def compression_distance(x,y,l_x=None,l_y=None):
if x==y:
return 0
x_b = x.encode('utf-8')
y_b = y.encode('utf-8')
if l_x is None:
l_x = len(lzma.compress(x_b))
l_y = len(lzma.compress(y_b))
l_xy = len(lzma.compress(x_b+y_b))
l_yx = len(lzma.compress(y_b+x_b))
dist = (min(l_xy,l_yx)-min(l_x,l_y))/max(l_x,l_y)
return dist
def compression_distance(x, y, l_x=None, l_y=None):
if x == y:
return 0
x_b = x.encode('utf-8')
y_b = y.encode('utf-8')
if l_x is None:
l_x = len(lzma.compress(x_b))
l_y = len(lzma.compress(y_b))
l_xy = len(lzma.compress(x_b + y_b))
l_yx = len(lzma.compress(y_b + x_b))
dist = (min(l_xy, l_yx) - min(l_x, l_y)) / max(l_x, l_y)
return dist
def xz_compress(data):
'''decompress xz `data` using backports.lzma, or if that's not
available then the commandline `xz --decompress` tool
'''
if xz is not None:
try:
data = xz.compress(data)
except:
logger.error('decompress of %s bytes failed', len(data))
raise
else:
## launch xz child
xz_child = subprocess.Popen(
['xz', '--compress'],
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
## use communicate to pass the data incrementally to the child
## while reading the output, to avoid blocking
data, errors = xz_child.communicate(data)
assert not errors, errors
return data
def open_db(self):
self.terms_ldb = leveldb.LevelDB(self.terms_fl)
self.docs_ldb = leveldb.LevelDB(self.docs_fl)
self.doc_buffer_size = 0
self.term_buffer_size = 0
#self.doc_flush_buffer = np.empty(self.max_doc_flush_buffer_size, dtype="S%d" % self.max_doc_size)
self.doc_flush_buffer = [None] * self.max_doc_flush_buffer_size
self.term_flush_buffer = np.empty(self.max_term_flush_buffer_size, dtype="S%d" % self.max_term_size)
self.doc_id_flush_buffer = np.empty(self.max_doc_flush_buffer_size, dtype=np.int64)
self.term_id_flush_buffer = np.empty(self.max_term_flush_buffer_size, dtype=np.int64)
if self.compression == COMPRESSION.NONE:
self.compress = lambda string: string
self.decompress = lambda string: string
elif self.compression == COMPRESSION.ZLIB:
import zlib
self.compress = lambda string: zlib.compress(string, self.compression_level)
self.decompress = lambda string: zlib.decompress(string)
elif self.compression == COMPRESSION.LZMA:
import backports.lzma as lzma
self.compress = lambda string: lzma.compress(bytearray(string), format=lzma.FORMAT_RAW)
self.decompress = lambda data: lzma.decompress(data, format=lzma.FORMAT_RAW)
elif self.compression == COMPRESSION.LZ4R:
import lz4
self.compress = lambda string: lz4.compress(string)
self.decompress = lambda string: lz4.decompress(string)
elif self.compression == COMPRESSION.LZ4H:
import lz4
self.compress = lambda string: lz4.compressHC(string)
self.decompress = lambda string: lz4.decompress(string)
else:
raise Exception("Wrong compression type %r" % self.compression)
def close(self):
"""Compress and write remaining data, and also write the index."""
