On Mon, Mar 22, 2004 at 02:53:50PM +0000, Tony S. Sykes wrote: > Help, I have my web limited by my company now, so I can't google till tomorrow. I > need help with parsing an xml file. I need to cut out a fax number from the file. > Easy enough I though just grep and cut the line, but the whole file is one line. I > know this is a simple command but I can't work it out on my own, and with out google > (strange how easy you get attached to things) I am stuck. Please help. > > Thanks, > > Tony.
Tony, Not sure if this'll help, but here's what I used to get all links to begin on a new line. sed s/'<a'/'\n<a'/g index.html | sed s/'\/a>'/'\/a>\n'/g > processed The "\n" gives you a new line, so I'm saying every time you see "<a" replace it with "\n<a" (starts on new line) and everytime you see "/a>" replace it with "/a>\n" So, assuming the xml fax line was something like <fax>123-456-7890</fax> you could do sed s/'<fax>'/'\n<fax>'/g file_with_fax_number.xml | sed s/'<\/fax>'/'<\/fax>\n'/g > some_output_file Then, you could grep some_output_file for "fax" or of course get fancy and cut a certain number of characters to pull out just the fax number. HTH, Todd
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com ____________________________________________________ Join the Club : http://www.mandrakeclub.com
