Browse Happy API detection bugs — at Initial Version
|Reported by:||nacin||Owned by:|
A few bugs we've noticed with the Browse Happy API so far:
- Mobile versions need to be labeled as such. Specifically, Mobile Safari, Opera Mini, and the Galaxy tablet, I believe. This should mean that the browser doesn't match with our list of five.
- For IE, we need to be looking at the Trident version numbers. Problem is, IE9 might be using a compatibility UA, which means it would get identified as IE7. Trident/5.0 is IE9, Trident/4.0 is using IE8.
I think the architecture of the API code needs some love. It served its purpose, but it's a bit inflexible for expansion. (See the Kindle hack and such.)
I'm going to open source this code onto http://code.trac.wordpress.org. It would be fantastic if some people can contribute to it and make it the best possible.