Dimitry Sibiryakov

> Don't you think there could be other replication related info items in the future? Statistics (number of replicated blocks / transactions / records), for example? So far I see...

Why I like to have them is errors' handling. If separate instances of Replicator is used for each transaction and something bad happen you can just kill one instance and...

Taking into account the usage even simple `` would be enough. If application can use only too old protocol at max - asking is useless, an error on attempt to...

And if you are going to have `IReplicator::getInfo()` call then separate items `minProtocol` and `maxProtocol` are better than single item with complex internal format.

The reason is simple: it is a very big task with many questions to be decided.

Item (3) is an ineffective workaround for problems created by item (2) so if you implement (2) you must somehow solve unpredictable and hard to diagnose problems with queries returning...

I doubt that a special SQL statement is needed here. Background validation can run as it is done for encryption now and start automatically on database startup if any not-valid-yet...

> It connect to a database, without password ! @moyzer By design Embedded mode doesn't need password.

Nevertheless update behavior is defined: fields in every calculations are seen in state they were before actual modifications. Shouldn't select follow the same conventions calculating whole returned record on fetch...

Is it actually possible? Adding a new function to TraceConnection will break both derived interfaces shifting pointers in pseudo-VMT.