def test_large_data_streaming(self, path):
ldb = LooseObjectDB(path)
string_ios = list() # list of streams we previously created
# serial mode
for randomize in range(2):
desc = (randomize and 'random ') or ''
print("Creating %s data ..." % desc, file=sys.stderr)
st = time()
size, stream = make_memory_file(self.large_data_size_bytes, randomize)
elapsed = time() - st
print("Done (in %f s)" % elapsed, file=sys.stderr)
string_ios.append(stream)
# writing - due to the compression it will seem faster than it is
st = time()
sha = ldb.store(IStream('blob', size, stream)).binsha
elapsed_add = time() - st
assert ldb.has_object(sha)
db_file = ldb.readable_db_object_path(bin_to_hex(sha))
fsize_kib = os.path.getsize(db_file) / 1000
size_kib = size / 1000
print("Added %i KiB (filesize = %i KiB) of %s data to loose odb in %f s ( %f Write KiB / s)" %
(size_kib, fsize_kib, desc, elapsed_add, size_kib / (elapsed_add or 1)), file=sys.stderr)
# reading all at once
st = time()
ostream = ldb.stream(sha)
shadata = ostream.read()
elapsed_readall = time() - st
stream.seek(0)
assert shadata == stream.getvalue()
print("Read %i KiB of %s data at once from loose odb in %f s ( %f Read KiB / s)" %
(size_kib, desc, elapsed_readall, size_kib / (elapsed_readall or 1)), file=sys.stderr)
# reading in chunks of 1 MiB
cs = 512 * 1000
chunks = list()
st = time()
ostream = ldb.stream(sha)
while True:
data = ostream.read(cs)
chunks.append(data)
if len(data) < cs:
break
# END read in chunks
elapsed_readchunks = time() - st
stream.seek(0)
assert b''.join(chunks) == stream.getvalue()
cs_kib = cs / 1000
print("Read %i KiB of %s data in %i KiB chunks from loose odb in %f s ( %f Read KiB / s)" %
(size_kib, desc, cs_kib, elapsed_readchunks, size_kib / (elapsed_readchunks or 1)), file=sys.stderr)
# del db file so we keep something to do
ostream = None # To release the file handle (win)
remove(db_file)
def test_base(self):
ldb = LooseObjectDB(os.path.join(self.gitrepopath, 'objects'))
for sha1 in ldb.sha_iter():
oinfo = ldb.info(sha1)
ostream = ldb.stream(sha1)
assert oinfo[:3] == ostream[:3]
assert len(ostream.read()) == ostream.size
assert ldb.has_object(oinfo.binsha)
# END for each sha in database
# assure we close all files
try:
del(ostream)
del(oinfo)
except UnboundLocalError:
pass
# END ignore exception if there are no loose objects
data = "my data".encode("ascii")
istream = IStream("blob", len(data), BytesIO(data))
# the object does not yet have a sha
assert istream.binsha is None
ldb.store(istream)
# now the sha is set
assert len(istream.binsha) == 20
assert ldb.has_object(istream.binsha)
def test_base(self):
ldb = LooseObjectDB(os.path.join(self.gitrepopath, "objects"))
for sha1 in ldb.sha_iter():
oinfo = ldb.info(sha1)
ostream = ldb.stream(sha1)
assert oinfo[:3] == ostream[:3]
assert len(ostream.read()) == ostream.size
assert ldb.has_object(oinfo.binsha)
# END for each sha in database
# assure we close all files
try:
del (ostream)
del (oinfo)
except UnboundLocalError:
pass
# END ignore exception if there are no loose objects
data = "my data".encode("ascii")
istream = IStream("blob", len(data), BytesIO(data))
# the object does not yet have a sha
assert istream.binsha is None
ldb.store(istream)
# now the sha is set
assert len(istream.binsha) == 20
assert ldb.has_object(istream.binsha)
def test_base(self):
ldb = LooseObjectDB(fixture_path("../../../.git/objects"))
for sha1 in ldb.sha_iter():
oinfo = ldb.info(sha1)
ostream = ldb.stream(sha1)
assert oinfo[:3] == ostream[:3]
assert len(ostream.read()) == ostream.size
assert ldb.has_object(oinfo.binsha)
# END for each sha in database
# assure we close all files
try:
del(ostream)
del(oinfo)
except UnboundLocalError:
pass
# END ignore exception if there are no loose objects
data = "my data"
istream = IStream("blob", len(data), StringIO(data))
# the object does not yet have a sha
assert istream.binsha is None
ldb.store(istream)
# now the sha is set
assert len(istream.binsha) == 20
assert ldb.has_object(istream.binsha)
# async operation
# Create a reader from an iterator
reader = IteratorReader(ldb.sha_iter())
# get reader for object streams
info_reader = ldb.stream_async(reader)
# read one
info = info_reader.read(1)[0]
# read all the rest until depletion
ostreams = info_reader.read()
# set the pool to use two threads
pool.set_size(2)
# synchronize the mode of operation
pool.set_size(0)
def test_writing(self, path): mdb = MemoryDB() # write data self._assert_object_writing_simple(mdb) # test stream copy ldb = LooseObjectDB(path) assert ldb.size() == 0 num_streams_copied = mdb.stream_copy(mdb.sha_iter(), ldb) assert num_streams_copied == mdb.size() assert ldb.size() == mdb.size() for sha in mdb.sha_iter(): assert ldb.has_object(sha) assert ldb.stream(sha).read() == mdb.stream(sha).read()
def test_writing(self, path):
mdb = MemoryDB()
# write data
self._assert_object_writing_simple(mdb)
# test stream copy
ldb = LooseObjectDB(path)
assert ldb.size() == 0
num_streams_copied = mdb.stream_copy(mdb.sha_iter(), ldb)
assert num_streams_copied == mdb.size()
assert ldb.size() == mdb.size()
for sha in mdb.sha_iter():
assert ldb.has_object(sha)
assert ldb.stream(sha).read() == mdb.stream(sha).read()