This a vim plugin that enables MATLAB-style cell mode execution for python scripts in vim, assuming an ipython interpreter running in screen (or tmux).
Blocks are delimited by ##. For example, say you have the following python script :
##
import numpy as np
print 'Hello' # (1)
np.zeros(3)
##
if True:
print 'Yay !' # (2)
print 'Foo' # (3)
##
If you put your cursor on the line marked with (1) and hit Ctrl-G, the 3 lines in the first cell will be sent to tmux. If you hit Ctrl-B, the same will happen but the cursor will move to the line after the ## (so you can chain Ctrl-B).
You can also select line(s) and hit Ctrl-C to send them to tmux. The plugin automatically deindent selected lines so that the first line has no indentation. So for example if you select the line marked (2) and (3), the print statements will be de-indented and sent to tmux and ipython will correctly run them.
On Linux, you need xclip if you want to use screen.
By default, the following mappings are enabled :
- C-c sends the currently selected lines to tmux
- C-g sends the current cell to tmux
- C-b sends the current cell to tmux, moving to the next one
You have to configure the target tmux/screen session/window/pane. By default, the following is used :
let g:tmux_sessionname='ipython'
let g:tmux_windowname='ipython'
let g:tmux_panenumber='0'
let g:screen_sessionname='ipython'
let g:screen_window='0'
This scripts relies on a temporary file to send text from vim to tmux. By default, tempname() is used to get a temporary file, but if you set g:cellmode_fname , this will be used instead.
To choose between tmux and screen, set g:cellmode_use_tmux=1 (or 0 if you want screen). Note that currently, CopyToScreen relies on OSX' pbcopy to set the paste buffer.
Note that if you want more advanced integration with IPython (using the new multi-client architecture), there is the vim-ipython project : https://github.com/ivanov/vim-ipython/
The main difference with vim-ipython is that this plugin simply emulate a paste as you would do it manually from vim to ipython. This allow to see the result of the execution directly in the ipython split whereas vim-ipython uses a separate vim buffer to show the results.