Perform identification on every page scanned to dynamically select different processing configurations.
Improve batch scanning functionality to allow different configurations to be identified without having to perform different scan jobs. We have multiple automatic identification configurations working fantastic but still have to perform each separate configuration as an entirely different scan. We already have these identifying perfectly each time we scan without selecting a configuration. Scan config 1 with multiple separators and index that entire batch at once. Then without changing any scan settings send config 2 batch through and it identifies and splits and index correctly. Why can we not mix the batches together and when it encounters identification page/config 1 use that config, then when it sees identification page/config 2 use that one. It doesn't make sense that we have to perform entirely separate scans and identification is only performed on the first encountered identifier.