LAE

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: ahines

So I am relatively new to regex, but I believe I am heading in the right direction. I have a column of data where there is many different types of strings. The vast majority I do not care about. I want to extract all the rows where that column (Text) contains the string 'CASH DEP O/S'. After that string there are 2 spaces and then a MM/DD date. I've been reading and trying different combinations for awhile, but I keep getting the same error when I try to to run:

brain.node.JRegex.cpp.147

I've ran a filter and tried removing all the '/', replacing them with spaces to make the pattern easier, but it's a no-go. Any regex experts have any ideas, or a good source of information on syntax for spaces, exact matches AND special characters (/)?

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: stonysmith

Is there a reason why you can't just use this?

emit * where myText.strfind("CASH DEP O/S") > -1

=====
The regex parser used in LAE is based upon a package named pcre. There are full docs at http://pcre.org

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: ahines

Stonysmith- that actually worked perfectly. I didn't think to use something quite so simple, as I had been on a "learning kick" to force myself into regex. Thanks for your help...Again!

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: ahines

Now that I think about it though- how would I incorporate that into a split filter? I would like the remaining fields in their own specific document for later use.

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: awilliams1024

You can set the 'PredicateExp' property of a Split node to myText.strFind("CASH DEP O/S") > -1

Note that Lavastorm operators are case sensitive so there should be a capital 'F' in strFind()

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: ahines

That worked great. Thanks, Guys- project complete!