Python vulnerabilities

`xml.parsers.expat` and `xml.etree.ElementTree` use insufficient entropy for Expat hash-flooding protection, which allows a crafted XML document to trigger hash flooding.\r\n\r\nFully mitigating this vulnerability requires both updating libexpat to 2.8.0 or later and applying this patch.

If `shutil.unpack_archive()` is given a ZIP archive with an absolute Windows path containing a drive (`C:\\...`) then the archive will be extracted outside the target directory which is different than other operating systems. Only Windows is affected by this vulnerability.

http.cookies.Morsel.js_output() returns an inline <script> snippet and only escapes " for JavaScript string context. It does not neutralize the HTML parser-sensitive sequence </script> inside the generated script element. Mitigation base64-encodes the cookie value to disallow escaping using cookie value.

The webbrowser.open() API would accept leading dashes in the URL which could be handled as command line options for certain web browsers. New behavior rejects leading dashes. Users are recommended to sanitize URLs prior to passing to webbrowser.open().

When an Expat parser with a registered ElementDeclHandler parses an inline document type definition containing a deeply nested content model a C stack overflow occurs.

The fix for CVE-2026-0672, which rejected control characters in http.cookies.Morsel, was incomplete. The Morsel.update(), |= operator, and unpickling paths were not patched, allowing control characters to bypass input validation. Additionally, BaseCookie.js_output() lacked the output validation applied to BaseCookie.output().

The "tarfile" module would still apply normalization of AREGTYPE (\x00) blocks to DIRTYPE, even while processing a multi-block member such as GNUTYPE_LONGNAME or GNUTYPE_LONGLINK. This could result in a crafted tar archive being misinterpreted by the tarfile module compared to other implementations.

When passing data to the b64decode(), standard_b64decode(), and urlsafe_b64decode() functions in the "base64" module the characters "+/" will always be accepted, regardless of the value of "altchars" parameter, typically used to establish an "alternative base64 alphabet" such as the URL safe alphabet. This behavior matches what is recommended in earlier base64 RFCs, but newer RFCs now recommend either dropping characters outside the specified base64 alphabet or raising an error. The old behavior has the possibility of causing data integrity issues. This behavior can only be insecure if your application uses an alternate base64 alphabet (without "+/"). If your application does not use the "altchars" parameter or the urlsafe_b64decode() function, then your application does not use an alternative base64 alphabet. The attached patches DOES NOT make the base64-decode behavior raise an error, as this would be a change in behavior and break existing programs. Instead, the patch deprecates the behavior which will be replaced with the newly recommended behavior in a future version of Python. Users are recommended to mitigate by verifying user-controlled inputs match the base64 alphabet they are expecting or verify that their application would not be affected if the b64decode() functions accepted "+" or "/" outside of altchars.

When building nested elements using xml.dom.minidom methods such as appendChild() that have a dependency on _clear_id_cache() the algorithm is quadratic. Availability can be impacted when building excessively nested documents.

When loading a plist file, the plistlib module reads data in size specified by the file itself, meaning a malicious file can cause OOM and DoS issues

When reading an HTTP response from a server, if no read amount is specified, the default behavior will be to use Content-Length. This allows a malicious server to cause the client to read large amounts of data into memory, potentially causing OOM or other DoS.

An issue was discovered in 5.1 before 5.1.14, 4.2 before 4.2.26, and 5.2 before 5.2.8. The methods `QuerySet.filter()`, `QuerySet.exclude()`, and `QuerySet.get()`, and the class `Q()`, are subject to SQL injection when using a suitably crafted dictionary, with dictionary expansion, as the `_connector` argument. Earlier, unsupported Django series (such as 5.0.x, 4.1.x, and 3.2.x) were not evaluated and may also be affected. Django would like to thank cyberstan for reporting this issue.

An issue was discovered in 5.1 before 5.1.14, 4.2 before 4.2.26, and 5.2 before 5.2.8. NFKC normalization in Python is slow on Windows. As a consequence, `django.http.HttpResponseRedirect`, `django.http.HttpResponsePermanentRedirect`, and the shortcut `django.shortcuts.redirect` were subject to a potential denial-of-service attack via certain inputs with a very large number of Unicode characters. Earlier, unsupported Django series (such as 5.0.x, 4.1.x, and 3.2.x) were not evaluated and may also be affected. Django would like to thank Seokchan Yoon for reporting this issue.

If the value passed to os.path.expandvars() is user-controlled a performance degradation is possible when expanding environment variables.

Trust boundary violation in Visual Studio Code - Python extension allows an unauthorized attacker to execute code locally.

Tornado is a Python web framework and asynchronous networking library. When Tornado's ``multipart/form-data`` parser encounters certain errors, it logs a warning but continues trying to parse the remainder of the data. This allows remote attackers to generate an extremely high volume of logs, constituting a DoS attack. This DoS is compounded by the fact that the logging subsystem is synchronous. All versions of Tornado prior to 6.5.0 are affected. The vulnerable parser is enabled by default. Upgrade to Tornado version 6.50 to receive a patch. As a workaround, risk can be mitigated by blocking `Content-Type: multipart/form-data` in a proxy.

h11 is a Python implementation of HTTP/1.1. Prior to version 0.16.0, a leniency in h11's parsing of line terminators in chunked-coding message bodies can lead to request smuggling vulnerabilities under certain conditions. This issue has been patched in version 0.16.0. Since exploitation requires the combination of buggy h11 with a buggy (reverse) proxy, fixing either component is sufficient to mitigate this issue.

PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution (RCE) vulnerability exists in PyTorch when loading a model using torch.load with weights_only=True. This issue has been patched in version 2.6.0.

Python JSON Logger is a JSON Formatter for Python Logging. Between 30 December 2024 and 4 March 2025 Python JSON Logger was vulnerable to RCE through a missing dependency. This occurred because msgspec-python313-pre was deleted by the owner leaving the name open to being claimed by a third party. If the package was claimed, it would allow them RCE on any Python JSON Logger user who installed the development dependencies on Python 3.13 (e.g. pip install python-json-logger[dev]). This issue has been resolved with 3.3.0.

Visual Studio Code Python Extension Remote Code Execution Vulnerability

A vulnerability has been found in the CPython `venv` module and CLI where path names provided when creating a virtual environment were not quoted properly, allowing the creator to inject commands into virtual environment "activation" scripts (ie "source venv/bin/activate"). This means that attacker-controlled virtual environments are able to run commands when the virtual environment is activated. Virtual environments which are not created by an attacker or which aren't activated before being used (ie "./venv/bin/python") are not affected.

There is a MEDIUM severity vulnerability affecting CPython. Regular expressions that allowed excessive backtracking during tarfile.TarFile header parsing are vulnerable to ReDoS via specifically-crafted tar archives.

There is a LOW severity vulnerability affecting CPython, specifically the 'http.cookies' standard library module. When parsing cookies that contained backslashes for quoted characters in the cookie value, the parser would use an algorithm with quadratic complexity, resulting in excess CPU resources being used while parsing the value.

The email module of Python through 3.11.3 incorrectly parses e-mail addresses that contain a special character. The wrong portion of an RFC2822 header is identified as the value of the addr-spec. In some applications, an attacker can bypass a protection mechanism in which application access is granted only after verifying receipt of e-mail to a specific domain (e.g., only @company.example.com addresses may be used for signup). This occurs in email/_parseaddr.py in recent versions of Python.

A flaw was found in Python, specifically in the FTP (File Transfer Protocol) client library in PASV (passive) mode. The issue is how the FTP client trusts the host from the PASV response by default. This flaw allows an attacker to set up a malicious FTP server that can trick FTP clients into connecting back to a given IP address and port. This vulnerability could lead to FTP client scanning ports, which otherwise would not have been possible.

Python 3.x through 3.10 has an open redirection vulnerability in lib/http/server.py due to no protection against multiple (/) at the beginning of URI path which may leads to information disclosure. NOTE: this is disputed by a third party because the http.server.html documentation page states "Warning: http.server is not recommended for production. It only implements basic security checks."

zlib before 1.2.12 allows memory corruption when deflating (i.e., when compressing) if the input has many distant matches.

A flaw was found in python. An improperly handled HTTP response in the HTTP client code of python may allow a remote attacker, who controls the HTTP server, to make the client script enter an infinite loop, consuming CPU time. The highest threat from this vulnerability is to system availability.

A flaw was found in Python, specifically within the urllib.parse module. This module helps break Uniform Resource Locator (URL) strings into components. The issue involves how the urlparse method does not sanitize input and allows characters like '\r' and '\n' in the URL path. This flaw allows an attacker to input a crafted URL, leading to injection attacks. This flaw affects Python versions prior to 3.10.0b1, 3.9.5, 3.8.11, 3.7.11 and 3.6.14.

There's a flaw in Python 3's pydoc. A local or adjacent attacker who discovers or is able to convince another local or adjacent user to start a pydoc server could access the server and use it to disclose sensitive information belonging to the other user that they would not normally be able to access. The highest risk of this flaw is to data confidentiality. This flaw affects Python versions before 3.8.9, Python versions before 3.9.3 and Python versions before 3.10.0a7.

The package python/cpython from 0 and before 3.6.13, from 3.7.0 and before 3.7.10, from 3.8.0 and before 3.8.8, from 3.9.0 and before 3.9.2 are vulnerable to Web Cache Poisoning via urllib.parse.parse_qsl and urllib.parse.parse_qs by using a vector called parameter cloaking. When the attacker can separate query parameters using a semicolon (;), they can cause a difference in the interpretation of the request between the proxy (running with default configuration) and the server. This can result in malicious requests being cached as completely safe ones, as the proxy would usually not see the semicolon as a separator, and therefore would not include it in a cache key of an unkeyed parameter.

Python 3.x through 3.9.1 has a buffer overflow in PyCArg_repr in _ctypes/callproc.c, which may lead to remote code execution in certain Python applications that accept floating-point numbers as untrusted input, as demonstrated by a 1e300 argument to c_double.from_param. This occurs because sprintf is used unsafely.

