public interface FuseFS extends Mountable
Most of these should work very similarly to the well known UNIX file system operations. A major exception is that instead of returning an error in 'errno', the operation should return the negated error value (-errno) directly.
All methods are optional, but some are essential for a useful filesystem (e.g. getattr). Open, flush, release, fsync, opendir, releasedir, fsyncdir, access, create, ftruncate, fgetattr, lock, init and destroy are special purpose methods, without which a full featured filesystem can still be implemented.
See http://fuse.sourceforge.net/wiki/ for more information.
Modifier and Type | Method and Description |
---|---|
int |
access(java.lang.String path,
int mask)
Check file access permissions
|
int |
bmap(java.lang.String path,
long blocksize,
long idx)
Map block index within file to block index within device
|
int |
chmod(java.lang.String path,
long mode)
Change the permission bits of a file
|
int |
chown(java.lang.String path,
long uid,
long gid)
Change the owner and group of a file
|
int |
create(java.lang.String path,
long mode,
FuseFileInfo fi)
Create and open a file
|
void |
destroy(jnr.ffi.Pointer initResult)
Clean up filesystem
|
int |
fallocate(java.lang.String path,
int mode,
long off,
long length,
FuseFileInfo fi)
Allocates space for an open file
|
int |
fgetattr(java.lang.String path,
FileStat stbuf,
FuseFileInfo fi)
Get attributes from an open file
|
int |
flock(java.lang.String path,
FuseFileInfo fi,
int op)
Perform BSD file locking operation
|
int |
flush(java.lang.String path,
FuseFileInfo fi)
Possibly flush cached data
|
int |
fsync(java.lang.String path,
int isdatasync,
FuseFileInfo fi)
Synchronize file contents
|
int |
fsyncdir(java.lang.String path,
FuseFileInfo fi)
Synchronize directory contents
|
int |
ftruncate(java.lang.String path,
long size,
FuseFileInfo fi)
Change the size of an open file
|
int |
getattr(java.lang.String path,
FileStat stat)
Get file attributes.
|
int |
getxattr(java.lang.String path,
java.lang.String name,
jnr.ffi.Pointer value,
long size)
Get the attribute NAME of the file pointed to by PATH to VALUE (which is
SIZE bytes long).
|
jnr.ffi.Pointer |
init(jnr.ffi.Pointer conn)
Initialize filesystem
|
int |
ioctl(java.lang.String path,
int cmd,
jnr.ffi.Pointer arg,
FuseFileInfo fi,
long flags,
jnr.ffi.Pointer data)
Ioctl
|
int |
link(java.lang.String oldpath,
java.lang.String newpath)
Create a hard link to a file
|
int |
listxattr(java.lang.String path,
jnr.ffi.Pointer list,
long size)
List extended attributes
|
int |
lock(java.lang.String path,
FuseFileInfo fi,
int cmd,
Flock flock)
Perform POSIX file locking operation
|
int |
mkdir(java.lang.String path,
long mode)
Create a directory
|
int |
mknod(java.lang.String path,
long mode,
long rdev)
Create a file node
|
int |
open(java.lang.String path,
FuseFileInfo fi)
File open operation
|
int |
opendir(java.lang.String path,
FuseFileInfo fi)
Open directory
|
int |
poll(java.lang.String path,
FuseFileInfo fi,
FusePollhandle ph,
jnr.ffi.Pointer reventsp)
Poll for IO readiness events
|
int |
read_buf(java.lang.String path,
jnr.ffi.Pointer bufp,
long size,
long off,
FuseFileInfo fi)
Store data from an open file in a buffer
|
int |
read(java.lang.String path,
jnr.ffi.Pointer buf,
long size,
long offset,
FuseFileInfo fi)
Read data from an open file
|
int |
readdir(java.lang.String path,
jnr.ffi.Pointer buf,
FuseFillDir filter,
long offset,
FuseFileInfo fi)
Read directory
|
int |
readlink(java.lang.String path,
jnr.ffi.Pointer buf,
long size)
Read the target of a symbolic link
|
int |
release(java.lang.String path,
FuseFileInfo fi)
Release an open file
|
int |
releasedir(java.lang.String path,
FuseFileInfo fi)
Release directory
|
int |
removexattr(java.lang.String path,
java.lang.String name)
Remove the attribute NAME from the file pointed to by PATH.
