rex question

sourceuno · Post by **sourceuno** » Fri May 25, 2001 4:33 pm

How can I use rex to retrieve any characters between these 2 strings, "delimeter" and "http://www.somesite.com"? Do I need to replace the '.' literals in the second string with '/.'?

Post by **mark** » Fri May 25, 2001 5:03 pm

It's not clear what you're really trying to do, but this might do what you asked:

>>delimiter\P=!http://www\.somesite\.com+\Fhttp://www\.somesite\.com

sourceuno · Post by **sourceuno** » Fri May 25, 2001 5:17 pm

Thanks, that almost did the job. Rex is now returning a string that includes "delimeter". How can I return anything between the last "delimeter" and "http://www.somesite.com"?

Post by **mark** » Fri May 25, 2001 5:23 pm

search in the other direction:
delimiter\P=!delimiter+>>\Fhttp://www\.somesite\.com

I don't know what your big picture is, but if you're extracting data from a delimited "record" you might want to look at <split> or <timport>.

sourceuno · Post by **sourceuno** » Fri May 25, 2001 7:15 pm

What I'm actually trying to do is retrieve the column names from an html table which are in the first row of the table. Can this be done using timport?

Post by **mark** » Sun May 27, 2001 10:53 pm

Probably. You would probably use "recexpr" and supply a single rex expression to match the entire "row". Then you can pickout the data from the various columns. See the timport manual for how to use recexpr.