pubmed_api 0.0.2 → 0.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: dd746c28ff6467d98cc9453f4099ebb3a51be2f5
-  data.tar.gz: 466279c7008a90605975055c48e1e011818b4ab8
+  metadata.gz: 19077723b1f29404e04bc59e2bb97f2c6d516b7c
+  data.tar.gz: 086727eb2181e56a9c95067e8707354addf23cf6
 SHA512:
-  metadata.gz: b886ca7fc3aa85f3a60dfd33653b6fff9c8152034550c6b78f2884f74a25dc391b9c00b6f14faf21feab79bb97460a9b4f728029621a7d10b1a2b2078e98b294
-  data.tar.gz: 93b6be04e8817a988b8c710d35c22db2c606e98fa6834ef9f08af98b33191c8aba6e4b4779a2a05e0b157245aa9ccbeaa0b69ee327baadc1696a5a2890e3752f
+  metadata.gz: 2dc9e9e7caca527a127804c35ca78d391f86c34dbffc120e3552a87fe156326be434a4c22d5f2eed962153fdfa1a0b00885336b80480b7cdaed1b5c7d1accbfe
+  data.tar.gz: d4b68d0317e3f78b70c02bade94a1360ae32c9ffa4a7f7c774723289012e85b8925618ad0ebd2c41442c6249c31fc8875f016f59ba046b7c43c22f0107081399

data/README.md

@@ -20,7 +20,21 @@ Or install it yourself as:
 
 ## Usage
 
-TODO: Write usage instructions here
+This is a work in progress. But you can use it 
+
+To search for papers:
+
+results = PubmedAPI::Interface.search("quantum physics") 
+results.pmids gives you a list of the matching pubmed ids
+
+To get a paper:
+
+strucs = PubmedAPI::Interface.fetch_papers([id])
+paper = struc[0]
+paper.title = "A paper title"
+paper.url = "http://alinktofulltext.com"
+
+Look in the spec for further examples 
 
 ## Contributing
 

data/lib/pubmed_api.rb

@@ -11,17 +11,18 @@ module PubmedAPI
  
 
     DEFAULT_OPTIONS = {:tool => 'ruby-pubmed-api',
-                       :database => 'pubmed', #which database eq pubmed/nlmcatalog
+                       :database => 'db=pubmed', #which database eq pubmed/nlmcatalog
                        :verb => 'search', #which API verb to use e.g. search/fetch
                        :email => '',
-                       #:reldate => 90, #How far back shall we go in days 
+                       #:reldate => 90, #How far back shall we go in days
+                       :add =>'', 
                        :retmax => 100000,
                        :retstart => 0,
                        :load_all_pmids => true }
                      
 
-    URI_TEMPLATE = 'http://eutils.ncbi.nlm.nih.gov/entrez/eutils/e{verb}.fcgi?db={database}&tool={tool}&email={email}'+
-                   '&reldate={reldate}&retmax={retmax}&retstart={retstart}&{query}&rettype=fasta&retmode=xml'
+    URI_TEMPLATE = 'http://eutils.ncbi.nlm.nih.gov/entrez/eutils/e{verb}.fcgi?{database}&tool={tool}&email={email}'+
+                   '&reldate={reldate}&retmax={retmax}&retstart={retstart}&{query}&retmode=xml&{add}'
 
     class << self
 
@@ -50,24 +51,44 @@ module PubmedAPI
       end
 
       def fetch_papers(ids)
-        xml = fetch_records(ids, 'pubmed')
+        xml = fetch_records(ids, {:verb => 'fetch',:database => 'db=pubmed'})
         parser = XMLParser.new
-        parser.parse_papers(xml)
+        papers = parser.parse_papers(xml)
+        lookup_hash = get_fulltext_links(ids)
+        
+        papers.each do |p|
+          if p.nil?
+             next
+          else
+            p.url =  lookup_hash[p.pmid].first.url 
+          end
+        end
       end
 
       def fetch_journals(nlmids)
         #Change the ids of those wierd journals 
         nlmids = nlmids.map { |e|  ((e.include? 'R') ? convert_odd_journal_ids(e) : e ) }
-        xml = fetch_records(nlmids, 'nlmcatalog')
+        xml = fetch_records(nlmids, {:verb => 'fetch',:database => 'db=nlmcatalog'})
         parser = XMLParser.new
         parser.parse_journals(xml)       
       end
 
-      def fetch_records(ids, database)
+     def get_fulltext_links(ids)
+       opts = {:verb => 'link',  :add => 'cmd=llinks', :database => 'dbfrom=pubmed'}
+       xml = fetch_records(ids, opts)
+
+       parser = XMLParser.new
+       lookup_hash = parser.parse_links(xml)
+       missing = (ids - lookup_hash.keys)
+       lookup_hash
+     end
+
+
+      def fetch_records(ids, opts={})
 
         xml_records = []
         
