You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Although the standard dwm patch for clickable status requires the use of BSD user signals, would it be worth programming an alternate interface and a custom dwm patch which uses some other RPC mechanism. For instance, a simple named pipe would do. We could have dwm write the index of the block into the FIFO. Then, all gocaudices has to do is have a goroutine waiting for the data (FIFOs are buffered) and send a signal over a channel when data is recieved. This would be much cleaner than all the cgo stuff you had to do for signals.
The downside of this is, obviously, that you can't use the stock dwm clickablestatus patch, but I think this could be well worth the cost of a separate patch to remove all the C signal handling.
I'll do some more research into named pipes and, if you want, start a draft PR.
The text was updated successfully, but these errors were encountered:
Funny you should bring this up, I've been thinking on using fifo specifically to make this patch, but I've lost a lot of time for programming and what little I have has been for zara, similar to this project except uses interfaces rather than external programs or scripts.
I say full send on this, I'll be working on it when I can but if you can get a working model, I'll merge it in. And tbh having a patch for dwm that handles status clicks through fifo rather than signals would be simpler, and might I say, suck less.
And I'd love nothing more than to get rid of the shitty cgo I had to write to get it working with the other patch.
Although the standard dwm patch for clickable status requires the use of BSD user signals, would it be worth programming an alternate interface and a custom dwm patch which uses some other RPC mechanism. For instance, a simple named pipe would do. We could have dwm write the index of the block into the FIFO. Then, all gocaudices has to do is have a goroutine waiting for the data (FIFOs are buffered) and send a signal over a channel when data is recieved. This would be much cleaner than all the cgo stuff you had to do for signals.
The downside of this is, obviously, that you can't use the stock dwm clickablestatus patch, but I think this could be well worth the cost of a separate patch to remove all the C signal handling.
I'll do some more research into named pipes and, if you want, start a draft PR.
The text was updated successfully, but these errors were encountered: