How does iDBQuery ingest a folder of files with OCR?

Question

iDBQuery · Accepted Answer

iDBQuery ingests a folder of files by processing every document in it — extracting text, running OCR on scanned pages, and capturing structure — so the whole folder becomes one queryable model. Ingestion runs in the background, and you can track each file's status as it's processed and made ready to query. Pointing iDBQuery at a folder is the fastest way to make a pile of files answerable. Instead of handling documents one at a time, you give it the folder and it works through everything inside.

For each file it:

- Detects the type — spreadsheet, native PDF, scanned image, and so on.
- Extracts content, running OCR on any page that's scanned or image-based so nothing is skipped.
- Structures the data so values and tables inside the files can be queried, not just searched.
- Reports progress per file, so you can see what's been ingested, what's still processing, and which files are ready.

Because this happens in the background, a large folder doesn't block you — you can start asking questions as files become available. Once ingestion finishes, the entire folder behaves like a single source in your live model: you ask one question and iDBQuery draws on every relevant file, citing the specific document and page behind each figure. It's ideal for batches of invoices, statements, reports or any archive of mixed documents you need to actually analyse rather than just store.

كيف يستوعب iDBQuery مجلد ملفات باستخدام التعرّف الضوئي على الحروف (OCR)؟

Related questions