File size: 2,341 Bytes
1f9c201
 
 
 
 
 
 
 
 
 
638f4a4
1f9c201
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
363aa87
1f9c201
363aa87
 
1f9c201
363aa87
1f9c201
 
 
638f4a4
1f9c201
 
363aa87
 
 
1f9c201
363aa87
1f9c201
 
 
363aa87
 
 
 
 
 
 
 
638f4a4
363aa87
1f9c201
 
 
638f4a4
1f9c201
363aa87
1f9c201
 
363aa87
 
 
1f9c201
 
 
638f4a4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
import gradio as gr
from bs4 import BeautifulSoup
from markdownify import MarkdownConverter
from playwright.sync_api import sync_playwright


def md(soup, **options):
    return MarkdownConverter(**options).convert_soup(soup)


def main_fn(url: str, check: list[str], request: gr.Request):
    user_agent = request.headers["user-agent"]

    with sync_playwright() as p:
        browser = p.chromium.launch(
            args=[
                "--single-process",
                "--no-zygote",
                "--no-sandbox",
                "--disable-gpu",
                "--disable-dev-shm-usage",
                "--headless=new",
            ]
        )
        context = browser.new_context(user_agent=user_agent)
        page = context.new_page()

        response = page.goto(url=url)
        status = response.status
        content = page.content()
        title = page.title()

        browser.close()

    soup = BeautifulSoup(content, features="html.parser")

    # Remove <script> e <style> corretamente
    for tag in ["script", "style"]:
        for t in soup.find_all(tag):
            t.decompose()

    # Converte conteúdo principal para Markdown
    body = soup.find("body")
    main = soup.find("main")

    strip_tags = check  # <- aplica o que o usuário selecionou

    if main:
        markdown = md(main, strip=strip_tags)
    else:
        markdown = md(body, strip=strip_tags)

    return f"{title}\n======\n\n{markdown}"


demo = gr.Interface(
    fn=main_fn,
    title="URL para Markdown V2",
    description="""
    <div style="width: fit-content; margin: 0 auto;">
        Este app acessa o HTML da URL informada e converte em Markdown.
        Utiliza o Playwright, então funciona com páginas dinâmicas como React.
    </div>
    <div style="width: fit-content; margin: 0 auto;">
        É possível ignorar links (<code>&lt;a&gt;</code>), imagens (<code>&lt;img&gt;</code>) e outros elementos.
    </div>""",
    inputs=[
        gr.Text(label="URL", placeholder="https://*****"),
        gr.CheckboxGroup(
            label="Ignorar tags no Markdown gerado",
            choices=["a", "img", "noscript"],
            value=[],
        ),
    ],
    outputs=[
        gr.TextArea(label="Markdown gerado", show_copy_button=True)
    ],
    allow_flagging="never",
)

demo.launch(server_name="0.0.0.0")