31

My Linux account in a remote machine comes with the following pre-defined .inputrc

"\e[1~": beginning-of-line
"\e[4~": end-of-line
“\e[5~”: history-search-backward
“\e[6~”: history-search-forward
"\e[3~": delete-char
"\e[2~": quoted-insert
"\e[5C": forward-word
"\e[5D": backward-word
"\e\e[C": forward-word
"\e\e[D": backward-word

set completion-ignore-case on
set bell-style visible
set expand-tilde on
set convert-meta off
set input-meta on
set output-meta on
set show-all-if-ambiguous on
set visible-stats on

I am reading the GNU bash documentation on the topic, but I haven't been able to find a section explaining what the character sequences \e[1~, \e[5C, \e\e[D, etc. mean.

I think I read somewhere that \e means the meta character (which I think is Alt in my keyboard), but don't understand what the other control characters [, ~, 5D etc. mean.

Anybody knows of a good reference on the topic?

Amelio Vazquez-Reina
  • 6,878
  • 21
  • 62
  • 83

1 Answers1

28

These are ANSI control-code escape sequences that are transmitted when various non alphanumeric keys are pressed on a "terminal" keyboard.

\e means the ASCII "ESCAPE" character (octal 033 hex 1B decimal 27). Which is part of a command sequence introduction (CSI).

Escape [ 2 ~ is a character sequence transmitted when you press the key labelled "Insert" on a VT220 (or later) terminal.

Many of these conventions have been adopted in software such as xterm and Linux consoles / shells - often extended in various, sometimes incompatible ways.

The use of these sequences in bash's READLINE function is described in the man pages for readline

       In  the  second  form,  "keyseq":function-name or macro, keyseq differs
       from keyname above in that strings denoting an entire key sequence  may
       be  specified  by  placing the sequence within double quotes.  Some GNU
       Emacs style key escapes can be used, as in the following example,  but
       the symbolic character names are not recognized.

          "\C-u": universal-argument
          "\C-x\C-r": re-read-init-file
          "\e[11~": "Function Key 1"

To get a list of which escape sequences correspond to which keyboard-keys, you can use a command such as infocmp -L -1 or infocmp -L -1 xterm

RedGrittyBrick
  • 81,981
  • 20
  • 135
  • 205
  • Excellent!! That's _exactly_ what I was looking for. Thanks @RedGrittyBrick! – Amelio Vazquez-Reina Apr 11 '11 at 15:41
  • +1. I was looking for `Ctrl` code in `inputrc` --> `ENTER` = `Ctrl` + `n`. – rendon Aug 17 '13 at 17:22
  • @user1133275: link updated but the command `man 3 readline` *on your own system* should give you the same information. The command `man 3 readline`also works in [WSL on WIndows 10](https://docs.microsoft.com/en-us/windows/wsl/install-win10). IF necessary you can ask a new question about the specific parts of `bind -P` output that are causing you difficulty. – RedGrittyBrick Mar 01 '19 at 09:42
  • I think this answer could use quite a bit more information. The output of neither `infocmp -L -1` or `infocmp -L -1 xterm` seems to contain most of the examples given in either the question or the answer. I just searched the output for both `\e[11~` and `\e[1~` neither of which were contained. – Kvothe Aug 24 '21 at 17:01
  • @Kvothe: try, for example, `infocmp -L -1 putty | grep key_f1` but I think the intent of my answer ten years ago was to explain the notation in general, not the specifics of every line in the question or man-page. – RedGrittyBrick Aug 24 '21 at 22:34
  • This is really helpful. The [man page](https://linux.die.net/man/3/readline) you mentioned hints that we can use a _keyname_ as well, such as `Control-u`. Do you know where we can find a comprehensive list of keyname identifiers? I'm particularly interested in a keyname for the up arrow key, instead of using the arcane `\e[A`. – Garret Wilson Apr 05 '23 at 15:20