Previously it was disciminated based on the entry mode.
For directories, the inode was the offset of the corresponding
goodbye tail mark while for all others it was the offset of the filename.
By simply using the start offset as calculated from the corresponding
goodbye table entry (which yields the archive offset of the filename),
the code is simplified and the more ambiguous read_directory_entry()
function can be used.
The disatvantage of this approach is the need to keep track of the
start and end offsets for each entry, as the end offset is needed in
order to access the goodbye table of directory entries.
The root node still has to be treated special, as it's inode is 1 as per fuse
definition and it has no filename as per the pxar file format definition.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Use Decoder::read_directory_entry() instead of Decoder::attributes() as this
already returns the needed DirectoryEntry.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
By not implementing readdir but only readdirplus, the FUSE_CAP_READDIRPLUS flag
is set while the FUSE_CAP_READDIRPLUS_AUTO flag is not set.
Thereby the kernel will issue only readdirplus calls.
Documentation at:
https://libfuse.github.io/doxygen/fuse-3_88_80_2include_2fuse__common_8h.html#a9b90333ad08d0e1c2ed0134d9305ee87
As the expensive part for accessing and reading the attributes is seeking and
decoding each directory entry, it is usefull to force readdirplus calls.
By this a struct `EntryParam` is returned for each entry, therebye avoiding a
subsequent lookup call.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
readdirplus returns the entries together with their `EntryParam`, so subsequent
lookups for each of the entries are avoided.
In order to reduce code duplication, the code for filling the reply buffer is
moved into a macro.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Remove the current caching of attrs and goodbye tables as it is broken anyway.
This will be replaced with a LRU cache.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
By moving the HashMap into `Context`, the use of lazy_static as well as the
additional Mutex can be avoided (`Context` is already guarded by a Mutex).
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
By reading and including xattrs and payload size in struct `DirectoryEntry`,
the tuple of return types is avoided and the code is simpler.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Limit the total number of entries and therefore the approximate memory
consumption instead of doing this on a per directory basis as it was previously.
This makes more sense as it limits not only the width but also the depth of the
directory tree.
Further, instead of hardcoding this value, allow to pass this information as
additional optional parameter 'entires-max'.
By this, creation of the archive with directories containing a large number of
entries is possible.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
If during creation of the archive, files/dirs with lacking read permissions are
encountered, the user is displayed a warning and the archive is created without
including the file/dir.
Previously this resulted in an error and the archive creation failed.
In order to implement this also for the .pxarexclude files, the Error type of
MatchPattern::from_file() and MatchPattern::from_line() was adopted accordingly.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
If nodes are excluded by feature flags, they must not appear in the goodbye table.
This is fixed by continuing with the next entry in the for loop.
Further the relative path buffer is now poped in order to correctly display the path.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The MatchPattern impl heavily used copies and therefore was inefficient regarding
memory management.
This patch intoduces MatchPatternSlice as struct to avoid copies and perform the
same pattern matching functionality.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The match_filename() in sequentail_decoder and encoder are moved to be static
functions of MatchPattern.
This allows to reuse the code also in the catalog find implementation.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Cache not only the goodbye table for the last directory but for each opened
directory.
The opendir fuse callback will fill the cache with the goodbye table and
releasedir will remove it from the cache.
This should reduce the number of chuncks fetched from the server in some cases.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The goodbye table is layed out as binary search tree based on the hash, so use
this to be more efficient when looking up a hash in the table for directories
with a large number of entries.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Decoders read must check if the file is a hardlink and read data from the
corresponding offset if so.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The returned filename should be the one of the file given at the offset, not of
the one the hardlink points to.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
And use an extra functzion set_callback() to configure that.
Also rewrite pxar/fuse.rs and implement a generic Session (will get
further cleanups with next patches).
Create the full target path and not fail if an intermediate directory does not
exist.
This is needed in order to restore multiple archives via the catalog, where the
target should further contain each archive name as subdir.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Allows to lookup an entry in a directory based on the provided `DirectoryEntry`.
