[Reproducible-commits] [debbindiff] 05/09: Add support for comparing zip and jar files

[Jérémy Bobbio]

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

lunar pushed a commit to branch master
in repository debbindiff.

commit 10967862602fea0ce76b045c2667c96bb0255be3
Author: Jérémy Bobbio <lunar at debian.org>
Date:   Tue Sep 30 12:07:07 2014 +0200

    Add support for comparing zip and jar files
---
 debbindiff/comparators/__init__.py |  3 ++
 debbindiff/comparators/zip.py      | 61 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 64 insertions(+)

diff --git a/debbindiff/comparators/__init__.py b/debbindiff/comparators/__init__.py
index 1e7dae7..ed44dd3 100644
--- a/debbindiff/comparators/__init__.py
+++ b/debbindiff/comparators/__init__.py
@@ -20,6 +20,7 @@
 import magic
 import os.path
 import re
+import sys
 from debbindiff import logger
 from debbindiff.difference import Difference, get_source
 from debbindiff.comparators.binary import compare_binary_files
@@ -33,6 +34,7 @@ from debbindiff.comparators.haskell import compare_hi_files
 from debbindiff.comparators.text import compare_text_files
 from debbindiff.comparators.tar import compare_tar_files
 from debbindiff.comparators.xz import compare_xz_files
+from debbindiff.comparators.zip import compare_zip_files
 
 def guess_mime_type(path):
     if not hasattr(guess_mime_type, 'mimedb'):
@@ -62,6 +64,7 @@ COMPARATORS = [
         (None,                                  r'\.mo$',              compare_mo_files),
         (r'^application/x-xz(;|$)',             r'\.xz$',              compare_xz_files),
         (r'^application/x-tar(;|$)',            r'\.tar$',             compare_tar_files),
+        (r'^application/zip(;|$)',              r'\.(zip|jar)$',       compare_zip_files),
         (r'^application/x-debian-package(;|$)', r'\.deb$',             compare_deb_files),
         (r'^application/x-gzip(;|$)',           r'\.gz$',              compare_gzip_files),
         (r'^application/x-bzip2(;|$)',          r'\.bzip2$',           compare_bzip2_files),
diff --git a/debbindiff/comparators/zip.py b/debbindiff/comparators/zip.py
new file mode 100644
index 0000000..ab0a83e
--- /dev/null
+++ b/debbindiff/comparators/zip.py
@@ -0,0 +1,61 @@
+# -*- coding: utf-8 -*-
+#
+# debbindiff: highlight differences between two builds of Debian packages
+#
+# Copyright © 2014 Jérémy Bobbio <lunar at debian.org>
+#
+# debdindiff is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# debbindiff is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with debbindiff.  If not, see <http://www.gnu.org/licenses/>.
+
+import os.path
+import subprocess
+from zipfile import ZipFile
+from debbindiff import logger
+from debbindiff.difference import Difference
+import debbindiff.comparators
+from debbindiff.comparators.utils import binary_fallback, make_temp_directory
+
+def get_zipinfo(path):
+    output = subprocess.check_output(['zipinfo', path], shell=False)
+    # the full path appears in the output, we need to remove it
+    return re.sub(re.escape(path), os.path.basename(path), output)
+
+ at binary_fallback
+def compare_zip_files(path1, path2, source=None):
+    differences = []
+    with ZipFile(path1, 'r') as zip1:
+        with ZipFile(path2, 'r') as zip2:
+            # look up differences in content
+            with make_temp_directory() as temp_dir1:
+                with make_temp_directory() as temp_dir2:
+                    for name in sorted(set(zip1.namelist()).intersection(zip2.namelist())):
+                        # skip directories
+                        if name.endswith('/'):
+                            continue
+                        logger.debug('extract member %s' % (name,))
+                        zip1.extract(name, temp_dir1)
+                        zip2.extract(name, temp_dir2)
+                        in_path1 = os.path.join(temp_dir1, name)
+                        in_path2 = os.path.join(temp_dir2, name)
+                        differences.extend(
+                            debbindiff.comparators.compare_files(
+                                in_path1, in_path2,
+                                source=name))
+                        os.unlink(in_path1)
+                        os.unlink(in_path2)
+            # look up differences in metadata
+            zipinfo1 = get_zipinfo(path1)
+            zipinfo2 = get_zipinfo(path2)
+            if zipinfo1 != zipinfo2:
+                differences.append(Difference(zipinfo1.splitlines(1), zipinfo2.splitlines(1), path1, path2, source="metadata"))
+    return differences

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