def test_url()

in link-verifier/verify-links.py [0:0]

• 38 lines of code
• 23 McCabe index (conditional complexity)

def test_url(url):
    """Tests a single url"""

    global use_gh_cache
    global main_repo_list
    global link_cache
    status = ''
    is_broken = False
    # Test if link was already tested before.
    if url in link_cache:
        return link_cache[url]
    # Test if link was cached in pre-fetched GitHub issues. If not, send a request for the link.
    if use_gh_cache:
        pr_match = re.search(PULL_REQUEST_SEARCH, url)
        issue_match = re.search(ISSUE_SEARCH, url)
        if pr_match is not None:
            repo_key = f'{pr_match.group(1)}/{pr_match.group(2)}'.lower()
            if repo_key in main_repo_list and PR_KEY in main_repo_list[repo_key]:
                if int(pr_match.group(3)) in main_repo_list[repo_key][PR_KEY]:
                    status = 'Good'
        elif issue_match is not None:
            repo_key = f'{issue_match.group(1)}/{issue_match.group(2)}'.lower()
            if repo_key in main_repo_list and ISSUE_KEY in main_repo_list[repo_key]:
                if int(issue_match.group(3)) in main_repo_list[repo_key][ISSUE_KEY]:
                    status = 'Good'
    if status != 'Good':
        try:
            r = requests.head(url, allow_redirects=True)
            # Some sites may return 404 for head but not get, e.g.
            # https://tls.mbed.org/kb/development/thread-safety-and-multi-threading
            if r.status_code >= 400:
                # Allow redirects is already enabled by default for GET.
                r = requests.get(url)
            # It's likely we will run into GitHub's rate-limiting if there are many links.
            if r.status_code == 429:
                time.sleep(int(r.headers['Retry-After']))
                r = requests.head(url, allow_redirects=True)
            if r.status_code >= 400:
                is_broken = True
            status = r.status_code
        # requests.exceptions.ConnectionError if URL does not exist, but we capture
        # all possible exceptions from trying the link to be safe.
        except Exception as e:
            print(str(e))
            is_broken = True
            status = 'Error'
    # Add result to cache so it won't be tested again.
    link_cache[url] = (is_broken, status)
    return is_broken, status