Readonly keepaliveControls the default timeout (in seconds) for sending keepalive probes.
Readonly keepaliveControls the default retransmission count for keepalive probes.
Readonly keepaliveControls the default interval (in seconds) between retrying keepalive probes.
Controls whether the socket transport service is offline. Setting it offline will cause non-local socket detachment.
Readonly sendcontrols the TCP sender window clamp
Optional aInstancePtr: objectA run time mechanism for interface discovery.
NS_OK if the interface is supported by the associated instance, NS_NOINTERFACE if it is not.
aInstancePtr must not be null.
[in] A requested interface IID
[out] A pointer to an interface pointer to receive the result.
Adds a new socket to the list of controlled sockets.
This will fail with the error code NS_ERROR_NOT_AVAILABLE if the maximum number of sockets is already reached. In this case, the notifyWhenCanAttachSocket method should be used.
Open file descriptor of the socket to control.
Socket handler that will receive notifications when the socket is ready or detached.
NOTE: this function may only be called from an event dispatch on the socket thread.
Creates a transport for a specified host and port.
NOTE: this function can be called from any thread
array of socket type strings. Empty array if using default socket type.
specifies the target hostname or IP address literal of the peer for this socket.
specifies the target port of the peer for this socket.
specifies the transport-layer proxy type to use. null if no proxy. used for communicating information about proxies like SOCKS (which are transparent to upper protocols).
Create a transport built on a Unix domain socket that uses abstract address name.
If abstract socket address isn't supported on System, this returns NS_ERROR_SOCKET_ADDRESS_NOT_SUPPORTED.
The name of abstract socket adress of the Unix domain socket to which we should connect.
Create a transport built on a Unix domain socket, connecting to the given filename.
Since Unix domain sockets are always local to the machine, they are not affected by the nsIIOService's 'offline' flag.
On systems that don't support Unix domain sockets at all, this returns NS_ERROR_SOCKET_ADDRESS_NOT_SUPPORTED.
The system-level socket API may impose restrictions on the length of the filename that are stricter than those of the underlying filesystem. If the file name is too long, this returns NS_ERROR_FILE_NAME_TOO_LONG.
The |aPath| parameter must specify an existing directory entry. Otherwise, this returns NS_ERROR_FILE_NOT_FOUND.
The program must have search permission on all components of the path prefix of |aPath|, and read and write permission on |aPath| itself. Without such permission, this returns NS_ERROR_CONNECTION_REFUSED.
The |aPath| parameter must refer to a unix-domain socket. Otherwise, this returns NS_ERROR_CONNECTION_REFUSED. (POSIX specifies ECONNREFUSED when "the target address was not listening for connections", and this is what Linux returns.)
The file name of the Unix domain socket to which we should connect.
if the number of sockets reaches the limit, then consumers can be notified when the number of sockets becomes less than the limit. the notification is asynchronous, delivered via the given nsIRunnable instance on the socket transport thread.
Event that will receive the notification when a new socket can be attached
NOTE: this function may only be called from an event dispatch on the socket thread.
Generated using TypeDoc
This is a private interface used by the internals of the networking library. It will never be frozen. Do not use it in external code.