luukvbaal

> Nice to have: option to have the same nnn buffer in all tabs. In nnn.nvim this was default, https://github.com/luukvbaal/nnn.nvim/commit/d4f5763da0dc4852b2d7c9449600383adf5f891d adds the option.

Didn't run into this when I tested it before but perhaps there has been a regression.

I think I was missing the `w` in `:wqa` when I tested this earlier. This seems to be a general issue with terminal buffers in neovim: https://github.com/neovim/neovim/issues/14061 https://github.com/neovim/neovim/issues/13549. Not sure...

Looks like Bram has taken it upon himself to start [implementing](https://github.com/vim/vim/commit/f6196f424474e2a9c160f2a995fc2691f82b58f9) this feature for vim as well. I was wondering if you are still up for porting/updating this PR if/when...

> Is there a way to add a condition, that something changed in the window layout, cursor position, etc? Actually, the current mechanism to only invalidate botline (w_valid flags VALID_BOTLINE...

> Just above the cursor is moved, thus removing VALID_CROW makes a lot of sense. Here we are actually just changing it temporarily(to scroll to bottom) and restoring it. So...

Latest commit lets the validate_cursor() call in win_new_height() handle the w_valid flags, except for VALID_TOPLINE because we update the cursor position in win_fix_cursor instead. I think this patch now makes...

Scratch that, I thought I was doing a bitwise OR. Looking into it once more...

This is the best I could come up with. It fixes the test case and I don't see any issues but I can't say I'm confident that w_valid flags are...

> would it be possible to define a variable or sth so the cursorline number can be treated special? Not sure it's necessary, can use `v:relnum == 0` for this.