From 4959a45064755bc30a935576bf3b96f656c9ebb2 Mon Sep 17 00:00:00 2001
From: "Antoine R. Dumont (@ardumont)" <antoine.romain.dumont@gmail.com>
Date: Sat, 15 Sep 2018 00:48:52 +0200
Subject: [PATCH] swh.model.hashutil: Mark hash_* function as deprecated
And revert to their original behavior (returning only digest as bytes)
---
swh/model/hashutil.py | 37 ++++++++++++++------------------
swh/model/tests/test_hashutil.py | 24 ---------------------
2 files changed, 16 insertions(+), 45 deletions(-)
diff --git a/swh/model/hashutil.py b/swh/model/hashutil.py
index 339b72fa..ee29b6e6 100644
--- a/swh/model/hashutil.py
+++ b/swh/model/hashutil.py
@@ -259,8 +259,10 @@ def _new_hash(algo, length=None):
def hash_file(fobj, length=None, algorithms=DEFAULT_ALGORITHMS,
- chunk_cb=None, hash_format='bytes'):
- """Hash the contents of the given file object with the given algorithms.
+ chunk_cb=None):
+ """(Deprecated) cf. MultiHash.from_file
+
+ Hash the contents of the given file object with the given algorithms.
Args:
fobj: a file-like object
@@ -280,10 +282,6 @@ def hash_file(fobj, length=None, algorithms=DEFAULT_ALGORITHMS,
hash_format is an unknown hash format
"""
- if hash_format not in HASH_FORMATS:
- raise ValueError('Unexpected hash format %s, expected one of %s' % (
- hash_format, HASH_FORMATS))
-
h = MultiHash(algorithms, length)
while True:
chunk = fobj.read(HASH_BLOCK_SIZE)
@@ -293,17 +291,15 @@ def hash_file(fobj, length=None, algorithms=DEFAULT_ALGORITHMS,
if chunk_cb:
chunk_cb(chunk)
- if hash_format == 'bytes':
- return h.digest()
- if hash_format == 'bytehex':
- return h.bytehexdigest()
- return h.hexdigest()
+ return h.digest()
def hash_path(path, algorithms=DEFAULT_ALGORITHMS, chunk_cb=None,
- hash_format='bytes', track_length=True):
- """Hash the contents of the file at the given path with the given
- algorithms.
+ track_length=True):
+ """(deprecated) cf. MultiHash.from_path
+
+ Hash the contents of the file at the given path with the given
+ algorithms.
Args:
path (str): the path of the file to hash
@@ -327,12 +323,13 @@ def hash_path(path, algorithms=DEFAULT_ALGORITHMS, chunk_cb=None,
algorithms = set(['length']).union(algorithms)
length = os.path.getsize(path)
with open(path, 'rb') as fobj:
- return hash_file(fobj, length, algorithms, chunk_cb=chunk_cb,
- hash_format=hash_format)
+ return hash_file(fobj, length, algorithms, chunk_cb=chunk_cb)
+
+def hash_data(data, algorithms=DEFAULT_ALGORITHMS):
+ """(deprecated) cf. MultiHash.from_data
-def hash_data(data, algorithms=DEFAULT_ALGORITHMS, hash_format='bytes'):
- """Hash the given binary blob with the given algorithms.
+ Hash the given binary blob with the given algorithms.
Args:
data (bytes): raw content to hash
@@ -350,9 +347,7 @@ def hash_data(data, algorithms=DEFAULT_ALGORITHMS, hash_format='bytes'):
hash_format is an unknown hash format
"""
- fobj = BytesIO(data)
- length = len(data)
- return hash_file(fobj, length, algorithms, hash_format=hash_format)
+ return MultiHash.from_data(data, hash_names=algorithms).digest()
def hash_git_data(data, git_type, base_algo='sha1'):
diff --git a/swh/model/tests/test_hashutil.py b/swh/model/tests/test_hashutil.py
index d288149b..92b3684a 100644
--- a/swh/model/tests/test_hashutil.py
+++ b/swh/model/tests/test_hashutil.py
@@ -147,16 +147,6 @@ class Hashutil(BaseHashutil):
self.assertIn('Unexpected hashing algorithm', cm.exception.args[0])
self.assertIn('unknown-hash', cm.exception.args[0])
- @istest
- def hash_data_unknown_hash_format(self):
- with self.assertRaises(ValueError) as cm:
- hashutil.hash_data(
- self.data, hashutil.DEFAULT_ALGORITHMS,
- hash_format='unknown-format')
-
- self.assertIn('Unexpected hash format', cm.exception.args[0])
- self.assertIn('unknown-format', cm.exception.args[0])
-
@istest
def hash_git_data(self):
checksums = {
@@ -181,20 +171,6 @@ class Hashutil(BaseHashutil):
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), hash_format='hex')
- self.assertEqual(checksums, self.hex_checksums)
-
- @istest
- def hash_file_bytehexdigest(self):
- fobj = io.BytesIO(self.data)
- checksums = hashutil.hash_file(
- fobj, length=len(self.data), hash_format='bytehex')
- self.assertEqual(checksums, self.bytehex_checksums)
-
@istest
def hash_file_missing_length(self):
fobj = io.BytesIO(self.data)
--
GitLab