"""Python 3.10+. Read-only; at most ten requests; no third-party dependencies.
Usage: python3 reference_history.py [dataset] [startDate] [endDate] [entity]
"""
import json
import os
import sys
from urllib.error import HTTPError
from urllib.parse import urlencode
from urllib.request import HTTPRedirectHandler, Request, build_opener


class NoRedirect(HTTPRedirectHandler):
    def redirect_request(self, req, fp, code, msg, headers, newurl):
        return None  # Never forward an API key to a redirect destination.


def main():
    args = sys.argv[1:]
    dataset = args[0] if args else "bis"
    filters = {
        "dataset": dataset,
        "startDate": args[1] if len(args) > 1 else "2020-01-01",
        "endDate": args[2] if len(args) > 2 else "2020-03-31",
    }
    if len(args) > 3:
        filters["entity"] = args[3]
    public_bis = dataset == "bis"
    base = os.environ.get("FILINGSIFT_API_URL", "https://api.filingsift.com/api/v1").rstrip("/")
    headers = {"Accept": "application/json"}
    if not public_bis:
        key = os.environ.get("FILINGSIFT_API_KEY")
        if not key:
            raise ValueError("Set FILINGSIFT_API_KEY with intelligence:read access.")
        headers["Authorization"] = "Bearer " + key
    path = "/reference/bis/history" if public_bis else "/account/reference-history"
    opener = build_opener(NoRedirect())
    pages, seen, cursor = [], set(), None
    for _ in range(10):
        query = dict(filters)
        if cursor is not None:
            query["cursor"] = cursor
        try:
            with opener.open(Request(base + path + "?" + urlencode(query), headers=headers), timeout=30) as response:
                envelope = json.load(response)
        except HTTPError as error:
            if error.code == 409:
                raise ValueError("Source snapshots changed: discard this run and restart without a cursor.") from error
            raise ValueError(f"HTTP {error.code}: check access, dates and coverage; for 429 honor Retry-After before another run.") from error
        data = envelope.get("data", {})
        pagination = data.get("pagination", {})
        if (data.get("schemaVersion") != "filingsift-reference-library-v1"
                or not isinstance(data.get("records"), list)
                or not isinstance(data.get("receipts"), list)
                or "nextCursor" not in pagination
                or not (pagination["nextCursor"] is None or isinstance(pagination["nextCursor"], str))):
            raise ValueError("Unexpected history response.")
        pages.append(envelope)  # Preserve exact decimal strings, nulls, units and receipts.
        cursor = pagination["nextCursor"]
        if cursor is None:
            break
        if not cursor or cursor in seen:
            raise ValueError("Repeated or empty cursor; stop and inspect the response.")
        seen.add(cursor)
    print(json.dumps({"request": filters, "complete": cursor is None, "nextCursor": cursor, "pages": pages}, indent=2))
    if cursor is not None:
        print("Stopped at ten pages. Output is partial; resume with the same filters and nextCursor, or narrow the dates/entity.", file=sys.stderr)


if __name__ == "__main__":
    try:
        main()
    except Exception as error:
        # Nothing is printed to stdout until the bounded run succeeds.
        print(str(error), file=sys.stderr)
        sys.exit(1)