https://github.com/python/cpython/commit/72e5b25efb580fb1f0fdfade516be90d90822164
commit: 72e5b25efb580fb1f0fdfade516be90d90822164
branch: main
author: Chris Markiewicz <markiew...@stanford.edu>
committer: zware <zachary.w...@gmail.com>
date: 2025-03-07T21:04:45-06:00
summary:
gh-128646: Implement GzipFile.readinto[1]() methods (GH-128647)
The new methods simply delegate to the underlying buffer, much like the
existing GzipFile.read[1] methods. This avoids extra allocations caused by the
BufferedIOBase.readinto implementation previously used.
This commit also factors out a common readability check rather than copying it
an additional two times.
files:
A Misc/NEWS.d/next/Library/2025-01-08-15-14-17.gh-issue-128647.GabglU.rst
M Lib/gzip.py
M Lib/test/test_gzip.py
diff --git a/Lib/gzip.py b/Lib/gzip.py
index 7e384f8a568c1c..d681ef6b488dad 100644
--- a/Lib/gzip.py
+++ b/Lib/gzip.py
@@ -325,11 +325,15 @@ def _write_raw(self, data):
return length
- def read(self, size=-1):
- self._check_not_closed()
+ def _check_read(self, caller):
if self.mode != READ:
import errno
- raise OSError(errno.EBADF, "read() on write-only GzipFile object")
+ msg = f"{caller}() on write-only GzipFile object"
+ raise OSError(errno.EBADF, msg)
+
+ def read(self, size=-1):
+ self._check_not_closed()
+ self._check_read("read")
return self._buffer.read(size)
def read1(self, size=-1):
@@ -337,19 +341,25 @@ def read1(self, size=-1):
Reads up to a buffer's worth of data if size is negative."""
self._check_not_closed()
- if self.mode != READ:
- import errno
- raise OSError(errno.EBADF, "read1() on write-only GzipFile object")
+ self._check_read("read1")
if size < 0:
size = io.DEFAULT_BUFFER_SIZE
return self._buffer.read1(size)
+ def readinto(self, b):
+ self._check_not_closed()
+ self._check_read("readinto")
+ return self._buffer.readinto(b)
+
+ def readinto1(self, b):
+ self._check_not_closed()
+ self._check_read("readinto1")
+ return self._buffer.readinto1(b)
+
def peek(self, n):
self._check_not_closed()
- if self.mode != READ:
- import errno
- raise OSError(errno.EBADF, "peek() on write-only GzipFile object")
+ self._check_read("peek")
return self._buffer.peek(n)
@property
diff --git a/Lib/test/test_gzip.py b/Lib/test/test_gzip.py
index 0940bb114df625..260fae5ae1b368 100644
--- a/Lib/test/test_gzip.py
+++ b/Lib/test/test_gzip.py
@@ -143,6 +143,38 @@ def test_read1(self):
self.assertEqual(f.tell(), nread)
self.assertEqual(b''.join(blocks), data1 * 50)
+ def test_readinto(self):
+ # 10MB of uncompressible data to ensure multiple reads
+ large_data = os.urandom(10 * 2**20)
+ with gzip.GzipFile(self.filename, 'wb') as f:
+ f.write(large_data)
+
+ buf = bytearray(len(large_data))
+ with gzip.GzipFile(self.filename, 'r') as f:
+ nbytes = f.readinto(buf)
+ self.assertEqual(nbytes, len(large_data))
+ self.assertEqual(buf, large_data)
+
+ def test_readinto1(self):
+ # 10MB of uncompressible data to ensure multiple reads
+ large_data = os.urandom(10 * 2**20)
+ with gzip.GzipFile(self.filename, 'wb') as f:
+ f.write(large_data)
+
+ nread = 0
+ buf = bytearray(len(large_data))
+ memview = memoryview(buf) # Simplifies slicing
+ with gzip.GzipFile(self.filename, 'r') as f:
+ for count in range(200):
+ nbytes = f.readinto1(memview[nread:])
+ if not nbytes:
+ break
+ nread += nbytes
+ self.assertEqual(f.tell(), nread)
+ self.assertEqual(buf, large_data)
+ # readinto1() should require multiple loops
+ self.assertGreater(count, 1)
+
@bigmemtest(size=_4G, memuse=1)
def test_read_large(self, size):
# Read chunk size over UINT_MAX should be supported, despite zlib's
diff --git
a/Misc/NEWS.d/next/Library/2025-01-08-15-14-17.gh-issue-128647.GabglU.rst
b/Misc/NEWS.d/next/Library/2025-01-08-15-14-17.gh-issue-128647.GabglU.rst
new file mode 100644
index 00000000000000..034a66b704dea8
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2025-01-08-15-14-17.gh-issue-128647.GabglU.rst
@@ -0,0 +1,4 @@
+Eagerly write to buffers passed to :class:`gzip.GzipFile`'s
+:meth:`~io.BufferedIOBase.readinto` and
+:meth:`~io.BufferedIOBase.readinto1` implementations,
+avoiding unnecessary allocations. Patch by Chris Markiewicz.
_______________________________________________
Python-checkins mailing list -- python-checkins@python.org
To unsubscribe send an email to python-checkins-le...@python.org
https://mail.python.org/mailman3/lists/python-checkins.python.org/
Member address: arch...@mail-archive.com
