BookmarkFS User Manual

This manual is for BookmarkFS, version 0.1.0.

Copyright © 2024 CismonX <admin@cismon.net>

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in the section entitled “GNU Free Documentation License”.

Short Table of Contents

Table of Contents


1 Overview

BookmarkFS is a FUSE-based pseudo-filesystem which provides an interface to the bookmark data of web browsers.

Currently, the following browsers (and their derivatives) are supported:

BookmarkFS is free software, distributed under the terms of the GNU General Public License, either version 3, or any later version of the license. You should have received a copy of the GNU General Public License along with BookmarkFS. If not, see https://www.gnu.org/licenses/.

BookmarkFS is hosted on Savannah. Write to the mailing lists for bug reports, feature requests, and other discussions.


1.1 Porting BookmarkFS

Currently, BookmarkFS only runs on GNU/Linux and FreeBSD.

Although BookmarkFS sticks hard to POSIX and avoids using platform-specific features, porting it to other operating systems is not trivial.

The major pitfall is the FUSE dependency. Generally speaking, FUSE is Linux-only. FreeBSD partially implements the FUSE protocol in its kernel, to the extent that BookmarkFS is mostly usable. However, that’s not the case for other operating systems.

For example, OpenBSD implements its own FUSE protocol, which is incompatible with the Linux one. While OpenBSD does provide a libfuse-compatible library, however, it only covers the high-level API, and BookmarkFS uses the low-level API. For a similar reason, WinFsp won’t work if you’re trying to port BookmarkFS to Microsoft Windows.


1.2 Limitations on FreeBSD

Currently, the FreeBSD fusefs(5) implementation does not support FUSE_IOCTL. All custom ioctl(2) calls on a FUSE filesystem fail with ENOTTY without the requests being sent to the FUSE server.

Thus, BookmarkFS features that depend on ioctl() do not work on FreeBSD, which includes:

Meanwhile, FreeBSD does not support FUSE_READDIRPLUS and directory entry caching, which makes listing directory entries less efficient.


1.3 Sandboxing

A BookmarkFS backend can be instructed to enter a sandboxed state, where it irrevocably relinquishes most access to the system resources that it’s not supposed to touch. For example:

  • access local files other than the bookmark storage
  • establish socket connections
  • execute other files

This mechanism reduces the attack surface for exploit, if a vulnerability is discovered in BookmarkFS and/or its dependencies. However, it only deals with untrusted input, and cannot help if the operating system has already been compromised.

Examples of what “untrusted input” may include:

  • Bookmark files that are not created by the user using a trusted program (e.g., a file obtained from some random person on the internet).
  • Filesystem calls from untrusted programs. The program itself may be isolated, but it has a chance to escape the isolated environment if it can exploit BookmarkFS.

On Linux, sandboxing is achieved using seccomp(2) and landlock(7). On FreeBSD, capsicum(4) is used.


1.4 The Utility Library

The BookmarkFS Utility Library implements various common utility functions. It is used internally by most of BookmarkFS’s components, including the backends (see Backends) and the mount.bookmarkfs program.

Typically, the library is built into a shared object, and installed as:

${libdir}/bookmarkfs_util${shlib_suffix}
${libdir}

Presumably ${prefix}/lib. See The Uniform Naming Scheme in GNU Automake.

${shlib_suffix}

The common filename extension for shared library files on the current platform (e.g., .so on GNU/Linux and FreeBSD).

Public headers are installed under ${pkgincludedir} (presumably ${prefix}/include/bookmarkfs). To use the library functions, include the following headers as needed:

hash.h

Non-cryptographic hash function.

hashmap.h

A simple hashtable implementation.

prng.h

Non-cryptographic pseudo-random number generator.

sandbox.h

A simple sandbox implementation. See Sandboxing.

version.h

Get version and feature information of the library.

watcher.h

Single-file filesystem watcher.

Usage of the library functions is rather simple and straightforward, thus there’s currently no dedicated manual sections. Refer to the comments in the corresponding header files for details.


2 Programs

BookmarkFS ships with programs that provide users with a command-line interface to create, mount, fix, and manage BookmarkFS filesystems.

Those programs do not work on their own, and usually require a “backend” for low-level functionalities. See Backends.


2.1 mount.bookmarkfs

The mount.bookmarkfs program mounts a BookmarkFS filesystem.

mount.bookmarkfs [options] src target
src

The bookmark storage, presumably the pathname of a regular file that contains bookmark data.

The exact interpretation of this option is backend-defined.

target

Pathname of the directory where the filesystem shall be mounted to (i.e., the “mountpoint”).

Files under the filesystem are assigned ownership according to the effective user ID and group ID of the calling process. On FreeBSD, you may wish to set the ‘vfs.usermountsysctl(8) to 1 for unprivileged mounts.

To unmount a BookmarkFS filesystem, run fusermount3(1) or umount(8) on target. The daemon process will automatically dismount the filesystem upon SIGINT or SIGTERM receipt, however, it only works in non-sandbox mode.

Options:

-o backend=name

The backend used by the filesystem (see Backends). This option is mandatory.

Alternatively, name could be specified in ‘lib_path:sym_name’ format, where:

lib_path 

Presumably the pathname to the backend library. Its exact interpretation is equivalent to the first argument for dlopen(3p).

sym_name

See Backend Symbol Name Format.

-o @key[=value]

A backend-specific option. This option can be provided multiple times.

-o accmode=mode

File access mode. Defaults to 0700.

This option applies to both directories and regular files. Execution bits on regular files are masked off.

Should be used in combination with -o allow_other for other users to access the files.

-o ctime

Maintain file change time, while modification time follows change time. If this option is not provided, maintain file modification time instead.

Usually, a bookmark’s “modification time” attribute behaves differently from both mtime and ctime. In Chromium, for instance, when a bookmark is renamed, neither itself nor the parent directory changes timestamp accordingly.

BookmarkFS do not follow browser behavior here, and instead try to stay compatible with POSIX. Since a bookmark has only one “modification time” attribute instead of two, the user has to choose which one to maintain:

modification time

ctime only updates when mtime does.

change time

ctime updates normally; mtime always updates when ctime does, even if the file content is not modified.

This behavior may be inefficient, but makes applications that depend on ctime less fragile.

The kernel may automatically update and cache ctime, making it more “correct” than what we expect. However, this behavior should not be relied upon.

-o eol

Add a newline (ASCII LF character) to the end of each file.

Before writing the file content back to the backend, a trailing newline is automatically removed (if one exists).

-o file_max=bytes

Max file size limit. Defaults to 32768.

This limit also applies to extended attribute values.

-o no_sandbox

Do not enable sandboxing features (see Sandboxing).

-o no_landlock

Do not use Landlock for sandboxing. This option is ignored on non-Linux platforms.

Without Landlock, sandboxing offers less security. However, Landlock is a rather new feature (requires kernel version 5.13 or later), thus we provide an option to disable it separately.

-F

Stay in the foreground, do not daemonize.

-h

Print help text, and then exit.

-V

Print version and feature information, and then exit.

Unrecognized options specified with -o are passed to libfuse (and subsequently to the kernel, if applicable) as-is. Notable options:

-o rw

Mount the filesystem read/write.

Warning: Always backup the bookmark storage before mounting it read/write, or risk losing your data!

By default, the filesystem is mounted read-only. This behavior won’t change in the future, due to the hackish nature of most BookmarkFS backends.

When mounted read/write, other process must not write to the underlying bookmark storage, otherwise data corruption may occur.

-o debug

Set libfuse log level to FUSE_LOG_DEBUG. Log messages related to each FUSE request will be printed to standard error.

-o fsname=name

Name for the filesystem. Defaults to the backend name.

This name is equivalent to the fs_spec field in fstab(5), and appears as the ‘SOURCE’ column in findmnt(8) output.

-o atime,diratime,relatime

These options (and other atime-related ones) are ignored.

BookmarkFS only supports noatime mounts, since the “access time” attribute of a bookmark necessarily means “the last time it was accessed from the browser”. As a bookmark management tool independent from the browser, BookmarkFS should never update that time automatically.

Nonetheless, the user is still allowed to explicitly update atime (e.g., with futimens(3p)).


2.2 fsck.bookmarkfs

The fsck.bookmarkfs program checks and optionally repairs a BookmarkFS filesystem.

fsck.bookmarkfs [options] pathname

Filesystem check on BookmarkFS has a different purpose compared to on-disk filesystems. See Filesystem Check.

Depending on the options specified, filesystem check works either in online or offline mode:

Online Mode

In online mode, fsck is performed on a mounted BookmarkFS filesystem using ioctl() (see Online Filesystem Check).

The pathname argument refers to the directory to operate on.

Offline Mode

In offline mode, fsck is performed directly on the bookmark storage via the corresponding backend.

