A method call is a request to run some procedure exported by the remote process. Procedures are identified by an (object_path, interface_name, method_name) tuple.

Construct a new MethodCall for the given object, interface, and method.

Use fields such as methodCallDestination and methodCallBody to populate a MethodCall.

{-# LANGUAGE OverloadedStrings #-}

methodCall "/" "org.example.Math" "Add"
    { methodCallDestination = Just "org.example.Calculator"
    , methodCallBody = [toVariant (1 :: Int32), toVariant (2 :: Int32)]
    }
 

The object path of the method call. Conceptually, object paths act like a procedural language's pointers. Each object referenced by a path is a collection of procedures.

The interface of the method call. Each object may implement any number of interfaces. Each method is part of at least one interface.

In certain cases, this may be Nothing, but most users should set it to a value.

The method name of the method call. Method names are unique within an interface, but might not be unique within an object.

The name of the application that sent this call.

Most users will just leave this empty, because the bus overwrites the sender for security reasons. Setting the sender manually is used for peer-peer connections.

Defaults to Nothing.

The name of the application to send the call to.

Most users should set this. If a message with no destination is sent to the bus, the bus will behave as if the destination was set to org.freedesktop.DBus. For peer-peer connections, the destination can be empty because there is only one peer.

Defaults to Nothing.

Set whether the bus should auto-start the remote

Defaults to True.

Set whether a reply is expected. This can save network and cpu resources by inhibiting unnecessary replies.

Defaults to True.

The arguments to the method call. See toVariant.

Defaults to [].

A method return is a reply to a method call, indicating that the call succeeded.

Construct a new MethodReturn, in reply to a method call with the given serial.

Use fields such as methodReturnBody to populate a MethodReturn.

The serial of the original method call. This lets the original caller match up this reply to the pending call.

The name of the application that is returning from a call.

Most users will just leave this empty, because the bus overwrites the sender for security reasons. Setting the sender manually is used for peer-peer connections.

Defaults to Nothing.

The name of the application that initiated the call.

Most users should set this. If a message with no destination is sent to the bus, the bus will behave as if the destination was set to org.freedesktop.DBus. For peer-peer connections, the destination can be empty because there is only one peer.

Defaults to Nothing.

Values returned from the method call. See toVariant.

Defaults to [].

A method error is a reply to a method call, indicating that the call received an error and did not succeed.

Construct a new MethodError, in reply to a method call with the given serial.

Use fields such as methodErrorBody to populate a MethodError.

The name of the error type. Names are used so clients can handle certain classes of error differently from others.

The serial of the original method call. This lets the original caller match up this reply to the pending call.

The name of the application that is returning from a call.

Most users will just leave this empty, because the bus overwrites the sender for security reasons. Setting the sender manually is used for peer-peer connections.

Defaults to Nothing.

The name of the application that initiated the call.

Most users should set this. If a message with no destination is sent to the bus, the bus will behave as if the destination was set to org.freedesktop.DBus. For peer-peer connections, the destination can be empty because there is only one peer.

Defaults to Nothing.

Additional information about the error. By convention, if the error body contains any items, the first item should be a string describing the error.

Get a human-readable description of the error, by returning the first item in the error body if it's a string.

Signals are broadcast by applications to notify other clients of some event.

Construct a new Signal for the given object, interface, and signal name.

Use fields such as signalBody to populate a Signal.

The path of the object that emitted this signal.

The name of this signal.

The interface that this signal belongs to.

The name of the application that emitted this signal.

Most users will just leave this empty, because the bus overwrites the sender for security reasons. Setting the sender manually is used for peer-peer connections.

Defaults to Nothing.

The name of the application to emit the signal to. If Nothing, the signal is sent to any application that has registered an appropriate match rule.

Defaults to Nothing.

Additional information about the signal, such as the new value or the time.

Defaults to [].

Not an actual message type, but a wrapper around messages received from the bus. Each value contains the message's Serial.

If casing against these constructors, always include a default case to handle messages of an unknown type. New message types may be added to the D-Bus specification, and applications should handle them gracefully by either ignoring or logging them.

No matter what sort of message was received, get its serial.

No matter what sort of message was received, get its sender (if provided).

No matter what sort of message was received, get its body (if provided).

Variants may contain any other built-in D-Bus value. Besides representing native VARIANT values, they allow type-safe storage and inspection of D-Bus collections.

Every variant is strongly-typed; that is, the type of its contained value is known at all times. This function retrieves that type, so that the correct cast can be used to retrieve the value.

Atomic types can be used as keys to dictionaries.

Users may not provide new instances of IsAtom because this could allow dictionaries to be created with invalid keys.

Value types can be used as items in containers, such as lists or dictionaries.

Users may not provide new instances of IsValue because this could allow containers to be created with items of heterogenous types.

Get the D-Bus type corresponding to the given Haskell value. The value may be undefined.

