"""Python 3 standard library example; no packages to install."""
import json
import os
from pathlib import Path
from urllib.parse import quote
from urllib.request import Request, urlopen

API = 'https://api.apify.com/v2'
ACTOR_INPUT = {
    'urls': ['https://storage.googleapis.com/cloud-samples-data/speech/commercial_mono.wav'],
    'timestamps': 'segment',
}


def main():
    token = os.environ.get('APIFY_TOKEN')
    if not token:
        raise RuntimeError('Set APIFY_TOKEN first')

    def request(path, body=None):
        headers = {}
        data = None
        if body is not None:
            data = json.dumps(body).encode()
            headers['Content-Type'] = 'application/json'
        req = Request(API + path, data=data, headers=headers)
        # A storage download may redirect; never forward the account token to that host.
        req.add_unredirected_header('Authorization', f'Bearer {token}')
        with urlopen(req, timeout=30) as response:
            return response.read(), response.headers

    def get_json(path, body=None):
        return json.loads(request(path, body)[0])

    def incomplete(item):
        billing = item.get('billing') or {}
        return item.get('error') or item.get('stop_processing') or billing.get('stop') or billing.get('status') in ('partial', 'unknown')

    # Send this POST once. A lost response requires checking Apify Console before another start.
    run = get_json('/actors/andronixmd~speech-to-text-transcriber/runs?maxTotalChargeUsd=0.1&memory=4096&timeout=180', ACTOR_INPUT)['data']
    print(f"Run: {run['id']}")
    terminal = ('SUCCEEDED', 'FAILED', 'TIMED-OUT', 'ABORTED')
    for _ in range(12):
        if run['status'] in terminal:
            break
        run = get_json(f"/actor-runs/{run['id']}?waitForFinish=20")['data']
    if run['status'] not in terminal:
        raise RuntimeError(f"Run {run['id']} is still {run['status']}; inspect it or continue GET polling, do not start again")
    if run['status'] != 'SUCCEEDED':
        raise RuntimeError(f"Run {run['id']}: {run['status']}; inspect its saved output")

    offset, failures = 0, 0
    while True:
        body, headers = request(f"/datasets/{run['defaultDatasetId']}/items?format=json&offset={offset}&limit=100")
        rows = json.loads(body)
        for index, row in enumerate(rows, start=offset):
            item = row
            if not incomplete(row) and row.get('result_record_key'):
                record = get_json(f"/key-value-stores/{run['defaultKeyValueStoreId']}/records/{quote(row['result_record_key'], safe='')}")
                item = record.get('value')  # Unwrap only; this example does not verify the signature.
            if item is None or incomplete(row) or incomplete(item) or not isinstance(item.get('text'), str) or item.get('artifact_errors'):
                print(f"Item {index} incomplete: {(item or row).get('error') or 'billing, content or subtitle publication requires inspection'}")
                failures += 1
                continue
            print(item['text'])
            for extension in ('srt', 'vtt'):
                filename = f'transcript-{index:03d}.{extension}'
                body, _ = request(f"/key-value-stores/{run['defaultKeyValueStoreId']}/records/{filename}")
                Path(filename).write_bytes(body)
                print(f'Saved {filename}')
        offset += len(rows)
        if offset >= int(headers['X-Apify-Pagination-Total']):
            break
        if not rows:
            raise RuntimeError('Dataset pagination stopped before all results were read')
    if not offset:
        raise RuntimeError('No results; inspect the run budget and saved output')
    if failures or offset != len(ACTOR_INPUT['urls']):
        raise RuntimeError('One or more results are incomplete; inspect this run before retrying')


if __name__ == '__main__':
    main()