Apache Beam python无法解析pubmed XML

Question

嗨，我写了一个beam管道来读取一个目录并使用pubmed_解析库解析下载的pubmed xml文件。该库通过标准python程序运行良好，但如果我将其转换为apache beam pipeline，则在解析过程中失败，并出现错误： 希望你能帮我解决这个问题

File "/home/micdsouz/venv/medline/data-preprocessing.py", line 19, in process
    pubmed_dict = pp.parse_pubmed_xml(element)
  File "/home/micdsouz/venv/local/lib/python2.7/site-packages/pubmed_parser/pubmed_oa_parser.py", line 112, in parse_pubmed_xml
    dict_article_meta = parse_article_meta(tree)
  File "/home/micdsouz/venv/local/lib/python2.7/site-packages/pubmed_parser/pubmed_oa_parser.py", line 60, in parse_article_meta
    pmid_node = article_meta.find('article-id[@pub-id-type="pmid"]')
AttributeError: 'NoneType' object has no attribute 'find' [while running 'ReadData']

---

^{pr2}$

Answer 1

这看起来是pubmed_parser库有问题。他们的跟踪器中有两个问题似乎与您的问题相符：

• https://github.com/titipata/pubmed_parser/issues/41
• https://github.com/titipata/pubmed_parser/issues/44