The pathname argument is the path to the bookmark storage, equivalent to the src argument given to mount.bookmarkfs.

Alternatively, pathname could be specified in ‘src:dir’ format, where dir refers to the directory to operate on, relative to the root directory of the subsystem.

Options:

-o backend=name

The backend used by the filesystem (see Backends).

Value of name could be specified in an alternative format. See Alternative Backend Name.

If this option is not provided, or name is empty, performs online fsck. Otherwise perform offline fsck.

-o @key[=value]

A backend-specific option. This option can be provided multiple times.

-o handler=name

The handler for resolving errors found during fsck (see Filesystem-Check Handlers).

Alternatively, name could be specified in ‘lib_path:sym_name’ format, where:

lib_path 

Presumably the pathname to the fsck handler library. Its exact interpretation is equivalent to the first argument for dlopen(3p).

sym_name

See Filesystem-Check Handler Symbol Name Format.

If this option is not provided, or name is empty, a built-in handler will be used. See Built-in Handler.

-o %key[=value]

A handler-specific option. This option can be provided multiple times.

-o repair

Attempt to repair errors found during fsck.

Warning: Always backup the bookmark storage before repairing, or risk losing your data!

-o rl_app=name

Readline application name in interactive mode. Defaults to ‘fsck.bookmarkfs’.

See Conditional Init Constructs in GNU Readline Library.

-o type=bookmark|tag|keyword

Subsystem type (see Filesystem Hierarchy). Defaults to ‘bookmark’.

This option is ignored when performing online fsck.

-i

Enable interactive mode.

-R

Perform fsck on subdirectories recursively.

This option is ignored when performing fsck on tags or keywords.

-o no_sandbox

Do not enable sandboxing features (see Sandboxing).

-o no_landlock

Do not use Landlock for sandboxing. This option is ignored on non-Linux platforms.

Also see Disabling Landlock.

-h

Print help text, and then exit.

-V

Print version and feature information, and then exit.


2.3 mkfs.bookmarkfs

The mkfs.bookmarkfs program creates a new BookmarkFS filesystem.

mkfs.bookmarkfs [options] pathname
pathname

The underlying bookmark storage for the new filesystem.

This option is equivalent to the src argument for mount.bookmarkfs.

Options:

-o backend=name

The backend used by the filesystem (see Backends). This option is mandatory.

Value of name could be specified in an alternative format. See Alternative Backend Name.

-o @key[=value]

A backend-specific option. This option can be provided multiple times.

-o force

If the file referred to by pathname already exists, overwrite it.

-h

Print help text, and then exit.

-V

Print version and feature information, and then exit.


2.4 bookmarkctl

The bookmarkctl program is a command-line wrapper for various I/O controls of a BookmarkFS filesystem.

bookmarkctl subcmd [args]

Sub-commands:

permd

Re-arranges the order of the directory entries obtained from readdir(). See Permute Directory Entries.

bookmarkctl permd pathname op name1 name2
pathname

Path to the directory.

name1
name2

Filename of entries under the directory.

op

Operation to perform on the directory:

swap

Exchange the positions of the directory entries represented by name1 and name2.

move-before

Move the directory entry represented by name1 to the position just before the one represented by name2.

move-after

Move the directory entry represented by name1 to the position just after the one represented by name2.

fsck

Check for errors within a BookmarkFS filesystem. See Filesystem Check.

bookmarkctl fsck pathname op
pathname

Path to the directory to perform checks on.

op

Operation to perform on the directory:

list

Display a list of errors found under the given directory. Will not recurse into subdirectories.

For the full fsck functionalities, see fsck.bookmarkfs.

help

Print help text, and then exit.

version

Print version information, and then exit.


3 The Filesystem

When a BookmarkFS filesystem is mounted using the mount.bookmarkfs program, a daemon process acts as a proxy between the kernel (which relays filesystem requests to FUSE requests) and the backend (which manipulates actual bookmark data, see Backends), thus providing POSIX (and platform-specific) filesystem API access to bookmarks.

BookmarkFS is designed in the hope that web browser bookmarks can be managed flexibly using a combination of existing software, without having to “reinvent the wheel”. However, like most other pseudo-filesystems, it cannot be considered fully POSIX-compliant. Users should be aware of the limitations when using BookmarkFS.


3.1 Filesystem Hierarchy

BookmarkFS has multiple subsystems. Each one appears as a directory under the mountpoint:

${mountpoint}/bookmarks
${mountpoint}/tags
${mountpoint}/keywords

If the backend does not support a subsystem, the corresponding directory does not exist.

Currently all subsystem definitions are hard-coded within the mount.bookmarkfs program, and cannot be extended by the backend.


3.1.1 Bookmarks

The “bookmarks” subsystem maintains the hierarchical structure, names, URLs and other information of a bookmark storage.

${mountpoint}/bookmarks/${bookmark_dir...}/${bookmark}

Each bookmark folder name appears as the filename for directory ${bookmark_dir}, and each ${bookmark} is a regular file that refers to a bookmark.

The name of a bookmark file is usually the “bookmark title”, which is the name that appears in the browser’s bookmark manager. The content of a bookmark file is usually the URL associated with the bookmark.

Not all bookmark names can be represented as a filename. For a bookmark or bookmark folder with an invalid name, the corresponding file is not visible to lookups and readdir() calls. To deal with such bookmarks, see Filesystem Check; or you can instruct the backend to identify bookmarks using GUIDs instead of titles (and then access the titles via extended attributes).

Some file attributes are used to represent bookmark metadata:

st_ino

ID of the bookmark (stored as lower bits).

st_size

Length of the bookmark URL in bytes. Always 0 for directories.

st_atim

Last access time of the bookmark.

st_mtim

Last modification time of the bookmark. See File Modification/Change Time.

Additional information of a bookmark or bookmark folder can be accessed via the extended attributes of the corresponding file, for backends that supports it (see Extended Attributes).


3.1.2 Tags

The “tags” subsystem maintains a many-to-many mapping between bookmarks and their alternative names.

${mountpoint}/tags/${tag_dir}/${bookmark}

Each tag name appears as the filename for directory ${tag_dir}, and ${bookmark} is a hard link to the bookmark file. A bookmark directory cannot be associated with a tag.

If multiple bookmark files with identical names are both associated with a tag, it is unspecified which one appears as an entry for the tag directory. However, consecutive lookups and readdir()s should produce consistent results for that file, provided that it is not renamed or deleted.

Tag files behave differently from traditional hard links. If the original bookmark file is renamed or deleted, it may also change accordingly. It may even link to another file that was previously shadowed. Applications should tread lightly if they wish to cache tag directory entries.

To associate a bookmark with a tag, use link(3p):

fd = open("tags/gnu/readline", O_CREAT | O_WRONLY, 0600);
// Oops, fd == -1, errno == EPERM

fd = link("bookmarks/other/readline", "tags/gnu/readline", 0);
// OK!

Make sure that the two files have identical names, otherwise link() fails with EPERM.


3.1.3 Keywords

The “keywords” subsystem maintains a one-to-one mapping between bookmarks and their alternative names, independent from tag names.

${mountpoint}/keywords/${keyword_name}

Each keyword name appears as the filename for regular file ${keyword_name}, which is a hard link to the bookmark file. A bookmark directory cannot be associated with a keyword.

To associate a bookmark with a keyword, use link() like we do with tags. If the original file is already associated with another keyword, link() fails with EEXIST.


3.2 Error Codes

When a filesystem operation fails, the kernel returns an error code for the system call. In addition to common error codes, there are a few more in BookmarkFS:

EPERM

Attempting to perform an unsupported operation. For example:

  • chmod(), chown(), and other operations that makes no sense for web browser bookmarks.
  • Moving a file across subsystems.
  • Creating a bookmark file with a name that is not valid UTF-8 (on Chromium backend).
EIO

An unexpected internal error occurred, likely due to a bug in BookmarkFS, or a corruption in the bookmark storage.

When this error occurs, a log message describing the situation may be printed to the standard error of the daemon process. Sometimes this error comes from the kernel-side FUSE implementation, and there’s no error message.

Once this error occurs, behavior of any further operations on the filesystem is undefined.

ESTALE

The file associated with the file descriptor no longer exists.

The error may occur when the underlying bookmark storage has been modified by another process (e.g., a web browser) after opening a file.

If the filesystem is mounted in exclusive mode, this error should not occur.

Other BookmarkFS-specific errors may occur. See the corresponding manual section for details.


3.3 Extended Attributes

BookmarkFS uses extended attributes to manage additional information associated with a bookmark.

Extended attributes is a platform-specific feature. On Linux, see xattr(7). On FreeBSD, see extattr(2).

All BookmarkFS extended attributes fall under the “user” namespace, which means they have a ‘user.’ name prefix on Linux, and should be accessed with EXTATTR_NAMESPACE_USER on FreeBSD. All attributes have a ‘bookmarkfs.’ name prefix.

