4:06:06beachHello slyrus. Since you often ask for SICL updates, I managed to connect the HIR interpreter to a primitive version of Clordane: metamodular.com/clordane.mp4 Only some simple stepping works for now.
6:43:32lokejackdaniel: about the delayed events when executing a comman asynchronously... Can you remind me, was the reason for this that the main event thread is blocked in XNextEvent()
6:44:10jackdanielI don't know what XNextEvent() is
6:46:08lokejackdaniel: It's the function that asks Xlib to wait for the next event.
6:46:21lokeThe event processing thread should be blocked itn it for most of the time.
6:46:58lokeOh wait... Sorry. The event processing thread and the application-frame's thread are different.
6:48:02jackdanielI still don't recall any function in clx called XNextEvent(), but you are correct, by default port has its own thread which fills the thread-safe fifo queue with events
6:51:44lokeah. in CLX it's called XLIB:PROCESS-EVENT
6:53:35lokeIn StumpWM, I had the need to get PROCESS-EVENT to break out early, and the only way I could think of was to have CLX listen to an extra file handle (a pipe), and then write a byte to that pipe in order to get it stop blocking.
6:53:53lokeIt's been a while so I can't recall all the details.
6:54:04jackdanielclx has support for a timeout argument
6:54:41lokeyeah, but that doesn't help if you want immediate asynchronous respone
6:54:51loke(without setting the timeout to zero and busy-wait)
6:56:20jackdanielI think that the correct way to handle problem of multiple input sources is a thread-safe fifo queue