PyDwarfs

A wrapper of DwarFS command line utilities.

DwarFS is stands for Deduplicating Warp-speed Advanced Read-only File System, it is a fast high compression read-only file system for Linux and Windows. See this project's homepage.

Note: This project currently does not provide support for Windows version of DwarFS command line utility.

Features / TODO

• Mount/unmount a DwarFS image file to specified mountpoint (dwarfs/mount.dwarfs)
• Extract a DwarFS image file (dwarfsextract, initially implemented)
• Create a DwarFS image file (mkdwarfs)
• Check a DwarFS image file (dwarfsck)
• Performance test (dwarfsbench)

Install

Just run the command: pip install pydwarfs.

Or you can download and install the wheel file from release page manually.

Examples

The following examples are for the more common uses of command line utilities.

Mount/unmount a DwarFS image file to specified mountpoint

Check if a path is a DwarFS mountpoint

>>> from pydwarfs.dwarfs import DwarFS
>>> DwarFS.isDwarFSMountPoint(this_is_dwarfs_mnt)
True
>>> DwarFS.isDwarFSMountPoint('/path/to/non-mnt')
False
>>> DwarFS.isDwarFSMountPoint('/proc')  # Obviously not a DwarFS mountpoint
False
>>>

List all of DwarFS mountpoint

>>> DwarFS.listAllDwarFSMountPoints()
[MountPoint(source='dwarfs', target=PosixPath('/path/to/dwarfs/image/mountpoint'), fstype='fuse.dwarfs', options=('rw', 'nosuid', 'nodev', 'relatime', 'user_id=1000', 'group_id=1000'), freq=0, passno=0), ...]
>>>

Create a DwarFS instance

In the default case, Dwarfs.init() will find the location of dwarfs command via shutil.which().

>>> from pydwarfs.dwarfs import DwarFS
>>> dwarfs = Dwarfs.init()
>>> dwarfs.executable
'/usr/bin/dwarfs'
>>>

If the DwarFS command line utilities are not installed, you can specify a location of dwarfs command instead.

>>> from pydwarfs.dwarfs import DwarFS
>>> dwarfs = Dwarfs.init('/path/to/valid/dwarfs')
>>> dwarfs.executable
'/path/to/valid/dwarfs'
>>>

A simple mount

>>> dwarfs.mount('/path/to/source.dwarfs', '/path/to/destination/directory')
>>>

To check if the mount was actually successful, exit the Python REPL and run the following command:

$ > findmnt /path/to/destination/directory
TARGET                         SOURCE FSTYPE      OPTIONS
/path/to/destination/directory dwarfs fuse.dwarfs rw,nosuid,nodev,relatime,user_id=1000,group_id=1000
$ >

Mount with some specified options

Use the dedicated attribute class DwarFSMountOptions:

>>> from pydwarfs.dwarfs import DwarFS, DwarFSMountOptions
>>> dwarfs = Dwarfs.init()
>>> options1 = DwarFSMountOptions(cachefile='32M', readonly=True, debuglevel='debug')
>>> dwarfs.mount('/path/to/source.1.dwarfs', '/path/to/destination/directory.1', options)
I 04:06:32.182380 [dwarfs_main.cpp:1328] dwarfs (v0.7.4, fuse version 35)
D 04:06:32.182909 [filesystem_v2.cpp:482] found valid section index
D 04:06:32.182927 [filesystem_v2.cpp:493] section BLOCK @ 64 [16,669,939 bytes]
...
D 04:06:32.191025 [filesystem_v2.cpp:515] read 734 blocks and 175,158 bytes of metadata
I 04:06:32.191041 [dwarfs_main.cpp:1144] file system initialized [8.624ms]
>>>

Or simply specify options by dict and **kwargs:

>>> from pydwarfs.dwarfs import DwarFS, DwarFSMountOptions
>>> dwarfs = Dwarfs.init()
>>> options2 = {'cachefile': '16M', 'readonly': True}
>>> dwarfs.mount('/path/to/source.2.dwarfs', '/path/to/destination/directory.2', options, debuglevel='debug')
I 04:14:01.981335 [dwarfs_main.cpp:1328] dwarfs (v0.7.4, fuse version 35)
D 04:14:01.983736 [filesystem_v2.cpp:482] found valid section index
D 04:14:01.983753 [filesystem_v2.cpp:493] section BLOCK @ 64 [67,097,028 bytes]
...
D 04:14:01.984644 [filesystem_v2.cpp:515] read 7 blocks and 2,755 bytes of metadata
I 04:14:01.984656 [dwarfs_main.cpp:1144] file system initialized [3.27ms]
>>>

Unmount

>>> dwarfs.unmount('/path/to/destination/directory')
>>>

Unmount by the method umount instead of the default fusermount:

>>> dwarfs.unmount('/path/to/destination/directory.1', method='umount')
>>>

You can forcely unmount by add the argument lazy_unmount=True:

>>> dwarfs.unmount('/path/to/destination/directory.2', lazy_unmount=True)
>>>

Extract a DwarFS image file

A simple extraction

>>> import os
>>> from pydwarfs.dwarfsextract import DwarFSExtract
>>> dwarfsextract = DwarFSExtract.init()
>>> output_dir = '/path/to/dest/directory'
>>> dwarfsextract.extract('/path/to/image.dwarfs', output_dir)
>>>
>>> os.listdir(output_dir)
['dataWin', 'Sakuna.exe', 'steam_appid.txt', 'xaudio2_9redist.dll', ...]
>>>

Note: The path pointed to by output_dir must be an existing directory.

Extract image file with progress

>>> another_output_dir = '/path/to/another/output/directory'
>>> for progress in dwarfsextract.extract('/path/to/image.dwarfs', another_output_dir, yield_progress=True):
...     print(progress)
...
100
99
98
...
2
1
0
>>> os.listdir(another_output_dir)
['dataWin', 'Sakuna.exe', 'steam_appid.txt', 'xaudio2_9redist.dll', ...]
>>>