media_aggregator.py 文件源码

def shorten_news(url, n = 5):
    from bs4 import BeautifulSoup as bs
    from summarizer import FrequencySummarizer as fs
    response = _req.get(url)
    if not response.ok:
        return False
    page = response.content
    soup = bs(page, "lxml")
    summary = fs().summarize("\n".join([x.text for x in soup.findAll("p") if len(x.text.split()) > 1]), n)
    summary.insert(0, soup.title.text)
    return ' '.join(summary)