コード例 #1
0
    def test_rsyncdelta_with_changes(self):
        changes_in_blocks = [
            (0, 0), (3, 2), (4, 0), (5, self.TEST_BLOCK_SIZE - 1),
            (math.ceil(len(self.TEST_FILE) / self.TEST_BLOCK_SIZE) - 1, 0)
        ]
        changed_blocks = [block for block, position in changes_in_blocks]

        with BytesIO(self.TEST_FILE) as changed_file:
            file_buffer = changed_file.getbuffer()

            for block, position in changes_in_blocks:
                file_buffer[block * self.TEST_BLOCK_SIZE + position] += 1

            changed_file_data = changed_file.getvalue()

            with BytesIO(self.TEST_FILE) as file_to:
                hashes = pyrsync2.blockchecksums(
                    file_to, blocksize=self.TEST_BLOCK_SIZE)

                delta = pyrsync2.rsyncdelta(changed_file,
                                            hashes,
                                            blocksize=self.TEST_BLOCK_SIZE,
                                            max_buffer=self.TEST_BLOCK_SIZE)

                for block, data in enumerate(delta):
                    if block in changed_blocks:
                        self.assertEqual(
                            self.get_block(changed_file_data, block), data)
                    else:
                        self.assertEqual(block, data)
コード例 #2
0
ファイル: PyRsyncTests.py プロジェクト: opprDev/SimpleCloud
	def testDeltaSameFile(self):
		with open("file1", "r+b") as file1:
			hashes1 = rsync.blockchecksums(file1)
			deltas = rsync.rsyncdelta(file1, hashes1)
			print("Printing deltas for same file:")
			print(deltas)
			for delta in deltas:
				print("d: "+str(delta))
コード例 #3
0
def common_rsync(patched_file, unpatched_file, resulting_file, blocksize):
	with open(unpatched_file, "rb") as unpatched, \
			open(patched_file, "rb") as patched, \
			open(resulting_file, "wb") as result:
		start = datetime.now()
		hashes = pyrsync2.blockchecksums(unpatched, blocksize)
		delta = pyrsync2.rsyncdelta(patched, hashes, blocksize)
		pyrsync2.patchstream(unpatched, result, delta, blocksize)
		duration = datetime.now() - start
	return duration
コード例 #4
0
    def get_delta(self, file):
        file_to = BytesIO(self.TEST_FILE)
        file_from = BytesIO(file)

        hashes = pyrsync2.blockchecksums(file_to,
                                         blocksize=self.TEST_BLOCK_SIZE)

        delta = pyrsync2.rsyncdelta(file_from,
                                    hashes,
                                    blocksize=self.TEST_BLOCK_SIZE)
        return list(delta)
コード例 #5
0
def syncFile(srcFile, dstFile):
    unpatched = open(dstFile, 'rb')
    hashes = pyrsync2.blockchecksums(unpatched)
    patchedFile = open(srcFile, 'rb')
    delta = pyrsync2.rsyncdelta(patchedFile, hashes)
    unpatched.seek(0)
    save_to = open(dstFile, 'wb')
    pyrsync2.patchstream(unpatched, save_to, delta)


# linkData('test', "D:/Clement Research/Test", "")
コード例 #6
0
ファイル: PyRsyncTests.py プロジェクト: opprDev/SimpleCloud
	def testCopying(self):
		with open("file1", "r+b") as file1, open("file2", "r+b") as file2:
			hashes2 = rsync.blockchecksums(file2)
			delta = rsync.rsyncdelta(file1, hashes2)
			print(hashes2)
			print(dir(hashes2))
			print(dir(delta))
			rsync.patchstream(file2, file2, delta)
		with open("file2", "r") as file2:
			line = file2.readline()
			self.assertEqual(TEXT, line)
			print(line)
コード例 #7
0
    def test_blockchecksums(self):
        with BytesIO(self.TEST_FILE) as file1:
            hashes = pyrsync2.blockchecksums(file1,
                                             blocksize=self.TEST_BLOCK_SIZE)

            for block, block_hash in enumerate(hashes):
                block_data = self.get_block(self.TEST_FILE, block)

                weaksum = pyrsync2.weakchecksum(block_data)[0]
                strongsum = hashlib.md5(block_data).digest()

