vterm overrides existing emacs cursor functions (C-n, C-p, C-a, C-e, etc.)
I like the speed of vterm but I'm unable to navigate the terminal like I can in term or shell-mode or e-shell. Is this a side effect of using the terminal emulator or am I missing something?
I think vterm-copy-mode (C-c C-t by default)may be what you are looking for, it may not be exactly how you would like to use vterm, but it does allow you to navigate and copy the portion of the buffer.
And of course, some better integration and adjustments could be made potentially to make vterm more emacs friendly.
At the moment, vterm behaves more or less like a real terminal. For most keys, vterm will just send them to the process that is currently running. So, C-a may be beginning-of-the-line in a shell, or the prefix key in a screen session.
If you enter vterm-copy-mode, the buffer will become a normal Emacs buffer (more less in fundamental-mode). You can use your navigation keys here. When you press RET, the region will be copied and you'll be brought back to a working terminal session. This is so that you can navigate, copy, and go back to a functioning command line.
Would prefer to have vanilla emacs behavior in vterm by default. As well as not having to switch between modes. (as they are context/state to keep in mind, and context-less seems the Emacs way)
Perhaps having vterm's 'special' keybindings under a prefix, such as M, would be a comfy option. E.g., M-C-p to send C-p via vterm. So the original C-p behaves vanilla.
Just an idea. Have little knowledge of emacs-vterm internals, maybe it makes no sense.