detecting substrings within strings

[SotonStu]

This is an awk script i have written, attempting to isolate all links in an html page and print out the frequency of each link:

#!/bin/gawk -f
# Print list of word frequencies

BEGIN{FS="\""}
NR == 1 { printf("%s\n%s", (NR==1) ? "" : &quot", FILENAME)}
/http/{
printf "\n"
for (i = 1; i <= NF; i++)
freq[$i]++
}

END {
for (word in freq)
printf &quot;%s\t%d\n&quot;, word, freq[word]
}

This gets the following output on a simple html file containing two links to google:

http://www.google.com

2
<A HREF = 1
from the word <A HREF = 1
>me!</A> 2

How can i get it to JUST add the links to the associative array and not all of the tags surroundng it.

Thanks again!

 

[SotonStu]

Sorry, my last thread has just been replied to which kinda makes my line of thinking here obsolete. looks liek i was barking up the wrong tree

 

[siidka]

Write an awk command file to do the following :

For each line of input, check that the number of brackets&quot;(&quot;and&quot&quot; in the line is balanced. print something out for each line as it's processed. At the end of input, print out the number of lines that had brackets in them, and the number where the number of brackets is unbalanced.
You just need to check the total numbers of &quot;(&quot;and&quot&quot; are correct, not that the ordering is feasible for balanced brackets. E.g. a+(5*2))( can be considered balanced