Ferreteria/v0.6/clade/Sys/InOut/Connx/run/Session/Remote/SSH2

This clade has been retired and is no longer in use.

clade: Sys\InOut\Connx\run\Session\Remote\SSH2

Clade Family Remote ⇒ SSH2 ⇒ (none) Clade Aliases Alias Clade AppAdmin \App Base* [ca,i] IO\Aspect\Connx\run\Session\Remote Canals* [c,i] IO\Aspect\Connx\aux\A\Canals CommandIface IO\Aspect\Connx\run\Command ConveyClass IO\Aspect\Connx\Canal EventOnMsg Sys\Events\Event\Loops\OnMessage EventOnReset IO\Aspect\Connx\Stream\aux\Event\Reset RExecTrait IO\Aspect\Connx\run\Session\common\HasRExec OpCmd* [c,i] IO\Aspect\Connx\run\Starter\ops\OpCmdOpCmdIface }; OpOpen* [c,i] Sys\Events\ItWent\connx\OpOpen OpShut* [c,i] Sys\Events\ItWent\connx\OpShut QRes* [c,i] Data\Mem\QVar\RW\Res SelfIface IO\Aspect\Connx\run\Session\Remote\SSH2 StreamClass Sys\InOut\Connx\Stream\Exec\SSH2 StreamIface Sys\InOut\Connx\Stream Subpages

History

date	what
2026-07-29	changed property-functions to actual properties changed base-class from Futilities App to Ferreteria App moved $RExec to common trait in IsRunning(), I think I was still confused about exec resources vs. stream resources: stream_select() expects arrays of stream-resources, and I was passing it RExec for the first two arrays; passing Stream resources now (but see 2026-08-01 for an update). This required making RNative (which should be renamed RStrm) public. #TODO: All of the code which uses PHP-native functions should be moved into Core clades so they can be swapped out for phpseclib if needed.
2026-08-01	...but actually, it turns out RExec is a stream, and it is used for setting up the Stream-objects (later, on demand). Possibly it should still be renamed $RStrm, to avoid confusion like this. Update: done. Note that $RExec is still used in IO\Aspect\Connx\run\Starter\Local\Proc (a Starter, not a Session) because it's a handle to the process rather than a primary stream. The corresponding resource-handle in IO\Aspect\Connx\run\Session\Remote\SSH2 is $RConn, via $QRConn -- maybe that should be renamed to $RExec? Note that there is a reason why the executable-resource is in the Starter for Local Proc but in the Session for Remote SSH2. It has something to do with how the underlying libraries work, and I'm pretty sure I documented it... somewhere.