Vim Tips Wiki

Delete words in a different way

1,612pages on
this wiki
Revision as of 06:27, July 13, 2012 by JohnBot (Talk | contribs)

Tip 1498 Printable Monobook Previous Next

created 2007 · complexity basic · author fomit · version 5.7

Often, you might want to delete a set of arguments within bash scripts in Unix environments as an admin.

So you might think dw (delete word) does the job, because it often does, actually. However, in Unix environments you might also interpret a word in a different way, e.g. a "/path/to/file" can be seen as just one word in terms of the number of arguments. In that case, "dw" will not delete "/path/to/file", instead it will interpret every slash "/" as a separate word, so you have to enter "dw" 6 times.

Instead, you can dynamically enter your own, currently needed delimiter without changing Vim's global behavior in vimrc by just typing "df " (d f space). That deletes from the cursor position to, and including, the next space.

You can easily adapt this style of deleting "words" by replacing space by your own character.

Maybe you want to delete a whole sentence, then you want to type "df.". There are many thinkable usages.


You can also use dW (capital W) to delete a WORD, which is a number of consecutive characters, such as "/usr/bin/perl". Also check out dt<char>, which deletes up to (not including) <char>.

If you need that "f " motion often, you could also map

:ono <Space> f<Space>

then delete with "d ", this is one key less.

When you use "df ", you simply tell Vim to delete all the chars until (and including) the next space character. This is no magic, and I think it needs to be clarified for novices. The command f<char> puts the cursor on the next occurrence of <char>. This is therefore a movement command. Any movement command, combined with the operator "d" instructs Vim to delete until the new cursor position.

Around Wikia's network

Random Wiki