soimort/you-get · error · Exception

Unknown pattern

Error message

Unknown pattern

What it means

Fallback path in miaopai_download: when the URL has no ?fid= and no /p/230444 marker, the extractor fetches the page with mobile headers and looks for r'"page_url"\s*:\s*"([^"]+)"' in the HTML. If that JSON key is absent (redesign, JS-rendered content, bot wall), it raises Exception('Unknown pattern') — the extractor has no remaining strategy to locate the video.

Source

Thrown at src/you_get/extractors/miaopai.py:175

    if re.match(r'^http[s]://(.+\.)?weibo\.com/(tv/)?show/(\d{4}:\w+)', url):
        return miaopai_download_h5api(url, info_only=info_only, output_dir=output_dir, merge=merge, **kwargs)

    if re.match(r'^http[s]://(.+\.)?weibo\.com/show\?fid=(\d{4}:\w+)', url):
        return miaopai_download_h5api(url, info_only=info_only, output_dir=output_dir, merge=merge, **kwargs)

    fid = match1(url, r'\?fid=(\d{4}:\w+)')
    if fid is not None:
        miaopai_download_by_fid(fid, output_dir, merge, info_only)
    elif '/p/230444' in url:
        fid = match1(url, r'/p/230444(\w+)')
        miaopai_download_by_fid('1034:'+fid, output_dir, merge, info_only)
        pass
    else:
        mobile_page = get_content(url, headers = fake_headers_mobile)
        hit = re.search(r'"page_url"\s*:\s*"([^"]+)"', mobile_page)
        if not hit:
            raise Exception('Unknown pattern')
        else:
            escaped_url = hit.group(1)
            miaopai_download(urllib.parse.unquote(escaped_url), output_dir=output_dir, merge=merge, info_only=info_only, **kwargs)


site_info = "miaopai"
download = miaopai_download
download_playlist = playlist_not_supported('miaopai')

View on GitHub (pinned to 049548f3f3)

Solutions

  1. Prefer the explicit fid form: append/extract ?fid= or use the /p/230444 URL form which take the earlier branches
  2. Inspect mobile_page content in a debugger to see what was actually returned (bot wall vs JS shell)
  3. Update the regex at src/you_get/extractors/miaopai.py:175 to the current embedded field name for the video URL
  4. Find the Weibo oid manually (browser devtools network tab) and call miaopai_download_by_fid directly

Example fix

# before
hit = re.search(r'"page_url"\s*:\s*"([^"]+)"', mobile_page)
if not hit:
    raise Exception('Unknown pattern')

# after
hit = (re.search(r'"page_url"\s*:\s*"([^"]+)"', mobile_page)
       or re.search(r'"video_url"\s*:\s*"([^"]+)"', mobile_page))
if not hit:
    raise Exception('Unknown pattern')
Defensive patterns

Strategy: try-catch

Validate before calling

import re

def miaopai_page_has_video_url(mobile_html):
    return re.search(r'"page_url"\s*:\s*"([^"]+)"', mobile_html) is not None

Try / catch

try:
    miaopai_download(url, ...)
except Exception as e:
    if str(e) == 'Unknown pattern':
        print('page layout changed or bot wall; extract fid manually')
    else:
        raise

Prevention

When it happens

Trigger: Fetching a miaopai URL whose served HTML contains no "page_url" field: dynamic pages built by JavaScript after load, an error/expired page, or an interstitial requiring login; also URLs of a new share format that skip both fid patterns.

Common situations: Site moving to client-side rendering so get_content sees a JS shell; region blocks serving a different page; share links rotting after Miaopai's Weibo integration changes.

Related errors

AI-assisted analysis of soimort/you-get@049548f3f3 (2026-08-15). Data as JSON: /api/errors/fca1070a1b841221. Report an issue: GitHub.