#!/usr/bin/env python3

from typing import NamedTuple

import argparse
import csv
import io
import logging
import operator
import re
import sys
from datetime import datetime
from datetime import UTC
from pathlib import Path
from urllib.request import urlopen

SCHEMES_URL = "https://www.iana.org/assignments/uri-schemes/uri-schemes-1.csv"
COUNTRY_URL = "https://data.iana.org/time-zones/data/iso3166.tab"
ZONE_URL = "https://data.iana.org/time-zones/data/zone.tab"

BOILER_PLATE = """# Generated by get_iana_data.py @ {current_date}

from typing import NamedTuple


class ZoneData(NamedTuple):
    key: str
    full_name: str
    short_name: str
    country_code: str
    country_name: str


def get_zone_data(key: str) -> ZoneData:
    return ZoneData(*ZONE_DATA[key])


"""

logging.basicConfig(level="INFO", format="%(levelname)s: %(message)s")


class ZoneData(NamedTuple):
    key: str
    full_name: str
    short_name: str
    country_code: str
    country_name: str


def download_file(url: str) -> str:
    logging.info("Download: %s", url)
    with urlopen(url) as f:
        content = f.read()
    return content.decode()


def parse_uri_schemes(content: str) -> list[str]:
    logging.info("Parse uri schemes")
    schemes: list[str] = []

    reader = csv.reader(io.StringIO(content), delimiter=",")
    next(reader)  # Skip header row
    for line in reader:
        scheme = line[0].lower().removesuffix(" (obsolete)")
        if not re.fullmatch("[a-z0-9+.-]+", scheme):
            logging.warning("unexpected scheme field contents: %s", scheme)
            continue

        schemes.append(scheme)

    return schemes


def parse_zone_names(content: str) -> list[tuple[str, str]]:
    logging.info("Parse tz data")

    zones: list[tuple[str, str]] = []

    data = csv.reader(io.StringIO(content), delimiter="\t")
    for row in data:
        if not row or row[0].startswith("#"):
            continue
        zones.append((row[0], row[2]))

    zones.sort(key=operator.itemgetter(1))
    return zones


def parse_country_names(content: str) -> dict[str, str]:
    logging.info("Parse country data")

    countries: dict[str, str] = {}

    data = csv.reader(io.StringIO(content), delimiter="\t")
    for row in data:
        if not row or row[0].startswith("#"):
            continue
        countries[row[0]] = row[1]

    return countries


def merge_zone_data(
    zones: list[tuple[str, str]], countries: dict[str, str]
) -> dict[str, ZoneData]:

    data: dict[str, ZoneData] = {}

    for country_code, key in zones:
        country_name = countries[country_code]
        full_name = key.replace("_", " ")
        short_name = full_name.rsplit("/", maxsplit=1)[1]
        data[key] = ZoneData(
            key=key,
            full_name=full_name,
            short_name=short_name,
            country_code=country_code,
            country_name=country_name,
        )

    return data


def generate_output(
    schemes: list[str], zones: dict[str, ZoneData], outpath: Path
) -> None:

    logging.info("Generate output")
    current_date = datetime.now(tz=UTC).isoformat()
    content = BOILER_PLATE.format(current_date=current_date)

    content += "ZONE_DATA = {\n"
    for key, zone_data in zones.items():
        zone_data_tuple = ", ".join([repr(d) for d in zone_data])
        content += f"    '{key}': ({zone_data_tuple}),\n"
    content += "}\n\n\n"

    content += "URI_SCHEMES = {\n"
    for scheme in schemes:
        content += f"    '{scheme}',\n"
    content += "}\n"

    outpath.write_text(content)
    logging.info("Wrote file to %s", outpath)


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Generate IANA data")
    parser.add_argument("out", help="Path to output file")
    args = parser.parse_args()

    outpath = Path(args.out)
    if outpath.is_dir():
        sys.exit("Output path is a directory")

    scheme_content = download_file(SCHEMES_URL)
    schemes = parse_uri_schemes(scheme_content)

    country_file_data = download_file(COUNTRY_URL)
    zone_file_data = download_file(ZONE_URL)

    country_data = parse_country_names(country_file_data)
    zone_data = parse_zone_names(zone_file_data)

    zones = merge_zone_data(zone_data, country_data)

    generate_output(schemes, zones, outpath)
    logging.info("Finished !")