Hi, My XML file has a content of

bookmark filepath="66286_toc.pdf" page="1" title="Table of Contents" type="GoToR" view="XYZ" view-left="-19" view-top="847" /

I want to extract only 66286_toc.pdf
I am not sure how to write a regex for this. Please help.

This would depend on exactly the structure of your data but you would probably want to extract numbers followed by an underscore followed by a letters then a dot then "pdf".

You may find this regular expression help file useful in formulating regular expressions:
http://geekswithblogs.net...s/235.aspx

Hi

Thanks for your quick reply. I read that but still cannot make it work. Maybe It's just because I don't understand it. Basically, I want to extract all pdf filenames between the bookmark tags in my XML files. Below are what I tried to do.

The first keyword is bookmark filepath=".
The second keyword is " page.

I tried ^{bookmark filepath=}*{" page}$

but it didn't work. Any thought?

Try this pattern

filepath="\d*_\w*.pdf"

Hi,

Did this work? I don't want to hijack the thread or anything but as it's so relevant I want to do something very similar but extracting everything in between the words option value and option.

This is to extract shopping cart info out of a biggish web site so it will need to search through all the sub-folders of the main site folder on my hard drive.

If anyone could help me I would very grateful, I'm a total noob at this coding stuff.

Thanks,

Mark

It worked in my test case Mark, I can't really say more than that!

Is it possible to create a pattern where the text in the web page looks like this.


Broker Name : A J Banford
Company Name : The Radford Company
Address:
12308 Ocean Gateway #5

ocean City, Maryland 21842
800-471-1108
530-546-7963 (fax)





Brokers By Name


I have thousands of web pages that all have the same formatted text in the middle. I need the program to copy from "Broker Name:" until "(fax)" including this text and put it into a text file, then skip a line or two, add the text from the next web page.

The result will a text file with all the names and addresses in it. I have another program that will parse the name and address and process them.

qpcth

I'm sure it would be, we do have a custom rule service if you're interested:
http://www.iconico.com/Da...mRule.aspx