def demo_get(service, params, compress=False):
url = build_url(service, params, compress)
print("REQ:")
print(url)
ua = UAResponse()
msg = MSG()
print()
print("RESP:")
if compress:
dst = io.BytesIO()
snappy.stream_decompress(urlopen(url), dst)
data = dst.getvalue()
else:
data = urlopen(url).read()
ua.ParseFromString(data)
if ua.Err == 0:
msg.ParseFromString(ua.Data)
# TODO: handle msg here
print(msg)
else:
print(ua)
def process_file(inpath, outpath):
"""Decompress snappy file into destination
"""
logging.debug("Parsing %s into %s", inpath, outpath)
with open(inpath, 'rb') as fin, open(outpath, 'wb') as fout:
snappy.stream_decompress(fin, fout)
return outpath
def decompress_to_file(blob, dest_file, compression):
"""
Decompresses the supplied blob of data into the dest_file file-like object using
the specified compression.
:param IOBase blob: A file-like object containing the compressed data.
:param IOBase dest_file: A file-like object into which the uncompressed data
should be written.
:param str compression: The compression algorithm to apply. Can be one of:
bzip2, gzip, snappy.
:rtype: None
"""
if compression == "snappy":
snappy = _try_import_snappy()
snappy.stream_decompress(blob, dest_file)
return
elif compression == "gzip":
source_file = gzip.GzipFile(fileobj=blob, mode="rb")
elif compression == "bzip2":
source_file = bz2.BZ2File(blob, "rb")
else:
raise ValueError("Unknown compression type: %s" % compression)
with source_file:
shutil.copyfileobj(source_file, dest_file)
def snappy_decompress(data):
if sz is None:
raise RuntimeError('Snappy decompression is not available')
## sz.decompress is broken per https://github.com/andrix/python-snappy/issues/28
fh = StringIO()
sz.stream_decompress(StringIO(data), fh)
return fh.getvalue()
def decompress_snappy(path):
directory_path = path[:-7]
with open(path, 'rb') as in_file:
with open(path_to_store, 'w') as out_file:
snappy.stream_decompress(in_file, out_file)
out_file.close()
in_file.close()
return directory_path
def stream_decompress(src_filename, dst_filename):
"""
文件解压
:param src_filename: 源文件文件名称
:param dst_filename: 压缩后的文件名称
:return:
"""
with open(src_filename, "rb") as rf, open(dst_filename, "wb") as wf:
stream_decompress(rf, wf)
def unpack(self):
out_labels = []
unpacked_files = []
# check if the file starts at offset 0. If not, carve the
# file first, as snappy tries to be smart and unpack
# all concatenated snappy data in a file.
havetmpfile = False
if not (self.offset == 0
and self.fileresult.filesize == self.unpacked_size):
temporary_file = tempfile.mkstemp(
dir=self.scan_environment.temporarydirectory)
havetmpfile = True
os.sendfile(temporary_file[0], self.infile.fileno(), self.offset,
self.unpacked_size)
os.fdopen(temporary_file[0]).close()
# determine the name of the output file
if self.fileresult.filename.suffix.lower() == '.sz':
file_path = pathlib.Path(self.fileresult.filename.stem)
else:
file_path = pathlib.Path("unpacked_from_snappy")
outfile_rel = self.rel_unpack_dir / file_path
outfile_full = self.scan_environment.unpack_path(outfile_rel)
os.makedirs(outfile_full.parent, exist_ok=True)
outfile = open(outfile_full, 'wb')
if havetmpfile:
infile = open(temporary_file[1], 'rb')
else:
infile = self.infile.infile
infile.seek(0)
try:
snappy.stream_decompress(infile, outfile)
outfile.close()
except Exception as e:
outfile.close()
if havetmpfile:
infile.close()
os.unlink(temporary_file[1])
raise UnpackParserException(e.args)
#return unpacked_files
if havetmpfile:
os.unlink(temporary_file[1])
fr = FileResult(self.fileresult, self.rel_unpack_dir / file_path,
set(out_labels))
unpacked_files.append(fr)
return unpacked_files
def _read_file(self, f: "pyarrow.NativeFile", path: str, **reader_args):
from pyarrow.fs import HadoopFileSystem
include_paths = reader_args.pop("include_paths", False)
if reader_args.get("compression") == "snappy":
import snappy
filesystem = reader_args.get("filesystem", None)
rawbytes = BytesIO()
if isinstance(filesystem, HadoopFileSystem):
snappy.hadoop_snappy.stream_decompress(src=f, dst=rawbytes)
else:
snappy.stream_decompress(src=f, dst=rawbytes)
data = rawbytes.getvalue()
else:
data = f.readall()
if include_paths:
return [(path, data)]
else:
return [data]
def decompress_snappy(data):
bio_in = BytesIO(data)
bio_out = BytesIO()
snappy.stream_decompress(bio_in, bio_out)
return bio_out.getvalue()
def decrypt_and_uncompress(data,
gpg_private=None,
tmp_dir=None,
compression='xz'):
'''Given a data buffer of bytes, if gpg_key_path is provided, decrypt
data using gnupg, and uncompress using `compression` scheme, which
defaults to "xz" and can also be "gz", "sz", or "".
:returns: a tuple of (logs, data), where `logs` is an array of
strings and data is a binary string
'''
if not data:
logger.error('decrypt_and_uncompress starting with empty data')
return ['no data'], None
_errors = []
if gpg_private is not None:
### setup gpg for decryption
gpg_dir = os.tempnam(tmp_dir, 'tmp-compress-and-encrypt-')
os.makedirs(gpg_dir)
try:
gpg_child = subprocess.Popen([
'gpg', '--no-permission-warning', '--homedir', gpg_dir,
'--import', gpg_private
],
stderr=subprocess.PIPE)
s_out, errors = gpg_child.communicate()
if errors:
_errors.append('gpg logs to stderr, read carefully:\n\n%s' %
errors)
## decrypt it, and free memory
## encrypt using the fingerprint for our trec-kba-rsa key pair
gpg_child = subprocess.Popen(
## setup gpg to decrypt with trec-kba private key
## (i.e. make it the recipient), with zero compression,
## ascii armoring is off by default, and --output - must
## appear before --decrypt -
[
'gpg', '--no-permission-warning', '--homedir', gpg_dir,
'--trust-model', 'always', '--output', '-', '--decrypt',
'-'
],
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
## communicate with child via its stdin
data, errors = gpg_child.communicate(data)
if errors:
_errors.append(errors)
finally:
## remove the gpg_dir
shutil.rmtree(gpg_dir, ignore_errors=True)
if not data:
logger.error('empty data after gpg decrypt')
_errors.append('gpg -> no data')
return _errors, None
if compression == 'xz':
data = xz_decompress(data)
elif compression == 'sz':
## sz.decompress is broken per https://github.com/andrix/python-snappy/issues/28
fh = StringIO()
sz.stream_decompress(StringIO(data), fh)
data = fh.getvalue()
elif compression == 'gz':
fh = StringIO(data)
gz_fh = gz.GzipFile(fileobj=fh, mode='r')
data = gz_fh.read(data)
elif compression == "" or compression is None:
## data is not compressed
pass
return _errors, data
