Archived project; service retired

WTG

The system collected publications, ran OCR and classification, then showed them to an operator for review. The service, its data and backups have been removed from the VPS. Anonymized screenshots, diagrams and measurements are kept here.

Documentation →Source code ↗

Code-map documents: June 2026; retained data history: June–July. Archiving and deletion: 7 September 2026.Version 2184148730 interface screenshots

01 / The task

Publication collection, OCR and classification. Archive

Each publication goes through several processing steps. The interface needed to show its source, OCR output, classification and the operator's decision. Errors and queues have separate screens.

Use case: Collecting publications and manually reviewing processed results.

Work done

Collection, queues, processing and the operator review screen.
Admin interface, data models and code map.
Saved anonymized screenshots and aggregates before removal. The service, production data and backups were deleted.

Screenshots

Screens, forms, errors

Screenshots, diagrams and source evidence retain their original language; captions and descriptions are translated.

02 / Technology

Stack

Python / FastAPI / Jinja

Operations admin interface, JSON API and orchestration.

PostgreSQL / SQLAlchemy / Alembic

Publications, jobs, decision versions, materials and migrations.

PaddleOCR

Separate OCR sidecar and worker.

Docker / Nginx / systemd

Former deployment of seven services and background jobs.

03 / Decisions

How the project works

Collection and long-running processing are separate jobs with leases and heartbeats.
OCR is separate from the web service, with its own queue, statuses and attempts.
Source publications, intermediate decisions and final materials have different roles.
The archive contains only anonymized images, documentation and aggregates; the working database, sessions and backups were not retained.

04 / Verification

What has been preserved

Browse anonymized schedules, groups, pipeline and materials.
Match a screen to its code-map entry and data diagram.
Open queue/history aggregates and read the measurement date and limitations.
Review the deletion record and disk/RAM comparison before and after decommissioning.

Check results

30 meaningful screenshots of the actual interface retained.
Each screenshot includes its date, scenario and code revision.
527,718 publications and 29,092 OCR jobs recorded; these are different entities, not quality metrics.
WTG was deleted after the local archive was verified; before cleanup, its directory occupied approximately 104 GiB.

Limitations

Archived project; service decommissioned. There is no WTG demo link or recovery option.
Job statuses and confidence values are not measurements of OCR or model accuracy.
Legacy noVNC and unavailable sections are not presented as working features.
Observed web memory growth to approximately 3.8 GiB is documented as an observation; the code-level root cause was not established.

Technical notes

Modules, API, data model, setup and check results.

Documentation →Materials ↓