Create mmng.py

mmng.py

@@ -0,0 +1,282 @@
+import os
+import re
+import time
+import json
+import requests
+import img2pdf
+import pdfkit
+from bs4 import BeautifulSoup
+from datetime import datetime, timezone
+from io import BytesIO
+
+# ===== Channel.io 設定 =====
+GROUP_ID = "551316"
+
+GET_URL = f"https://desk-api.channel.io/desk/channels/200605/groups/{GROUP_ID}/messages"
+POST_URL = GET_URL
+
+PARAMS = {
+    "sortOrder": "desc",
+    "limit": 36,
+    "logFolded": "false",
+}
+
+X_ACCOUNT = os.getenv("dmsendertoken")
+if not X_ACCOUNT:
+    raise RuntimeError("環境変数 dmsendertoken が設定されていません")
+
+HEADERS_GET = {
+    "accept": "application/json",
+    "accept-language": "ja",
+    "x-account": X_ACCOUNT,
+}
+
+HEADERS_POST = {
+    "accept": "application/json",
+    "accept-language": "ja",
+    "content-type": "application/json",
+    "x-account": X_ACCOUNT,
+}
+
+# ===== Utils =====
+
+def parse_updated_at(value):
+    if isinstance(value, (int, float)):
+        return datetime.fromtimestamp(value / 1000, tz=timezone.utc)
+    elif isinstance(value, str):
+        return datetime.fromisoformat(value.replace("Z", "+00:00"))
+    return None
+
+
+def extract_url(text):
+    url_pattern = r"https?://[^\s]+"
+    m = re.search(url_pattern, text)
+    if m:
+        return m.group(0)
+    return None
+
+
+def is_mmnga_magazine(url):
+    return re.search(r"momon-ga\.com/magazine/", url)
+
+
+# ===== PDF作成 =====
+
+def create_pdf_from_images(image_urls):
+
+    image_bytes = []
+
+    for url in image_urls:
+        r = requests.get(url, timeout=60)
+        r.raise_for_status()
+        image_bytes.append(r.content)
+
+    pdf_bytes = img2pdf.convert(image_bytes)
+
+    return pdf_bytes
+
+
+def create_fullpage_pdf(url):
+
+    pdf = pdfkit.from_url(url, False)
+
+    return pdf
+
+
+# ===== mmnga処理 =====
+
+def get_mmnga_images(url):
+
+    r = requests.get(url, timeout=60)
+    r.raise_for_status()
+
+    soup = BeautifulSoup(r.text, "html.parser")
+
+    post = soup.find("div", id="post")
+
+    if not post:
+        raise Exception("post divが見つかりません")
+
+    imgs = post.find_all("img")
+
+    image_urls = []
+
+    for img in imgs:
+        src = img.get("src")
+        if not src:
+            continue
+
+        if src.startswith("//"):
+            src = "https:" + src
+
+        if src.startswith("/"):
+            src = "https://momon-ga.com" + src
+
+        image_urls.append(src)
+
+    if not image_urls:
+        raise Exception("画像が見つかりません")
+
+    return image_urls
+
+
+# ===== Channel送信 =====
+
+def send_to_channel(text):
+
+    payload = {
+        "requestId": f"desk-web-{int(time.time()*1000)}",
+        "blocks": [
+            {"type": "text", "value": text}
+        ],
+    }
+
+    r = requests.post(
+        POST_URL,
+        headers=HEADERS_POST,
+        data=json.dumps(payload),
+        timeout=30
+    )
+
+    r.raise_for_status()
+
+
+def upload_file_to_channel(file_bytes):
+
+    upload_url = f"https://media.channel.io/cht/v1/pri-file/200605/groups/{GROUP_ID}/message/send_pdf_file.pdf"
+
+    headers = {
+        "x-account": X_ACCOUNT,
+        "Content-Type": "application/pdf",
+        "Content-Length": str(len(file_bytes)),
+    }
+
+    r = requests.post(
+        upload_url,
+        headers=headers,
+        data=file_bytes,
+        timeout=300
+    )
+
+    r.raise_for_status()
+
+    return r.json()
+
+
+def send_pdf_message(file_json):
+
+    payload = {
+        "requestId": f"desk-web-{int(time.time()*1000)}",
+        "blocks": [
+            {"type": "text", "value": "PDFプレビュー"}
+        ],
+        "files": [file_json],
+    }
+
+    r = requests.post(
+        POST_URL,
+        headers=HEADERS_POST,
+        data=json.dumps(payload),
+        timeout=30
+    )
+
+    r.raise_for_status()
+
+
+# ===== Main =====
+
+def main():
+
+    processed_messages = set()
+
+    while True:
+
+        try:
+
+            res = requests.get(
+                GET_URL,
+                headers=HEADERS_GET,
+                params=PARAMS,
+                timeout=30
+            )
+
+            res.raise_for_status()
+
+            messages = res.json().get("messages", [])
+
+            latest_msg = None
+            latest_time = None
+
+            for msg in messages:
+
+                msg_id = msg.get("id")
+                text = msg.get("plainText")
+                updated_at = msg.get("updatedAt")
+
+                if not text:
+                    continue
+
+                if msg_id in processed_messages:
+                    continue
+
+                t = parse_updated_at(updated_at)
+
+                if not t:
+                    continue
+
+                if latest_time is None or t > latest_time:
+                    latest_time = t
+                    latest_msg = msg
+
+            if not latest_msg:
+                time.sleep(10)
+                continue
+
+            url = extract_url(latest_msg["plainText"])
+
+            if not url:
+                processed_messages.add(latest_msg["id"])
+                time.sleep(10)
+                continue
+
+            send_to_channel("PDF生成を開始します")
+
+            try:
+
+                if is_mmnga_magazine(url):
+
+                    send_to_channel("mmnga漫画ページを検出しました")
+
+                    image_urls = get_mmnga_images(url)
+
+                    pdf_bytes = create_pdf_from_images(image_urls)
+
+                else:
+
+                    send_to_channel("通常ページをPDF化します")
+
+                    pdf_bytes = create_fullpage_pdf(url)
+
+                send_to_channel("PDFをアップロードしています")
+
+                file_json = upload_file_to_channel(pdf_bytes)
+
+                send_pdf_message(file_json)
+
+                send_to_channel("完了しました")
+
+            except Exception as e:
+
+                send_to_channel(f"エラー: {e}")
+
+            processed_messages.add(latest_msg["id"])
+
+        except Exception as e:
+
+            print("error:", e)
+
+        time.sleep(15)
+
+
+if __name__ == "__main__":
+    main()