Lib/zipfile.py in Python through 3.7.2 allows remote attackers to cause a denial of service (resource consumption) via a ZIP bomb.

Lib/webbrowser.py in Python through 3.6.3 does not validate strings before launching the program specified by the BROWSER environment variable, which might allow remote attackers to conduct argument-injection attacks via a crafted URL. NOTE: a software maintainer indicates that exploitation is impossible because the code relies on subprocess.Popen and the default shell=False setting

CPython (aka Python) up to 2.7.13 is vulnerable to an integer overflow in the PyString_DecodeEscape function in stringobject.c, resulting in heap-based buffer overflow (and possible arbitrary code execution)

Array index error in the scanstring function in the _json module in Python 2.7 through 3.5 and simplejson before 2.6.1 allows context-dependent attackers to read arbitrary process memory via a negative index value in the idx argument to the raw_decode function.

XML External Entity vulnerability in libexpat 2.2.0 and earlier (Expat XML Parser Library) allows attackers to put the parser in an infinite loop using a malformed external entity definition from an external DTD.

The DES and Triple DES ciphers, as used in the TLS, SSH, and IPSec protocols and other protocols and products, have a birthday bound of approximately four billion blocks, which makes it easier for remote attackers to obtain cleartext data via a birthday attack against a long-duration encrypted session, as demonstrated by an HTTPS session using Triple DES in CBC mode, aka a "Sweet32" attack.

Untrusted search path vulnerability in python.exe in Python through 3.5.0 on Windows allows local users to gain privileges via a Trojan horse readline.pyd file in the current working directory. NOTE: the vendor says "It was determined that this is a longtime behavior of Python that cannot really be altered at this point."

Multiple integer overflows in the XML_GetBuffer function in Expat through 2.1.0, as used in Google Chrome before 44.0.2403.89 and other products, allow remote attackers to cause a denial of service (heap-based buffer overflow) or possibly have unspecified other impact via crafted XML data, a related issue to CVE-2015-2716.

The HTTP clients in the (1) httplib, (2) urllib, (3) urllib2, and (4) xmlrpclib libraries in CPython (aka Python) 2.x before 2.7.9 and 3.x before 3.4.3, when accessing an HTTPS URL, do not (a) check the certificate against a trust store or verify that the server hostname matches a domain name in the subject's (b) Common Name or (c) subjectAltName field of the X.509 certificate, which allows man-in-the-middle attackers to spoof SSL servers via an arbitrary valid certificate.

Race condition in the _get_masked_mode function in Lib/os.py in Python 3.2 through 3.5, when exist_ok is set to true and multiple threads are used, might allow local users to bypass intended file permissions by leveraging a separate application vulnerability before the umask has been set to the expected value.

Integer overflow in bufferobject.c in Python before 2.7.8 allows context-dependent attackers to obtain sensitive information from process memory via a large size and offset in a "buffer" function.

OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCipherSpec messages, which allows man-in-the-middle attackers to trigger use of a zero-length master key in certain OpenSSL-to-OpenSSL communications, and consequently hijack sessions or obtain sensitive information, via a crafted TLS handshake, aka the "CCS Injection" vulnerability.

Python 2.7 before 3.4 only uses the last eight bits of the prefix to randomize hash values, which causes it to compute hash values without restricting the ability to trigger hash collisions predictably and makes it easier for context-dependent attackers to cause a denial of service (CPU consumption) via crafted input to an application that maintains a hash table. NOTE: this vulnerability exists because of an incomplete fix for CVE-2012-1150.

Python before 3.3.4 RC1 allows remote attackers to cause a denial of service (infinite loop and CPU consumption) via a file size value larger than the size of the zip file to the (1) ZipExtFile.read, (2) ZipExtFile.read(n), (3) ZipExtFile.readlines, (4) ZipFile.extract, or (5) ZipFile.extractall function.

The audioop module in Python 2.7 and 3.2 does not verify the relationships between size arguments and byte string lengths, which allows context-dependent attackers to cause a denial of service (memory corruption and application crash) via crafted arguments, as demonstrated by a call to audioop.reverse with a one-byte string, a different vulnerability than CVE-2010-1634.

Multiple integer overflows in audioop.c in the audioop module in Python 2.6, 2.7, 3.1, and 3.2 allow context-dependent attackers to cause a denial of service (application crash) via a large fragment, as demonstrated by a call to audioop.lin2lin with a long string in the first argument, leading to a buffer overflow. NOTE: this vulnerability exists because of an incorrect fix for CVE-2008-3143.5.

Multiple buffer overflows in the RLE decoder in the rgbimg module in Python 2.5 allow remote attackers to have an unspecified impact via an image file containing crafted data that triggers improper processing within the (1) longimagedata or (2) expandrow function.

Integer overflow in rgbimgmodule.c in the rgbimg module in Python 2.5 allows remote attackers to have an unspecified impact via a large image that triggers a buffer overflow. NOTE: this vulnerability exists because of an incomplete fix for CVE-2008-3143.12.