Hi Roman,
I've been shown your KWin work last week - thanks for picking it up!
The tested parts of input_method_v2 are the parts which text-input-v3 is using
[0]. There's an application [1] and a compositor [0] with examples [2,3] that
you can use for verifying the KWin implementation. I think
Hi,
what's the current state of this patch?
I have a KWin patch up for review to implement the text-input v3
protocol, so I would be interested in pushing forward the input_method
v2 here.
If there are already some test applications speaking both ends of the
pipeline I could try it out in KWin.
This protocol is based on v1, current text-input-v3, and wl_keyboard version 6.
The pieces passing data relevant to the application on the other side of the
compositor are a mirror copy of text-input-v3 events and requests.
Compared to input-method-v1:
- assumes that once preedit is displayed,
