https://github.com/python/cpython/commit/28eac9a7263ad8dcfa9b536aa238549131857e0f
commit: 28eac9a7263ad8dcfa9b536aa238549131857e0f
branch: main
author: Serhiy Storchaka <[email protected]>
committer: serhiy-storchaka <[email protected]>
date: 2026-05-22T12:17:34+03:00
summary:

gh-137571: Protect against possible UnboundLocalError in 
gzip._GzipReader.read() (GH-150222)

This has not been observed in practice, but we cannot be 100% sure that
it will not happen with some weird gzip data.

files:
M Lib/gzip.py

diff --git a/Lib/gzip.py b/Lib/gzip.py
index 8720acc4db9976..0713b922522ee1 100644
--- a/Lib/gzip.py
+++ b/Lib/gzip.py
@@ -610,10 +610,10 @@ def read(self, size=-1):
             # Read a chunk of data from the file
             if self._decompressor.needs_input:
                 buf = self._fp.read(READ_BUFFER_SIZE)
-                uncompress = self._decompressor.decompress(buf, size)
             else:
-                uncompress = self._decompressor.decompress(b"", size)
+                buf = b""
 
+            uncompress = self._decompressor.decompress(buf, size)
             if self._decompressor.unused_data != b"":
                 # Prepend the already read bytes to the fileobj so they can
                 # be seen by _read_eof() and _read_gzip_header()

_______________________________________________
Python-checkins mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://mail.python.org/mailman3//lists/python-checkins.python.org
Member address: [email protected]

Reply via email to