|
int |
rename(java.lang.String oldpath,
java.lang.String newpath)
Rename a file
|
int |
rmdir(java.lang.String path)
Remove a directory
|
int |
setxattr(java.lang.String path,
java.lang.String name,
jnr.ffi.Pointer value,
long size,
int flags)
Set the attribute NAME of the file pointed to by PATH to VALUE (which
is SIZE bytes long).
|
int |
statfs(java.lang.String path,
Statvfs stbuf)
Get file system statistics
|
int |
symlink(java.lang.String oldpath,
java.lang.String newpath)
Create a symbolic link
|
int |
truncate(java.lang.String path,
long size)
Change the size of a file
|
int |
unlink(java.lang.String path)
Remove a file
|
int |
utimens(java.lang.String path,
Timespec[] timespec)
Change the access and modification times of a file with
nanosecond resolution
|
int |
write_buf(java.lang.String path,
FuseBufvec buf,
long off,
FuseFileInfo fi)
Write contents of buffer to an open file
|
int |
write(java.lang.String path,
jnr.ffi.Pointer buf,
long size,
long offset,
FuseFileInfo fi)
Write data to an open file
|
int getattr(java.lang.String path, FileStat stat)
Similar to stat(). The 'st_dev' and 'st_blksize' fields are ignored. The 'st_ino' field is ignored except if the 'use_ino' mount option is given.
int readlink(java.lang.String path, jnr.ffi.Pointer buf, long size)
The buffer should be filled with a null terminated string. The buffer size argument includes the space for the terminating null character. If the linkname is too long to fit in the buffer, it should be truncated. The return value should be 0 for success.
int mknod(java.lang.String path, long mode, long rdev)
This is called for creation of all non-directory, non-symlink nodes. If the filesystem defines a create() method, then for regular files that will be called instead.
mode
- The argument mode specifies the permissions to use in case a new file
is created. @see ru.serce.jnrfuse.struct.FileStat flagsint mkdir(java.lang.String path, long mode)
Note that the mode argument may not have the type specification bits set, i.e. S_ISDIR(mode) can be false. To obtain the correct directory type bits use mode|S_IFDIR
mode
- The argument mode specifies the permissions to use in case a new file
is created. @see ru.serce.jnrfuse.struct.FileStat flagsint unlink(java.lang.String path)
int rmdir(java.lang.String path)
int symlink(java.lang.String oldpath, java.lang.String newpath)
int rename(java.lang.String oldpath, java.lang.String newpath)
int link(java.lang.String oldpath, java.lang.String newpath)
int chmod(java.lang.String path, long mode)
mode
- The argument mode specifies the permissions to use in case a new file
is created. @see ru.serce.jnrfuse.struct.FileStat flagsint chown(java.lang.String path, long uid, long gid)
int truncate(java.lang.String path, long size)
int open(java.lang.String path, FuseFileInfo fi)
No creation (O_CREAT, O_EXCL) and by default also no truncation (O_TRUNC) flags will be passed to open(). If an application specifies O_TRUNC, fuse first calls truncate() and then open(). Only if 'atomic_o_trunc' has been specified and kernel version is 2.6.24 or later, O_TRUNC is passed on to open.
Unless the 'default_permissions' mount option is given, open should check if the operation is permitted for the given flags. Optionally open may also return an arbitrary filehandle in the fuse_file_info structure, which will be passed to all file operations.
OpenFlags
int read(java.lang.String path, jnr.ffi.Pointer buf, long size, long offset, FuseFileInfo fi)
Read should return exactly the number of bytes requested except on EOF or error, otherwise the rest of the data will be substituted with zeroes. An exception to this is when the 'direct_io' mount option is specified, in which case the return value of the read system call will reflect the return value of this operation.
int write(java.lang.String path, jnr.ffi.Pointer buf, long size, long offset, FuseFileInfo fi)
Write should return exactly the number of bytes requested except on error. An exception to this is when the 'direct_io' mount option is specified (see read operation).
int statfs(java.lang.String path, Statvfs stbuf)
The 'f_frsize', 'f_favail', 'f_fsid' and 'f_flag' fields are ignored
int flush(java.lang.String path, FuseFileInfo fi)
BIG NOTE: This is not equivalent to fsync(). It's not a request to sync dirty data.
