You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
82 lines
3.2 KiB
Python
82 lines
3.2 KiB
Python
"""add 8-K parsing: parsed_status/items_json + sec_filing_events
|
|
|
|
Revision ID: g8a9b0c1d2e3
|
|
Revises: f7a8b9c0d1e2
|
|
Create Date: 2026-04-08
|
|
"""
|
|
from typing import Sequence, Union
|
|
|
|
from alembic import op
|
|
import sqlalchemy as sa
|
|
from sqlalchemy.dialects import postgresql
|
|
|
|
revision: str = "g8a9b0c1d2e3"
|
|
down_revision: Union[str, Sequence[str], None] = "f7a8b9c0d1e2"
|
|
branch_labels: Union[str, Sequence[str], None] = None
|
|
depends_on: Union[str, Sequence[str], None] = None
|
|
|
|
|
|
def upgrade() -> None:
|
|
conn = op.get_bind()
|
|
|
|
# Add parsed_status + items_json to sec_filings
|
|
col_names = [row[0] for row in conn.execute(
|
|
sa.text("SELECT column_name FROM information_schema.columns WHERE table_name='sec_filings'")
|
|
)]
|
|
if "parsed_status" not in col_names:
|
|
op.add_column(
|
|
"sec_filings",
|
|
sa.Column("parsed_status", sa.String(20), server_default="pending", nullable=True),
|
|
)
|
|
if "items_json" not in col_names:
|
|
op.add_column(
|
|
"sec_filings",
|
|
sa.Column("items_json", postgresql.JSON(), nullable=True),
|
|
)
|
|
|
|
# Mark all existing 8-K filings as 'pending' so the parser can backfill them
|
|
conn.execute(
|
|
sa.text(
|
|
"UPDATE sec_filings SET parsed_status = 'pending' "
|
|
"WHERE form_type IN ('8-K', '8-K/A') AND parsed_status IS NULL"
|
|
)
|
|
)
|
|
|
|
# Create sec_filing_events table
|
|
if not conn.dialect.has_table(conn, "sec_filing_events"):
|
|
op.create_table(
|
|
"sec_filing_events",
|
|
sa.Column("id", postgresql.UUID(as_uuid=True), primary_key=True),
|
|
sa.Column("ticker", sa.String(10), nullable=False),
|
|
sa.Column("accession_number", sa.String(30), nullable=False),
|
|
sa.Column("form_type", sa.String(20), nullable=False),
|
|
sa.Column("filing_date", postgresql.TIMESTAMP(timezone=True), nullable=False),
|
|
sa.Column("item_number", sa.String(10), nullable=False),
|
|
sa.Column("event_type", sa.String(50), nullable=False),
|
|
sa.Column("title", sa.String(512), nullable=True),
|
|
sa.Column("summary", sa.Text(), nullable=True),
|
|
sa.Column("content_source", sa.String(20), nullable=True),
|
|
sa.Column("created_at", postgresql.TIMESTAMP(timezone=True), nullable=True),
|
|
sa.Column("updated_at", postgresql.TIMESTAMP(timezone=True), nullable=True),
|
|
sa.UniqueConstraint("accession_number", "item_number", name="uq_filing_event"),
|
|
)
|
|
op.create_index("ix_filing_events_ticker", "sec_filing_events", ["ticker"])
|
|
op.create_index("ix_filing_events_accession", "sec_filing_events", ["accession_number"])
|
|
op.create_index(
|
|
"idx_filing_events_ticker_date",
|
|
"sec_filing_events",
|
|
["ticker", "filing_date"],
|
|
)
|
|
op.create_index(
|
|
"idx_filing_events_ticker_type",
|
|
"sec_filing_events",
|
|
["ticker", "event_type"],
|
|
)
|
|
op.create_index("idx_filing_events_date", "sec_filing_events", ["filing_date"])
|
|
|
|
|
|
def downgrade() -> None:
|
|
op.drop_table("sec_filing_events")
|
|
op.drop_column("sec_filings", "items_json")
|
|
op.drop_column("sec_filings", "parsed_status")
|