How do I apply a set of keystrokes to every occurrence of a word?

Question

If I want to delete every word, that is easy, because I would do something like this:

:s%/WORD//g

But what if I want to do xyz to every occurrence of "WORD"? For example, let's say I want to delete every line that contains "WORD". I can think of a couple ways to do this, but none of them seem ideal.

1)

/WORD

and then manually press ddnddnddnddn... until there are no more occurrences. But this is time consuming if there are many lines to do this on.

2)

qd/WORD<enter>ddq

and then guess how many times WORD occurs, and then, for example:

500@d 

But this will either A) miss a bunch of lines or B) delete of bunch of extra lines.

Is there a better way to do this?

Answer 1

You can use the :global and :normal commands for this. The :global command has the following syntax:

:global/{pattern}/{command}

This will run "command" on every occurrence of "pattern". You can then use the :normal command to run a series of keystrokes on each occurrence of a word using :global. Like so:

:g/WORD/normal {keystrokes}

If those keystrokes involve special keys like <Ctrl>, you can combine it with the :execute command like so:

:g/WORD/exe "normal {keystrokes}"

Do note that you must put a backslash before each occurrence of a special key.

For more on these commands, see :help :normal and :help :global.

Answer 2

Just for completeness, you can use :substitute, if you match - in your example (delete the whole line that has WORD) - the entire line containing WORD, e.g.:

:%s/^.*\<WORD\>.*\n

This has an implicit "replace match with nothing", i.e. it's the shorter form of:

:%s/^.*\<WORD\>.*\n//

Combined with \zs, \ze, \(-\) (grouping), and \= (in the replacement part) with submatch(), :substitute can be very powerful.

Answer 3

It seems you are looking for :global.

:g/WORD/d_