Looking into this service as an improvement to an existing service. Got the samples, done some testing and am rather impressed, however:
On my home dashboard it originally said 50 Pages / 250 Fields, I've since processed some samples and some of my own documents. Each time the request has been to process the whole document using processImage(), some as xml output some as searchable PDF output (I'm using PHP).
My dashboard now says 47 Pages / 235 Fields.
Question: Why has the number of fields gone down?
I haven't used processField() and haven't asked the engine to recognize any, furthermore I cannot see any way of asking the engine not to try and recognize any.
I've checked and the fields count goes down when using the profile of documentConversion or textExtraction. I've never used the fieldLevelRecognition profile.
Now this is only a concern because I'm really not interested in field recognition at all, only in getting my pages OCRed. I can't see in the documentation what happens when the pages haven't run out but the fields have. If I have 4 pages / 0 fields left, and I try and process a document, what happens?