1

I have a chunck of text in a file:

<tr bgcolor="#F9F9F9">
     <td align="left">8/7/2012 11:23:42 AM</td>
     <td align="left"><em>Here is the text I want to parse out</em></td>
     <td class="ra">9.00</td>
     <td class="ra">297.00</td>
     <td class="ra">0.00</td>
     <td class="ra">0.00</td>
     <td class="ra">$0.00</td>
     <td class="ra">$0.50</td>
     <td class="ra"></td>
 </tr>

using grep I would like to end up with the result being

Here is the text I want to parse out

Working on the code now I have 

cat file.txt | grep -m 1 -oP '<em>[^</em>]*'

but that does not work... thanks for your help!

Greg Alexander
  • 1,192
  • 3
  • 12
  • 25
  • 2
    `cat file.txt | grep ...` can be simplified to `grep ... file.txt`. – Asaph Aug 07 '12 at 17:13
  • 2
    Do note that while what you want to do is possible (as demonstrated in answers below), [regex is generally not the right tool to parse XML](http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags/1732454#1732454). For more robust solutions, use tools such as [xmlstarlet](http://xmlstar.sourceforge.net/) or a language that gives you access to a proper XML parser. – Shawn Chin Aug 07 '12 at 17:20

1 Answers1

4

A correct regex would be (?<=<em>).*?(?=</em>).

So, try: 

grep -m 1 -oP '(?<=<em>).*?(?=</em>)' file.txt
Lev Levitsky
  • 63,701
  • 20
  • 147
  • 175
  • That gives me this Here is the text I want to parse out 9.00 297.00 0.00 0.00 $0.00 $0.50 – Greg Alexander Aug 07 '12 at 18:13
  • OK, so what it is doing is going to the last which is in another text block below it... should have mentioned that, so I need the end to be the first occurrence of ... make sense? – Greg Alexander Aug 07 '12 at 18:19
  • @GregAlexander That could happen if the XML was all in one line, rather than nicely formatted as you show. Try to add a `?` after `*` as I did in the edit. – Lev Levitsky Aug 07 '12 at 18:22