Big changes

alembic/versions/20260220_0012_add_data_repair_jobs.py

@@ -0,0 +1,106 @@
+"""Add data repair job audit table.
+
+Revision ID: 20260220_0012
+Revises: 20260220_0011
+Create Date: 2026-02-20 23:35:00.000000
+"""
+
+from collections.abc import Sequence
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+
+revision: str = "20260220_0012"
+down_revision: str | Sequence[str] | None = "20260220_0011"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def _table_names() -> set[str]:
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+    return set(inspector.get_table_names())
+
+
+def _create_data_repair_jobs_table() -> None:
+    op.create_table(
+        "data_repair_jobs",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("job_name", sa.String(length=64), nullable=False),
+        sa.Column("requested_by", sa.String(length=255), nullable=True),
+        sa.Column(
+            "scope",
+            postgresql.JSONB(astext_type=sa.Text()),
+            nullable=False,
+            server_default=sa.text("'{}'::jsonb"),
+        ),
+        sa.Column("dry_run", sa.Boolean(), nullable=False, server_default=sa.text("true")),
+        sa.Column(
+            "status",
+            sa.String(length=16),
+            nullable=False,
+            server_default=sa.text("'planned'"),
+        ),
+        sa.Column(
+            "summary",
+            postgresql.JSONB(astext_type=sa.Text()),
+            nullable=False,
+            server_default=sa.text("'{}'::jsonb"),
+        ),
+        sa.Column("error_text", sa.Text(), nullable=True),
+        sa.Column("started_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("finished_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+        sa.CheckConstraint(
+            "status IN ('planned', 'running', 'completed', 'failed')",
+            name="data_repair_jobs_status_valid",
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_data_repair_jobs")),
+    )
+
+
+def _create_data_repair_jobs_indexes() -> None:
+    op.create_index(
+        "ix_data_repair_jobs_created_at",
+        "data_repair_jobs",
+        ["created_at"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_data_repair_jobs_job_name_created_at",
+        "data_repair_jobs",
+        ["job_name", "created_at"],
+        unique=False,
+    )
+
+
+def _drop_data_repair_jobs_indexes() -> None:
+    op.drop_index("ix_data_repair_jobs_job_name_created_at", table_name="data_repair_jobs")
+    op.drop_index("ix_data_repair_jobs_created_at", table_name="data_repair_jobs")
+
+
+def upgrade() -> None:
+    if "data_repair_jobs" in _table_names():
+        return
+    _create_data_repair_jobs_table()
+    _create_data_repair_jobs_indexes()
+
+
+def downgrade() -> None:
+    if "data_repair_jobs" not in _table_names():
+        return
+    _drop_data_repair_jobs_indexes()
+    op.drop_table("data_repair_jobs")

alembic/versions/20260221_0013_add_publication_pdf_jobs.py

@@ -0,0 +1,179 @@
+"""Add persistent publication PDF job tracking tables.
+
+Revision ID: 20260221_0013
+Revises: 20260220_0012
+Create Date: 2026-02-21 12:25:00.000000
+"""
+
+from collections.abc import Sequence
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision: str = "20260221_0013"
+down_revision: str | Sequence[str] | None = "20260220_0012"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def _table_names() -> set[str]:
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+    return set(inspector.get_table_names())
+
+
+def _create_publication_pdf_jobs_table() -> None:
+    op.create_table(
+        "publication_pdf_jobs",
+        sa.Column("publication_id", sa.Integer(), nullable=False),
+        sa.Column(
+            "status",
+            sa.String(length=16),
+            nullable=False,
+            server_default=sa.text("'queued'"),
+        ),
+        sa.Column("attempt_count", sa.Integer(), nullable=False, server_default=sa.text("0")),
+        sa.Column("queued_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("last_attempt_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("resolved_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("last_failure_reason", sa.String(length=64), nullable=True),
+        sa.Column("last_failure_detail", sa.Text(), nullable=True),
+        sa.Column("last_source", sa.String(length=32), nullable=True),
+        sa.Column("last_requested_by_user_id", sa.Integer(), nullable=True),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+        sa.CheckConstraint(
+            "status IN ('queued', 'running', 'resolved', 'failed')",
+            name="publication_pdf_jobs_status_valid",
+        ),
+        sa.ForeignKeyConstraint(
+            ["last_requested_by_user_id"],
+            ["users.id"],
+            ondelete="SET NULL",
+        ),
+        sa.ForeignKeyConstraint(
+            ["publication_id"],
+            ["publications.id"],
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("publication_id", name=op.f("pk_publication_pdf_jobs")),
+    )
+
+
+def _create_publication_pdf_jobs_indexes() -> None:
+    op.create_index(
+        "ix_publication_pdf_jobs_status",
+        "publication_pdf_jobs",
+        ["status"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_publication_pdf_jobs_updated_at",
+        "publication_pdf_jobs",
+        ["updated_at"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_publication_pdf_jobs_queued_at",
+        "publication_pdf_jobs",
+        ["queued_at"],
+        unique=False,
+    )
+
+
+def _create_publication_pdf_job_events_table() -> None:
+    op.create_table(
+        "publication_pdf_job_events",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("publication_id", sa.Integer(), nullable=False),
+        sa.Column("user_id", sa.Integer(), nullable=True),
+        sa.Column("event_type", sa.String(length=32), nullable=False),
+        sa.Column("status", sa.String(length=16), nullable=True),
+        sa.Column("source", sa.String(length=32), nullable=True),
+        sa.Column("failure_reason", sa.String(length=64), nullable=True),
+        sa.Column("message", sa.Text(), nullable=True),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            nullable=False,
+            server_default=sa.func.now(),
+        ),
+        sa.ForeignKeyConstraint(
+            ["publication_id"],
+            ["publications.id"],
+            ondelete="CASCADE",
+        ),
+        sa.ForeignKeyConstraint(
+            ["user_id"],
+            ["users.id"],
+            ondelete="SET NULL",
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_publication_pdf_job_events")),
+    )
+
+
+def _create_publication_pdf_job_events_indexes() -> None:
+    op.create_index(
+        "ix_publication_pdf_job_events_publication_created",
+        "publication_pdf_job_events",
+        ["publication_id", "created_at"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_publication_pdf_job_events_created_at",
+        "publication_pdf_job_events",
+        ["created_at"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_publication_pdf_job_events_event_type",
+        "publication_pdf_job_events",
+        ["event_type"],
+        unique=False,
+    )
+
+
+def _drop_publication_pdf_jobs_indexes() -> None:
+    op.drop_index("ix_publication_pdf_jobs_queued_at", table_name="publication_pdf_jobs")
+    op.drop_index("ix_publication_pdf_jobs_updated_at", table_name="publication_pdf_jobs")
+    op.drop_index("ix_publication_pdf_jobs_status", table_name="publication_pdf_jobs")
+
+
+def _drop_publication_pdf_job_events_indexes() -> None:
+    op.drop_index("ix_publication_pdf_job_events_event_type", table_name="publication_pdf_job_events")
+    op.drop_index("ix_publication_pdf_job_events_created_at", table_name="publication_pdf_job_events")
+    op.drop_index(
+        "ix_publication_pdf_job_events_publication_created",
+        table_name="publication_pdf_job_events",
+    )
+
+
+def upgrade() -> None:
+    tables = _table_names()
+    if "publication_pdf_jobs" not in tables:
+        _create_publication_pdf_jobs_table()
+        _create_publication_pdf_jobs_indexes()
+    if "publication_pdf_job_events" not in tables:
+        _create_publication_pdf_job_events_table()
+        _create_publication_pdf_job_events_indexes()
+
+
+def downgrade() -> None:
+    tables = _table_names()
+    if "publication_pdf_job_events" in tables:
+        _drop_publication_pdf_job_events_indexes()
+        op.drop_table("publication_pdf_job_events")
+    if "publication_pdf_jobs" in tables:
+        _drop_publication_pdf_jobs_indexes()
+        op.drop_table("publication_pdf_jobs")