compressed = lzma.compress(self._queue)
self._outfile.write(compressed)
size = ((len(self._index) - 1) * self.in_block_size) + len(self._queue)
self._header.magic = b'ZEEX'
self._header.block_size = self.in_block_size
self._header.data_length = size
self._header.cdata_length = len(compressed) + self._last_out_pos
if USEBUFFER:
self._outfile.write(buffer(ctypes.c_uint32(len(self._index)))[:])
else:
self._outfile.write(ctypes.c_uint32(len(self._index)))
#sys.stderr.write("Index: \n")
for idx, i in enumerate(self._index):
#sys.stderr.write("\t %d: %d\n"% (idx,i))
if USEBUFFER:
self._outfile.write(buffer(ctypes.c_uint64(i))[:])
else:
self._outfile.write(ctypes.c_uint64(i))
self._outfile.seek(0)
if USEBUFFER:
self._outfile.write(buffer(self._header)[:])
else:
self._outfile.write(self._header)
self._outfile.close()
def xz_compress(data):
'''decompress xz `data` using backports.lzma, or if that's not
available then the commandline `xz --decompress` tool
'''
if xz is not None:
try:
data = xz.compress(data)
except:
logger.error('decompress of %s bytes failed', len(data))
raise
else:
## launch xz child
xz_child = subprocess.Popen(['xz', '--compress'],
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
## use communicate to pass the data incrementally to the child
## while reading the output, to avoid blocking
data, errors = xz_child.communicate(data)
assert not errors, errors
return data
def sync(package, source):
# update local package with source
# add package -> mirror url in
deps = source.find_deps(package)
all_pkgs = [package] + list(deps)
index_file = source.export_index(all_pkgs)
# save file into local
# if you use deb http://[hostname]/ubuntu main/, then use this
index_folder = mirror_root + '/main'
# else you use deb http://[hostname]/ubuntu bionic main, then use this
# index_folder = mirror_root + '/' + source.get_index_path()
os.system('mkdir -p ' + index_folder)
index_full_path = index_folder + '/' + 'Packages'
with open(index_full_path, 'w') as f:
f.write(index_file)
index_compress_full_path = index_folder + '/' + 'Packages.xz'
with open(index_compress_full_path, 'w') as f:
data = lzma.compress(index_file)
f.write(data)
release_full_path = index_folder + '/' + 'Release'
with open(release_full_path, 'w') as f:
f.write(
source.export_release(index_full_path, index_compress_full_path))
download_map = source.export_download_map(all_pkgs)
for key in download_map:
bin_full_path = mirror_root + '/' + key
bin_folder = mirror_root + '/' + '/'.join(key.split('/')[:-1])
print(bin_folder)
os.system('mkdir -p ' + bin_folder)
size = 0
try:
size = os.stat(bin_full_path).st_size
except:
pass
if size == download_map[key]['size']:
continue
if http_proxy == "":
command = 'wget ' + download_map[key][
'url'] + ' -O ' + bin_full_path
else:
command = 'wget -e use_proxy=yes -e http_proxy=' + http_proxy + ' ' + download_map[
key]['url'] + ' -O ' + bin_full_path
print(command)
os.system(command)
def write(self, data):
"""Feeds Data to the compressor. The data buffer will not be compressed till either it reaches the block size or the "close" funciton is called on the writer object.
@param data
Data to be fed to the writer
WARNING: These methods are made to resemble File class methods only for convenience. They may not behave intricately like one.
"""
self._queue += data
while (len(self._queue) >= self.in_block_size):
data = self._queue[0:self.in_block_size]
self._queue = self._queue[self.in_block_size:]
compressed = lzma.compress(data)
self._outfile.write(compressed)
cur_out_pos = self._last_out_pos + len(compressed)
self._index.append(cur_out_pos)
self._last_out_pos = cur_out_pos
def open_db(self):
self.terms_ldb = leveldb.LevelDB(self.terms_fl)
self.docs_ldb = leveldb.LevelDB(self.docs_fl)
self.doc_buffer_size = 0
self.term_buffer_size = 0
#self.doc_flush_buffer = np.empty(self.max_doc_flush_buffer_size, dtype="S%d" % self.max_doc_size)
self.doc_flush_buffer = [None] * self.max_doc_flush_buffer_size
self.term_flush_buffer = np.empty(self.max_term_flush_buffer_size,
dtype="S%d" % self.max_term_size)
self.doc_id_flush_buffer = np.empty(self.max_doc_flush_buffer_size,
dtype=np.int64)
self.term_id_flush_buffer = np.empty(self.max_term_flush_buffer_size,
dtype=np.int64)
if self.compression == COMPRESSION.NONE:
self.compress = lambda string: string
self.decompress = lambda string: string
elif self.compression == COMPRESSION.ZLIB:
import zlib
self.compress = lambda string: zlib.compress(
string, self.compression_level)
self.decompress = lambda string: zlib.decompress(string)
elif self.compression == COMPRESSION.LZMA:
import backports.lzma as lzma
self.compress = lambda string: lzma.compress(
bytearray(string), format=lzma.FORMAT_RAW)
self.decompress = lambda data: lzma.decompress(
data, format=lzma.FORMAT_RAW)
elif self.compression == COMPRESSION.LZ4R:
import lz4
self.compress = lambda string: lz4.compress(string)
self.decompress = lambda string: lz4.decompress(string)
elif self.compression == COMPRESSION.LZ4H:
import lz4
self.compress = lambda string: lz4.compressHC(string)
self.decompress = lambda string: lz4.decompress(string)
else:
raise Exception("Wrong compression type %r" % self.compression)
def dumps(obj):
"""Serialize Python object via pickle into a compressed string."""
