UADetector is a library to identify over 190 different desktop and mobile browsers and 130 other User-Agents like feed readers, email clients and multimedia players. In addition, even more than 400 robots like BingBot, Googlebot or Yahoo Bot can be identified.
What's the main reason for string matching robots instead of using regex as is currently done for browser, os, device detection? The performance is likely better, but it seems that keeping the uas database up to date with the latest versions/variations of robot user agent strings may make a regex version an option worth considering.
There is little real robots, robots are a lot of false (ie false google bots). Therefore, it does not use the regex.
I do not say better, than say an untruth.
What's the main reason for string matching robots instead of using regex as is currently done for browser, os, device detection? The performance is likely better, but it seems that keeping the uas database up to date with the latest versions/variations of robot user agent strings may make a regex version an option worth considering.