SQLite Database with Eight Tables | Generated by AI
The project uses SQLite — there’s a dedicated db.py module. The database lives at data/tt.db (not the empty tt.db in the project root).
Here are the 8 tables in the database:
users — staff registry (staff_id, staff_name, registered_at)
session — current logged-in user (single-row, id=1)
client_timesheets — weekly timesheet records (staff_id, month, week, hours)
client_timesheet_days— daily breakdown per timesheet (FK → client_timesheets)
vendor_punch_cards — daily punch card records (50+ columns for attendance data)
vendor_timesheets — daily vendor timesheet records (am/pm/hours)
references — record type references (essential_keys, raw_example)
contractor_cache — cached contractor data (name, extracted_at, data as JSON blob)
images — uploaded image registry (filename, md5, added_at)
The data/tt.db database is the real one. The root-level tt.db is empty/unused — probably a stale leftover from an earlier iteration.