Does the mira command have a shell-like history mechanism for commands typed at the terminal? I'm constantly hitting up-arrow to correct a minor typo, but it doesn't work. I'm using Miranda 2.044 for Mac OS X.
Unfortunately it does not seem so.
You can use $$ as shorthand for the last expression evaluated however, and !! for the last shell command.
See: https://www.cs.kent.ac.uk/people/staff/dat/miranda/manual/5.html
Try this Python 2 oneliner, which redirects standard input to Mira through a shell:
python -c "import readline,subprocess as s;p=s.Popen('mira',shell=True,stdin=s.PIPE);[p.stdin.write(raw_input()+'\n') for _ in iter(int,1)]"
It also allows you to use backspace to correct typos.
Alternatively, try the rlwrap utility:
rlwrap mira
Related
I noticed when in vi mode in bash (i.e. the mode enabled with "set -o vi"), that some commands, such as "diw", that work in vim but not in vi, don't work on the bash command line. Is there an easy way to configure bash so that its keybindings will support vim commands? I would like to be able to enter vim commands on the command line without having to actually start the vim program, as described in this question.
The best way of doing this that I know of would be to use athame.
It can be a surprisingly powerful experience in some cases. I particularly like it for interacting with a repl.
Athame patches your shell to add full Vim support by routing your keystrokes through an actual Vim process. Athame can currently be used to patch readline (used by bash, gdb, python, etc) and/or zsh (which doesn't use readline).
Alternatively I find spacemacs with the eshell to be a reasonably functional if strange solution.
Teach vi-command-mode diw to any software that uses readline (such as bash) by adding this to your ~/.inputrc:
set keymap vi-command
"diw": "bde"
First, the "vi mode" you get with set -o vi is not vi itself. It's an incomplete approximation of vi's behavior built into readline, the command-line editing library used by bash under the hood.
Second, because it is incomplete there's no reason whatsoever to expect every vi command to work.
Third, no, there's no "vim mode" so even less reason to expect any vim commands to work.
Fourth, if you absolutely want to edit the current command-line with Vim-like commands, why don't you go all the way and… actually use Vim:
<C-x><C-e>
That said, $ man readline tells you everything you need to customize its behavior and add bindings.
You can use ble.sh. Which is a command line editor written in pure Bash which replaces the default GNU Readline. I like to call it "zsh for bash".
Appart from vim-style navigation it gives you:
Enhanced completion
Syntax highlighting
Other interesting features
After installation don't forget to add (recommended) to your .bashrc:
# at the start of your .bashrc file
[[ $- == *i* ]] && source /usr/share/blesh/ble.sh --noattach
...
#for vim-style mode
set -o vi
...
#at the end of your .bashrc file
[[ ${BLE_VERSION-} ]] && ble-attach
or you can just:
# at the start of your .bashrc file
source /usr/share/blesh/ble.sh
but this may not work as expected - read this.
I'm trying to use vi mode in bash. via the .inputrc (on OSX):
set editing-mode vi
In vi insert mode, the right arrow key moves the cursor to the right, but it stops on the last character in the line. If the cursor is past the end of the line, it moves the cursor to the left. So, in sum, the farthest right you can go is to the last character in the line.
$ cd /usr/locl/bin
# Move the cursor to the middle of the line, and fix something there
$ cd /usr/local/bin
# Now move the cursor back to the end, and write a character (/)
$ cd /usr/local/bi/n
As shown above this means you cannot edit the end of the line without going into command mode (and using 'a').
I found an article which seems to indicate the version of readline/bash might be the problem. However I used brew to upgrade bash (GNU bash, version 4.3.42), and even tried to install and link readline (6.3.8), as recommend by that site. But no luck. Its possible the upgrade was done incorrectly.
This means I'm looking for one of the following:
The proper way to upgrade bash and readline in OSX terminal
A way to check the versions of bash/readline that OSX is actually using
Another fix for this bug (somehow passing in the virtualedit=onemore option in the inputrc)
Indications that the latest versions of readline might have re-introduced the bug, and solutions.
Also: if theres a similar thing with editrc
Note: I'm looking to make readline in bash act near identical to the default (for others who periodically use my terminal), but allow me to use vim mode. This means I don't need workarounds, but fixes.
Answer here: https://unix.stackexchange.com/a/222506/198846
Apparently it's a bug in the bash version shipped with OSX (3.2), it's fixed in 4.3 according to that answer. Use bash --version to check your bash version.
To fix, update bash, e.g.
macports: sudo port install bash
brew: brew install bash
Once installed bash --version will show it's updated (assuming standard brew/macports paths at the start of your $PATH)
You then have to change your default login shell (in System Preferences->Users & Groups->right click your user->advanced options->login shell) to the path of the new bash:
(default) macports: /opt/local/bin/bash
(default) brew: /usr/local/bin/bash
Changing the default login shell step is required even if which bash shows the macports/brew one.
All credit to the answer linked above.
Can't you edit .bashrc itself instead to use vi mode? The command set -o vi does it in my case (you are in edit mode initially). Also remove the .inputrc edit as the two may actually interact nefariously afterwards.
