provide a way to silence "failed to read..." matron spam (in repl)

[pq]

periodically, i see streams like this coming from matron.

norns/matron/src/hardware/battery.c

Line 62 in 484caad

fprintf(stderr, "failed to read battery percentage\n");

is it adding any value? if not, it'd be great to have a way to filter this out.

assuming these messages should keep going to stderr, it may be easiest (but not best?) to do this in maiden? (cc @ngwese)

dunno.

[simonvanderveldt]

Ideally we'd fix the cause of these failures, so far not sure what's causing them though :(
See monome/norns-image#43

[pq]

yeah. not great to just drop them on the floor. on the other hand, it's a bit noisy in the repl and could make folks think there's something up with the device.

[tehn]

i'm fine to just mute these battery warnings for now and open a ticket to investigate.

the code should also handle the fail more gracefully, ie, don't corrupt the previous read.

[simonvanderveldt]

Where does the repl get these logline from? Does it show all messages sent to stderr?

[pq]

@simonvanderveldt: i believe so. @ngwese can confirm?

[ngwese]

the repl prints whatever ws-wrapper sends. i don’t remember if it captures both stdout and stderr or just stdout (from matron).

[pq]

@ngwese : is there any way to know whether we're getting (or sending) stderr? (i'm thinking it would be great to be able to render the output differently.)

EDIT: i'm sure this is a way; i guess i should rephrase and ask for your thoughts on how we might differentiate. looking at ws-wrapper now...

[pq]

an interesting possible data point regarding the cause of the errors: i've only noticed the errors when my norns has been powered (specifically by my macbook). i don't think i've seen the errors when self-powered...

[ngwese]

@pq regarding separation of stderr from stdout in the repl- there’s only one stream sent over the websocket right now. ws-wrapper would need to change and a simple protocol introduced.

[pq]

an interesting possible data point regarding the cause of the errors: i've only noticed the errors when my norns has been powered (specifically by my macbook). i don't think i've seen the errors when self-powered...

seeing this correlation again. also fairly certain that the battery is fully charged when it occurs FWIW.

[markwheeler]

Had this happen today whilst norns was running on mains power, no other connections.