grep second instance of same string

Shell Programming and Scripting
1

Hi all, i am new to unix scripting in ksh or any shell for that matter. I have downloaded a xml file from a website and saved on my local harddrive. inside the xml, the same tag is listed multiple times.

<title>Tonight</title>
<title>Thursday</title>
<title>Friday</title>
<title>Saturday</title>

i was able to find the first occurrence of the line with:

ls file.xml | while read line; do
    fileTitle=`grep -m 1 "<title>" "$line" | sed s/".*\<title\>"//g | sed s/"\<\/title\>.*"//g`

Which my end result was to output the string "Tonight" all by itself.

However, i am choking on trying to find the 2nd occurrence of <title> and 3rd and so on, so i can print each day of the week as its own variable :mad: :wall:

I am in dire help and cannot figure this out, any help will be greatly appreciated.

2

How about this:

SEQ=3
awk -vN=$SEQ -F'[<>]' '$2=="title" && !--N { print $3 ; exit }' infile.xml
3

that didn't work :frowning:

here is the extent of my script so far

#!/bin/sh

basefolder=/s/batchtmp/weather_tmp

cd $basefolder

ls file.xml | while read line; do
fileTitle=`grep -m 1 "<title>" "$line" | sed s/".*\<title\>"//g | sed s/"\<\/title\>.*"//g`
SEQ=3
fileTitle2-`awk -vN=$SEQ -F'[<>]' '$2=="title" && !--N { print $3 ; exit }' file.xml`

echo $fileTitle
echo $fileTitle2

done

This is what i get from the terminal

computer:~ me$ . test.sh 
awk: invalid -v option

Tonight
4

You may need to use nawk if your on Solaris:

ls file.xml | while read line; do
fileTitle=`grep -m 1 "<title>" "$line" | sed s/".*\<title\>"//g | sed s/"\<\/title\>.*"//g`
SEQ=3
fileTitle2=`nawk -vN=$SEQ -F'[<>]' '$2=="title" && !--N { print $3 ; exit }' $line`
 
echo $fileTitle
echo $fileTitle2
 
done
5

Im on Mac OS X

-bash: nawk: command not found

i guess it doesn't like nawk, ugh

6

OK, the awk on Mac OS seems a little strange, perhaps best to avoid using -v at all try this:

ls file.xml | while read line; do
fileTitle=`grep -m 1 "<title>" "$line" | sed s/".*\<title\>"//g | sed s/"\<\/title\>.*"//g`
SEQ=3
fileTitle2=`awk -F'[<>]' '$2=="title" && !--N { print $3 ; exit }' N=$SEQ $line`
 
echo $fileTitle
echo $fileTitle2
 
done
7

That worked! Thank you very much!