Working with Whitespace

Any thought-leadering piece, any conference talk, any tweet from an outspoken nerd; if there’s one common theme — one opinion for which there is no counterargument — it’s that there’s no excuse for careless, messy whitespace in a codebase. Here’s how I handle whitespace with my weapon of choice, Vim.

Always show whitespace

Here are a couple of lines that should be in everyone’s .vimrc. The first displays invisible characters, and the second defines which symbols should be used to represent the different kinds of invisible characters. I mostly borrowed the second line from Steve Losh’s vimrc.

set list
set lcs=tab:▸\ ,extends:❯,precedes:❮,nbsp:.,trail:·,eol:¬

Cleaning trailing whitespace

This is another gem I found in Steve Losh’s vimrc. I don’t care to learn exactly how it works just yet. At the moment, I have it mapped to <leader>w, but it might be a good idea to automatically run the command anytime you write to a file. I think it really depends on how you feel this will affect your version control workflow.

nnoremap <leader>w mz:%s/\s\+$//<cr>:let @/=''<cr>`z

Normalising blank lines

Unless you’re drawing ASCII art, I don’t think there’s ever a good reason to have contiguous blank lines. Any group of rules or logic in a codebase should be separated by one single blank line, not more. Before I took the nosedive into Vim and Unix, I would clean these blank lines manually or force myself to ignore the disorder. Neither of those are any fun.

Vim provides us with filters, which pipe the contents of a buffer to some external Unix command and read the results of the command back into the buffer, replacing the original text. If you pipe a file to cat -s, it gives it back to you with the blank lines normalised. To do this straight from Vim:

:%!cat -s

The percent symbol references the entire buffer (which means filters also work with ranges), and the bang tells Vim to drop to the shell.