Current Server Time: 05:24 (Central Europe)

#apertus IRC Channel Logs

2020/05/17

Timezone: UTC


02:45
Bertl_oO
off to bed now ... have a good one everyone!
02:45
Bertl_oO
changed nick to: Bertl_zZ
04:04
Spirit532
left the channel
04:04
Spirit532
joined the channel
09:03
se6ast1an
good day
09:07
BAndiT1983|away
changed nick to: BAndiT1983
12:30
Bertl_zZ
changed nick to: Bertl
12:30
Bertl
morning folks!
14:05
panintended
Hi all
14:09
panintended
se6ast1an BAndiT1983 and everyone else: For the visualiser, what is the expected outcome in the below scenario when it comes to BUTTON_UP/DOWN events received?
14:09
panintended
a button is clicked, then mouse button is held but then released outside of the button area
14:09
panintended
(this is for T1185 btw)
14:12
BAndiT1983
hi panintended
14:12
BAndiT1983
how is it going?
14:13
panintended
not too bad, yourself? :)
14:13
BAndiT1983
am fine too
14:13
BAndiT1983
for this scenario i would assume that the user has decided not to click, so nothing should happen usually
14:13
panintended
I have a prototype for T1185 but want to see what the expected behavior for the above is
14:14
BAndiT1983
but this is difficult as the highlighting is already on maybe
14:14
panintended
<BAndiT1983 "for this scenario i would assume"> so no BUTTON_DOWN or BUTTON_UP?
14:15
BAndiT1983
i would say yes, but how could button_down be avoided then?
14:16
BAndiT1983
maybe just check if release and fire the event, there were discussions about clicking outside in imgui issues somewhere
14:19
panintended
so, to give a concrete example:
14:19
panintended
upon and a click-and-release of Button 1 (both click and release within the Button 1 area), a Button::BUTTON_1_DOWN will be sent, directly followed by Button::BUTTON_1_UP
14:23
BAndiT1983
yep, that's the usual scenario
14:24
BAndiT1983
as you can't avoid click on the real board it should always fire button_up, even if released the button
14:24
BAndiT1983
sorry, i mean even if the mouse left button area
14:28
panintended
hmm, now I'm confused :)
14:28
panintended
you mentioned before: "for this scenario i would assume that the user has decided not to click, so nothing should happen usually
14:28
panintended
"
14:31
BAndiT1983
yes, this is what i would expect in normal application UI, but then i've remembered real board and there you can't undo the "click"
14:31
panintended
ah, ok then
14:32
panintended
yes, that's what my thought was too and that's why I thought to ask first :)
14:34
BAndiT1983
would always implement how the real board is behaving, it's still a simulator of some sort
14:34
BAndiT1983
what is the approach in imgui for button release events?
14:35
panintended
<BAndiT1983 "would always implement how the r"> yes, my thought exactly
14:35
panintended
so to wrap it up:
14:35
panintended
1. click and release both within the button area -> BUTTON_X_DOWN and then BUTTON_X_UP in the next itreation
14:35
panintended
2. click within button area, release outside button area -> no BUTTON_X_* events sent
14:37
BAndiT1983
second case wouldn't be possible, as the highlighting should be tested while holding the button down, just fire button up always
14:42
panintended
but if we send a BUTTON_UP, then on the firmware side this will be interpreted as an actual button press (and e.g. enter the corresponding menu etc. - this is how it's currently done in RenderUI). do you mean BUTTON_DOWN maybe?
14:43
BAndiT1983
no, i mean button up, what i see in the firmware usually, which is still done manually, but will be turned into event handlers and other nice stuff, is the button down which highlights the button in the UI and the action executes at button up
14:51
panintended
ok, I'll share a GIF with the new behavior based on the above to get some feedback
15:03
olalomar
joined the channel
15:04
olalomar
left the channel
15:05
olalomar
joined the channel
15:16
olalomar24
joined the channel
15:23
olalomar
left the channel
15:33
olalomar24
left the channel
15:35
mumptai
joined the channel
16:04
lexano
left the channel
16:17
lexano
joined the channel
17:21
panintended
