Files
signal-platform/app/services/ticker_universe_service.py
T
dennisthiessenandClaude Opus 5 1d4ed39fd2 fix(tickers): close the delisting review findings
Detection could retire an actively traded symbol — silently, since it then
vanishes from every signal. Three causes:

- Form 25 is filed per security class. An issuer removing its notes, preferred
  or warrants files one while the common keeps trading. The filing's own
  descriptionClassSecurity distinguishes them, so the primary document is now
  fetched and read; anything not recognisably common equity is rejected, as is
  anything unreadable (pre-2009 filings have no primary_doc.xml). Fail closed.
- Form 15 ends a reporting obligation and is no evidence trading stopped. The
  whole family is dropped.
- A historical filing for a long-gone class could retire a symbol whose bars ran
  years later, stamping the old date. Filings before the last bar (less a 30-day
  lead for the exchange) are now ignored.

Rule 12d2-2 makes removal effective ten days after filing, so delisted_on is the
effective date rather than the filing date.

bootstrap_universe(prune_missing=True) still ran a cascading delete over
delisted rows, undoing the retention this branch exists for; it now skips them
and reports kept_delisted so the count is explicable.

clear_delisted had no route, which made "safe to automate because it is
reversible" false — reversal needed SQL. POST/DELETE /tickers/{symbol}/delisting
now mark and un-mark, giving an operator a non-destructive alternative to the
cascading DELETE that was the only option.

Shared-CIK siblings (GOOG/GOOGL) stay safe by construction: the probe is
per-symbol and gated on that symbol's own staleness, so a class that still
trades is never probed.

Not addressed: pruning a symbol merely dropped from the index still destroys its
history — the same survivorship problem in a different costume, needing a
tracked/membership state separate from delisting.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-11 16:58:51 +02:00

403 lines
15 KiB
Python

"""Ticker universe discovery and bootstrap service.
Provides a minimal, provider-backed way to populate tracked tickers from
well-known universes (S&P 500, NASDAQ-100, NASDAQ All).
"""
from __future__ import annotations
import asyncio
import json
import logging
import os
import re
from collections.abc import Iterable
from datetime import datetime, timezone
from pathlib import Path
import httpx
from sqlalchemy import delete, select
from sqlalchemy.ext.asyncio import AsyncSession
from app.config import settings
from app.exceptions import ProviderError, ValidationError
from app.models.ticker import Ticker
from app.services import settings_store
logger = logging.getLogger(__name__)
SUPPORTED_UNIVERSES = {"sp500", "nasdaq100", "nasdaq_all"}
_SYMBOL_PATTERN = re.compile(r"^[A-Z0-9-]{1,10}$")
_SEED_UNIVERSES: dict[str, list[str]] = {
"sp500": [
"AAPL", "MSFT", "NVDA", "AMZN", "META", "GOOGL", "GOOG", "BRK-B", "TSLA", "JPM",
"V", "MA", "UNH", "XOM", "LLY", "AVGO", "COST", "PG", "JNJ", "HD", "MRK", "BAC",
"ABBV", "PEP", "KO", "ADBE", "NFLX", "CRM", "CSCO", "WMT", "AMD", "TMO", "MCD",
"ORCL", "ACN", "CVX", "LIN", "DHR", "ABT", "QCOM", "TXN", "PM", "DIS", "INTU",
],
"nasdaq100": [
"AAPL", "MSFT", "NVDA", "AMZN", "META", "GOOGL", "GOOG", "TSLA", "AVGO", "COST",
"NFLX", "ADBE", "CSCO", "AMD", "INTU", "QCOM", "AMGN", "TXN", "INTC", "BKNG", "GILD",
"ISRG", "MDLZ", "ADP", "LRCX", "ADI", "PANW", "SNPS", "CDNS", "KLAC", "MELI", "MU",
"SBUX", "CSX", "REGN", "VRTX", "MAR", "MNST", "CTAS", "ASML", "PYPL", "AMAT", "NXPI",
],
"nasdaq_all": [
"AAPL", "MSFT", "NVDA", "AMZN", "META", "GOOGL", "TSLA", "AMD", "INTC", "QCOM", "CSCO",
"ADBE", "NFLX", "PYPL", "AMAT", "MU", "SBUX", "GILD", "INTU", "BKNG", "ADP", "CTAS",
"PANW", "SNPS", "CDNS", "LRCX", "KLAC", "MELI", "ASML", "REGN", "VRTX", "MDLZ", "AMGN",
],
}
_CA_BUNDLE = os.environ.get("SSL_CERT_FILE", "")
if not _CA_BUNDLE or not Path(_CA_BUNDLE).exists():
_CA_BUNDLE_PATH: str | bool = True
else:
_CA_BUNDLE_PATH = _CA_BUNDLE
# Wikipedia often returns 403 to non-browser UAs; use a normal browser-like
# identity for constituent scrapes (no cookies/login).
_HTTP_HEADERS = {
"User-Agent": (
"Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
"AppleWebKit/537.36 (KHTML, like Gecko) "
"Chrome/126.0.0.0 Safari/537.36"
),
"Accept": "text/html,application/xhtml+xml;q=0.9,*/*;q=0.8",
"Accept-Language": "en-US,en;q=0.9",
}
# Modern Wikipedia S&P/Nasdaq tables use exchange templates (NyseSymbol /
# NasdaqSymbol) rather than a plain <td><a>SYMBOL</a></td>. Prefer quote URLs
# and template params; keep the legacy cell pattern as a last resort.
_WIKI_SYMBOL_PATTERNS: tuple[re.Pattern[str], ...] = (
re.compile(r"nyse\.com/quote/XNYS:([A-Za-z0-9.-]{1,10})", re.IGNORECASE),
re.compile(
r"nasdaq\.com/market-activity/stocks/([A-Za-z0-9.-]{1,10})",
re.IGNORECASE,
),
# {{NyseSymbol|BNY}} / {{NasdaqSymbol|AAPL}} rendered data-mw params
re.compile(
r'"target":\{"wt":"(?:Nyse|Nasdaq)Symbol"[^}]*\}.*"wt":"([A-Z][A-Z0-9.-]{0,9})"',
re.IGNORECASE,
),
re.compile(r"<td>\s*<a[^>]*>([A-Z.]{1,10})</a>\s*</td>", re.IGNORECASE),
)
def _extract_wiki_symbols(html: str) -> list[str]:
"""Pull ticker symbols out of a Wikipedia constituents page."""
found: list[str] = []
for pattern in _WIKI_SYMBOL_PATTERNS:
found.extend(pattern.findall(html))
return found
def _validate_universe(universe: str) -> str:
normalised = universe.strip().lower()
if normalised not in SUPPORTED_UNIVERSES:
supported = ", ".join(sorted(SUPPORTED_UNIVERSES))
raise ValidationError(f"Unsupported universe '{universe}'. Supported: {supported}")
return normalised
def _normalise_symbols(symbols: Iterable[str]) -> list[str]:
deduped: set[str] = set()
for raw_symbol in symbols:
symbol = raw_symbol.strip().upper().replace(".", "-")
if not symbol:
continue
if _SYMBOL_PATTERN.fullmatch(symbol) is None:
continue
deduped.add(symbol)
return sorted(deduped)
async def _fetch_wiki_constituent_symbols(
client: httpx.AsyncClient,
url: str,
) -> tuple[list[str], str | None]:
try:
response = await client.get(url, headers=_HTTP_HEADERS)
except httpx.HTTPError as exc:
return [], f"{url}: network error ({type(exc).__name__}: {exc})"
if response.status_code != 200:
return [], f"{url}: HTTP {response.status_code}"
matches = _extract_wiki_symbols(response.text)
if not matches:
