soimort/you-get · error · Exception

Failed

Error message

Failed

What it means

Raised as plain Exception in icourses_download (src/you_get/extractors/icourses.py:39). The code tries up to 5 times to probe the generated video URL with url_info() to learn the file size; if every attempt raises urllib's HTTPError (logged as 'Failed to fetch the video file! Retrying...'), size stays None and the download is aborted.

Source

Thrown at src/you_get/extractors/icourses.py:39

        raise Exception('You can download it with -l flag')
    icourses_parser = ICousesExactor(url=url)
    icourses_parser.basic_extract()
    title = icourses_parser.title
    size = None
    for i in range(5):
        try:
            # use this url only for size
            size_url = icourses_parser.generate_url(0)
            _, type_, size = url_info(size_url, headers=fake_headers)
        except error.HTTPError:
            logging.warning('Failed to fetch the video file! Retrying...')
            sleep(random.Random().randint(2, 5))  # Prevent from blockage
        else:
            print_info(site_info, title, type_, size)
            break

    if size is None:
        raise Exception("Failed")

    if not kwargs['info_only']:
        real_url = icourses_parser.update_url(0)
        headers = fake_headers.copy()
        headers['Referer'] = url
        download_urls_icourses(real_url, title, 'flv',total_size=size, output_dir=output_dir, max_size=15728640, dyn_callback=icourses_parser.update_url)
    return


def get_course_title(url, course_type, page=None):
    if page is None:
        try:
            # shard course page could be gbk but with charset="utf-8"
            page = get_content(url, decoded=False).decode('gbk')
        except UnicodeDecodeError:
            page = get_content(url, decoded=False).decode('utf8')

    if course_type == 'shared_old':

View on GitHub (pinned to 049548f3f3)

Solutions

1. Wait several minutes and retry — the failure mode is usually temporary blocking.
2. Verify the generated size_url in a browser/curl; a persistent 4xx means the URL construction (flashvars parsing) is broken — upgrade you-get or fix ICousesExactor.
3. Reduce request frequency in scripts (honor the existing randomized sleeps) to avoid tripping the block.

import time
for attempt in range(3):
    try:
        icourses_download(url, output_dir)
        break
    except Exception as e:
        if str(e) == 'Failed':
            time.sleep(300)  # let the server-side block cool down
            continue
        raise

When it happens

Trigger: ICousesExactor.generate_url(0) yields a URL that url_info() rejects with HTTPError on 5 consecutive attempts: the icourses server blocks/throttles the request (the code sleeps 2-5s between tries specifically to 'prevent blockage'), or the generated URL is invalid (bad uuid/IService flashvars) so the server answers 4xx/5xx every time.

Common situations: Aggressive repeated downloads getting rate-limited/blocked by icourses.cn; expired or wrongly parsed player tokens producing dead URLs; site-side changes to the video API.

Related errors

• Server refused to provide download link! (Errno:%s)
• Share not found or canceled: %s
• You can download it with -l flag
• Unknown url pattern
• No video found in this playlist

AI-assisted analysis of soimort/you-get@049548f3f3 (2026-08-15). Data as JSON: /api/errors/76d00d887923e27b. Report an issue: GitHub.