return lzma.compress(pickle.dumps(obj))
## could check something about the FCs
#from dossier.fc.feature_collection import FeatureCollectionChunk as FCChunk
#fcc = FCChunk(path)
#for fc in fcc:
# print fc['feature_name']
## assume the incoming data is some long-term archival stuff in XZ
## compressed format, and you want to see how slow XZ is:
xz_data = open(args.path).read()
start = time.time()
data = xz.decompress(xz_data)
decompression_time = time.time() - start
start = time.time()
xz_data2 = xz.compress(data)
assert xz_data2 == xz_data
compression_time = time.time() - start
def report(rec):
rec['MB'] = rec['bytes'] / 2**20
rec['ratio'] = rec['bytes'] / len(data)
ctime = rec.get('compression_time')
rec['compression_rate'] = ctime and rec['MB'] / ctime or float('inf')
dtime = rec.get('decompression_time')
rec['decompression_rate'] = dtime and rec['MB'] / dtime or 0
print(
'%(name)s:\t%(MB)d MB of FC data, %(ratio).3f compression, %(compression_time).3f seconds --> %(compression_rate).3f MB/sec compression, '
'%(decompression_time).3f seconds --> %(decompression_rate).3f MB/sec decompression'
% rec)
def Z(contents) : return len(lzma.compress(contents, format=lzma.FORMAT_RAW, filters= lzma_filters))
def dumps(obj):
"""Serialize Python object via pickle into a compressed string."""
return lzma.compress(pickle.dumps(obj))
## could check something about the FCs
#from dossier.fc.feature_collection import FeatureCollectionChunk as FCChunk
#fcc = FCChunk(path)
#for fc in fcc:
# print fc['feature_name']
## assume the incoming data is some long-term archival stuff in XZ
## compressed format, and you want to see how slow XZ is:
xz_data = open(args.path).read()
start = time.time()
data = xz.decompress(xz_data)
decompression_time = time.time() - start
start = time.time()
xz_data2 = xz.compress(data)
assert xz_data2 == xz_data
compression_time = time.time() - start
def report(rec):
rec['MB'] = rec['bytes'] / 2**20
rec['ratio'] = rec['bytes'] / len(data)
ctime = rec.get('compression_time')
rec['compression_rate'] = ctime and rec['MB'] / ctime or float('inf')
dtime = rec.get('decompression_time')
rec['decompression_rate'] = dtime and rec['MB'] / dtime or 0
print('%(name)s:\t%(MB)d MB of FC data, %(ratio).3f compression, %(compression_time).3f seconds --> %(compression_rate).3f MB/sec compression, '
'%(decompression_time).3f seconds --> %(decompression_rate).3f MB/sec decompression' % rec)
sys.stdout.flush()
raw_rec = dict(name='raw', bytes=len(data), compression_time=0, decompression_time=0)
if sdist_formats == 'xztar':
tarxz_path = os.path.join(release_dir, 'deluge-%s.tar.xz' % version)
else:
# Compress release archive with xz
tar_path = os.path.join(release_dir, 'deluge-%s.tar' % version)
tarxz_path = tar_path + '.xz'
print('Compressing tar (%s) with xz' % tar_path)
try:
from backports import lzma
except ImportError:
print('backports.lzma not installed, falling back to xz shell command')
call(['xz', '-e9zkf', tar_path])
else:
with open(tar_path, 'rb') as tar_file, open(tarxz_path, 'wb') as xz_file:
xz_file.write(
lzma.compress(bytes(tar_file.read()), preset=9 | lzma.PRESET_EXTREME)
)
# Calculate shasum and add to sha256sums.txt
with open(tarxz_path, 'rb') as _file:
sha256sum = '%s %s' % (
sha256(_file.read()).hexdigest(),
os.path.basename(tarxz_path),
)
with open(os.path.join(release_dir, 'sha256sums.txt'), 'w') as _file:
_file.write(sha256sum + '\n')
print('Complete: %s' % release_dir)
def compress(data, *args, **kwargs):
return lzma.compress(data, *args, **kwargs)
def compress_image(image):
compressed = lzma.compress(bytearray(image))
return compressed.__sizeof__()
def put(self, items):
zitems = lzma.compress(cPickle.dumps(items))
self.bigcall('put', '/q/%s' % self.queue, data=zitems)
def dbput(self, table, doc): # doc may be a list of docs
zdoc = lzma.compress(cPickle.dumps(doc))
self.bigcall('put', '/db/%s' % table, zdoc)