return [], f"{url}: no symbols parsed"
return list(matches), None
async def _fetch_nasdaq_trader_symbols(
client: httpx.AsyncClient,
) -> tuple[list[str], str | None]:
url = "https://www.nasdaqtrader.com/dynamic/SymDir/nasdaqlisted.txt"
try:
response = await client.get(url, headers=_HTTP_HEADERS)
except httpx.HTTPError as exc:
return [], f"{url}: network error ({type(exc).__name__}: {exc})"
if response.status_code != 200:
return [], f"{url}: HTTP {response.status_code}"
symbols: list[str] = []
for line in response.text.splitlines():
if not line or line.startswith("Symbol|") or line.startswith("File Creation Time"):
continue
parts = line.split("|")
if not parts:
continue
symbol = parts[0].strip()
test_issue = parts[6].strip() if len(parts) > 6 else "N"
if test_issue == "Y":
continue
symbols.append(symbol)
if not symbols:
return [], f"{url}: no symbols parsed"
return symbols, None
async def _fetch_universe_symbols_from_public(universe: str) -> tuple[list[str], list[str], str | None]:
failures: list[str] = []
sp500_url = "https://en.wikipedia.org/wiki/List_of_S%26P_500_companies"
nasdaq100_url = "https://en.wikipedia.org/wiki/Nasdaq-100"
async with httpx.AsyncClient(timeout=30.0, verify=_CA_BUNDLE_PATH) as client:
if universe == "sp500":
symbols, error = await _fetch_wiki_constituent_symbols(client, sp500_url)
if error:
failures.append(error)
else:
return symbols, failures, "wikipedia_sp500"
if universe == "nasdaq100":
symbols, error = await _fetch_wiki_constituent_symbols(client, nasdaq100_url)
if error:
failures.append(error)
else:
return symbols, failures, "wikipedia_nasdaq100"
if universe == "nasdaq_all":
symbols, error = await _fetch_nasdaq_trader_symbols(client)
if error:
failures.append(error)
else:
return symbols, failures, "nasdaq_trader"
return [], failures, None
async def _read_cached_symbols(db: AsyncSession, universe: str) -> list[str]:
key = f"ticker_universe_cache_{universe}"
setting = await settings_store.get_setting(db, key)
if setting is None:
return []
try:
payload = json.loads(setting.value)
except (TypeError, ValueError):
return []
if isinstance(payload, dict):
symbols = payload.get("symbols", [])
elif isinstance(payload, list):
symbols = payload
else:
symbols = []
if not isinstance(symbols, list):
return []
return _normalise_symbols([str(symbol) for symbol in symbols])
async def _write_cached_symbols(
db: AsyncSession,
universe: str,
symbols: list[str],
source: str,
) -> None:
key = f"ticker_universe_cache_{universe}"
payload = {
"symbols": symbols,
"source": source,
"updated_at": datetime.now(timezone.utc).isoformat(),
}
await settings_store.upsert_setting(db, key, json.dumps(payload))
await db.commit()
async def fetch_universe_symbols(
db: AsyncSession,
universe: str,
) -> tuple[list[str], str]:
"""Fetch and normalise symbols for a supported universe with fallbacks.
Fallback order:
1) Free public sources (Wikipedia/NASDAQ trader)
2) Cached snapshot in SystemSetting
3) Built-in seed symbols
Returns ``(symbols, source_label)`` so bootstrap UI can show where the
list came from (important when the public source fails and a stale cache
still lists BK instead of BNY).
The seeds are representative, not complete, so a *fresh* install whose
public source is down bootstraps a partial universe. A warm instance is
unaffected — it falls through to its cached snapshot.
"""
normalised_universe = _validate_universe(universe)
failures: list[str] = []
public_symbols, public_failures, public_source = await _fetch_universe_symbols_from_public(normalised_universe)
failures.extend(public_failures)
cleaned_public = _normalise_symbols(public_symbols)
if cleaned_public:
await _write_cached_symbols(db, normalised_universe, cleaned_public, public_source or "public")
return cleaned_public, public_source or "public"
cached_symbols = await _read_cached_symbols(db, normalised_universe)
if cached_symbols:
logger.warning(
"Using cached universe symbols for %s because live fetch failed: %s",
normalised_universe,
"; ".join(failures[:3]),
)
return cached_symbols, "cache"
seed_symbols = _normalise_symbols(_SEED_UNIVERSES.get(normalised_universe, []))
if seed_symbols:
logger.warning(
"Using built-in seed symbols for %s because live/cache fetch failed: %s",
normalised_universe,
"; ".join(failures[:3]),
)
return seed_symbols, "seed"
