Martin Hoffmann

But of course! We should probably use different terms to distinguish this from serde. `DeserializeRdata` can probably just be `Deserialize` (or whatever name we choose) and be used for records...

They sort of do indirectly through the record data. The ‘header’ of a record (i.e., domain name, TTL, class, record type) also needs to parse tokens, so using the same...

They are just string tokens (or ‘words’ in the current scanner’s parlance) that are then converted to values by the `Scan` impl for the type. You will need to have...

This should be possible. Ideally, it should be possible to make the current `Scanner` implement the new `Scanner` trait, so that everything keeps mostly working. Most work will probably be...

The format is defined in [RFC 1035](https://tools.ietf.org/html/rfc1035). The definition could be more clear, but the parser follows it as closely as possible.

Not sure if you are still interested in this, but with the serde traits now implemented in #116, you can indeed parse record data from the command line using a...

Can you add another `-v` to your command to get more output? (I.e., `routinator -vv vrps`). My guess would be that you have networking issues, which the extra `-v` should...

I’ve realized (while looking at #770) that a stored publication point can have multiple states that we need to distinguish between if we want to create a correct data set...

I’m starting to think this is getting way to complex and making sure that we don’t accidentally produce an incomplete data set from just the store is too tricky as...

Hi and thanks for the report! The error message means that file mentioned doesn’t contain all the data expected. Quickest way to fix this by deleting the file or –...