"""Add integrity_status + integrity_checked_at to image_record.

Tracks per-image structural verification state so corrupt files (truncated
downloads, broken containers, partial archives) can be flagged and excluded
from random/showcase/ML/suggestion paths instead of blowing up downstream
preprocessing.

Status values:
  - unknown: never verified (default for existing rows + freshly imported
    until the verifier runs)
  - ok: passed marker check + format-level verify
  - truncated: missing trailing bytes (PIL EOI/IEND, ffprobe end-of-stream)
  - unreadable: header/structure invalid; can't even open the file
  - missing: filepath doesn't exist on disk

Revision ID: l26042501
Revises: k26042201
Create Date: 2026-04-25
"""
from __future__ import annotations

from alembic import op
import sqlalchemy as sa


revision = 'l26042501'
down_revision = 'k26042201'
branch_labels = None
depends_on = None


def upgrade():
    op.add_column(
        'image_record',
        sa.Column(
            'integrity_status',
            sa.String(16),
            nullable=False,
            server_default='unknown',
        ),
    )
    op.add_column(
        'image_record',
        sa.Column(
            'integrity_checked_at',
            sa.DateTime(timezone=True),
            nullable=True,
        ),
    )
    # Partial index so the "list flagged rows" report and the
    # exclude-from-random/ML filters stay cheap on a million-row table.
    op.create_index(
        'image_record_integrity_status_idx',
        'image_record',
        ['integrity_status'],
        postgresql_where=sa.text("integrity_status <> 'ok'"),
    )


def downgrade():
    op.drop_index('image_record_integrity_status_idx', table_name='image_record')
    op.drop_column('image_record', 'integrity_checked_at')
    op.drop_column('image_record', 'integrity_status')