Extracting from a string with various lengths of character on each side

[Dogo23]

Hi all,

I posted a similar question before only to realize the need was different that what I originally thought. I have a string with an ID that I need to extract. There are varying lengths of characters on either side of this ID making it difficult for me to specify the location to work from. I have noticed that there are always seven underscores before the the ID is found, but I'm not sure if that helps at all with locating the ID. Not sure if there's some sort of SUBSTR function that could be used for this in proc sql.

Here's an example of the data from my table named "tracking" with the ID that I need in bold: 

hraccact_email_acquisition_NA_ENG_20170222_gpsubscrip_HIL17TRANSS10EN1T33_A-509469751O

hrmessage_email_announcenews_NA_ENG_20170302_hrsubscrip_H0042100144510381EN_A-533270247X

hresum_email_HSUM0117_AS_ENG_20170124_stspecial_H0040055123110381EN_A-532246836EN?stop_mobi=yes

Thanks for any help!

[Tom · Accepted solution]

Did you try the SCAN() function?

scan(string,8,'_')