Flush is called on each close() of a file descriptor. So if a filesystem wants to return write errors in close() and the file has cached dirty data, this is a good place to write back data and return any errors. Since many applications ignore close() errors this is not always useful.
NOTE: The flush() method may be called more than once for each open(). This happens if more than one file descriptor refers to an opened file due to dup(), dup2() or fork() calls. It is not possible to determine if a flush is final, so each flush should be treated equally. Multiple write-flush sequences are relatively rare, so this shouldn't be a problem.
Filesystems shouldn't assume that flush will always be called after some writes, or that if will be called at all.
int release(java.lang.String path, FuseFileInfo fi)
Release is called when there are no more references to an open file: all file descriptors are closed and all memory mappings are unmapped.
For every open() call there will be exactly one release() call with the same flags and file descriptor. It is possible to have a file opened more than once, in which case only the last release will mean, that no more reads/writes will happen on the file. The return value of release is ignored.
int fsync(java.lang.String path, int isdatasync, FuseFileInfo fi)
If the datasync parameter is non-zero, then only the user data should be flushed, not the meta data.
int setxattr(java.lang.String path, java.lang.String name, jnr.ffi.Pointer value, long size, int flags)
flags
- @see XAttrConstants
int getxattr(java.lang.String path, java.lang.String name, jnr.ffi.Pointer value, long size)
int listxattr(java.lang.String path, jnr.ffi.Pointer list, long size)
The retrieved list is placed in list, a caller-allocated buffer whose size (in bytes) is specified in the argument size. The list is the set of (null-terminated) names, one after the other. Names of extended attributes to which the calling process does not have access may be omitted from the list. The length of the attribute name list is returned
int removexattr(java.lang.String path, java.lang.String name)
int opendir(java.lang.String path, FuseFileInfo fi)
Unless the 'default_permissions' mount option is given, this method should check if opendir is permitted for this directory. Optionally opendir may also return an arbitrary filehandle in the fuse_file_info structure, which will be passed to readdir, closedir and fsyncdir.
int readdir(java.lang.String path, jnr.ffi.Pointer buf, FuseFillDir filter, long offset, FuseFileInfo fi)
This supersedes the old getdir() interface. New applications should use this.
The filesystem may choose between two modes of operation:
1) The readdir implementation ignores the offset parameter, and passes zero to the filler function's offset. The filler function will not return '1' (unless an error happens), so the whole directory is read in a single readdir operation. This works just like the old getdir() method.
2) The readdir implementation keeps track of the offsets of the directory entries. It uses the offset parameter and always passes non-zero offset to the filler function. When the buffer is full (or an error happens) the filler function will return '1'.
int releasedir(java.lang.String path, FuseFileInfo fi)
int fsyncdir(java.lang.String path, FuseFileInfo fi)
If the datasync parameter is non-zero, then only the user data should be flushed, not the meta data
jnr.ffi.Pointer init(jnr.ffi.Pointer conn)
The return value will passed in the private_data field of fuse_context to all file operations and as a parameter to the destroy() method.
void destroy(jnr.ffi.Pointer initResult)
Called on filesystem exit.
int access(java.lang.String path, int mask)
This will be called for the access() system call. If the 'default_permissions' mount option is given, this method is not called.
This method is not called under Linux kernel versions 2.4.x
mask
- see @{link ru.serce.jnrfuse.flags.AccessConstants}int create(java.lang.String path, long mode, FuseFileInfo fi)
If the file does not exist, first create it with the specified mode, and then open it.
If this method is not implemented or under Linux kernel versions earlier than 2.6.15, the mknod() and open() methods will be called instead.
mode
- The argument mode specifies the permissions to use in case a new file
is created. See ru.serce.jnrfuse.struct.FileStat flagsint ftruncate(java.lang.String path, long size, FuseFileInfo fi)
This method is called instead of the truncate() method if the truncation was invoked from an ftruncate() system call.
If this method is not implemented or under Linux kernel versions earlier than 2.6.15, the truncate() method will be called instead.
int fgetattr(java.lang.String path, FileStat stbuf, FuseFileInfo fi)
This method is called instead of the getattr() method if the file information is available.