For example, to get the GUID of a bookmark file (Firefox backend):

// Linux
len = fgetxattr(fd, "user.bookmarkfs.guid", buf, sizeof(buf));

// FreeBSD
len = extattr_get_fd(fd, EXTATTR_NAMESPACE_USER, "bookmarkfs.guid",
        buf, sizeof(buf));

BookmarkFS does not define any common attributes, neither can users create arbitrary ones. The backend decides which attributes are available during initialization, and all bookmark files share the same set of attributes.


3.4 Directory Entries

The ‘.’ and ‘..’ entries

POSIX consider ‘.’ and ‘..’ as “special” filenames, which must refer to the current and parent directory, if they exist.

These entries are optional, and BookmarkFS does not support them, for the sake of simplicity. Additionally, bookmarks with such names are hidden from the filesystem until fixed with fsck (see Filesystem Check).

Ordering of directory entries

POSIX does not specify the ordering of the directory entries retrieved from the directory stream using readdir(3p). It only guarantees that if an entry is not added or removed from the directory after the most recent call to opendir(3p) or rewinddir(3p), that entry is returned once and only once.

This allows filesystem implementations to organize directory entries in a more relaxed manner. There could be extra overhead to maintain a predictable ordering of directory entries, since they may not have a linear structure on modern on-disk filesystems (e.g., ext4 uses htree for large directories).

As for users of a filesystem, the order of directory entries generally does not matter. If they care, they can add a prefix to the filename, and let the application do the sorting.

However, the order of which a bookmark entry appears in the web browser sometimes does matter. In BookmarkFS, it is guaranteed to be equivalent to the directory traversal order. New entries are appended to the end; removed entries do not affect the order of other entries.

To change the ordering of directory entries, see Permute Directory Entries.


3.4.1 Permute Directory Entries

BookmarkFS provides an I/O control for rearranging directory entries:

#include <bookmarkfs/ioctl.h>

int ioctl (int dirfd, BOOKMARKFS_IOC_PERMD,
        struct bookmarkfs_permd_data const *argp);

The bookmarkfs_permd_data structure is defined as:

struct bookmarkfs_permd_data {
    enum bookmarkfs_permd_op op;

    char name1[NAME_MAX + 1];
    char name2[NAME_MAX + 1];
};

The op field denotes the operation to perform on the directory:

BOOKMARKFS_PERMD_OP_SWAP

Exchange the positions of the directory entries represented by name1 and name2.

BOOKMARKFS_PERMD_OP_MOVE_BEFORE

Move the directory entry represented by name1 to the position just before the one represented by name2.

BOOKMARKFS_PERMD_OP_MOVE_AFTER

Move the directory entry represented by name1 to the position just after the one represented by name2.

On success, ioctl() returns 0. Otherwise, it returns -1 and sets errno:

EACCES

Write or search permission is denied for the directory.

EINVAL

op is not one of the values defined in enum bookmarkfs_permd_op.

EINVAL

name1 or name2 is not a valid filename (e.g., empty string; contains ‘/’ character).

ENOENT

The directory does not contain entries named name1 or name2.

ENOTTY

The kernel does not support FUSE_IOCTL. See Limitations on FreeBSD.

EPERM

The backend does not support rearranging entries for this directory.

To ensure that the order change is visible to further readdir() calls, fsync() or close() the directory.


3.5 Filesystem Check

On-disk filesystems may suffer from data corruption due to power loss or hardware failures, thus they usually provide a “filesystem check” mechanism to detect and fix those problems.

As a pseudo-filesystem, BookmarkFS does not check for data integrity, and “filesystem check” is given a new purpose: To check if a bookmark name is valid as a filename, and “repair” (rename) it if it isn’t.

A POSIX-compliant filesystem has various restrictions regarding filenames:

  • must not contain ‘/’ characters
  • must not be empty or longer than NAME_MAX
  • must not duplicate with another file in the same directory
  • .’ and ‘..’ must refer to the current and parent directory

It is commonplace for bookmark names to not meet such criteria, thus a filesystem check is often necessary when switching to BookmarkFS from another bookmark management software.


3.5.1 Online Filesystem Check

To perform filesystem check on a mounted BookmarkFS filesystem, use the following I/O controls:

#include <bookmarkfs/ioctl.h>

int ioctl (int dirfd, BOOKMARKFS_IOC_FSCK_NEXT,
        struct bookmarkfs_fsck_data *argp);

int ioctl (int dirfd, BOOKMARKFS_IOC_FSCK_APPLY,
        struct bookmarkfs_fsck_data *argp);

int ioctl (int dirfd, BOOKMARKFS_IOC_FSCK_REWIND);

The bookmarkfs_fsck_data structure is defined as:

struct bookmarkfs_fsck_data {
    uint64_t id;
    uint64_t extra;
    char     name[NAME_MAX + 1];
};

Filesystem-check commands:

BOOKMARKFS_IOC_FSCK_NEXT

Find the next bookmark with invalid name under the directory.

On success, ioctl() updates argp, and returns one of the values defined in enum bookmarkfs_fsck_result:

BOOKMARKFS_FSCK_RESULT_END

There are no more bookmarks with invalid name under the directory.

Fields in argp have unspecified values.

BOOKMARKFS_FSCK_RESULT_NAME_DUPLICATE

The bookmark name duplicates with another bookmark which appears earlier in the directory stream.

Value of extra is the ID of the other bookmark.

BOOKMARKFS_FSCK_RESULT_NAME_BADCHAR

The bookmark contains a bad character (i.e., the ASCII ‘/’ character).

Value of extra is the byte offset where the bad character first appears in name.

BOOKMARKFS_FSCK_RESULT_NAME_BADLEN

The bookmark name is either longer than NAME_MAX, or an empty string.

Value of extra is the length of the bookmark name, and name is truncated if longer than NAME_MAX.

BOOKMARKFS_FSCK_RESULT_NAME_DOTDOT

The bookmark name is either ‘.’ or ‘..’.

Value of extra is unspecified.

On failure, ioctl() returns -1, and sets errno.

BOOKMARKFS_IOC_FSCK_APPLY

“Repair” a bookmark by renaming it.

The id field must be set to a value previously obtained from BOOKMARKFS_IOC_FSCK_NEXT with the same dirfd, otherwise the behavior is undefined.

The name field should be set to the new name for the bookmark. The extra field is unused.

On success, ioctl() updates argp, and returns one of the values defined in enum bookmarkfs_fsck_result, like with BOOKMARKFS_IOC_FSCK_NEXT. Additionally, it may also return:

BOOKMARKFS_FSCK_RESULT_NAME_INVALID

The new name is not a valid bookmark name.

Value of extra is a backend-specific reason code explaining why the bookmark name is invalid. It may equal to one of the following predefined values:

BOOKMARKFS_NAME_INVALID_REASON_NOTUTF8

The name is not valid UTF-8.

On failure, ioctl() returns -1, and sets errno:

EACCES

Write or search permission is denied for the directory.

To ensure that the rename is visible to further lookups and readdir() calls, fsync() or close() the directory.

BOOKMARKFS_IOC_FSCK_REWIND

Reset the fsck state for the directory.

Further BOOKMARKFS_IOC_FSCK_NEXT requests will start from the beginning of the directory stream.

On success, ioctl() returns 0. Otherwise, it returns -1, and sets errno.

Common error codes for all filesystem-check ioctls:

ENOTTY

The kernel does not support FUSE_IOCTL. See Limitations on FreeBSD.

EPERM

The backend does not support fsck for this directory.


4 Backends

In BookmarkFS, each backend provides a way to manipulate a certain kind of application bookmarks.

Typically, backends are built into shared libraries, and are installed as:

${pkglibdir}/backend_${name}${shlib_suffix}
${pkglibdir}

Presumably ${prefix}/lib/bookmarkfs. See The Uniform Naming Scheme in GNU Automake.

${name}

The backend name, equivalent to the value given to the -o backend=name option of frontend programs.

${shlib_suffix}

The common filename extension for shared library files on the current platform (e.g., .so on GNU/Linux and FreeBSD).


4.1 Firefox Backend

The Firefox backend provides access to the bookmark data of the web browser Mozilla Firefox and its derivatives, notably Tor Browser and Librewolf.

Firefox bookmarks are stored in a SQLite database under the profile directory:

~/.mozilla/firefox/${profile_name}/places.sqlite

When mounting the filesystem, this pathname shall be passed as the src argument (see mount.bookmarkfs). Actual path for the profile directories may differ across distributions.

Backend name: ‘firefox’.

Backend-specific options (mount.bookmarkfs only):

filename=title|guid

Whether to use the bookmark title or GUID as the bookmark file name. Defaults to ‘title’.

