bpo-43650: Fix MemoryError on zip.read in shutil._unpack_zipfile for …

…large files (GH-25058)

`shutil.unpack_archive()` tries to read the whole file into memory, making no use of any kind of smaller buffer. Process crashes for really large files: I.e. archive: ~1.7G, unpacked: ~10G. Before the crash it can easily take away all available RAM on smaller systems. Had to pull the code form `zipfile.Zipfile.extractall()` to fix this

Automerge-Triggered-By: GH:gpshead
(cherry picked from commit f32c795)

Co-authored-by: Igor Bolshakov <ibolsch@gmail.com>

Lib/shutil.py

@@ -1159,20 +1159,16 @@ def _unpack_zipfile(filename, extract_dir):
  if name.startswith('/') or '..' in name:
  continue
 
- target = os.path.join(extract_dir, *name.split('/'))
- if not target:
+ targetpath = os.path.join(extract_dir, *name.split('/'))
+ if not targetpath:
  continue
 
- _ensure_directory(target)
+ _ensure_directory(targetpath)
  if not name.endswith('/'):
  # file
- data = zip.read(info.filename)
- f = open(target, 'wb')
- try:
- f.write(data)
- finally:
- f.close()
- del data
+ with zip.open(name, 'r') as source, \
+ open(targetpath, 'wb') as target:
+ copyfileobj(source, target)
  finally:
  zip.close()
 

Misc/NEWS.d/next/Library/2021-03-29-00-23-30.bpo-43650.v01tic.rst

@@ -0,0 +1,2 @@
+Fix :exc:`MemoryError` in :func:`shutil.unpack_archive` which fails inside
+:func:`shutil._unpack_zipfile` on large files. Patch by Igor Bolshakov.