Multiple plugins can be mounted into the key database (KDB). On every access to the key data base they are executed and thus can change the functionality and behavior.
Unlike Elektra's core the plugins have all kinds of dependencies. It is the responsibility of the plugin to find and check its dependencies using CMake. If a dependency cannot be found, the plugin will automatically disable itself.
Elektra has a wide range of different plugins. The plugin folders should contain a README.md with further information. (Or follow links below.) The plugins are:
Overview Plugins
All plugins implement the same interface:
kdbOpen()
calls elektraPluginOpen()
of every plugin to let them do their initialisation.
kdbGet()
requests elektraPluginGet()
of every plugin in the queried backends to return a key set.
kdbSet()
usually calls elektraPluginSet()
of every plugin in the queried backends to store the configuration.
kdbSet()
also calls elektraPluginError()
for every plugin when an error happens. Because of elektraPluginError()
, plugins are guaranteed to have their chance for necessary cleanups.
kdbClose()
makes sure that plugins can finally free their own resources in elektraPluginClose()
.
Furthermore, plugins might export symbols:
checkconf
can be called during mounting to ensure a plugin has valid configuration.
genconf
can be called to produce all valid configurations of a plugin.
For an easy introduction, see this tutorial how to write a storage plugin. For more background information of the plugins framework, continue here. Otherwise, you can visit the the API documentation.
Before configuration is actually written, the file name needs to be determined (resolvers will be automatically added by kdb mount):
- blockresolver resolves tagged blocks inside config files
- curlget fetches configuration file from a remote host
- gitresolver checks out and commits files to a local git repository and afterwards the configuration file must be synced with harddisc (recommended to add at every kdb mount):
- multifile
- noresolver does not resolve, but can act as one
- resolver uses advanced POSIX APIs to handle conflicts gracefully
- sync uses POSIX APIs to sync configuration files with the hard disk
- wresolver minimalistic resolver for non-POSIX systems
Are responsible for reading writing the configuration to configuration files.
Read and write everything a KeySet might contain:
- dump makes a dump of a KeySet in an Elektra-specific format
- ini supports a range of INI file formats.
- quickdump uses binary portable format based on dump, but more efficient
Read (and write) standard config files:
- augeas reads/writes many different configuration files using the augeas library
- hosts reads/writes hosts files
- kconfig reads/writes KConfig ini files
- line reads/writes any file line by line
- yajl reads/writes JSON.
Using semi-structured data for config files, mainly suitable for spec-namespace (put a focus on having nice syntax for metadata):
- ni parses INI files based on (including metadata) ni.
- tcl-like config files (including metadata).
Only suited for import/export:
- mini dependency free, line based key-value storage plugin.
- simpleini line-based key-value pairs with configurable format (without sections)
- xerces uses XML (without a specific schema).
- xmltool uses XML in the deprecated Elektra XML schema for importing configuration from Elektra 0.7.
- yamlsmith exports key sets in the YAML format
Plugins that just show some functionality, (currently) not intended for productive use:
- c writes Elektra C-structures (
ksNew(.. keyNew(...
)
- csvstorage for csv files
- dpkg reads /var/lib/dpkg/{available,status}
- file reads and writes a file from/to a single key
- fstab for fstab files.
- mmapstorage uses binary, not portable memory mapped file for a high performance storage
- mozprefs for Mozilla preference files
- passwd for passwd files
- specload calls an external application to request its specification, depends on quickdump
- toml reads and writes data using a parser generated by Flex and Bison
- yamlcpp reads and writes data in the YAML format using yaml-cpp
- yanlr reads data using a parser generated by ANTLR
Information compiled in Elektra:
- version is a built-in plugin directly within the core so that it cannot give wrong version information
- constants various constants, including version information
- desktop contains information which desktop is currently running
Providing information found on the system not available in persistent files:
- uname information from the uname syscall.
Filter plugins process keys and their values in both directions. In one direction they undo what they do in the other direction. Most filter plugins available now encode and decode values. Storage plugins that use characters to separate key names, values or metadata will not work without them.
Rewrite unwanted characters with different techniques:
- base64 using the Base64 encoding scheme (RFC4648)
- ccode using the technique from arrays in the programming language C
- hexcode using hex codes
- directoryvalue converts directory values to leaf values
- hexnumber converts between hexadecimal and decimal
- keytometa transforms keys to metadata
- rename renames keys according to different rules
- crypto encrypts / decrypts confidential values
- fcrypt encrypts / decrypts entire backend files
- gpgme encrypts / decrypts confidential values (with GPGME)
- hidden hides keys whose names start with a
.
.
- iconv makes sure the configuration will have correct character encoding
- null takes care of null values and other binary specialities
Log/Send out all changes to configuration to:
- dbus sends notifications for every change via dbus
notification
- dbusrecv receives notifications via dbus
notification
- journald logs key database changes to journald
- logchange prints the change of every key on the console
- syslog logs key database changes to syslog
- zeromqsend sends notifications for every change via ZeroMQ sockets
notification
- zeromqrecv receives notifications via ZeroMQ sockets
notification
Notification of key changes:
Trace everything that happens within KDB:
Copies metadata to keys:
- glob using globbing techniques (needed by some plugins)
- spec copies metadata from spec namespace (the standard way)
Plugins that check if values are valid based on metadata (typically copied by the spec
plugin just before):
Value Validation
- conditionals by using if-then-else like statements
- date validates date and time data
- ipaddr checks IP addresses using regular expressions
- mathcheck by mathematical expressions using key values as operands
- network by using network APIs
- macaddr checks if MAC addresses are valid and normalizes them
- path by checking files on file system
- range checks if a value is within a given range
- reference checks if a value is a valid reference to another key
- rgbcolor validates and normalizes hexcolors
- type type checking (CORBA types) with enum functionality
- unit validates and normalizes units of memory (e.g. 20KB to 20000 Bytes)
- validation by using regex
Other Validation
These plugins start an interpreter and allow you to execute a script in an interpreted language whenever Elektra’s key database gets accessed. Note that they depend on the presence of the respective binding during run-time.
- jni java plugins started by jni, works with jna plugins
- lua Lua plugins
- python Python 3 plugins
- ruby Ruby plugins
- shell executes shell commandos
- cache caches keysets from previous
kdbGet()
calls
- cpptemplate a template for C++ based plugins
- doc contains the documentation of the plugin interface
- error yields errors as described in metadata (handy for test purposes)
- gopts global plugin to automatically call
elektraGetOpts
- iterate iterate over all keys and run exported functions on tagged keys
- list loads other plugins
- process proxy plugin that executes other plugins in a separate process
- profile links profile keys
- template to be copied for new plugins