Extracts from a string, @mentions, #tags and/or any other [utf8 symbol][keyword]. The approach uses a tokenizer instead of the common regex, beacuse of this, it has better performance and works well for large amout of text.
go get github.com/jmsegrev/notations