Upload a document

Document number, revision and purpose of issue are recorded before anything is read. A revision the project already holds is refused rather than accepted alongside it.

Demonstration project - no figure here describes a real facilityChanges are not saved

Why

The file type, size, document number and revision are validated by the ingestion service. The bytes have nowhere to go: no object store is reachable, and the form says so rather than appearing to accept a file it discards.

Writes are validated against the engineering rules and then refused, because no database has been configured for this environment.

Document details

A revision already held is refused

Uploading the revision that is already current produces two current revisions of one drawing, and from then on nobody can say which one a deliverable cited. The check names what is held and when it was issued, so you can tell whether the file is wrong or the revision code is.

Revisions already held

What a new upload would be checked against.

Revisions already held
Document	Revisions held	Current
ZRF-PID-1201	C2, C3	C3
ZRF-PID-1202	A, B	B
ZRF-PID-1101	B	B
ZRF-PID-3101	A	A
ZRF-PID-3201	A	A
ZRF-PFD-1200	B	B
ZRF-PFD-1100	A	A
ZRF-SLD-6001	A	A
ZRF-DS-P-1201A	B	B
ZRF-DS-PT-1201	A	A
ZRF-DS-PSV-1210	A	A
ZRF-REG-EQ-001	C1	C1
ZRF-REG-LL-001	C2	C2
ZRF-REG-II-001	C1	C1
ZRF-REG-VR-001	C1	C1
ZRF-REG-CS-001	B	B
ZRF-REG-MDR-001	C3	C3
ZRF-SPC-3300	C1	C1
ZRF-SPC-4100	B	B
ZRF-CTR-APX-A	0	0
ZRF-CTR-APX-D	0	0
ZRF-CTR-APX-F	0	0
ZRF-VND-P-1201	0	0
ZRF-SCH-0001	C4	C4
ZRF-PID-9001	A	A

What would happen next

A PDF is classified, then read for native text, rasterised and passed to symbol detection. A workbook is classified and read as a structured register. An XER file is imported as a schedule. Nothing is extracted as verified: a reading is a reading until a person checks it.