
results = check_vulnerabilities(requirements, base_url)
if output_path is not None:
new_path = modify_file_name(output_path)
save_to_file(new_path, results)
print(f"Vulnerability scan complete. Results saved to {output_path}")
print(f"Requirements scan complete. Results saved to {new_path}")
else:
print_separator("\nVulnerability Report", "=", 40, 5)
print(results)