A bookmark GUID is a base64url-encoded 128-bit string uniquely associated with a bookmark or bookmark folder.

When creating a new file:

title

The GUID is randomly generated by the backend.

guid

The filename must be a valid GUID, and must not duplicate with other files on the same filesystem, otherwise open() or mkdir() fails with EPERM.

Also set the GUID string as the bookmark title.

lock=exclusive|normal

The database connection locking mode for the bookmark storage. Defaults to ‘normal’ when the filesystem is mounted read-only, ‘exclusive’ otherwise.

This option corresponds to the locking_mode pragma on SQLite. With lock=exclusive, other process cannot access the bookmark storage until the filesystem is dismounted.

The Firefox browser holds an exclusive lock on the database by default. If you wish to mount the bookmarks while keeping the browser session open, set the storage.sqlite.exclusiveLock.enabled browser preference to false.

assume_title_distinct

If this options is provided, the backend assumes that bookmark names are distinct under the same bookmark folder. This option is ignored with filename=guid.

This option may improve readdir() performance, however, making a false assumption results in a directory entry with duplicate names. It is recommended to perform a full filesystem check (see Filesystem Check) on the bookmark storage before mounting with this option.

If launched from fsck.bookmarkfs, all backend-specific options are ignored, and always enforces the lock=exclusive option.

Backend-specific options (mkfs.bookmarkfs only):

date_added=timestamp

File creation time for the bookmark root directories. Defaults to the current time.

Format of timestamp is equivalent to the ‘date_added’ extended attribute (see below).

Extended attributes:

title

The bookmark title. Only available with filename=guid.

This value is allowed be set to any string that does not contain a NUL character, and does not have to be valid as a filename.

guid

The bookmark GUID. Only available with filename=title.

description

An arbitrary text associated with the bookmark.

Usually, when bookmarking a page in the browser, this value is set to the content attribute of the <meta> element whose name is ‘description’.

date_added

The bookmark creation time.

Value is a decimal integer representing number of microseconds since the Unix epoch.


4.2 Chromium Backend

The Chromium backend provides access to the bookmark data of the web browser Chromium and its derivatives, notably ungoogled-chromium.

Chromium bookmarks are stored in a text file (in JSON format) under the profile directory:

~/.config/chromium/${profile_name}/Bookmarks

When mounting the filesystem, this pathname shall be passed as the src argument (see mount.bookmarkfs). Actual path for the profile directories may differ across distributions.

Backend name: ‘chromium’.

Backend-specific options (mount.bookmarkfs only):

filename=title|guid

Whether to use the bookmark title or GUID as the bookmark file name. Defaults to ‘title’.

A bookmark GUID is a hex-encoded 128-bit string uniquely associated with a bookmark or bookmark folder. The GUID string has a “8-4-4-4-12” format, while all alphabetic characters are in lowercase. For example:

0bc5d13f-2cba-5d74-951f-3f233fe6c908

When creating a new file:

title

The GUID is randomly generated by the backend. It is guaranteed to be a valid version 4 UUID as specified by RFC 4122.

guid

The filename must be a valid GUID, and must not duplicate with other files on the same filesystem, otherwise open() or mkdir() fails with EPERM.

Also set the GUID string as the bookmark title.

watcher=native|fallback|none

The file watcher to use for the bookmark storage. Defaults to ‘native’ when the filesystem is mounted read-only, ‘none’ otherwise.

native

Watch for file changes using platform-specific features:

Linux

fanotify(7) is used. Requires kernel version 5.13 or later for unprivileged users.

inotify(7) does not have this limitation, however, it is incompatible with our sandboxing design.

FreeBSD

kevent(2) with EVFILT_VNODE is used.

fallback

Watch for file changes by checking st_ino and st_mtim attributes with fstatat() periodically.

Less efficient than “native” implementations, but should work on any POSIX-compatible system.

none

Do not watch for file changes.

With watcher=none, changes on the bookmark storage are not visible to the filesystem.

If launched from fsck.bookmarkfs, all backend-specific options are ignored, and always enforces the watcher=none option.

Backend-specific options (mkfs.bookmarkfs only):

date_added=timestamp

File creation time for the bookmark root directories. Defaults to the current time.

Format of timestamp is equivalent to the ‘date_added’ extended attribute (see below).

Extended attributes:

title

The bookmark title. Only available with filename=guid.

This value is allowed be set to any UTF-8 string that does not contain a NUL character, and does not have to be valid as a filename.

guid

The bookmark GUID. Only available with filename=title.

date_added

The bookmark creation time.

Value is a decimal integer representing number of microseconds since the Windows FILETIME epoch (134774 days ahead of the Unix epoch).

Notable limitations:

  • Does not scale well with large bookmark storage, since everything is kept in memory, and the entire JSON file has to be parsed whenever a reload is required.
  • No support for tags (see Tags) and keywords (see Keywords), since Chromium does not have such concepts.

4.3 Backend API

The Backend API specifies how a BookmarkFS backend communicates with a frontend program (e.g., mount.bookmarkfs).

Warning: Currently BookmarkFS is experimental. The Backend API may change drastically without prior notice.

To implement the Backend API, a backend library should expose a symbol with the following name:

bookmarkfs_backend_${name}

Where ${name} is equivalent to the value given to the -o backend=name option of frontend programs.

The symbol should name a data object identifier of the following type:

#include <bookmarkfs/backend.h>

struct bookmarkfs_backend {
    bookmarkfs_backend_create_func  *backend_create;
    bookmarkfs_backend_destroy_func *backend_destroy;
    bookmarkfs_backend_info_func    *backend_info;
    bookmarkfs_backend_init_func    *backend_init;
    bookmarkfs_backend_mkfs_func    *backend_mkfs;
    bookmarkfs_backend_sandbox_func *backend_sandbox;

    bookmarkfs_bookmark_get_func    *bookmark_get;
    bookmarkfs_bookmark_list_func   *bookmark_list;
    bookmarkfs_bookmark_lookup_func *bookmark_lookup;

    bookmarkfs_bookmark_create_func  *bookmark_create;
    bookmarkfs_bookmark_delete_func  *bookmark_delete;
    bookmarkfs_bookmark_fsck_func    *bookmark_fsck;
    bookmarkfs_bookmark_permute_func *bookmark_permute;
    bookmarkfs_bookmark_rename_func  *bookmark_rename;
    bookmarkfs_bookmark_set_func     *bookmark_set;
    bookmarkfs_bookmark_sync_func    *bookmark_sync;

    bookmarkfs_object_free_func *object_free;
};

Each field is a function pointer provided by the backend. Some can be optional and set to NULL, if the backend does not support the corresponding features.


4.3.1 Initialize Backend

The backend_init function is called by the frontend program before calling any other functions (except for backend_info). If not NULL, it is guaranteed to be called exactly once throughout the lifetime of the process.

typedef int (bookmarkfs_backend_init_func) (
    uint32_t flags
);

Function arguments:

flags

A bit array of the following flags:

BOOKMARKFS_BACKEND_LIB_READY

Indicates that the utility library is already initialized. See The Utility Library.

Some frontend programs use the utility library. If they do, they always initialize it before initializing the backend. The utility library must not be initialized multiple times, otherwise the behavior is undefined.

BOOKMARKFS_FRONTEND_FSCK
BOOKMARKFS_FRONTEND_MOUNT
BOOKMARKFS_FRONTEND_MKFS

Denotes the frontend program that launches the backend. These flags are mutually exclusive.

The function should return 0 on success, and -1 on error.


4.3.2 Get Backend Information

If not NULL, the backend_info function is called when the user instructs the frontend program to print information about the backend.

When this function is called, the backend should write a human-readable message of the corresponding information to standard output.

typedef void (bookmarkfs_backend_info_func) (
    uint32_t flags
);

Function arguments:

flags

A bit array of the following flags:

BOOKMARKFS_BACKEND_INFO_HELP

Indicates that the backend should print a help message. Mutually exclusive with BOOKMARKFS_BACKEND_INFO_VERSION.

The message should contain a brief description of this backend, as well as a list of backend-specific options.

BOOKMARKFS_BACKEND_INFO_VERSION

Indicates that the backend should print a version message. Mutually exclusive with BOOKMARKFS_BACKEND_INFO_HELP.

In addition to the version number, the version message may also contain dependency versions, compile-time options, and other information that can be used to determine a specific version of the backend.

BOOKMARKFS_FRONTEND_FSCK
BOOKMARKFS_FRONTEND_MOUNT
BOOKMARKFS_FRONTEND_MKFS

Denotes the frontend program that launches the backend. These flags are mutually exclusive.


4.3.3 Create Backend Context

A backend context maintains internal states for manipulating a specific bookmark storage.