Currently this is only called after the create() method if that is implemented (see above). Later it may be called for invocations of fstat() too.
int lock(java.lang.String path, FuseFileInfo fi, int cmd, Flock flock)
The cmd argument will be either F_GETLK, F_SETLK or F_SETLKW.
For the meaning of fields in 'struct flock' see the man page for fcntl(2). The l_whence field will always be set to SEEK_SET.
For checking lock ownership, the 'fuse_file_info->owner' argument must be used.
For F_GETLK operation, the library will first check currently held locks, and if a conflicting lock is found it will return information without calling this method. This ensures, that for local locks the l_pid field is correctly filled in. The results may not be accurate in case of race conditions and in the presence of hard links, but it's unlikely that an application would rely on accurate GETLK results in these cases. If a conflicting lock is not found, this method will be called, and the filesystem may fill out l_pid by a meaningful value, or it may leave this field zero.
For F_SETLK and F_SETLKW the l_pid field will be set to the pid of the process performing the locking operation.
Note: if this method is not implemented, the kernel will still allow file locking to work locally. Hence it is only interesting for network filesystems and similar.
cmd
- see Fcntl
int utimens(java.lang.String path, Timespec[] timespec)
This supersedes the old utime() interface. New applications should use this.
See the utimensat(2) man page for details.
int bmap(java.lang.String path, long blocksize, long idx)
Note: This makes sense only for block device backed filesystems mounted with the 'blkdev' option
idx
- block index within fileblocksize
- unit of block indexint ioctl(java.lang.String path, int cmd, jnr.ffi.Pointer arg, FuseFileInfo fi, long flags, jnr.ffi.Pointer data)
flags will have FUSE_IOCTL_COMPAT set for 32bit ioctls in 64bit environment. The size and direction of data is determined by _IOC_*() decoding of cmd. For _IOC_NONE, data will be NULL, for _IOC_WRITE data is out area, for _IOC_READ in area and if both are set in/out area. In all non-NULL cases, the area is of _IOC_SIZE(cmd) bytes.
flags
- See IoctlFlags
int poll(java.lang.String path, FuseFileInfo fi, FusePollhandle ph, jnr.ffi.Pointer reventsp)
Note: If ph is non-NULL, the client should notify when IO readiness events occur by calling fuse_notify_poll() with the specified ph.
Regardless of the number of times poll with a non-NULL ph is received, single notification is enough to clear all. Notifying more times incurs overhead but doesn't harm correctness.
The callee is responsible for destroying ph with fuse_pollhandle_destroy() when no longer in use.
reventsp
- A pointer to a bitmask of the returned events satisfied.int write_buf(java.lang.String path, FuseBufvec buf, long off, FuseFileInfo fi)
Similar to the write() method, but data is supplied in a generic buffer. Use fuse_buf_copy() to transfer data to the destination.
IMPORTANT: Is not enabled by default, for enabling use ru.serce.jnrfuse.AbstractFuseFS#isBufOperationsImplemented()
int read_buf(java.lang.String path, jnr.ffi.Pointer bufp, long size, long off, FuseFileInfo fi)
Similar to the read() method, but data is stored and returned in a generic buffer.
No actual copying of data has to take place, the source file descriptor may simply be stored in the buffer for later data transfer.
The buffer must be allocated dynamically and stored at the location pointed to by bufp. If the buffer contains memory regions, they too must be allocated using malloc(). The allocated memory will be freed by the caller.
IMPORTANT: Is not enabled by default, for enabling use ru.serce.jnrfuse.AbstractFuseFS#isBufOperationsImplemented()
int flock(java.lang.String path, FuseFileInfo fi, int op)
The op argument will be either LOCK_SH, LOCK_EX or LOCK_UN
Nonblocking requests will be indicated by ORing LOCK_NB to the above operations
For more information see the flock(2) manual page.
Additionally fi->owner will be set to a value unique to this open file. This same value will be supplied to ->release() when the file is released.
Note: if this method is not implemented, the kernel will still allow file locking to work locally. Hence it is only interesting for network filesystems and similar.
op
- see Flock
int fallocate(java.lang.String path, int mode, long off, long length, FuseFileInfo fi)
This function ensures that required space is allocated for specified file. If this function returns success then any subsequent write request to specified range is guaranteed not to fail because of lack of space on the file system media.