From 5ca5dce7216f5a2f3e3cfcf24b5e69270a577487 Mon Sep 17 00:00:00 2001
From: "Antoine R. Dumont (@ardumont)" <antoine.romain.dumont@gmail.com>
Date: Fri, 14 Sep 2018 00:21:33 +0200
Subject: [PATCH] hashutil: Allow option to require hexdigest instead of binary
digest
Related T421
---
swh/model/hashutil.py | 12 +++++++++---
swh/model/tests/test_hashutil.py | 7 +++++++
2 files changed, 16 insertions(+), 3 deletions(-)
diff --git a/swh/model/hashutil.py b/swh/model/hashutil.py
index 33551616..6675b5cb 100644
--- a/swh/model/hashutil.py
+++ b/swh/model/hashutil.py
@@ -162,7 +162,8 @@ def _new_hash(algo, length=None):
return _new_hashlib_hash(algo)
-def hash_file(fobj, length=None, algorithms=DEFAULT_ALGORITHMS, chunk_cb=None):
+def hash_file(fobj, length=None, algorithms=DEFAULT_ALGORITHMS, chunk_cb=None,
+ hexdigest=False):
"""Hash the contents of the given file object with the given algorithms.
Args:
@@ -171,11 +172,14 @@ def hash_file(fobj, length=None, algorithms=DEFAULT_ALGORITHMS, chunk_cb=None):
git-specific algorithms)
algorithms: the hashing algorithms to be used, as an iterable over
strings
+ hexdigest (bool): False returns the hash as binary, otherwise
+ returns as hex
- Returns: a dict mapping each algorithm to a bytes digest.
+ Returns: a dict mapping each algorithm to a digest (bytes by default).
Raises:
ValueError if algorithms contains an unknown hash algorithm.
+
"""
hashes = {algo: _new_hash(algo, length) for algo in algorithms}
@@ -188,6 +192,8 @@ def hash_file(fobj, length=None, algorithms=DEFAULT_ALGORITHMS, chunk_cb=None):
if chunk_cb:
chunk_cb(chunk)
+ if hexdigest:
+ return {algo: hash.hexdigest() for algo, hash in hashes.items()}
return {algo: hash.digest() for algo, hash in hashes.items()}
@@ -209,7 +215,7 @@ def hash_path(path, algorithms=DEFAULT_ALGORITHMS, chunk_cb=None):
"""
length = os.path.getsize(path)
with open(path, 'rb') as fobj:
- hash = hash_file(fobj, length, algorithms, chunk_cb)
+ hash = hash_file(fobj, length, algorithms, chunk_cb=chunk_cb)
hash['length'] = length
return hash
diff --git a/swh/model/tests/test_hashutil.py b/swh/model/tests/test_hashutil.py
index da49af99..fabbf16b 100644
--- a/swh/model/tests/test_hashutil.py
+++ b/swh/model/tests/test_hashutil.py
@@ -95,6 +95,13 @@ class Hashutil(unittest.TestCase):
checksums = hashutil.hash_file(fobj, length=len(self.data))
self.assertEqual(checksums, self.checksums)
+ @istest
+ def hash_file_hexdigest(self):
+ fobj = io.BytesIO(self.data)
+ checksums = hashutil.hash_file(fobj, length=len(self.data),
+ hexdigest=True)
+ self.assertEqual(checksums, self.hex_checksums)
+
@istest
def hash_file_missing_length(self):
fobj = io.BytesIO(self.data)
--
GitLab