I'm using W32 GNU Emacs 24.1.50.1 with Cygwin bash 3.2.51. I know that this combination is very troublesome (see EmacsWiki and Cygwin list). However the problem I want to solve now is not confined to Cygwin. I got the issue with MinGW as well. So, the problem is:
In Emacs start a shell via M-x shell provided you have set up the variables shell-file-name and explicit-bash-args appropriately (see e.g. here). Then issue a command which expects stdin input, e.g. cat. To cancel out of cat you can usually type Ctrl-D once and you return to the shell prompt. In W32 emacs when you do this (or even when you issue explicitly M-x comint-send-eof) the cat command gets exited and the bash shell quits as well.
I tried to increases bash's "resilience' by setting the environment variable IGNOREEOF to, say, 5, but then a single Ctrl-D results in the following output
$ Use "exit" to leave the shell.
$ Use "exit" to leave the shell.
$ Use "exit" to leave the shell.
$ Use "exit" to leave the shell.
$ Use "exit" to leave the shell.
$ exit
Process shell finished
and, of course, the bash has exited again. This problem only happens when the shell runs in W32 emacs - Cygwin bash through the Cygwin tty window and MinGW bash through its own MinTTY work fine.
Why is this happening in W32 Emacs and how can I just exit the running subcommand (e.g. cat) when using W32 Emacs without exiting the bash shell?
The notion of tty like you expect it to behave doesn't really exist in the W32 worl. If you want to see this Unixy behavior, you need not only to run a Cygwin program but that Cygwin program needs to talk to another Cygwin program (Emacs, terminal, you name it). IOW you'll only get that with the Cygwin version of Emacs.
The GNU Emacs FAQ for MS Windows notes that this issue exists for any sub-process in the shell buffer. So it affects the DOS command line as well (but of course you don't usually use Ctrl-D in DOS, so it's not so noticeable). No workaround is suggested in the FAQ, so I'm guessing that there's no easy fix.
Usually you can use Ctrl-Z as well as Ctrl-D. Take a look at this note:
http://ignaciopp.wordpress.com/2009/07/02/emacs-shell-workaround-for-killing-input-in-windows-version/
Just compiled MacVim with homebrew. Here's what it looks like when I :sh and then type ls:
http://cloud.jtmkrueger.com/image/2N0S0T3k3l1J
As you can see, it's just character codes.
UPDATE
I run oh-my-zsh
Tried installing the plugin named here:
http://vim.1045645.n5.nabble.com/ANSI-colors-td1219411.html
Didn't seem to help
ANOTHER UPDATE
Upon removing my zsh syntax highlighting plugin It seems to work ok. Is there a way to turn off zsh plugins when using oh-my-zsh only when it's a vim 'dumb terminal'?
When you do :sh in GVim or MacVim, you don't get a real terminal emulator.
It's "dumb" and there's no way to make it understand those escape sequences. You better get used to it or ask (with convincing arguments and a ready-made patch) on the vim-dev mailing list.
You might want to try the ConqueTerm plugin which does its best to interpret ANSI sequences, even inside MacVim.
Just for the reference, :h guioptions now support the following flag:
'!' External commands are executed in a terminal window. Without
this flag the MS-Windows GUI will open a console window to
execute the command. The Unix GUI will simulate a dumb
terminal to list the command output.
The terminal window will be positioned at the bottom, and grow
upwards as needed.
Set :set go+=!, run :sh, and be surprised :).
What you see is actually not just character codes, but your usual shell prompt which contains color codes. You can probably disable it by redefining PS1 or remove your modified definition in ~/.bashrc.
If you would like to use a color prompt on the command line, but not in MacVim you can fix this in ~/.bashrc by setting PS1 differently when inside vim (from here)
if [ $VIM ]; then
export PS1='\h:\w\$ '
fi
You could try the following, instead of ls, type command ls; it shouldn't show the escapes codes.
If it works you can simple create a new file in a folder in your path, say vls, with the following contents:
#!/bin/sh
command ls $#
after that chmod +x vls and again, if it is in your path, you should be able to use that from vim.
I just started using a Solaris 10 (Sparc) box where I telnet in and get confronted with a very unfriendly interface (compared to the standard bash shell I use in cygwin or linux) --- the arrow keys do not work as I expect them to. Being an NIS system, changing the shell is not as easy as using the "chsh" command. And setting the SHELL environment variable in ~/.login and ~/.profile is not working for me. So I'm thinking that I may need to write a script to determine if bash is running the script and starting bash if the answer is no. My first attempt, trying to invoke /bin/bash from ~/.profile seems to work but kind of doesn't feel right. Other suggestions? And how do I tell programmatically which shell is actually executing?
You can tell what shell is running with echo $0. For example:
$ echo $0
-bash
If you're changing shell you probably want to replace the current shell process rather than be a child of it, so use exec.
Also, you want to pass bash the -l flag so it acts as if it has been called as part of the login process.
So you'll want something like:
exec bash -l
You are probably running with ksh(1) on Solaris. You have several options, read the manpage for ksh and configure it or install another shell you're more familiar with like bash. I'd personnaly recommend zsh.