It will happen in the future that we have a volume document in the db that has been worked on by the annotators already, and we want to import a new better version of the OCR for the volume. In the simple case, texts start / end between pages so that's easy and that can be transferred from one version of the OCR to the next very easily. In the case where there is a text break in the middle of a page though, this will mean doing a bit of diffs / context search to map the breaks between the old and new versions. This should probably be done at import time
It will happen in the future that we have a volume document in the db that has been worked on by the annotators already, and we want to import a new better version of the OCR for the volume. In the simple case, texts start / end between pages so that's easy and that can be transferred from one version of the OCR to the next very easily. In the case where there is a text break in the middle of a page though, this will mean doing a bit of diffs / context search to map the breaks between the old and new versions. This should probably be done at import time