Sha256: 5d8cbd6877611ffd54e84ac3b769da6835bf1678ce0efb508e45fe9111430801
Contents?: true
Size: 1.57 KB
Versions: 65
Compression:
Stored size: 1.57 KB
Contents
module Onebox
module Engine
class PubmedOnebox
include Engine
include LayoutSupport
matches_regexp Regexp.new("^https?://(?:(?:\\w)+\\.)?(www.ncbi.nlm.nih)\\.gov(?:/)?/pubmed/\\d+")
private
def get_xml
doc = Nokogiri::XML(open(URI.join(@url, "?report=xml&format=text")))
pre = doc.xpath("//pre")
Nokogiri::XML("<root>" + pre.text + "</root>")
end
def authors_of_xml(xml)
initials = xml.css("Initials").map { |x| x.content }
last_names = xml.css("LastName").map { |x| x.content }
author_list = (initials.zip(last_names)).map { |i, l| i + " " + l }
if author_list.length > 1 then
author_list[-2] = author_list[-2] + " and " + author_list[-1]
author_list.pop
end
author_list.join(", ")
end
def date_of_xml(xml)
date_arr = (xml.css("PubDate").children).map { |x| x.content }
date_arr = date_arr.select { |s| !s.match(/^\s+$/) }
date_arr = (date_arr.map { |s| s.split }).flatten
date_arr.sort.reverse.join(" ") # Reverse sort so month before year.
end
def data
xml = get_xml()
{
title: xml.css("ArticleTitle").text,
authors: authors_of_xml(xml),
journal: xml.css("Title").text,
abstract: xml.css("AbstractText").text,
date: date_of_xml(xml),
link: @url,
pmid: match[:pmid]
}
end
def match
@match ||= @url.match(%r{www\.ncbi\.nlm\.nih\.gov/pubmed/(?<pmid>[0-9]+)})
end
end
end
end
Version data entries
65 entries across 65 versions & 1 rubygems