Jean issues a challenge:

[ok, now count up the number of times i used the word "think" in this entry]

I respond:

%perl -p -e s/think/think\n/g; < jean-thinks.txt > jean-thinks2.txt

%grep -c think jean-thinks2.txt
14

4 Responses to “Replace, Grep”

1. stran says:

   5/6/2005 at 9:36 am

   %grep ‘think’ jean-thinks.txt | wc -l
   14

   BOOYAH!
   :^)

   Reply
2. Elliott Bäck says:

   5/6/2005 at 10:35 am

   Sorry stran, that doesn’t work. That will give you:

   %grep ‘think’ jean-thinks.txt | wc -l
   6

   See, grep counts lines, and there are only 6 lines with ‘think’ in them. You really need to split think onto seperate lines first, although the slashes got lost in the post…

   Reply
3. Xander says:

   5/8/2005 at 5:48 pm

   Why not use sed? Perl seems like an overkill. And no need for a temporary file if u pipe it.

   sed s/think/think\n/g jean-thinks.txt | grep think | wc -l

   Reply
4. Elliott Bäck says:

   5/8/2005 at 8:08 pm

   I used temporary files because for some reason piping this stuff on windows doesn’t work properly (i.e., like unix). If I were on linux, Xander, you got it

   Reply