viuer 0.11.0
When there are multiple input readers active (for example a async input reader for a TUI and then viuer), then viuer may not get the complete response and so go into a infinite loop, similar to #70.
In our project where this issue first came up, only viuer 0.11.0 with the kitty protocol is affected (tested on wezterm), due to viuer waiting for a response in the local case.
All-in-all, i dont know what viuer could do to properly fix this instead of doing workarounds.
This issue originally came up in tramhao/termusic#654, which has been introduced in #90 and i also asked for suggestions in crossterm-rs/crossterm#1039 due to the original project where this came up it being the place that "ate" the viuer response.
The only workarounds for viuer i can think of:
- allow configuring kitty's protocol path to force the use of
remote (due to that not waiting for a response); note that this would fix the print paths, but the issue would still occur if the "check support" functions run while another reader is active
- completely unload the crossterm EventStream in our downstream project (even then i am not sure if that would work, due to crossterm likely keeping the poller around)
viuer 0.11.0
When there are multiple input readers active (for example a async input reader for a TUI and then viuer), then viuer may not get the complete response and so go into a infinite loop, similar to #70.
In our project where this issue first came up, only viuer 0.11.0 with the kitty protocol is affected (tested on wezterm), due to viuer waiting for a response in the
localcase.All-in-all, i dont know what viuer could do to properly fix this instead of doing workarounds.
This issue originally came up in tramhao/termusic#654, which has been introduced in #90 and i also asked for suggestions in crossterm-rs/crossterm#1039 due to the original project where this came up it being the place that "ate" the viuer response.
The only workarounds for viuer i can think of:
remote(due to that not waiting for a response); note that this would fix theprintpaths, but the issue would still occur if the "check support" functions run while another reader is active