Wednesday, April 4, 2012

Re: vimgrep

Ben, Thanks for your prompt response.
Somehow I am getting 3 copies of your premature, and complete replies.

Yes. You're right \*\* was one of the approaches I used. Unfortunately * is treated as a metacharacter/special character/keyword by grep/vimgrep/c language and this where the difficulty is originating from.

I tried your suggestion, \*\@<!\*\*\*\@!, and I get a no pattern found reply while I stirring at a line that has **.

How would you do this in grep on cygwin command line?

Thanks again.


From: Ben Fritz <fritzophrenic@gmail.com>
To: vim_use@googlegroups.com
Cc: "vim@vim.org" <vim@vim.org>; hilal Adam <hilaldm@yahoo.com>
Sent: Wednesday, April 4, 2012 12:28 PM
Subject: Re: vimgrep

On Wednesday, April 4, 2012 12:22:19 PM UTC-5, Ben Fritz wrote:
> On Wednesday, April 4, 2012 12:12:28 PM UTC-5, hilal Adam wrote:
> > Sorry if not correct platform for this question.</div>
> > Need to use vimgrep/grep to find a particular string in c code.  I am trying to find all occurrences of '**' (pointer to pointer). But I get hundreds of lines of output for all comment lines which include a minimum of 2 '*'s.</div>
> > Any help is appreciated.</div>
> >
> > </div>
> > HA
> > </div></div></div>
>
> vimgrep, just like '/' searching in Vim, uses regular expressions. In a regular expression, * means "as many as possible of the preceding search atom" which normally means "as many as possible of the preceding character".
>
> You don'

Dammit, I accidentally hit some key combination which Google Groups used to post before I was done.

You don't give the exact search you used, so I was guessing here that you searched with ** as your search pattern, which will match zero or more '*' characters, but I realize now that this will match absolutely anything, because "as many as possible" means zero or more, so ** will match every line in every file.

Probably, then, you searched for \*\*, which will match 2 '*' characters in a row. However, it will match ANY two * characters, even if followed by another character which is not a * character.

You need to specify that you only want to match where the preceding character and the next character are not also *.

To do this, I'd use a pattern like:

\*\@<!\*\*\*\@!

or simpler, with "very not magic":

\V*\@<!***\@!

See :help /\@<! and :help /\@!

--
You received this message from the "vim_use" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php


No comments: