If I understand the code correctly, the scoring algorithm is just based on a line number and it will prefer smaller numbers, i.e. it seems to assume that more precise matches are higher up in the browscap.ini file.
This is not always the case - here's an example from the latest 6022 version with the user agent;
Mozilla/5.0 (Linux; Android 6.0; Infinix HOT 4 Build/MRA58K) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/57.0.2987.132 Mobile Safari/537.36
The correct match (as it's returned from the browscap.org/ua-lookup page) would be:
mozilla/5.0 (*linux*android?6.0*) applewebkit* (*khtml*like*gecko*) chrome/57.*safari/*
...but bgo finds this instead (the platform version is not matched):
Mozilla/5.0 (*Linux*Android*) applewebkit* (*khtml*like*gecko*) Chrome/57.*Safari/*
It is also possible that I'm misunderstanding something, in which case I'm sorry for causing a fuss.
Hi!
If I understand the code correctly, the scoring algorithm is just based on a line number and it will prefer smaller numbers, i.e. it seems to assume that more precise matches are higher up in the browscap.ini file.
This is not always the case - here's an example from the latest 6022 version with the user agent;
Mozilla/5.0 (Linux; Android 6.0; Infinix HOT 4 Build/MRA58K) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/57.0.2987.132 Mobile Safari/537.36
The correct match (as it's returned from the browscap.org/ua-lookup page) would be:
mozilla/5.0 (*linux*android?6.0*) applewebkit* (*khtml*like*gecko*) chrome/57.*safari/*
...but bgo finds this instead (the platform version is not matched):
Mozilla/5.0 (*Linux*Android*) applewebkit* (*khtml*like*gecko*) Chrome/57.*Safari/*
It is also possible that I'm misunderstanding something, in which case I'm sorry for causing a fuss.
Cheers!