                self.assertEqual(block_hash, (weaksum, strongsum))
コード例 #8
0
    def test_rsyncdelta_same_file(self):
        with BytesIO(self.TEST_FILE) as file_to:
            hashes = pyrsync2.blockchecksums(file_to,
                                             blocksize=self.TEST_BLOCK_SIZE)

            with BytesIO(self.TEST_FILE) as file_from:
                delta = pyrsync2.rsyncdelta(file_from,
                                            hashes,
                                            blocksize=self.TEST_BLOCK_SIZE)

                for index, block in enumerate(delta):
                    self.assertEqual(index, block)
コード例 #9
0
ファイル: PyRsyncTests.py プロジェクト: opprDev/SimpleCloud
	def testSingleDeltaForBigFile(self):
		# Create output file
		with open(BIG_OUT, "wb"):
			pass
		start = time.time()
		with open(BIG_OUT, "r+b") as outstream, open(BIG_IN, "rb") as instream:
			hashes = rsync.blockchecksums(outstream)
			deltas = rsync.rsyncdelta(instream, hashes)
			rsync.patchstream(outstream, outstream, deltas)
		finish = time.time()
		elapsed = finish - start
		print("Took " + str(elapsed) + " seconds")
		self.assertTrue(filecmp.cmp(BIG_IN, BIG_OUT, shallow=False))
コード例 #10
0
    def test_blockchecksums(self):
        with BytesIO(self.TEST_FILE) as file1:
            hashes = pyrsync2.blockchecksums(
                file1,
                blocksize=self.TEST_BLOCK_SIZE
            )

            for block, block_hash in enumerate(hashes):
                block_data = self.get_block(self.TEST_FILE, block)

                weaksum = pyrsync2.weakchecksum(block_data)[0]
                strongsum = hashlib.md5(block_data).digest()

                self.assertEqual(block_hash, (weaksum, strongsum))
コード例 #11
0
ファイル: index.py プロジェクト: georgyangelov/syncall
    def get_block_checksums(self, file_name, block_size):
        with self.fs_access_lock:
            if file_name not in self._index:
                return []

            file_data = self._get_index_unsafe(file_name)
            if 'deleted' in file_data and file_data['deleted']:
                return []

            with open(self.get_file_path(file_name), 'rb') as file:
                block_checksums = list(
                    pyrsync2.blockchecksums(file, blocksize=block_size))

        return block_checksums
コード例 #12
0
    def get_delta(self, file):
        file_to = BytesIO(self.TEST_FILE)
        file_from = BytesIO(file)

        hashes = pyrsync2.blockchecksums(
            file_to,
            blocksize=self.TEST_BLOCK_SIZE
        )

        delta = pyrsync2.rsyncdelta(
            file_from,
            hashes,
            blocksize=self.TEST_BLOCK_SIZE
        )
        return list(delta)
コード例 #13
0
    def test_rsyncdelta_same_file(self):
        with BytesIO(self.TEST_FILE) as file_to:
            hashes = pyrsync2.blockchecksums(
                file_to,
                blocksize=self.TEST_BLOCK_SIZE
            )

            with BytesIO(self.TEST_FILE) as file_from:
                delta = pyrsync2.rsyncdelta(
                    file_from, hashes,
                    blocksize=self.TEST_BLOCK_SIZE
                )

                for index, block in enumerate(delta):
                    self.assertEqual(index, block)
コード例 #14
0
ファイル: index.py プロジェクト: stormbreakerbg/syncall
    def get_block_checksums(self, file_name, block_size):
        with self.fs_access_lock:
            if file_name not in self._index:
                return []

            file_data = self._get_index_unsafe(file_name)
            if 'deleted' in file_data and file_data['deleted']:
                return []

            with open(self.get_file_path(file_name), 'rb') as file:
                block_checksums = list(pyrsync2.blockchecksums(
                    file,
                    blocksize=block_size
                ))