BAndiT1983: I've posted a GIF of the updated behavior, can you have a look please? https://lab.apertus.org/T1185
17:21
panintended
thanks
17:26
BAndiT1983
panintended: looks good
17:27
BAndiT1983
please try it with button 5 or 6, to see if highlighting is working while you press the button
17:29
panintended
OK will do. Why would 5 or 6 behave differently though?
17:30
BAndiT1983
as the buttons are switched manually for now with switch/case and not event handlers
17:31
BAndiT1983
but this should change of course, but it probably requires also the split of button IDs and events like button down and up, currently there are just different IDs for that, which is limiting
17:32
panintended
just to make sure: what you're referring to is still within the context of the visualizer, right? Not the firmware
17:34
BAndiT1983
nope, i'm talking about the firmware
17:35
BAndiT1983
check MainPage, there you will see the manual intervention, which should be removed by handlers and automatic highlighting etc.
17:35
BAndiT1983
that's why not all buttons are behaving the same
17:35
BAndiT1983
https://github.com/apertus-open-source-cinema/AXIOM-Remote/blob/f5109c4a7b1947ed2c7e4ecfe4e3ea33f08f0539/Firmware/UI/Screens/MainPage.cpp#L90
17:36
BAndiT1983
*by using handlers
17:37
panintended
ah, so what you mean is that visualizer button higlighting will be triggered all the way back from the firmware? (e.g. with callbacks)
17:38
BAndiT1983
as visualizer is only sending the signals which also real buttons trigger, the firmware should behave like it's doing on the hardware
17:39
panintended
understood :)
17:40
panintended
Off for now, I join you people later tonight
17:40
BAndiT1983
have fun, see you
17:48
BAndiT1983
changed nick to: BAndiT1983|away
19:15
BAndiT1983|away
changed nick to: BAndiT1983
20:20
lexano
left the channel
20:41
comradekingu
left the channel
20:46
comradekingu
joined the channel
21:31
panintended
BAndiT1983:
21:32
BAndiT1983
hi panintended
21:32
panintended
please have a look at my fork, and pull to test if needed. I'm waiting for your feedback to iterate on the solution
21:33
BAndiT1983
ok, will do quickly, have missed somehow that you wanted a review there
21:34
panintended
FYI (1) the knob button is not included in this (2) Ah, and also if you notice unexpected behavior (e.g. when selecting the 'Exit' option in the MENU) this is not linked to the changes. I reverted and the issue exists even without them
21:34
panintended
No worries, I just pushed now actually :)
21:38
BAndiT1983
you mean the crash on button 11?
21:39
BAndiT1983
we have 2xbutton 10, strange, probably a typo
21:40
panintended
I remember I had a few crashes, but also when going into the menu (button 4) and then selecting the "exit" option I got the LCD brightness dialog instead of returning to the main screen
21:40
BAndiT1983
it crashed for me in the settings menu, when confirming exit with button 11
21:41
BAndiT1983
haven't looked at functionality lately, as i'm occupied with a little bit different development right now
21:41
BAndiT1983
have you checked if the highlighting works when button 6 is held down?
21:41
panintended
oops, the double 10 buttons is my bad
21:43
BAndiT1983
just a small thing, would have been detected probably while checking PR
21:43
BAndiT1983
but at the moment it's not behaving like the board, when mouse pressed then "button down should be sent"
21:43
panintended
aaah, I just now realized what you meant earlier! You mean highlighting on the display
21:43
BAndiT1983
yep
21:44
panintended
ok, cool. I'll continue on that tomorrow then
21:44
BAndiT1983
great, we can chat about it tomorrow, have to leave for today, work awaits in the morning
21:44
panintended
that's why something felt out of place :)
21:44
panintended
yeah, likewise
21:44
BAndiT1983
step by step you will get the things, we also have remote connection to real board
21:44
panintended
good night everyone
21:45
BAndiT1983
maybe you are also interested to help to get framebuffer transfer from real one to the visualizer and button clicks to real one#
21:45
BAndiT1983
off for today, good night
21:45
BAndiT1983
changed nick to: BAndiT1983|away