Currently we have millions of documents stored in a collection that use an email field with index. We would like to efficiently search through the collection for either a specific email address, or a domain name. The docs indicate that a query should start with a carrot for the regex to be used efficiently. We would love to be able to do the 2 following searches:

/jdoe@example.com$/
/@example.com$/

Both could easily use the same index if they could be optimized.