Drive sync and ZIP unpacking – system description

Data model

Table names carry the installation's own prefix in the database.

Drive sync

1. drivesync: a job row (addDocJob) and a one-off doc_drive_sync queue row for the chosen client (a super admin without client: everyone); the answer gives the row number and – with system/job list visibility – the link to the job list.
2. doc_drive_sync (hourly too, immediately via the button): walks the folders of the client(s)' Drive storage rules in full depth and creates a record with document/document::addDoc() for every file without one (the file stays on the Drive); Google's own formats are skipped; the file of a trashed document counts as known; chunked run, one notification at the end.

ZIP unpacking

1. unpackzip: zip extension only; for a ZIP on the Drive by its Drive id, for a local ZIP via a temporary copy (≤300 MB), it starts a doc_zip_unpack job; selectable target root (account/drive/folder or local directory + subfolder); the ZIP is kept (keep_archive).
2. doc_zip_unpack (one-off; started by upload or row action): within a time budget, entry by entry, loads the PDF into the shared addDocument() with the chosen storage rule, queues it for recognition, then starts a doc_recognise job; the rest continues as a new row; at the end the ZIP is deleted (unless keep_archive); the user is notified.

Permissions

• drivesync (UNMAPPED, missing from the catalogue list) + core modify; a restricted user must choose a client (error_drivesync_company), within scope.
• unpackzip (UNMAPPED) + core modify + scope.