gh-41668: Add partial data recovery support for truncated gzip streams

Lib/compression/_common/_streams.py

@@ -115,8 +115,13 @@ def readall(self):
         # sys.maxsize means the max length of output buffer is unlimited,
         # so that the whole input buffer can be decompressed within one
         # .decompress() call.
-        while data := self.read(sys.maxsize):
-            chunks.append(data)
+        try:
+            while data := self.read(sys.maxsize):
+                chunks.append(data)
+        except EOFError as exc:
+            if chunks:
+                exc.partial_data = b"".join(chunks)
+            raise
 
         return b"".join(chunks)
 

Lib/gzip.py

@@ -677,9 +677,10 @@ def compress(data, compresslevel=_COMPRESS_LEVEL_TRADEOFF, *, mtime=0):
     return header + gzip_data[10:]
 
 
-def decompress(data):
+def decompress(data, *, partial=False):
     """Decompress a gzip compressed string in one shot.
     Return the decompressed string.
+    If partial is True, returns data decompressed prior to an unexpected EOF.
     """
     decompressed_members = []
     while True:
@@ -691,8 +692,14 @@ def decompress(data):
         # Read all the data except the header
         decompressed = do.decompress(data[fp.tell():])
         if not do.eof or len(do.unused_data) < 8:
-            raise EOFError("Compressed file ended before the end-of-stream "
+            if partial and decompressed:
+                decompressed_members.append(decompressed)
+                return b"".join(decompressed_members)
+            exc = EOFError("Compressed file ended before the end-of-stream "
                            "marker was reached")
+            if decompressed:
+                exc.partial_data = decompressed
+            raise exc
         crc, length = struct.unpack("<II", do.unused_data[:8])
         if crc != zlib.crc32(decompressed):
             raise BadGzipFile("CRC check failed")

Lib/test/test_gzip.py

@@ -827,6 +827,17 @@ def test_decompress_missing_trailer(self):
         compressed_data = gzip.compress(data1)
         self.assertRaises(EOFError, gzip.decompress, compressed_data[:-8])
 
+    def test_decompress_truncated_partial(self):
+        compressed_data = gzip.compress(data1)
+        truncated = compressed_data[:-8]
+        self.assertEqual(gzip.decompress(truncated, partial=True), data1)
+        try:
+            gzip.decompress(truncated)
+        except EOFError as exc:
+            self.assertEqual(getattr(exc, "partial_data", None), data1)
+        else:
+            self.fail("EOFError was not raised")
+
     def test_truncated_header(self):
         truncated_headers = [
             b"\x1f\x8b\x08\x00\x00\x00\x00\x00\x00",             # Missing OS byte

Misc/NEWS.d/next/Library/2026-10-28-01-44-00.gh-issue-41668.gzip-partial.rst

@@ -0,0 +1 @@
+Add partial data recovery support in :func:`gzip.decompress` via ``partial=True`` and attach ``partial_data`` attribute to :exc:`EOFError` on truncated gzip streams.