A signature is a list of D-Bus types, obeying some basic rules of validity.

The rules of signature validity are complex: see http://dbus.freedesktop.org/doc/dbus-specification.html#message-protocol-signatures for details.

Convert a list of types into a valid signature.

Returns Nothing if the given types are not a valid signature.

Convert a list of types into a valid signature.

Throws an exception if the given types are not a valid signature.

Get the list of types in a signature. The inverse of signature.

Convert a signature into a signature string. The inverse of parseSignature.

Parse a signature string into a valid signature.

Returns Nothing if the given string is not a valid signature.

Object paths are special strings, used to identify a particular object exported from a D-Bus application.

Object paths must begin with a slash, and consist of alphanumeric characters separated by slashes.

See http://dbus.freedesktop.org/doc/dbus-specification.html#message-protocol-marshaling-object-path for details.

Interfaces are used to group a set of methods and signals within an exported object. Interface names consist of alphanumeric characters separated by periods.

See http://dbus.freedesktop.org/doc/dbus-specification.html#message-protocol-names-interface for details.

Member names are used to identify a single method or signal within an interface. Method names consist of alphanumeric characters.

See http://dbus.freedesktop.org/doc/dbus-specification.html#message-protocol-names-member for details.

Error names are used to identify which type of error was returned from a method call. Error names consist of alphanumeric characters separated by periods.

See http://dbus.freedesktop.org/doc/dbus-specification.html#message-protocol-names-error for details.

Bus names are used to identify particular clients on the message bus. A bus name may be either unique or well-known, where unique names start with a colon. Bus names consist of alphanumeric characters separated by periods.

See http://dbus.freedesktop.org/doc/dbus-specification.html#message-protocol-names-bus for details.

A D-Bus Structure is a container type similar to Haskell tuples, storing values of any type that is convertable to IsVariant. A Structure may contain up to 255 values.

Most users can use the IsVariant instance for tuples to extract the values of a structure. This type is for very large structures, which may be awkward to work with as tuples.

A D-Bus Array is a container type similar to Haskell lists, storing zero or more values of a single D-Bus type.

Most users can use the IsVariant instance for lists or vectors to extract the values of an array. This type is for advanced use cases, where the user wants to convert array values to Haskell types that are not instances of IsValue.

A D-Bus Dictionary is a container type similar to Haskell maps, storing zero or more associations between keys and values.

Most users can use the IsVariant instance for maps to extract the values of a dictionary. This type is for advanced use cases, where the user wants to convert dictionary items to Haskell types that are not instances of IsValue.

When a D-Bus server must listen for connections, or a client must connect to a server, the listening socket's configuration is specified with an address. An address contains the method, which determines the protocol and transport mechanism, and parameters, which provide additional method-specific information about the address.

Try to convert a method string and parameter map to an Address.

Returns Nothing if the method or parameters are invalid.

Convert an address to a string in the format expected by parseAddress.

Convert a list of addresses to a string in the format expected by parseAddresses.

Try to parse a string containing one valid address.

An address string is in the format method:key1=val1,key2=val2. There are some limitations on the characters allowed within methods and parameters; see the D-Bus specification for full details.

Try to parse a string containing one or more valid addresses.

Addresses are separated by semicolons. See parseAddress for the format of addresses.

Returns the address in the environment variable DBUS_SYSTEM_BUS_ADDRESS, or unix:path=/var/run/dbus/system_bus_socket if DBUS_SYSTEM_BUS_ADDRESS is not set.

Returns Nothing if DBUS_SYSTEM_BUS_ADDRESS contains an invalid address.

Returns the address in the environment variable DBUS_SESSION_BUS_ADDRESS, which must be set.

Returns Nothing if DBUS_SYSTEM_BUS_ADDRESS is unset or contains an invalid address.

Returns the address in the environment variable DBUS_STARTER_ADDRESS, which must be set.

Returns Nothing if DBUS_STARTER_ADDRESS is unset or contains an invalid address.

Convert a Message into a ByteString. Although unusual, it is possible for marshaling to fail; if this occurs, an error will be returned instead.

Parse a ByteString into a ReceivedMessage. The result can be inspected to see what type of message was parsed. Unknown message types can still be parsed successfully, as long as they otherwise conform to the D-Bus standard.

A value used to uniquely identify a particular message within a session. Serials are 32-bit unsigned integers, and eventually wrap.

Get the first serial in the sequence.

Get the next serial in the sequence. This may wrap around to firstSerial.

A D-Bus UUID is 128 bits of data, usually randomly generated. They are used for identifying unique server instances to clients.

Older versions of the D-Bus spec also called these values GUIDs.

D-Bus UUIDs are not the same as the RFC-standardized UUIDs or GUIDs.

Format a D-Bus UUID as hex-encoded ASCII.

Generate a random D-Bus UUID. This value is suitable for use in a randomly-allocated address, or as a listener's socket address "guid" parameter.