- Performant and efficient
- Read only the necessary parts of the file
- Stop as soon as all desired tags are found
- Clean and healthy code base
- Reasonable abstraction: Don't let the heavy stuff (find / parse tags) get into the way of simpler stuff
- Extendable
- Fail-safe
- Get as much tags as possible - don't fail the whole process if a single tag cannot be parsed
- Simple usage
- Sensible default options