Hi,

 

For my bachelor thesis i have a rather large dataset which i have chosen to arrange and "clean" in filemaker, before further analysis.

 

The data origins from a forum, with 36.000 topic replies, including the topic starter itself, in the same form as the replies.

 

I've found out how to separate topic starters from the others, via a dummy variable, returning 0 if topic text has RE: as the first 3 chars, and 1 if not.

However, as my data seem to have been damaged, I have a lot of empty topic fields, but with the original URL, looking like this:

http://www.lydmaskinen.dk/viewtopic.php?f=1&t=63898&p=553224#p553224

The first f=1 is the ID of the subforum the entire dataset is taken from, and the &t=63898 is ID of the topic. The last two must be page#, but i am actually not sure of that.

 

However, is it possible to make a calculation, using one of the text functions in Filemaker, to get the value of the &t=63898 ?

First off, i would have used Left / Right functions, but the number of digits varies from 4 to 6, which makes it very complicated...

 

Is there some sort of search function that returns everything between "&t=" and "&p=" or in some of the cases nothing at the end..

 

 

Any help will be much appreciated!

Try =

Let ( [
text = OriginalURL & "&" ;
pos = Position ( text ; "&t=" ; 1 ; 1 ) ;
start = pos + 3 ;
end = Position ( text ; "&" ; start ; 1 )
] ;
Case ( pos ; Middle ( text ; start ; end - start ) )
)

This extracts the text between "&t=" (if it exists) and the subsequent "&", if any (otherwise until the end of the OriginalURL text).

Thank you so much!   It worked, perfectly.