I have a User struct with ID and LoginName fields and I want this struct to be accessible by either of these fields with single call to the DB. I know BoltDB is not supposed to handle arbitrary field indexing etc. (unlike SQL) but this case is a little different as I happen to know in advance the additional field to b used as index.
So is there some kind of secondary key or multiple key indexing? or maybe some strategy that I fail to see? If not then I'll just implement it with two calls, I just prefer a "cleaner" solution...
Thanks!
No, it's not there. BoltDB is a lot like Go. Clean and simple. And building a layer on top is easy. BoltDB even allows update transactions to be trivially implemented so two more more buckets can be updated, or not, atomically. So creating an update transaction that keeps two or more buckets in sync is easy. But it sounds like you know that and just wanted to check that you aren't missing something.