def process_one_file(bibcode, fname, provider):
ext = fname.split('.')[-1]
d = {
CONSTANTS["BIBCODE"]: bibcode,
CONSTANTS["PROVIDER"]: provider,
CONSTANTS["FORMAT"]: ext,
CONSTANTS["FILE_SOURCE"]: fname
}
extractor = StandardFileExtract.StandardExtractorXML(d)
extractor.open_xml()
xml = extractor.parse_xml()
sections = xml.xpath(sections_xpath) or xml.xpath(paragraphs_xpath)
summary = sections[-1].text_content()
sys.stderr.write("summary is of type {}\n".format(type(summary)))
text = TextCleaner(text=summary)
if sections:
summary = unicode(sections[-1].text_content())
if summary:
text = TextCleaner(text=summary).run()
return text