Using awk for displaying of row

gregarion · 01-24-2010, 10:02 AM

Code:

Database
rina lives:fatin:20:20:20
rina:fatin:20:20:20

i am having a small problem in extracting out the information from the database. For example, if i were to input a book titled rina into my search , i would only want it to display the row which has the title rina, like this

Code:

rina:fatin:20:20:20

but when i use a coding like this :

Code:

grep -i "$Title" fruit| awk -F":" '{ print $0}'`

its displays both titles and their information

Code:

rina lives:fatin:20:20:20
rina:fatin:20:20:20

for some reason, it will read both rows as both contains the word "rina". does anybody know how i can modify my statement to solve this problem?

pixellany · 01-24-2010, 10:15 AM

Your grep statement returns any line containing the value of "title"--in this case I assume it was "rina". So grep is performing as expected.

awk '{print $0}' seems to mean "print the whole line"

To get the line containing ONLY rina in the first field, I would use sed addressing---eg:

sed -n '/rina:/p' filename

gregarion · 01-24-2010, 10:27 AM

does sed recognize the -f":" also?

gregarion · 01-24-2010, 10:30 AM

hmmm, sed dosnt seem to read the FS command like awk right? so how do i go about solving the problem if all my rows are defined in that way?

pixellany · 01-24-2010, 10:35 AM

Don't try to apply AWK options to SED---read the man pages or one of the various tutorials.

In my SED example, the address includes the ":", thereby accomplishing the same purpose as specifying the field separator in AWK.

Really good tutorials here:
http://www.grymoire.com/Unix/

gregarion · 01-24-2010, 10:49 AM

okay, just to check, i tried using awk for this problem. so my coding was like this .

Code:

Title=
echo "read this:"
read Title
awk '/'$Title'/' fruit

but it also displays both strings. why does awk have the same problem also?

Disillusionist · 01-24-2010, 12:27 PM

Code:

awk -F":" ' $1 == "rina" ' fruit

pixellany · 01-24-2010, 12:37 PM

Quote:

Originally Posted by gregarion

okay, just to check, i tried using awk for this problem. so my coding was like this .

Code:

Title=
echo "read this:"
read Title
awk '/'$Title'/' fruit

but it also displays both strings. why does awk have the same problem also?

I believe you are missing a key concept. AWK displays both line because they both contain the keyword. That is why I used the ":" in my SED example.

With any regular expression (which includes keywords in plain english), the match is made if the pattern is found---it does not matter if there is also something else there.

Disillusionist · 01-24-2010, 03:40 PM

The important thing is to work out exactly what you are after.

Do you want to search for Titles containing the input?
Do you want to search for Titles exactly matching the input?
Do you want to search anywhere in the file for the input?

To search for Titles containing the input:

Code:

#!/bin/bash

Title=
read -p "Enter title: " Title

awk -F":" ' $1 ~ TITLE' TITLE="Title" fruit

To search for Titles exactly matching the input:

Code:

#!/bin/bash

Title=
read -p "Enter title: " Title

awk -F":" ' $1 == TITLE' TITLE="Title" fruit

To search anywhere in the file for the input:

Code:

#!/bin/bash

Title=
read -p "Enter title: " Title

awk -F":" ' $0 ~ TITLE' TITLE="Title" fruit

I don't believe line 3 is really required, but put it in as it was in your original code.