Bash scripting help needed

[vehemoth]

Two things, first; how do I set a variable from a previously escaped line
and second


#!/bin/bash
#RSS Setup
badlines=6
URI=http://quotes4all.net/rss/240010110/quotes.xml #URI of RSS Feed
firstline=description #Top line to output
extrachar="- " #characters at start of second line
secondline=title #second line to output
quotes=temp

#Work Start
wget -q -O - $URI | sed -e 's/[ \t]*//' |\
sed ':a;N;$!ba;s/<description>\n/<description>/g' | sed ':a;N;$!ba;s/\n<\/description>/<\/description>/g' |\
grep -A 1 '<title>' |\
sed -e :a -e 's/<[^>]*>//g;/</N' |\
head -n $(($badlines+2)) |\
tail -n $((1))

Can I replace this line


sed ':a;N;$!ba;s/<description>\n/<description>/g' | sed ':a;N;$!ba;s/\n<\/description>/<\/description>/g' |\

with something that works with more than just the description tag (it finds the tag and end tag and deletes the new line before or after if there is no contents on that line)
Thanks for your help :)

[mikejonesey]

previously escaped line... example?

[sisco311]

You can't realistically parse tag-based markup languages like HTML and XML using Bash or utilities such as grep, sed or cut. If you just want to dump/render HTML, see (links|links2|lynx|w3m) -dump, html2text, vilistextum. For parsing out pieces of data, see tidy+(xmlstarlet|xmlgawk|xpath|xml2), or learn xslt. See: http://www.codinghorror.com/blog/archives/001311.html

[Lars Noodén]

As mentioned, learn XSLT.

Or become familiar with XML parsers. CPAN's XML:: Parser (http://search.cpan.org/~msergeant/XML-Parser-2.36/Parser.pm) or HTML:: Parser (http://search.cpan.org/~gaas/HTML-Parser-3.68/Parser.pm) are modules to start with in Perl. Other languages have similar tools.