Each BookmarkFS filesystem mounted by mount.bookmarkfs is backed by a backend context. The context is created before mount, and destroyed after dismount. Operations on the filesystem, if applicable, are fulfilled by invoking the corresponding backend functions on the context.

To create a backend context, the backend_create function is called. It must not be NULL.

typedef int (bookmarkfs_backend_create_func) (
    struct bookmarkfs_backend_conf const  *conf,
    struct bookmarkfs_backend_create_resp *resp
);

Function arguments:

conf

Backend configuration items.

The bookmarkfs_backend_conf structure is defined as:

struct bookmarkfs_backend_conf {
    uint32_t  version;
    uint32_t  flags;
    char     *store_path;

    struct bookmarkfs_conf_opt *opts;
};
version

Version number of the frontend program, equivalent to the BOOKMARKFS_VERNUM macro in bookmarkfs/version.h.

flags

A bit array of the following flags:

BOOKMARKFS_BACKEND_READONLY

Indicates that the filesystem is mounted read-only, and the frontend program will not call functions that may write to the bookmark storage.

BOOKMARKFS_BACKEND_CTIME

Indicates that the -o ctime option is given to mount.bookmarkfs.

BOOKMARKFS_BACKEND_NO_SANDBOX

Indicates that the -o no_sandbox option is given to mount.bookmarkfs.

BOOKMARKFS_BACKEND_NO_LANDLOCK

Indicates that the -o no_landlock option is given to mount.bookmarkfs.

BOOKMARKFS_BACKEND_FSCK_ONLY

Indicates that the backend is launched from fsck.bookmarkfs.

The backend must claim exclusive access to the bookmark storage, and only the bookmark_lookup, bookmark_list, bookmark_fsck and bookmark_sync functions will be called on the bookmarks.

store_path

Path to the bookmark storage, equivalent to the src argument passed to mount.bookmarkfs.

The string is guaranteed to be NUL-terminated, and valid throughout the lifetime of the current backend. The backend is free to modify the string, however, it must not write past the string boundary.

opts

Linked list of backend-specific options, NULL if there are none.

The bookmarkfs_conf_opt structure is defined as:

struct bookmarkfs_conf_opt {
    char *key;
    char *val;

    struct bookmarkfs_conf_opt *next;
};
key
val

Key and value of the current option. If the option does not have a value (no ‘=’ character present), val is NULL.

When not NULL, the strings are guaranteed to be NUL-terminated, and valid throughout the lifetime of the current backend context. The backend is free to modify the strings, however, it must not write past the string boundary.

next

Pointer to the next option, NULL if there are no more options.

The backend must not re-assign the fields in opt.

resp

Information of the new backend context.

The initial content of this argument is unspecified. When the backend context is successfully created, the backend should populate this argument with appropriate values.

The bookmarkfs_backend_create_resp structure is defined as:

struct bookmarkfs_backend_create_resp {
    char const *name;
    void       *backend_ctx;
    uint64_t    bookmarks_root_id;
    uint64_t    tags_root_id;
    char const *bookmark_attrs;
    uint32_t    flags;
};
name

Name of the backend context, used as default value for the -o fsname=name option of mount.bookmarkfs.

Must be a NUL-terminated string valid at least until the next function call on this backend context.

backend_ctx

An opaque pointer referring to the backend context.

The pointer will be passed to further function calls on this context.

bookmarks_root_id

ID of the bookmark root directory (i.e., ${mountpoint}/bookmarks). Must not be greater than BOOKMARKFS_MAX_ID.

If sandboxing is requested, and the ID cannot be determined in a safe way before entering sandbox, the backend may leave this field as-is or set it to UINT64_MAX.

tags_root_id

ID of the tags root directory (i.e., ${mountpoint}/tags). Must not be greater than BOOKMARKFS_MAX_ID.

This field should be left as-is or set to UINT64_MAX, if one of the following conditions is met:

  • Sandboxing is requested, and the ID cannot be determined in a safe way before entering sandbox.
  • The backend does not support tags for this context (see Tags).
bookmark_attrs

Names of extended attributes (see Extended Attributes) supported for this backend context.

Must be a concatenation of NUL-terminated strings valid throughout the lifetime of this backend context. An empty string denotes the end of list.

flags

A bit array of the following flags:

BOOKMARKFS_BACKEND_EXCLUSIVE

Indicates that the backend claims exclusive access to the bookmark storage.

In exclusive mode, modifications to the bookmark storage only come from the frontend, and the frontend program may perform optimizations based on this assumption.

BOOKMARKFS_BACKEND_HAS_KEYWORD

Indicates that the backend supports keywords for this context (see Keywords).

The function should return 0 on success, and -1 on error.


4.3.4 Destroy Backend Context

When a backend context is no longer used, the backend_destroy function is called. It must not be NULL.

The backend should release all system resources associated with this context. If changes have been made to the bookmark storage, the backend should try to persist them.

typedef void (bookmarkfs_backend_destroy_func) (
    void *backend_ctx
);

Function arguments:

backend_ctx

The pointer referring to the backend context.


4.3.5 Lookup Bookmark

The bookmark_lookup function is called to obtain the attributes of a bookmark, or a bookmark-related object. It must not be NULL.

typedef int (bookmarkfs_bookmark_lookup_func) (
    void                            *backend_ctx,
    uint64_t                         id,
    char const                      *name,
    uint32_t                         flags,
    struct bookmarkfs_bookmark_stat *stat_buf
);

Function arguments:

backend_ctx

The pointer referring to the backend context.

id

The subsystem object ID to be used for lookup. It could either be a bookmark ID or a tag ID, depending on the value of flags.

name

Name of the subsystem object to lookup. It could be a bookmark name, tag name or keyword name, depending on the value of flags.

When not NULL, name is guaranteed to be a valid filename, which should be used to lookup an object under the directory referred to by id. Otherwise, the function should operate on the object referred to by id itself.

flags

A bit array of the following flags:

BOOKMARKFS_BOOKMARK_TYPE_MASK

The subsystem to operate on.

Equals to one of the following values (after shifting):

BOOKMARKFS_BOOKMARK_TYPE_BOOKMARK

Indicates that the function should operate on the “bookmarks” subsystem. See Bookmarks.

id and name refers to bookmark ID and bookmark name.

BOOKMARKFS_BOOKMARK_TYPE_TAG

Indicates that the function should operate on the “tags” subsystem. See Tags.

id and name refers to tag ID and tag name.

BOOKMARKFS_BOOKMARK_TYPE_KEYWORD

Indicates that the function should operate on the “keywords” subsystem. See Keywords.

name refers to keyword name (must not be NULL). The value of id is unspecified.

stat_buf

Attributes of the object to lookup.

The initial content of this argument is unspecified. When the object is successfully found, the backend should populate this argument with appropriate values.

The bookmarkfs_bookmark_stat structure is defined as:

struct bookmarkfs_bookmark_stat {
    uint64_t id;
    ssize_t  value_len;

    struct timespec atime;
    struct timespec mtime;
};
id

ID of the object.

value_len

Length of the object value in bytes. -1 if the object refers to a directory.

atime
mtime

Last access and modification time of the object. Must not be earlier than the Unix epoch.

On success, the function should return 0. Otherwise, it should return a negated errno indicating the error encountered.

Generally, the error code should follow filesystem conventions, but the backend may return other error codes which it sees fit. Also see Error Codes.


5 Filesystem-Check Handlers

The filesystem-check handler is responsible for instructing the fsck.bookmarkfs program on what to do with each invalid bookmark name.

Like backends, filesystem-check handlers are typically built into shared libraries, and are installed as:

${pkglibdir}/fsck_handler_${name}${shlib_suffix}
${pkglibdir}

Presumably ${prefix}/lib/bookmarkfs. See The Uniform Naming Scheme in GNU Automake.

${name}

The handler name, equivalent to the value given to the -o handler=name option of fsck.bookmarkfs.

${shlib_suffix}

The common filename extension for shared library files on the current platform (e.g., .so on GNU/Linux and FreeBSD).


5.1 Built-in Handler

The built-in filesystem-check handler is linked into the fsck.bookmarkfs program. It has minimal functionalities, but is capable enough to handle most common fsck scenarios.

Handler-specific options:

prompt=str

Readline prompt string. Defaults to ‘% ’.

This option is ignored in non-interactive mode.

translit=char

Transliterate bad (‘/’) characters into char (must be a single ASCII character). Defaults to ‘_’.

For each bookmark entry, the built-in handler prints a message to standard output explaining why the bookmark name is invalid.

When the -o repair option is given, the handler renames the bookmark according to the following rules:

BOOKMARKFS_FSCK_RESULT_NAME_DUPLICATE

Append a ‘_${counter}’ suffix to the name, where ${counter} is a self-incrementing 32-bit integer in decimal format.

If appending the suffix would exceed max name length, truncate the name first.