        return block_checksums
コード例 #15
0
ファイル: PyRsyncTests.py プロジェクト: opprDev/SimpleCloud
	def testMultipleDeltasForBigFile(self):
		# Create output file
		with open(BIG_OUT, "wb"):
			pass
		num_deltas = 0
		start = time.time()
		with open(BIG_OUT, "r+b") as outstream, open(BIG_IN, "rb") as instream:
			hashes = rsync.blockchecksums(outstream)
			deltas = rsync.rsyncdelta(instream, hashes)
			for delta in deltas:
				num_deltas += 1
				#print("delta: "+str(delta))
				rsync.patchstream(outstream, outstream, [delta])
		finish = time.time()
		elapsed = finish - start
		print("Took " + str(elapsed) + " seconds and "+str(num_deltas)+" individual deltas")
		self.assertTrue(filecmp.cmp(BIG_IN, BIG_OUT, shallow=False))
コード例 #16
0
    def test_rsyncdelta_with_changes(self):
        changes_in_blocks = [
            (0, 0),
            (3, 2),
            (4, 0),
            (5, self.TEST_BLOCK_SIZE - 1),
            (math.ceil(len(self.TEST_FILE) / self.TEST_BLOCK_SIZE) - 1, 0)
        ]
        changed_blocks = [block for block, position in changes_in_blocks]

        with BytesIO(self.TEST_FILE) as changed_file:
            file_buffer = changed_file.getbuffer()

            for block, position in changes_in_blocks:
                file_buffer[block * self.TEST_BLOCK_SIZE + position] += 1

            changed_file_data = changed_file.getvalue()

            with BytesIO(self.TEST_FILE) as file_to:
                hashes = pyrsync2.blockchecksums(
                    file_to,
                    blocksize=self.TEST_BLOCK_SIZE
                )

                delta = pyrsync2.rsyncdelta(
                    changed_file,
                    hashes,
                    blocksize=self.TEST_BLOCK_SIZE,
                    max_buffer=self.TEST_BLOCK_SIZE
                )

                for block, data in enumerate(delta):
                    if block in changed_blocks:
                        self.assertEqual(
                            self.get_block(changed_file_data, block),
                            data
                        )
                    else:
                        self.assertEqual(block, data)
コード例 #17
0
ファイル: PyRsyncTests.py プロジェクト: opprDev/SimpleCloud
	def testDeltaDifferentFiles(self):
		with open("file1", "r+b") as file1, open("file2", "r+b") as file2:
			hashes2 = rsync.blockchecksums(file2)
			delta = rsync.rsyncdelta(file1, hashes2)
			for i in delta:
				print("i: "+ str(i))
コード例 #18
0
def archive_files(source,
                  target,
                  compress=False,
                  rsync=False,
                  filetype=None,
                  limit=0,
                  verbose=False):
    verboseprint = print if verbose else lambda *a, **k: None
    target_filter = lambda fileinfo: filter_files(fileinfo, filetype, limit,
                                                  verbose)
    tardir, tarfn = os.path.split(target)
    target_iteration = list_archives(
        tardir,
        # Each month is an iteration
        re.compile(
            re.match('{0}-\d{{4}}-\d{{2}}'.format(socket.gethostname()),
                     tarfn).group(0)))
    try:
        if rsync and target_iteration:
            with io.BytesIO() as target_buffer, \
                    open(target_iteration[0], 'rb') as target_father_fid, \
                    open(target + '.diff', 'wb') as diff_fid:
                # Open TAR buffer for writing
                with tarfile.open(
                        fileobj=target_buffer,
                        mode='w|xz' if compress else 'w|') as target_fid:
                    for root in source:
                        target_fid.add(root, filter=target_filter)
                target_buffer.seek(0)

                # Write diff file based on rsync algorithm
                verboseprint('Making the diff file')
                hashes = pyrsync2.blockchecksums(target_father_fid)
                delta = pyrsync2.rsyncdelta(target_buffer,
                                            hashes,
                                            max_buffer=65535)
                for element in delta:
                    if isinstance(element, int):
                        diff_fid.write(b'\x00\x00')
                        diff_fid.write(element.to_bytes(8, byteorder='big'))
                    else:
                        verboseprint('Saving {} bytes'.format(len(element)))
                        diff_fid.write(
                            len(element).to_bytes(2, byteorder='big'))
                        diff_fid.write(element)
        else:
            # Open TAR file for writing
            with tarfile.open(target,
                              'w:xz' if compress else 'w') as target_fid:
                for root in source:
                    target_fid.add(root, filter=target_filter)

    except FileNotFoundError as not_found:
        if os.path.dirname(not_found.filename) == tardir:
            verboseprint('Making directory {}'.format(tardir))
            os.makedirs(tardir)
            return archive_files(source, target, compress, rsync, filetype,
                                 limit, verbose)
        else:
            print(not_found.filename + ' not found')
            return False

    else:
        return True