Question

I have a OPML file that I would like to parse the links and names in order to create a HTML formatted list.

<outline text="Wired Features" type="rss" xmlUrl="http://downloads.wired.com/podcasts/xml/features.xml?_kip_ipx=1854665749-1310493405" htmlUrl="http://www.wired.com" />
<outline text="ArcSight Podcasts" type="rss" xmlUrl="http://www.arcsight.com/podcasts/itunes/" htmlUrl="http://www.arcsight.com" />

Using SED or something similar I would like the print the items in a respective HTML output, i.e

<a href="http://downloads.wired.com/podcasts/xml/features.xml?_kip_ipx=1854665749-1310493405" title="http://www.wired.com">Wired Features</a>

Answer 1

perl -nle'
  ($text)  = /text="(.*?)"/   ;
  ($url)   = /xmlUrl=(".*?")/ ;
  ($title) = /htmlUrl=(".*?")/;
  /./ and printf "<a href=%s title=%s>%s</a>\n",
     $url, $title, $text; 
  ' infile

Assuming there are no embedded newlines in the interested parts.

With XMLgawk:

xgawk -lxml 'XMLSTARTELEM  {
  printf "<a href=%s title=>%s>%s</a>\n",
    q XMLATTR["xmlUrl"] q, q XMLATTR["htmlUrl"] q, XMLATTR["text"]
  }' q=\" infile

Edit: the Perl solution could be rewritten with a single regex:

perl -nle'
  /text="(.*?)".*xmlUrl=(".*?").*htmlUrl=(".*?")/
    and printf "<a href=%s title=%s>%s</a>\n",
     $2, $3, $1; 
  ' infile