BOOKMARKFS_FSCK_RESULT_NAME_BADCHAR

Transliterate the bad character into another character.

BOOKMARKFS_FSCK_RESULT_NAME_BADLEN

If the name is too long, truncate it to NAME_MAX bytes.

If the name is empty, see below:

BOOKMARKFS_FSCK_RESULT_NAME_DOTDOT
BOOKMARKFS_FSCK_RESULT_NAME_INVALID

Rename to ‘fsck-${id}’, where ${id} is the bookmark ID in decimal format.

In interactive mode, the handler prompts the user before applying the rename. The user may issue a command to indicate what to do with each entry:

p

Print the ID and new name of current entry.

a[-]

Apply the proposed rename for the current entry.

e[-] new_name

Change the proposed rename to new_name and then apply.

c

Continue to the next entry.

s[-]

Skip current directory.

S[-]

Skip current directory and all subdirectories.

r[-]

Rewind current directory.

R[-]

Rewind all.

w[-]

Save applied changes.

q

Save applied changes and quit.

The optional ‘-’ suffix inhibits the default behavior of continuing to the next entry, after the command completes successfully.


5.2 Tcl-Based Handler

The Tcl-based filesystem-check handler allows fsck entries to be handled via Tcl scripting.

Handler name: ‘tcl’.

Handler-specific options:

script=path

Path to the Tcl script file. This option is mandatory.

The script is evaluated once after interpreter initialization. The evaluation result will be used as the command name for later executions.

The following variables are set before script evaluation:

$::bookmarkfs::fsck::isInteractive

Equals to 1 if the -i option is given to fsck.bookmarkfs, 0 otherwise.

$::bookmarkfs::fsck::isReadonly

Equals to 0 if the -o repair option is given to fsck.bookmarkfs, 1 otherwise.

unsafe

Enable unsafe Tcl interpreter features.

Should be used in combination with the -o no_sandbox option if you wish to access extra system resources (e.g., open local files).

Without this option, the Tcl interpreter has limited functionalities as if created with interp create -safe. See Safe Interpreters.

Each time fsck.bookmarkfs finds an entry, or completes a handler-initiated operation, the command returned from the script is executed with a single list argument. The first element of the list is an integer indicating the reason for this handler call:

$::bookmarkfs::fsck::result::nameDuplicate
$::bookmarkfs::fsck::result::nameBadChar
$::bookmarkfs::fsck::result::nameBadLen
$::bookmarkfs::fsck::result::nameDotDot
$::bookmarkfs::fsck::result::nameInvalid

See Filesystem-Check Result Code for the meaning of each value.

The second element is a list of information about the current entry:

  1. The bookmark ID
  2. Extra information regarding the invalid name, equivalent to the extra field in structure bookmarkfs_fsck_data.
  3. Name of the bookmark
  4. ID of the parent directory
-1

If the command is being executed for the first time, or the previous execution returns $::bookmarkfs::fsck::handler::next, this value never appears.

If the previous execution of the command returns $::bookmarkfs::fsck::handler::userInput, the second element of the list is a string of user input.

Otherwise, this value indicates that the previous operation is successfully performed on the entry.

The command should return a list indicating the operation to perform on the entry. First element of the list should be one of the following values:

$::bookmarkfs::fsck::handler::next

Continue to the next entry.

$::bookmarkfs::fsck::handler::apply

Apply change for the current entry. Not available in read-only mode.

Second element of the list should be a string for the new name of the bookmark.

$::bookmarkfs::fsck::handler::userInput

Request for user input. Only available in interactive mode.

Second element of the list should be a string for Readline prompt, equivalent to the prompt=str option of the built-in handler.

$::bookmarkfs::fsck::handler::save

Save applied changes.

$::bookmarkfs::fsck::handler::stop

Save applied changes and quit.

Once this value is returned, the command will never be executed again.

$::bookmarkfs::fsck::handler::rewind

Rewind current directory.

$::bookmarkfs::fsck::handler::skip

Skip current directory.

$::bookmarkfs::fsck::handler::skipChildren

Skip current directory and all subdirectories.

$::bookmarkfs::fsck::handler::reset

Rewind all.

Here is an example Tcl script that simply prints each fsck entry (requires the unsafe option):

chan configure stdout -encoding utf-8
coroutine whatever apply {{} {
    set args [yield [info coroutine]]
    while 1 {
        lassign $args why data
        if {$why > -1} {
            lassign $data id extra name parent
            puts "id: $id, extra: $extra, name: $name, parent: $parent"
        }
        set args [yield [list $::bookmarkfs::fsck::handler::next]]
    }
}}

5.3 Handler API

The Filesystem-Check Handler API specifies how a fsck handler communicates with fsck.bookmarkfs.

Warning: Currently BookmarkFS is experimental. The Handler API may change drastically without prior notice.

To implement the Filesystem-Check Handler API, a fsck handler library should expose a symbol with the following name:

bookmarkfs_fsck_handler_${name}

Where ${name} is equivalent to the value given to the -o handler=name option of fsck.bookmarkfs.

The symbol should name a data object identifier of the following type:

#include <bookmarkfs/fsck_handler.h>

struct bookmarkfs_fsck_handler {
    bookmarkfs_fsck_handler_info_func *info;

    bookmarkfs_fsck_handler_create_func  *create;
    bookmarkfs_fsck_handler_destroy_func *destroy;

    bookmarkfs_fsck_handler_run_func *run;
};

Each field is a function pointer provided by the fsck handler. Some can be optional and set to NULL, if the backend does not support the corresponding features.


5.3.1 Get Handler Information

If not NULL, the info function is called when the user instructs fsck.bookmarkfs to print information about the handler.

When this function is called, the handler should write a human-readable message of the corresponding information to standard output.

typedef void (bookmarkfs_fsck_handler_info_func) (
    uint32_t flags
);

Function arguments:

flags

A bit array of the following flags:

BOOKMARKFS_FSCK_HANDLER_INFO_HELP
BOOKMARKFS_FSCK_HANDLER_INFO_VERSION

Indicates that the backend should print a help/version message.

These flags are analogous to the corresponding flags in the backend_info function of the backend API. See Get Backend Information.


5.3.2 Create Handler Context

A filesystem-check handler context maintains internal states for handling fsck entries.

To create a handler context, the create function is called. It must not be NULL.

typedef int (bookmarkfs_fsck_handler_create_func) (
    struct bookmarkfs_conf_opt const  *opts,
    uint32_t                           flags,
    void                             **handler_ctx_ptr
);

Function arguments:

opts

Linked list of handler-specific options, NULL if there are none.

Handler-specific option should be processed in the same way as backend-specific options. See Backend-Specific Options.

flags

A bit array of the following flags:

BOOKMARKFS_FSCK_HANDLER_INTERACTIVE

Indicates that the -i option is given to fsck.bookmarkfs.

BOOKMARKFS_FSCK_HANDLER_READONLY

Indicates that the -o repair option is not given to fsck.bookmarkfs.

handler_ctx_ptr

Pointer to an opaque pointer referring to the handler context.

The initial value of that pointer is unspecified. It should be set by the handler when the handler context is successfully created.

The function should return 0 on success, and -1 on error.


5.3.3 Destroy Handler Context

When a handler context is no longer used, the destroy function is called. It must not be NULL.

The handler should release all system resources associated with this context.

typedef void (bookmarkfs_fsck_handler_destroy_func) (
    void *handler_ctx
);

Function arguments:

handler_ctx

The pointer referring to the handler context.


5.3.4 Run Handler

Each time fsck.bookmarkfs finds and entry, or completes a handler-initiated operation, the run function is called. It must not be NULL.

typedef int (bookmarkfs_fsck_handler_run_func) (
    void                               *handler_ctx,
    int                                 why,
    union bookmarkfs_fsck_handler_data *data
);

Function arguments:

handler_ctx

The pointer referring to the handler context.

why

Reason code for this handler call:

Non-negative value

The fsck.bookmarkfs program has found an entry.

See Filesystem-Check Result Code for possible values. However, BOOKMARKFS_FSCK_RESULT_END is handled by fsck.bookmarkfs and never appears.

-1

Indicates that an operation instructed by the previous return value of this function is successfully performed.

If run is being called for the first time, or the previous invocation returns BOOKMARKFS_FSCK_NEXT, this value never appears.

data

Information for this handler call.

The bookmarkfs_fsck_handler_data union is defined as:

union bookmarkfs_fsck_handler_data {
    struct bookmarkfs_fsck_handler_entry entry;
    char *str;
};
entry

Information for the bookmark entry. This field is set when why is non-negative.

The bookmarkfs_fsck_handler_entry structure is defined as:

struct bookmarkfs_fsck_handler_entry {
    uint64_t parent_id;

    struct bookmarkfs_fsck_data data;
};
parent_id

ID of the bookmark folder that contains this bookmark entry.

data

See Online Filesystem Check for the definition of the bookmarkfs_fsck_data structure.

str

A NUL-terminated string.