This is needed to navigate the filesystem based on `DirectoryEntry`s and similar
to the find_goodbye_entry() function in src/pxar/fuse.rs
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
As it turns out the original implementation was correct and the start in
`DirectoryEntry` points to the `PxarEntry` and not as wrongly stated to the
filename.
This reverts the incorrect code and adds comments to the fields clarifying this.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
`Decoder::restore()` calls the `SequentialDecoder::restore()` which expects to
encounter a `PxarEntry` at first. But the start of `DirectoryEntry` points to the
filename (except for the root dir), so skip over it.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The expensive call to Decoder::read_directory_entry() can be omitted as
Decoder::attributes() returns all the information the fuse response needs.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
This exposes the option to pass a list of exclude MatchPattern via the
'--exclude' option.
The list is encoded as file '.pxarexclude-cli' in the archives root directory.
If such a file is present in the filesystem, it is skipped and not included in
the archive in order to avoid conflicting information.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
This provides the functionality needed to encode MatchPatterns passed on the cli
in the root directory.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
and add Session::from_decoder() in order to be able to create a fuse session
with a `Decoder` given as argument.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
This patch introduces `Context` to hold the decoder, ino_offset and caches for
the attributes and the goodbye table.
By caching, certain callbacks can be handled without the need to read additional
data via the decoder, which improves performance.
The searching of the goodbye table is refactored as well, avoiding recursive
function calls in case of a hash collision.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The hash of the filename in the goodbye table items allows to quickly compare to
a hashed filename.
Unfortunately, a matching hash is no garantee for matching filenames as hash
collisions are possible.
This patch fixes such possible collisions by further checking the filenames once
a matching hash has been found.
This introduces no significant extra cost (except for the filename comparison)
for cases with matching hashes, as the lookup call has to seek and read the file
attributes (including the filename) anyway.
In cases with hash collision, the next matching item is read and treaded
analogously (what means we need at least one extra seek).
As collisions should be not that frequent, this should be an acceptable penalty.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The functionality of stat is split into smaller sub-functions, which allows
to reuse them more flexible, as the code flow is similar but not always the same.
By this, the ugly and incorrect re-setting of the i-node in the lookup callback
function is avoided.
The correct i-node is now calculated beforehand and stat simply creates a
`libc::stat` struct from the provided parameters.
Also, this fixes incorrect i-node assignments in the readdir callback function.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The lookup call checks if the given filename is found in the directory referenced
by the i-node by calclulating the filenames hash and looking it up in the
directories goodbye table.
If found, the entries parameters are returned.
In order to be able to lookup the parent offset by a given file offset in the
readdir callback, this also stores the corresponding values in a HashMap.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
In order to read the contents of the goodbye table while keeping the
functionality of list_dir in place as is.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
... and thereby allow it to read a single directory entry based on the
start and end archive offsets.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
... as they are not needed with the latest iteration of the fuse callback
function impl (which never made it into the repos).
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
The previous implementation simply skipped over `size` bytes, which is not
correct as size includes also the header.
By relying on `SequentailDecoder`s read_filename function, this is correctly
taken care of plus some more integrity checks.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Also, removes an unused println statement in the decoder callback function and
fixes a typo.
Further, use ABI compatible Option<&T> for FFI to avoid use of raw pointers.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Implements functions attributes, open, read, read_link and get_dir
to be used by the fuse implementation which uses file offsets within the archive
as inodes to reference the archives items.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
By this it is possible to read and check just the first u64 of the corresponding
structs in order to identify the items.
This is needed for the fuse implementation in order to get entries based on the
archive offset, used as inode.
Directories are referenced by the offset to the goodbye tail while other items
are referenced by the offset of the filename followed by the entry.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
... and use std::mem::MaybeUninit or proxmox::tools::vec::uninitialized() instead.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
By taking ownership it is easier to move the decoder into another struct,
e.g. into a session context in fuse.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
This adds the basic code in order to create a fuse session and mount an archive.
It adds libfuse3-3 as runtime dependency and libfuse3-dev as build dependency.
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
