app.py

import os
import re
import shutil
import requests
import gradio as gr


def download_file(url, video_id, cache_dir="./__pycache__"):
    if os.path.exists(cache_dir):
        shutil.rmtree(cache_dir)

    os.makedirs(cache_dir)
    local_file = f"{cache_dir}/{video_id}.mp4"
    response = requests.get(url, stream=True)
    if response.status_code == 200:
        with open(local_file, "wb") as file:
            for chunk in response.iter_content(chunk_size=8192):
                file.write(chunk)

    return local_file


def extract_fst_url(text):
    url_pattern = r'(https?://[^\s"]+)'
    match = re.search(url_pattern, text)
    if match:
        out_url = match.group(1).split("?")[0]
        if out_url[-1] == "/":
            out_url = out_url[:-1]

        return out_url

    else:
        return f"https://www.bilibili.com/video/{text}"


def get_real_url(short_url):
    return requests.get(
        short_url,
        headers={
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/53.0.2785.116 Safari/537.36"
        },
        allow_redirects=True,
        timeout=10,
    ).url.split("/?")[0]


def infer(video_url: str, p: int):
    title = cover = desc = dur = video = author = avatar = None
    if not video_url:
        title = "Empty video link!"
        return title, cover, video, desc, dur, avatar, author

    video_url = extract_fst_url(video_url)
    try:
        if "b23.tv" in video_url:
            video_url = get_real_url(video_url)

        response = requests.get(os.getenv("api"), params={"url": video_url})
        response_json = response.json()
        retcode = response_json["code"]
        if retcode == 1:
            title = response_json["title"]
            cover = response_json["imgurl"]
            desc = response_json["desc"]

            response_data = response_json["data"][int(p) - 1]
            dur = response_data["duration"]
            video_id = video_url.split("/")[-1]
            video = download_file(response_data["video_url"], video_id)

            author_data = response_json["user"]
            author = author_data["name"]
            avatar = author_data["user_img"]

        else:
            title = f"Failed to call API, error code: {retcode}"

    except Exception as e:
        title = f"Failed to parse video: {e}"

    return title, cover, video, desc, dur, avatar, author


if __name__ == "__main__":
    gr.Interface(
        fn=infer,
        inputs=[
            gr.Textbox(
                label="Please input Bilibili video link",
                placeholder="https://www.bilibili.com/video/*",
                show_copy_button=True,
            ),
            gr.Slider(label="Part", minimum=1, maximum=1000, step=1, value=1),
        ],
        outputs=[
            gr.Textbox(label="Video title", show_copy_button=True),
            gr.Image(label="Video cover"),
            gr.Video(label="Download video", show_download_button=True),
            gr.TextArea(label="Video introduction", show_copy_button=True),
            gr.Textbox(label="Video duration(s)", show_copy_button=True),
            gr.Image(label="Uploader avatar"),
            gr.Textbox(label="Uploader nickname", show_copy_button=True),
        ],
        title="Bilibili video parser",
        description="This site does not provide any video storage services, only to provide the most basic resolution services",
        flagging_mode="never",
        examples=[
            ["BV1Dt4y1o7bU", 1],
            ["https://b23.tv/LuTAbzj", 1],
            ["https://www.bilibili.com/video/BV1Dt4y1o7bU", 1],
        ],
        cache_examples=False,
    ).launch()