This field is set to the user input string when why is -1, and the previous invocation of run returns BOOKMARKFS_FSCK_USER_INPUT.

The string is only guaranteed to be valid until the function returns.

On success, the function should return one of the following values, indicating the operation to perform:

BOOKMARKFS_FSCK_NEXT

Continue to the next entry.

BOOKMARKFS_FSCK_APPLY

Apply change for the current entry. Not available in read-only mode.

The handler should copy the new name for the bookmark to data->entry.data.name.

BOOKMARKFS_FSCK_USER_INPUT

Request for user input. Only available in interactive mode.

The handler should set data->str to a prompt string for Readline. The string must be valid until the next function call on this handler context.

BOOKMARKFS_FSCK_SAVE

Save applied changes.

BOOKMARKFS_FSCK_STOP

Save applied changes and quit.

The run function is guaranteed not to be called again for this context.

BOOKMARKFS_FSCK_REWIND

Rewind current directory.

BOOKMARKFS_FSCK_SKIP

Skip current directory.

BOOKMARKFS_FSCK_SKIP_CHILDREN

Skip current directory and all subdirectories.

BOOKMARKFS_FSCK_RESET

Rewind all.

On error, the function should return -1.


Appendix A General Index


Appendix B GNU Free Documentation License

Version 1.3, 3 November 2008
Copyright © 2000, 2001, 2002, 2007, 2008 Free Software Foundation, Inc.
https://fsf.org/

Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
  1. PREAMBLE

    The purpose of this License is to make a manual, textbook, or other functional and useful document free in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it, with or without modifying it, either commercially or noncommercially. Secondarily, this License preserves for the author and publisher a way to get credit for their work, while not being considered responsible for modifications made by others.

    This License is a kind of “copyleft”, which means that derivative works of the document must themselves be free in the same sense. It complements the GNU General Public License, which is a copyleft license designed for free software.

    We have designed this License in order to use it for manuals for free software, because free software needs free documentation: a free program should come with manuals providing the same freedoms that the software does. But this License is not limited to software manuals; it can be used for any textual work, regardless of subject matter or whether it is published as a printed book. We recommend this License principally for works whose purpose is instruction or reference.

  2. APPLICABILITY AND DEFINITIONS

    This License applies to any manual or other work, in any medium, that contains a notice placed by the copyright holder saying it can be distributed under the terms of this License. Such a notice grants a world-wide, royalty-free license, unlimited in duration, to use that work under the conditions stated herein. The “Document”, below, refers to any such manual or work. Any member of the public is a licensee, and is addressed as “you”. You accept the license if you copy, modify or distribute the work in a way requiring permission under copyright law.

    A “Modified Version” of the Document means any work containing the Document or a portion of it, either copied verbatim, or with modifications and/or translated into another language.

    A “Secondary Section” is a named appendix or a front-matter section of the Document that deals exclusively with the relationship of the publishers or authors of the Document to the Document’s overall subject (or to related matters) and contains nothing that could fall directly within that overall subject. (Thus, if the Document is in part a textbook of mathematics, a Secondary Section may not explain any mathematics.) The relationship could be a matter of historical connection with the subject or with related matters, or of legal, commercial, philosophical, ethical or political position regarding them.

    The “Invariant Sections” are certain Secondary Sections whose titles are designated, as being those of Invariant Sections, in the notice that says that the Document is released under this License. If a section does not fit the above definition of Secondary then it is not allowed to be designated as Invariant. The Document may contain zero Invariant Sections. If the Document does not identify any Invariant Sections then there are none.

    The “Cover Texts” are certain short passages of text that are listed, as Front-Cover Texts or Back-Cover Texts, in the notice that says that the Document is released under this License. A Front-Cover Text may be at most 5 words, and a Back-Cover Text may be at most 25 words.

    A “Transparent” copy of the Document means a machine-readable copy, represented in a format whose specification is available to the general public, that is suitable for revising the document straightforwardly with generic text editors or (for images composed of pixels) generic paint programs or (for drawings) some widely available drawing editor, and that is suitable for input to text formatters or for automatic translation to a variety of formats suitable for input to text formatters. A copy made in an otherwise Transparent file format whose markup, or absence of markup, has been arranged to thwart or discourage subsequent modification by readers is not Transparent. An image format is not Transparent if used for any substantial amount of text. A copy that is not “Transparent” is called “Opaque”.

    Examples of suitable formats for Transparent copies include plain ASCII without markup, Texinfo input format, LaTeX input format, SGML or XML using a publicly available DTD, and standard-conforming simple HTML, PostScript or PDF designed for human modification. Examples of transparent image formats include PNG, XCF and JPG. Opaque formats include proprietary formats that can be read and edited only by proprietary word processors, SGML or XML for which the DTD and/or processing tools are not generally available, and the machine-generated HTML, PostScript or PDF produced by some word processors for output purposes only.

    The “Title Page” means, for a printed book, the title page itself, plus such following pages as are needed to hold, legibly, the material this License requires to appear in the title page. For works in formats which do not have any title page as such, “Title Page” means the text near the most prominent appearance of the work’s title, preceding the beginning of the body of the text.

    The “publisher” means any person or entity that distributes copies of the Document to the public.

    A section “Entitled XYZ” means a named subunit of the Document whose title either is precisely XYZ or contains XYZ in parentheses following text that translates XYZ in another language. (Here XYZ stands for a specific section name mentioned below, such as “Acknowledgements”, “Dedications”, “Endorsements”, or “History”.) To “Preserve the Title” of such a section when you modify the Document means that it remains a section “Entitled XYZ” according to this definition.

    The Document may include Warranty Disclaimers next to the notice which states that this License applies to the Document. These Warranty Disclaimers are considered to be included by reference in this License, but only as regards disclaiming warranties: any other implication that these Warranty Disclaimers may have is void and has no effect on the meaning of this License.

  3. VERBATIM COPYING

    You may copy and distribute the Document in any medium, either commercially or noncommercially, provided that this License, the copyright notices, and the license notice saying this License applies to the Document are reproduced in all copies, and that you add no other conditions whatsoever to those of this License. You may not use technical measures to obstruct or control the reading or further copying of the copies you make or distribute. However, you may accept compensation in exchange for copies. If you distribute a large enough number of copies you must also follow the conditions in section 3.

    You may also lend copies, under the same conditions stated above, and you may publicly display copies.

  4. COPYING IN QUANTITY

    If you publish printed copies (or copies in media that commonly have printed covers) of the Document, numbering more than 100, and the Document’s license notice requires Cover Texts, you must enclose the copies in covers that carry, clearly and legibly, all these Cover Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on the back cover. Both covers must also clearly and legibly identify you as the publisher of these copies. The front cover must present the full title with all words of the title equally prominent and visible. You may add other material on the covers in addition. Copying with changes limited to the covers, as long as they preserve the title of the Document and satisfy these conditions, can be treated as verbatim copying in other respects.

    If the required texts for either cover are too voluminous to fit legibly, you should put the first ones listed (as many as fit reasonably) on the actual cover, and continue the rest onto adjacent pages.

    If you publish or distribute Opaque copies of the Document numbering more than 100, you must either include a machine-readable Transparent copy along with each Opaque copy, or state in or with each Opaque copy a computer-network location from which the general network-using public has access to download using public-standard network protocols a complete Transparent copy of the Document, free of added material. If you use the latter option, you must take reasonably prudent steps, when you begin distribution of Opaque copies in quantity, to ensure that this Transparent copy will remain thus accessible at the stated location until at least one year after the last time you distribute an Opaque copy (directly or through your agents or retailers) of that edition to the public.

    It is requested, but not required, that you contact the authors of the Document well before redistributing any large number of copies, to give them a chance to provide you with an updated version of the Document.

