processors.py 文件源码

def extract_image_url(text):
    text = _strip_url(text)
    imgurl = None
    if text:
        # check if the text is style content
        match = _CSS_IMAGERE.search(text)
        text = match.groups()[0] if match else text
        parsed = urlparse(text)
        path = None
        match = _IMAGE_PATH_RE.search(parsed.path)
        if match:
            path = match.group()
        elif parsed.query:
            match = _GENERIC_PATH_RE.search(parsed.path)
            if match:
                path = match.group()
        if path is not None:
            parsed = list(parsed)
            parsed[2] = path
            imgurl = urlunparse(parsed)
        if not imgurl:
            imgurl = text
    return imgurl