summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorLars Gustäbel <lars@gustaebel.de>2009-03-22 20:34:29 (GMT)
committerLars Gustäbel <lars@gustaebel.de>2009-03-22 20:34:29 (GMT)
commit42e0091208254a8af8e7a96c94559c000370a3b7 (patch)
treec21cd0243e0c55b497b89b1212df5f987d788d64
parentdb08306bc2f170921e58f0edfd158bae05926c18 (diff)
downloadcpython-42e0091208254a8af8e7a96c94559c000370a3b7.zip
cpython-42e0091208254a8af8e7a96c94559c000370a3b7.tar.gz
cpython-42e0091208254a8af8e7a96c94559c000370a3b7.tar.bz2
Merged revisions 70523 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk ........ r70523 | lars.gustaebel | 2009-03-22 21:09:33 +0100 (Sun, 22 Mar 2009) | 5 lines Issue #5068: Fixed the tarfile._BZ2Proxy.read() method that would loop forever on incomplete input. That caused tarfile.open() to hang when used with mode 'r' or 'r:bz2' and a fileobj argument that contained no data or partial bzip2 compressed data. ........
-rw-r--r--Lib/tarfile.py9
-rw-r--r--Lib/test/test_tarfile.py25
-rw-r--r--Misc/NEWS5
3 files changed, 34 insertions, 5 deletions
diff --git a/Lib/tarfile.py b/Lib/tarfile.py
index 4264da3..3a3d2c9 100644
--- a/Lib/tarfile.py
+++ b/Lib/tarfile.py
@@ -639,12 +639,11 @@ class _BZ2Proxy(object):
def read(self, size):
x = len(self.buf)
while x < size:
- try:
- raw = self.fileobj.read(self.blocksize)
- data = self.bz2obj.decompress(raw)
- self.buf += data
- except EOFError:
+ raw = self.fileobj.read(self.blocksize)
+ if not raw:
break
+ data = self.bz2obj.decompress(raw)
+ self.buf += data
x += len(data)
buf = self.buf[:size]
diff --git a/Lib/test/test_tarfile.py b/Lib/test/test_tarfile.py
index affbeeb..f4c0035 100644
--- a/Lib/test/test_tarfile.py
+++ b/Lib/test/test_tarfile.py
@@ -1121,6 +1121,30 @@ class Bz2WriteTest(WriteTest):
class Bz2StreamWriteTest(StreamWriteTest):
mode = "w|bz2"
+class Bz2PartialReadTest(unittest.TestCase):
+ # Issue5068: The _BZ2Proxy.read() method loops forever
+ # on an empty or partial bzipped file.
+
+ def _test_partial_input(self, mode):
+ class MyBytesIO(io.BytesIO):
+ hit_eof = False
+ def read(self, n):
+ if self.hit_eof:
+ raise AssertionError("infinite loop detected in tarfile.open()")
+ self.hit_eof = self.tell() == len(self.getvalue())
+ return super(MyBytesIO, self).read(n)
+
+ data = bz2.compress(tarfile.TarInfo("foo").tobuf())
+ for x in range(len(data) + 1):
+ tarfile.open(fileobj=MyBytesIO(data[:x]), mode=mode)
+
+ def test_partial_input(self):
+ self._test_partial_input("r")
+
+ def test_partial_input_bz2(self):
+ self._test_partial_input("r:bz2")
+
+
def test_main():
if not os.path.exists(TEMPDIR):
os.mkdir(TEMPDIR)
@@ -1178,6 +1202,7 @@ def test_main():
Bz2StreamReadTest,
Bz2WriteTest,
Bz2StreamWriteTest,
+ Bz2PartialReadTest,
]
try:
diff --git a/Misc/NEWS b/Misc/NEWS
index 0d3137f..460a36e 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -32,6 +32,11 @@ Core and Builtins
Library
-------
+- Issue #5068: Fixed the tarfile._BZ2Proxy.read() method that would loop
+ forever on incomplete input. That caused tarfile.open() to hang when used
+ with mode 'r' or 'r:bz2' and a fileobj argument that contained no data or
+ partial bzip2 compressed data.
+
- Issue #2110: Add support for thousands separator and 'n' type
specifier to Decimal.__format__