  5. MODIFICATIONS

    You may copy and distribute a Modified Version of the Document under the conditions of sections 2 and 3 above, provided that you release the Modified Version under precisely this License, with the Modified Version filling the role of the Document, thus licensing distribution and modification of the Modified Version to whoever possesses a copy of it. In addition, you must do these things in the Modified Version:

    1. Use in the Title Page (and on the covers, if any) a title distinct from that of the Document, and from those of previous versions (which should, if there were any, be listed in the History section of the Document). You may use the same title as a previous version if the original publisher of that version gives permission.
    2. List on the Title Page, as authors, one or more persons or entities responsible for authorship of the modifications in the Modified Version, together with at least five of the principal authors of the Document (all of its principal authors, if it has fewer than five), unless they release you from this requirement.
    3. State on the Title page the name of the publisher of the Modified Version, as the publisher.
    4. Preserve all the copyright notices of the Document.
    5. Add an appropriate copyright notice for your modifications adjacent to the other copyright notices.
    6. Include, immediately after the copyright notices, a license notice giving the public permission to use the Modified Version under the terms of this License, in the form shown in the Addendum below.
    7. Preserve in that license notice the full lists of Invariant Sections and required Cover Texts given in the Document’s license notice.
    8. Include an unaltered copy of this License.
    9. Preserve the section Entitled “History”, Preserve its Title, and add to it an item stating at least the title, year, new authors, and publisher of the Modified Version as given on the Title Page. If there is no section Entitled “History” in the Document, create one stating the title, year, authors, and publisher of the Document as given on its Title Page, then add an item describing the Modified Version as stated in the previous sentence.
    10. Preserve the network location, if any, given in the Document for public access to a Transparent copy of the Document, and likewise the network locations given in the Document for previous versions it was based on. These may be placed in the “History” section. You may omit a network location for a work that was published at least four years before the Document itself, or if the original publisher of the version it refers to gives permission.
    11. For any section Entitled “Acknowledgements” or “Dedications”, Preserve the Title of the section, and preserve in the section all the substance and tone of each of the contributor acknowledgements and/or dedications given therein.
    12. Preserve all the Invariant Sections of the Document, unaltered in their text and in their titles. Section numbers or the equivalent are not considered part of the section titles.
    13. Delete any section Entitled “Endorsements”. Such a section may not be included in the Modified Version.
    14. Do not retitle any existing section to be Entitled “Endorsements” or to conflict in title with any Invariant Section.
    15. Preserve any Warranty Disclaimers.

    If the Modified Version includes new front-matter sections or appendices that qualify as Secondary Sections and contain no material copied from the Document, you may at your option designate some or all of these sections as invariant. To do this, add their titles to the list of Invariant Sections in the Modified Version’s license notice. These titles must be distinct from any other section titles.

    You may add a section Entitled “Endorsements”, provided it contains nothing but endorsements of your Modified Version by various parties—for example, statements of peer review or that the text has been approved by an organization as the authoritative definition of a standard.

    You may add a passage of up to five words as a Front-Cover Text, and a passage of up to 25 words as a Back-Cover Text, to the end of the list of Cover Texts in the Modified Version. Only one passage of Front-Cover Text and one of Back-Cover Text may be added by (or through arrangements made by) any one entity. If the Document already includes a cover text for the same cover, previously added by you or by arrangement made by the same entity you are acting on behalf of, you may not add another; but you may replace the old one, on explicit permission from the previous publisher that added the old one.

    The author(s) and publisher(s) of the Document do not by this License give permission to use their names for publicity for or to assert or imply endorsement of any Modified Version.

  6. COMBINING DOCUMENTS

    You may combine the Document with other documents released under this License, under the terms defined in section 4 above for modified versions, provided that you include in the combination all of the Invariant Sections of all of the original documents, unmodified, and list them all as Invariant Sections of your combined work in its license notice, and that you preserve all their Warranty Disclaimers.

    The combined work need only contain one copy of this License, and multiple identical Invariant Sections may be replaced with a single copy. If there are multiple Invariant Sections with the same name but different contents, make the title of each such section unique by adding at the end of it, in parentheses, the name of the original author or publisher of that section if known, or else a unique number. Make the same adjustment to the section titles in the list of Invariant Sections in the license notice of the combined work.

    In the combination, you must combine any sections Entitled “History” in the various original documents, forming one section Entitled “History”; likewise combine any sections Entitled “Acknowledgements”, and any sections Entitled “Dedications”. You must delete all sections Entitled “Endorsements.”

  7. COLLECTIONS OF DOCUMENTS

    You may make a collection consisting of the Document and other documents released under this License, and replace the individual copies of this License in the various documents with a single copy that is included in the collection, provided that you follow the rules of this License for verbatim copying of each of the documents in all other respects.

    You may extract a single document from such a collection, and distribute it individually under this License, provided you insert a copy of this License into the extracted document, and follow this License in all other respects regarding verbatim copying of that document.

  8. AGGREGATION WITH INDEPENDENT WORKS

    A compilation of the Document or its derivatives with other separate and independent documents or works, in or on a volume of a storage or distribution medium, is called an “aggregate” if the copyright resulting from the compilation is not used to limit the legal rights of the compilation’s users beyond what the individual works permit. When the Document is included in an aggregate, this License does not apply to the other works in the aggregate which are not themselves derivative works of the Document.

    If the Cover Text requirement of section 3 is applicable to these copies of the Document, then if the Document is less than one half of the entire aggregate, the Document’s Cover Texts may be placed on covers that bracket the Document within the aggregate, or the electronic equivalent of covers if the Document is in electronic form. Otherwise they must appear on printed covers that bracket the whole aggregate.

  9. TRANSLATION

    Translation is considered a kind of modification, so you may distribute translations of the Document under the terms of section 4. Replacing Invariant Sections with translations requires special permission from their copyright holders, but you may include translations of some or all Invariant Sections in addition to the original versions of these Invariant Sections. You may include a translation of this License, and all the license notices in the Document, and any Warranty Disclaimers, provided that you also include the original English version of this License and the original versions of those notices and disclaimers. In case of a disagreement between the translation and the original version of this License or a notice or disclaimer, the original version will prevail.

    If a section in the Document is Entitled “Acknowledgements”, “Dedications”, or “History”, the requirement (section 4) to Preserve its Title (section 1) will typically require changing the actual title.

  10. TERMINATION

    You may not copy, modify, sublicense, or distribute the Document except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense, or distribute it is void, and will automatically terminate your rights under this License.

    However, if you cease all violation of this License, then your license from a particular copyright holder is reinstated (a) provisionally, unless and until the copyright holder explicitly and finally terminates your license, and (b) permanently, if the copyright holder fails to notify you of the violation by some reasonable means prior to 60 days after the cessation.

    Moreover, your license from a particular copyright holder is reinstated permanently if the copyright holder notifies you of the violation by some reasonable means, this is the first time you have received notice of violation of this License (for any work) from that copyright holder, and you cure the violation prior to 30 days after your receipt of the notice.

    Termination of your rights under this section does not terminate the licenses of parties who have received copies or rights from you under this License. If your rights have been terminated and not permanently reinstated, receipt of a copy of some or all of the same material does not give you any rights to use it.

  11. FUTURE REVISIONS OF THIS LICENSE

    The Free Software Foundation may publish new, revised versions of the GNU Free Documentation License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. See https://www.gnu.org/licenses/.

    Each version of the License is given a distinguishing version number. If the Document specifies that a particular numbered version of this License “or any later version” applies to it, you have the option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. If the Document does not specify a version number of this License, you may choose any version ever published (not as a draft) by the Free Software Foundation. If the Document specifies that a proxy can decide which future versions of this License can be used, that proxy’s public statement of acceptance of a version permanently authorizes you to choose that version for the Document.

  12. RELICENSING

    “Massive Multiauthor Collaboration Site” (or “MMC Site”) means any World Wide Web server that publishes copyrightable works and also provides prominent facilities for anybody to edit those works. A public wiki that anybody can edit is an example of such a server. A “Massive Multiauthor Collaboration” (or “MMC”) contained in the site means any set of copyrightable works thus published on the MMC site.

    “CC-BY-SA” means the Creative Commons Attribution-Share Alike 3.0 license published by Creative Commons Corporation, a not-for-profit corporation with a principal place of business in San Francisco, California, as well as future copyleft versions of that license published by that same organization.

    “Incorporate” means to publish or republish a Document, in whole or in part, as part of another Document.

    An MMC is “eligible for relicensing” if it is licensed under this License, and if all works that were first published under this License somewhere other than this MMC, and subsequently incorporated in whole or in part into the MMC, (1) had no cover texts or invariant sections, and (2) were thus incorporated prior to November 1, 2008.

    The operator of an MMC Site may republish an MMC contained in the site under CC-BY-SA on the same site at any time before August 1, 2009, provided the MMC is eligible for relicensing.

ADDENDUM: How to use this License for your documents

To use this License in a document you have written, include a copy of the License in the document and put the following copyright and license notices just after the title page:

  Copyright (C)  year  your name.
  Permission is granted to copy, distribute and/or modify this document
  under the terms of the GNU Free Documentation License, Version 1.3
  or any later version published by the Free Software Foundation;
  with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
  Texts.  A copy of the license is included in the section entitled ``GNU
  Free Documentation License''.

If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts, replace the “with…Texts.” line with this:

    with the Invariant Sections being list their titles, with
    the Front-Cover Texts being list, and with the Back-Cover Texts
    being list.

If you have Invariant Sections without Cover Texts, or some other combination of the three, merge those two alternatives to suit the situation.

If your document contains nontrivial examples of program code, we recommend releasing these examples in parallel under your choice of free software license, such as the GNU General Public License, to permit their use in free software.