Issue #17429: platform.linux_distribution() now decodes files from the UTF-8

encoding with the surrogateescape error handler, instead of decoding from the
locale encoding in strict mode. It fixes the function on Fedora 19 which is
probably the first major distribution release with a non-ASCII name. Patch
written by Toshio Kuratomi.

4 files changed, 30 insertions, 2 deletions

diff --git a/Lib/platform.py b/Lib/platform.py
index ec5df85..030ef2a 100755
--- a/Lib/platform.py
+++ b/Lib/platform.py
@@ -129,6 +129,10 @@ except AttributeError:
         # Standard Unix uses /dev/null
         DEV_NULL = '/dev/null'
 
+# Directory to search for configuration information on Unix.
+# Constant used by test_platform to test linux_distribution().
+_UNIXCONFDIR = '/etc'
+
 ### Platform specific APIs
 
 _libc_search = re.compile(b'(__libc_init)'
@@ -315,7 +319,7 @@ def linux_distribution(distname='', version='', id='',
 
     """
     try:
-        etc = os.listdir('/etc')
+        etc = os.listdir(_UNIXCONFDIR)
     except os.error:
         # Probably not a Unix system
         return distname,version,id
@@ -331,7 +335,8 @@ def linux_distribution(distname='', version='', id='',
         return _dist_try_harder(distname,version,id)
 
     # Read the first line
-    with open('/etc/'+file, 'r') as f:
+    with open(os.path.join(_UNIXCONFDIR, file), 'r',
+              encoding='utf-8', errors='surrogateescape') as f:
         firstline = f.readline()
     _distname, _version, _id = _parse_release_file(firstline)
 
diff --git a/Lib/test/test_platform.py b/Lib/test/test_platform.py
index 0dcfe05..8a5408e 100644
--- a/Lib/test/test_platform.py
+++ b/Lib/test/test_platform.py
@@ -1,7 +1,10 @@
+from unittest import mock
+import contextlib
 import os
 import platform
 import subprocess
 import sys
+import tempfile
 import unittest
 import warnings
 
@@ -295,6 +298,19 @@ class PlatformTest(unittest.TestCase):
                     returncode = ret >> 8
                 self.assertEqual(returncode, len(data))
 
+    def test_linux_distribution_encoding(self):
+        # Issue #17429
+        with tempfile.TemporaryDirectory() as tempdir:
+            filename = os.path.join(tempdir, 'fedora-release')
+            with open(filename, 'w', encoding='utf-8') as f:
+                f.write('Fedora release 19 (Schr\xf6dinger\u2019s Cat)\n')
+
+            with mock.patch('platform._UNIXCONFDIR', tempdir):
+                distname, version, distid = platform.linux_distribution()
+
+            self.assertEqual(distname, 'Fedora')
+            self.assertEqual(version, '19')
+            self.assertEqual(distid, 'Schr\xf6dinger\u2019s Cat')
 
 def test_main():
     support.run_unittest(
diff --git a/Misc/ACKS b/Misc/ACKS
index 2c957e3..486e2af 100644
--- a/Misc/ACKS
+++ b/Misc/ACKS
@@ -689,6 +689,7 @@ Steven Kryskalla
 Andrew Kuchling
 Dave Kuhlman
 Jon Kuhn
+Toshio Kuratomi
 Vladimir Kushnir
 Erno Kuusela
 Ross Lagerwall
diff --git a/Misc/NEWS b/Misc/NEWS
index 5eacf9c..3596b4e 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -18,6 +18,12 @@ Core and Builtins
 Library
 -------
 
+- Issue #17429: platform.linux_distribution() now decodes files from the UTF-8
+  encoding with the surrogateescape error handler, instead of decoding from the
+  locale encoding in strict mode. It fixes the function on Fedora 19 which is
+  probably the first major distribution release with a non-ASCII name. Patch
+  written by Toshio Kuratomi.
+
 - Issue #19929: Call os.read with 32768 within subprocess.Popen.communicate
   rather than 4096 for efficiency.  A microbenchmark shows Linux and OS X
   both using ~50% less cpu time this way.