野獣大百科の編集方針及びテンプレート集

from twikit import Client
import json
import re
from datetime import datetime, timezone, timedelta
import requests
import asyncio

def extract_tweet_id(url: str) -> str:
    match = re.search(r"status/(\d+)", url)
    if not match:
        raise ValueError()
    return match.group(1)

def format_datetime(created_at):
    dt_utc = datetime.strptime(
        created_at,
        "%a %b %d %H:%M:%S %z %Y"
    )
    dt_jst = dt_utc.astimezone(timezone(timedelta(hours=9)))
    hour = dt_jst.hour
    minute = dt_jst.minute

    if hour < 12:
        ampm = "午前"
        display_hour = hour if hour != 0 else 12
    else:
        ampm = "午後"
        display_hour = hour - 12 if hour > 12 else 12
        
    return f"{ampm}{display_hour}:{minute:02d} · {dt_jst.year}年{dt_jst.month}月{dt_jst.day}日"

def extractData(entry, root=False):
    if root:
        tweet_result = entry["content"]["itemContent"]["tweet_results"]["result"]
    else:
        tweet_result = entry["item"]["itemContent"]["tweet_results"]["result"]
    tweet_legacy = tweet_result["legacy"]
    user_legacy = tweet_result["core"]["user_results"]["result"]["legacy"]
    
    return {
        "name": user_legacy["name"],
        "screen_name": user_legacy["screen_name"],
        "created_at": format_datetime(tweet_legacy["created_at"]),
        "full_text": tweet_legacy["full_text"],
        "favorite_count": tweet_legacy["favorite_count"],
        "retweet_count": tweet_legacy["retweet_count"],
        "quote_count": tweet_legacy["quote_count"],
        "reply_count": tweet_legacy["reply_count"],
    }

def get_archive_today_short_url(target_url, timeout=10):
    archive_check_url = f"https://archive.ph/timemap/{target_url}"

    try:
        resp = requests.get(
            archive_check_url,
            allow_redirects=False,
            timeout=timeout,
            headers={
                "User-Agent": "Mozilla/5.0"
            }
        )
    except requests.RequestException:
        return None

    if resp.status_code == 302:
        return f"https://archive.ph/timegate/{target_url}"

    return None

async def main():
    client = Client('ja-JP')
    client.load_cookies("./cookies_twikit.json")

    tweet_url = "TARGET_URL"
    tweet_id = extract_tweet_id(tweet_url)

    response, _ = await client.gql.tweet_detail(str(tweet_id), None)
    entries = response["data"]["threaded_conversation_with_injections_v2"]["instructions"][1]["entries"]
    tweets = []

    print(json.dumps(entries))

    for entry in entries:
        if entry["content"].get("items"):
            for entry_reply in entry["content"].get("items"):
                if entry_reply["item"]["itemContent"].get("cursorType"):
                    break
                tweets.append(extractData(entry_reply))
        else:
            if entry["content"].get("cursorType"):
                break
            tweets.append(extractData(entry, root=True))

    archived = get_archive_today_short_url(tweet_url)
    if archived:
        em_archive = f"""（<a href="{archived}">魚拓</a>）"""
    else:
        em_archive = "（<del>魚拓</del>）"

    li_items = []

    for tweet in tweets:
        li_items.append(f"""        <li class="list-group-item">
            <p>{tweet["name"]}<br><small>@{tweet["screen_name"]}</small></p>
            <p>{tweet["full_text"].replace("\n", "<br>")}</p>
            <p><small>{tweet["created_at"]}</small></p>
        </li>""")
    li_html = "\n".join(li_items)

    html = f"""<div class="card h6 m-4">
    <ul class="list-group list-group-flush">
        <li class="list-group-item bg-light">
            <a href="{tweet_url}" target="_blank" rel="noopener noreferrer">
                {tweet_url}
            </a>
            {em_archive}
        </li>
        {li_html}
    </ul>
</div>"""

    with open(f"./{tweet_id}.txt", "w", encoding="utf-8") as f:
        f.write(html)

if __name__ == "__main__":
    asyncio.run(main())