[Reproducible-commits] [diffoscope] 01/21: Open files in binary mode when we read/write raw bytes

[Jérémy Bobbio]

This is an automated email from the git hooks/post-receive script.

lunar pushed a commit to branch master
in repository diffoscope.

commit 22e1ebfe37be8c24e79f293f20997c91bc776633
Author: Jérémy Bobbio <lunar at debian.org>
Date:   Fri Sep 18 15:34:11 2015 +0200

    Open files in binary mode when we read/write raw bytes
    
    This changes is required for Python 3 as files opened in text mode read and
    write unicode strings when files opened in binary mode read and write raw
    bytes.
---
 diffoscope/comparators/binary.py     | 6 +++---
 diffoscope/comparators/deb.py        | 2 +-
 diffoscope/comparators/libarchive.py | 2 +-
 diffoscope/comparators/png.py        | 2 +-
 tests/comparators/test_text.py       | 2 +-
 5 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/diffoscope/comparators/binary.py b/diffoscope/comparators/binary.py
index 6294620..89ff4d1 100644
--- a/diffoscope/comparators/binary.py
+++ b/diffoscope/comparators/binary.py
@@ -49,7 +49,7 @@ def xxd(path):
 
 def hexdump_fallback(path):
     hexdump = ''
-    with open(path) as f:
+    with open(path, 'rb') as f:
         for buf in iter(lambda: f.read(32), b''):
             hexdump += u'%s\n' % hexlify(buf)
     return hexdump
@@ -120,7 +120,7 @@ class File(object):
             with self.get_content():
                 # tlsh is not meaningful with files smaller than 512 bytes
                 if os.stat(self.path).st_size >= 512:
-                    self._fuzzy_hash = tlsh.hash(open(self.path).read())
+                    self._fuzzy_hash = tlsh.hash(open(self.path, 'rb').read())
                 else:
                     self._fuzzy_hash = None
         return self._fuzzy_hash
@@ -162,7 +162,7 @@ class File(object):
         my_size = os.path.getsize(self.path)
         other_size = os.path.getsize(other.path)
         if my_size == other_size and my_size <= SMALL_FILE_THRESHOLD:
-            if open(self.path).read() == open(other.path).read():
+            if open(self.path, 'rb').read() == open(other.path, 'rb').read():
                 return True
 
         return 0 == subprocess.call(['cmp', '--silent', self.path, other.path],
diff --git a/diffoscope/comparators/deb.py b/diffoscope/comparators/deb.py
index 0d5cffb..157aafb 100644
--- a/diffoscope/comparators/deb.py
+++ b/diffoscope/comparators/deb.py
@@ -47,7 +47,7 @@ class ArContainer(Archive):
         member = self.archive.getmember(member_name)
         dest_path = os.path.join(dest_dir, os.path.basename(member_name))
         member.seek(0)
-        with open(dest_path, 'w') as fp:
+        with open(dest_path, 'wb') as fp:
             for buf in iter(lambda: member.read(AR_EXTRACTION_BUFFER_SIZE), b''):
                 fp.write(buf)
         return dest_path
diff --git a/diffoscope/comparators/libarchive.py b/diffoscope/comparators/libarchive.py
index 702324a..e0b8960 100644
--- a/diffoscope/comparators/libarchive.py
+++ b/diffoscope/comparators/libarchive.py
@@ -127,7 +127,7 @@ class LibarchiveContainer(Archive):
             for entry in archive:
                 if entry.pathname == member_name:
                     logger.debug('entry found, writing %s', dest_path)
-                    with open(dest_path, 'w') as f:
+                    with open(dest_path, 'wb') as f:
                         for buf in entry.get_blocks():
                             f.write(buf)
                     return dest_path
diff --git a/diffoscope/comparators/png.py b/diffoscope/comparators/png.py
index 22f81c0..1eceacd 100644
--- a/diffoscope/comparators/png.py
+++ b/diffoscope/comparators/png.py
@@ -32,7 +32,7 @@ class Sng(Command):
 
     def feed_stdin(self, stdin):
         try:
-            with open(self.path) as f:
+            with open(self.path, 'rb') as f:
                 for buf in iter(partial(f.read, 32768), b''):
                     stdin.write(buf)
         finally:
diff --git a/tests/comparators/test_text.py b/tests/comparators/test_text.py
index cf71108..2bd222c 100644
--- a/tests/comparators/test_text.py
+++ b/tests/comparators/test_text.py
@@ -69,7 +69,7 @@ def test_difference_between_iso88591_and_unicode(iso8859, unicode1):
 
 def test_difference_between_iso88591_and_unicode_only(iso8859, tmpdir):
     utf8_path = str(tmpdir.join('utf8'))
-    with open(utf8_path, 'w') as f:
+    with open(utf8_path, 'wb') as f:
         f.write(codecs.open(os.path.join(os.path.dirname(__file__), '../data/text_iso8859'), encoding='iso8859-1').read().encode('utf-8'))
     utf8 = specialize(FilesystemFile(utf8_path))
     difference = iso8859.compare(utf8)

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/reproducible/diffoscope.git