def dest_size(path):
"""
Retrieves size of backed up data.
"""
try:
with bz2.open(path, 'rt') as input:
return json_loads(input.read())
except Exception:
return 0
def dest_size(path):
"""
Retrieves size of backed up data.
"""
try:
with bz2.open(path, 'rt') as input:
return json_loads(input.read())
except Exception:
return 0
def test_freshbz2(self):
print_msg('Running {} test'.format(self.test_name))
os.system('../../src/mongob --config config.yaml' +
' --progress-file current_progress.yaml' +
' --log backup.log')
print_msg('Checking result for {}'.format(self.test_name))
with open(self.test_info['dataset_file'], 'rt') as input:
data_from_file = json_loads(input.read())
data_from_file.sort(key=lambda x: x['_id'])
data_from_src = list(self.coll_src.find().sort('_id', 1))
with bz2.open(self.db_dest, 'rt') as input:
data_from_dest = json_loads(input.read())
data_from_dest.sort(key=lambda x: x['_id'])
self.assertEqual(data_from_file, data_from_src)
self.assertEqual(data_from_file, data_from_dest)
def setup_dataset(coll, dataset_file):
"""
Setting up data set for a test by dropping the existing collection then
loading data from a JSON file.
"""
print_msg("Loading dataset for {}".format(coll.name))
coll.drop()
with open(dataset_file, 'r') as input:
coll.insert_many(json_loads(input.read()))
def setup_dataset(coll, dataset_file):
"""
Setting up data set for a test by dropping the existing collection then
loading data from a JSON file.
"""
print_msg("Loading dataset for {}".format(coll.name))
coll.drop()
with open(dataset_file, "r") as input:
coll.insert_many(json_loads(input.read()))
def test_freshbz2(self):
print_msg('Running {} test'.format(self.test_name))
os.system(
'../../src/mongob --config config.yaml'
+ ' --progress-file current_progress.yaml'
+ ' --log backup.log'
)
print_msg('Checking result for {}'.format(self.test_name))
with open(self.test_info['dataset_file'], 'rt') as input:
data_from_file = json_loads(input.read())
data_from_file.sort(key=lambda x: x['_id'])
data_from_src = list(self.coll_src.find().sort('_id', 1))
with bz2.open(self.db_dest, 'rt') as input:
data_from_dest = json_loads(input.read())
data_from_dest.sort(key=lambda x: x['_id'])
self.assertEqual(data_from_file, data_from_src)
self.assertEqual(data_from_file, data_from_dest)
def test_freshrun(self):
self.coll_dest.drop()
print_msg("Running {} test".format(self.test_name))
os.system(
"../../src/mongob --config config.yaml" + " --progress-file current_progress.yaml" + " --log backup.log"
)
print_msg("Checking result for {}".format(self.test_name))
with open(self.test_info["dataset_file"], "r") as input:
data_from_file = json_loads(input.read())
data_from_file.sort(key=lambda x: x["_id"])
data_from_src = list(self.coll_src.find().sort("_id", 1))
data_from_dest = list(self.coll_dest.find().sort("_id", 1))
self.assertEqual(data_from_file, data_from_src)
self.assertEqual(data_from_file, data_from_dest)
def insert_docs(path, docs):
"""
Inserts all docs into a destination file and returns IDs of the inserted
docs. This function makes no attempt at making sure there is no
duplication.
"""
# Not used:
# As we write one list each time we open destination while the desire output
# should be a single list, we need to do the list concatenation manually by
# removing the closing square bracket ("]") and add a comma to separate the
# currently inserted items.
# Destination file is created for the first time
if not os.path.isfile(path):
with bz2.open(path, 'wt') as output:
output.write(json_dumps(docs))
else:
# with bz2.open(path, 'r') as output:
# # Remove last ]
# output.seek(-1, os.SEEK_END)
# while output.peek(1)[:1] != b']' and output.tell() > 0:
# print(output.peek(1)[:1])
# output.seek(-1, os.SEEK_CUR)
# output.truncate()
# # Write , as list separator
# output.write(b', ')
# # Now, write docs without the beginning [
# output.write(json_dumps(docs)[1:].encode('utf-8'))
with bz2.open(path, 'rt') as input:
current_docs = json_loads(input.read())
current_docs.extend(docs)
with bz2.open(path, 'wt') as output:
output.write(json_dumps(current_docs))
return [doc['_id'] for doc in current_docs]
def insert_docs(path, docs):
"""
Inserts all docs into a destination file and returns IDs of the inserted
docs. This function makes no attempt at making sure there is no
duplication.
"""
# Not used:
# As we write one list each time we open destination while the desire output
# should be a single list, we need to do the list concatenation manually by
# removing the closing square bracket ("]") and add a comma to separate the
# currently inserted items.
# Destination file is created for the first time
if not os.path.isfile(path):
with bz2.open(path, 'wt') as output:
output.write(json_dumps(docs))
else:
# with bz2.open(path, 'r') as output:
# # Remove last ]
# output.seek(-1, os.SEEK_END)
# while output.peek(1)[:1] != b']' and output.tell() > 0:
# print(output.peek(1)[:1])
# output.seek(-1, os.SEEK_CUR)
# output.truncate()
# # Write , as list separator
# output.write(b', ')
# # Now, write docs without the beginning [
# output.write(json_dumps(docs)[1:].encode('utf-8'))
with bz2.open(path, 'rt') as input:
current_docs = json_loads(input.read())
current_docs.extend(docs)
with bz2.open(path, 'wt') as output:
output.write(json_dumps(current_docs))
return [doc['_id'] for doc in current_docs]
def fixmeta(meta):
return json_loads(json.dumps(meta))
