finding git content directly by hash

The GitWorkingCopy.content method was iterating over all commits of
the master branch before to find the commit with the specified hash.
When the commit was not on the master branch, that would fail. Instead,
it is not iterated over a branch anymore, but the commit is directly
accessed through GitPython.

sumatra/versioncontrol/_git.py

@@ -106,23 +106,21 @@ def diff(self):
     def content(self, digest, file=None):
         """Get the file content from repository."""
         repo = git.Repo(self.path)
-        for item in repo.iter_commits('master'):
-            if item.hexsha == digest:
-                curtree = item.tree
-                if file is None:
-                    return curtree.blobs[0].data_stream.read() # Get the latest added file content.
-                dirname,filename = os.path.split(file)
-                if dirname != '':
-                    for dname in dirname.split(os.path.sep):
-                        for subtree in curtree.trees:
-                            if subtree.name == dname:
-                                curtree = subtree
-                                break
-                for blob in curtree.blobs:
-                    if blob.name == filename:
-                        expected_encoding = sys.getfilesystemencoding()
-                        file_content = blob.data_stream.read().decode(expected_encoding)
-                        return file_content
+        curtree = repo.commit(digest).tree
+        if file is None:
+            return curtree.blobs[0].data_stream.read() # Get the latest added file content.
+        dirname, filename = os.path.split(file)
+        if dirname != '':
+            for dname in dirname.split(os.path.sep):
+                for subtree in curtree.trees:
+                    if subtree.name == dname:
+                        curtree = subtree
+                        break
+        for blob in curtree.blobs:
+            if blob.name == filename:
+                expected_encoding = sys.getfilesystemencoding()
+                file_content = blob.data_stream.read().decode(expected_encoding)
+                return file_content
         return 'File content not found.'
 
     def contains(self, path):