reason = "; ".join(failures[:6]) if failures else "no provider returned symbols"
raise ProviderError(f"Universe '{normalised_universe}' returned no valid symbols. Attempts: {reason}")
async def _fetch_alpaca_asset_names() -> dict[str, str]:
"""One Alpaca Trading-API call → {internal_symbol: company_name} for all US
equities. Tries paper and live endpoints so it works with either key type."""
if not settings.alpaca_api_key or not settings.alpaca_api_secret:
raise ValidationError("Alpaca API credentials are required to backfill names")
from alpaca.trading.client import TradingClient
from alpaca.trading.enums import AssetClass, AssetStatus
from alpaca.trading.requests import GetAssetsRequest
req = GetAssetsRequest(status=AssetStatus.ACTIVE, asset_class=AssetClass.US_EQUITY)
last_err: Exception | None = None
for paper in (True, False):
try:
client = TradingClient(settings.alpaca_api_key, settings.alpaca_api_secret, paper=paper)
assets = await asyncio.to_thread(client.get_all_assets, req)
names: dict[str, str] = {}
for asset in assets:
sym = getattr(asset, "symbol", None)
nm = getattr(asset, "name", None)
if sym and nm:
names[sym.replace(".", "-").upper()] = nm # BRK.B → BRK-B
if names:
return names
except Exception as exc: # noqa: BLE001 — try the other endpoint
last_err = exc
raise ProviderError(f"Failed to fetch asset names from Alpaca: {last_err}")
async def backfill_ticker_names(db: AsyncSession, *, only_missing: bool = True) -> dict[str, int]:
"""Fill Ticker.name from Alpaca in a single request for the whole universe."""
result = await db.execute(select(Ticker))
tickers = list(result.scalars().all())
targets = [t for t in tickers if not t.name] if only_missing else tickers
if not targets:
return {"updated": 0, "checked": 0, "unmatched": 0}
names = await _fetch_alpaca_asset_names()
updated = 0
for ticker in targets:
nm = names.get(ticker.symbol.upper())
if nm and nm != ticker.name:
ticker.name = nm[:120]
updated += 1
await db.commit()
return {"updated": updated, "checked": len(targets), "unmatched": len(targets) - updated}
async def bootstrap_universe(
db: AsyncSession,
universe: str,
*,
prune_missing: bool = False,
) -> dict[str, int | str]:
"""Upsert ticker universe into tracked tickers.
Returns summary counts for added/existing/deleted symbols.
"""
normalised_universe = _validate_universe(universe)
symbols, source = await fetch_universe_symbols(db, normalised_universe)
existing_rows = await db.execute(select(Ticker.symbol))
existing_symbols = set(existing_rows.scalars().all())
target_symbols = set(symbols)
symbols_to_add = sorted(target_symbols - existing_symbols)
symbols_to_delete = sorted(existing_symbols - target_symbols) if prune_missing else []
for symbol in symbols_to_add:
db.add(Ticker(symbol=symbol))
deleted_count = 0
skipped_delisted: list[str] = []
if symbols_to_delete:
# A delisted row was retained on purpose — its price history is exactly
# what a survivorship-honest backtest needs, and the delete cascades it
# away. Pruning must not undo that. (Pruning a symbol that is merely no
# longer an index constituent still destroys history; that needs a
# tracked/membership state separate from delisting.)
protected = (
await db.execute(
select(Ticker.symbol).where(
Ticker.symbol.in_(symbols_to_delete),
Ticker.delisted_on.is_not(None),
)
)
).scalars().all()
skipped_delisted = sorted(protected)
deletable = [s for s in symbols_to_delete if s not in set(protected)]
if deletable:
result = await db.execute(delete(Ticker).where(Ticker.symbol.in_(deletable)))
deleted_count = int(result.rowcount or 0)
await db.commit()
# Best-effort: fill company names for any tickers still missing one. Never let
# a name-fetch hiccup fail the bootstrap itself.
try:
await backfill_ticker_names(db, only_missing=True)
except Exception: # noqa: BLE001
logger.warning("Ticker name backfill failed during bootstrap", exc_info=True)
return {
"universe": normalised_universe,
"source": source,
"total_universe_symbols": len(symbols),
"added": len(symbols_to_add),
"already_tracked": len(target_symbols & existing_symbols),
"deleted": deleted_count,
"added_symbols": symbols_to_add[:50],
# Delisted rows a prune declined to destroy, so the caller can see the
# count did not match what they asked to remove.
"kept_delisted": skipped_delisted[:50],
"kept_delisted_count": len(skipped_delisted),
}