ytdl-org/youtube-dl · error · ExtractorError

Unable to extract media URL

Error message

Unable to extract media URL

What it means

Raised by the Metacafe extractor during URL discovery: earlier probes (mediaURL query param, <video src> tag) found nothing, a flashvars parameter string exists, but parsing it with compat_parse_qs yields no 'mediaData' key. Not expected=True, so it surfaces as an extraction failure.

Source

Thrown at youtube_dl/extractor/metacafe.py:194

            mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
            if mobj is None:
                video_url = mediaURL
            else:
                gdaKey = mobj.group(1)
                video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
        if video_url is None:
            mobj = re.search(r'<video src="([^"]+)"', webpage)
            if mobj:
                video_url = mobj.group(1)
                video_ext = 'mp4'
        if video_url is None:
            flashvars = self._search_regex(
                r' name="flashvars" value="(.*?)"', webpage, 'flashvars',
                default=None)
            if flashvars:
                vardict = compat_parse_qs(flashvars)
                if 'mediaData' not in vardict:
                    raise ExtractorError('Unable to extract media URL')
                mobj = re.search(
                    r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
                if mobj is None:
                    raise ExtractorError('Unable to extract media URL')
                mediaURL = mobj.group('mediaURL').replace('\\/', '/')
                video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
                video_ext = determine_ext(video_url)
        if video_url is None:
            player_url = self._search_regex(
                r"swfobject\.embedSWF\('([^']+)'",
                webpage, 'config URL', default=None)
            if player_url:
                config_url = self._search_regex(
                    r'config=(.+)$', player_url, 'config URL')
                config_doc = self._download_xml(
                    config_url, video_id,
                    note='Downloading video config')
                smil_url = config_doc.find('.//properties').attrib['smil_file']

View on GitHub (pinned to 956b8c5855)

Solutions

1. Update youtube-dl / yt-dlp — Metacafe page structure changes are fixed in newer releases.
2. Run with -v to see which of the URL-probing branches executed and dump the flashvars content.
3. Try the alternate URL form (adding a family-friendly or byId variant) in case the canonical page renders differently.
4. If reproducible, report the URL to the issue tracker so the flashvars parsing can be adjusted.

# Pre-flight: confirm the page carries a known media marker
import urllib.request
html = urllib.request.urlopen(url).read().decode('utf-8', 'replace')
assert ('mediaURL=' in html or '<video src=' in html or 'mediaData' in html), 'No known Metacafe media marker'

def flashvars_has_mediadata(vardict):
    return isinstance(vardict, dict) and 'mediaData' in vardict

try:
    info = ydl.extract_info(url, download=False)
except ExtractorError as e:
    if 'Unable to extract media URL' in str(e):
        try_alternative_extractor_or_manual_download(url)  # page shape changed; fall back
    else:
        raise

When it happens

Trigger: Webpage has neither (?m)&mediaURL= / &videoURL= nor a <video src="..."> tag, but does contain name="flashvars" value="..."; the parsed flashvars dict lacks 'mediaData'.

Common situations: Metacafe changed its flash player embedding (renamed the flashvars key); the video page is a redirect/login shell with placeholder flashvars; very old or very new page templates that never carried mediaData.

Related errors

• Unsupported video type
• Could not find video iframe in any portlets

AI-assisted analysis of ytdl-org/youtube-dl@956b8c5855 (2026-08-14). Data as JSON: /api/errors/ea8a95a390c656fb. Report an issue: GitHub.