jaq.py 文件源码

def process_content(html,item_dict):
    root = Xhtml.fromstring(html)
    # ??????
    try:
        content = root.xpath('//*[@class="article-content"]')[0]
    except IndexError:
        return ''
    #
    item_dict['cover'] = None
    imgs = root.xpath('//img[@src]')
    if imgs:
        for img in imgs:
            src=img.attrib['src'].strip()
            if (not item_dict['cover']) and  src[-3:].lower() in ['jpg','png','gif'] :
                item_dict['cover']='http:'+src
                # ????
                coverelement = Element('img')
                coverelement.set('src', item_dict['cover'])
                content.insert(0, coverelement)
            elif src[:22]=="data:image/png;base64,":
                img.set("src","")
            else:
                pass


    item_dict['content'] = Xhtml.tostring(content, encoding='unicode')
    #
    print "++++\tGet jaq items\t++++"
    print item_dict['cover']
    print item_dict['created']
    print item_dict['title']
    print item_dict['desc']
    print item_dict['link']
    return item_dict