finding only whole word in grep

[bcdixit]

How do I find only specific word in a file using grep.
for eg.
I want to find the word 'PRODUCT' (case insenstive)

I used the -i switch to ignore case and -w to match the whole word 'product'.

But I find matches that are
product
PRODUCT
PRODUCT_..

I think the '_'(underscore character) is considered as a word character. How do I make grep to igonre the underscore character.

Please Help.
Thanks
bcdixit

 

[PHV]

A crude workaround:
tr '_' 'A' < /path/to/file | grep -i -w product

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[bcdixit]

thanks..but i am actually using grep in a find command. i.e. i am looking for mutiple files in a directory with that particular keyword

 

[PHV]

Any chance you could post your actual code ?

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[bcdixit]

find . -name "*" -exec grep -qwi "product" '{}' \; -print>result.txt

basically, i want the list of file names in a file result.txt which have the keyword only "product".

some of the files have "product_" or "_product" etc etc

 

[PHV]

find . -type f | while read f; do tr '_' 'A' < $f | (grep -qwi product && echo $f); done > result.txt

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[bcdixit]

can you explain this code for me before I go and try it out .
thanks

 

[Annihilannic]

Strange, none of my greps seem to match "product_" with a grep -w.

Annihilannic.