alembic/versions/20260221_0014_add_scholar_publication_favorite_state.py

@@ -0,0 +1,44 @@
+"""add scholar publication favorite state
+
+Revision ID: 20260221_0014
+Revises: 20260221_0013
+Create Date: 2026-02-21 14:35:00
+"""
+
+from __future__ import annotations
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "20260221_0014"
+down_revision = "20260221_0013"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "scholar_publications",
+        sa.Column(
+            "is_favorite",
+            sa.Boolean(),
+            nullable=False,
+            server_default=sa.text("false"),
+        ),
+    )
+    op.create_index(
+        "ix_scholar_publications_is_favorite",
+        "scholar_publications",
+        ["is_favorite"],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    op.drop_index(
+        "ix_scholar_publications_is_favorite",
+        table_name="scholar_publications",
+    )
+    op.drop_column("scholar_publications", "is_favorite")

alembic/versions/20260221_0015_enforce_request_delay_floor.py

@@ -0,0 +1,80 @@
+"""enforce request delay floor at two seconds
+
+Revision ID: 20260221_0015
+Revises: 20260221_0014
+Create Date: 2026-02-21 18:10:00
+"""
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision: str = "20260221_0015"
+down_revision: str | Sequence[str] | None = "20260221_0014"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+TABLE_NAME = "user_settings"
+CHECK_NAME_MIN_1 = "request_delay_seconds_min_1"
+CHECK_NAME_MIN_2 = "request_delay_seconds_min_2"
+LEGACY_CHECK_NAME_MIN_1 = "ck_user_settings_request_delay_seconds_min_1"
+LEGACY_CHECK_NAME_MIN_2 = "ck_user_settings_request_delay_seconds_min_2"
+
+
+def _check_constraints() -> set[str]:
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+    return {
+        str(item.get("name"))
+        for item in inspector.get_check_constraints(TABLE_NAME)
+        if item.get("name")
+    }
+
+
+def _drop_check_if_exists(name: str) -> None:
+    if name not in _check_constraints():
+        return
+    op.drop_constraint(
+        name,
+        TABLE_NAME,
+        type_="check",
+    )
+
+
+def upgrade() -> None:
+    op.execute(
+        sa.text(
+            "UPDATE user_settings SET request_delay_seconds = 2 "
+            "WHERE request_delay_seconds < 2"
+        )
+    )
+
+    _drop_check_if_exists(CHECK_NAME_MIN_1)
+    _drop_check_if_exists(LEGACY_CHECK_NAME_MIN_1)
+
+    existing = _check_constraints()
+    if CHECK_NAME_MIN_2 not in existing and LEGACY_CHECK_NAME_MIN_2 not in existing:
+        op.create_check_constraint(
+            CHECK_NAME_MIN_2,
+            TABLE_NAME,
+            "request_delay_seconds >= 2",
+        )
+
+
+def downgrade() -> None:
+    _drop_check_if_exists(CHECK_NAME_MIN_2)
+    _drop_check_if_exists(LEGACY_CHECK_NAME_MIN_2)
+
+    existing = _check_constraints()
+    if CHECK_NAME_MIN_1 not in existing and LEGACY_CHECK_NAME_MIN_1 not in existing:
+        op.create_check_constraint(
+            CHECK_NAME_MIN_1,
+            TABLE_NAME,
+            "request_delay_seconds >= 1",
+        )