-        options = DEFAULT_OPTIONS
+        options = DEFAULT_OPTIONS.merge(opts)
 
         #dice array into reasonable length chunks for download
         n_length = 500
@@ -76,14 +97,18 @@ module PubmedAPI
       
           #Turn string to something html friendly 
           id_string = slice.join(",")
-          doc = make_api_request(options.merge({:verb => 'fetch',:database => database, :query => 'id='+id_string}))
+          doc = make_api_request(options.merge({ :query => 'id='+id_string}))
           records = doc.xpath('./*/*')
-          xml_records << records
+          xml_records += records
 
         end
-        xml_records.flatten
+
+        xml_records
       end
 
+      
+
+
       #Maked the HTTP request and return the responce
       #TODO handle failures
       def make_api_request(options)
@@ -96,7 +121,7 @@ module PubmedAPI
       def convert_odd_journal_ids(id)
         
         new_id = nil
-        results = search(id, {:database => 'nlmcatalog'})
+        results = search(id, {:database => 'db=nlmcatalog'})
         if results.pmids.length ==1
           new_id = results.pmids[0]
         else
@@ -111,7 +136,7 @@ module PubmedAPI
         id = nil
         term = issn + "[ISSN]+AND+ncbijournals[filter]"
 
-        results = search(term, {:database => 'nlmcatalog'})
+        results = search(term, {:database => 'db=nlmcatalog'})
         if results.pmids.length ==1
           id = results.pmids[0]
         else
@@ -122,6 +147,9 @@ module PubmedAPI
       end
 
 
+ 
+
+
       # 300ms minimum wait.
       def wait
         sleep WAIT_TIME 

data/lib/pubmed_api/parsers.rb

@@ -13,7 +13,7 @@ module PubmedAPI
 
   	  results.count = doc.xpath('/eSearchResult/Count').first.content.to_i
 
-  	  doc.xpath('/eSearchResult/IdList/Id').each {|n| results.pmids << n.content.to_i}
+  	  doc.xpath('/eSearchResult/IdList/Id').each {|n| results.pmids << n.content.to_s}
   	      
   	  doc.xpath('/eSearchResult/TranslationStack/TermSet/Term').each do |n|
   	    if n.content =~ /"(.*)"\[MeSH Terms\]/
@@ -134,6 +134,33 @@ module PubmedAPI
       pmid
     end
 
+    
+    LinkStruct = Struct.new( :url, :pub_id, :pub_name, :cat)
+
+    def parse_links(links_xml)
+      
+      l_struc_arr = []
+      link_arr = []
+      lookup_hash = Hash.new{ |a,b| a[b] = Array.new }
+      
+      links_xml.each do |node|
+        
+        node.css('IdUrlList/IdUrlSet').each do |links|
+          id = links.xpath('Id').text
+
+          links.css('ObjUrl').each do |l|
+            l_struc = LinkStruct.new(l.xpath('Url').text, l.xpath('Provider/Id').text, l.xpath('Name').text,
+                                        l.xpath('Category').text)
+ 
+            lookup_hash[id] << l_struc
+          end
+        end
+      end      
+      
+      lookup_hash
+    end
+
+
 
     AuthorStruct = Struct.new( :fore_name, :initials, :last_name)
     

data/lib/pubmed_api/version.rb

@@ -1,3 +1,3 @@
 module PubmedAPI
-  VERSION = "0.0.2"
+  VERSION = "0.0.3"
 end

data/spec/lib/pubmed_api_spec.rb

@@ -2,10 +2,13 @@ require 'spec_helper'
 
 describe PubmedAPI do
 
+  before(:all)  do 
+    @q_results = PubmedAPI::Interface.search("quantum physics") 
+  end
+
 
   it "should perform a search" do
-    results = PubmedAPI::Interface.search("quantum physics", {:reldate => 90})    
-    expect(results.pmids.length).to be > 10
+    expect(@q_results.pmids.length).to be > 10
   end
 
   it "should handle phrases not found" do
@@ -29,10 +32,12 @@ describe PubmedAPI do
   it "should fetch a paper" do
     id = '25554862'
     title = "Completing the picture for the smallest eigenvalue of real Wishart matrices."
+    url = "http://link.aps.org/abstract/PRL/v113/p250201"
     strucs = PubmedAPI::Interface.fetch_papers([id])
     paper = strucs[0]
     expect(paper.title).to eql(title)
     expect(paper.pmid).to eql(id)
+    expect(paper.url).to eql(url)
   end 
 
   it "should fetch a journal" do
@@ -54,4 +59,14 @@ describe PubmedAPI do
      expect(fixed).to eql('100381')
   end
 
+  it "should get fulltext urls for ids" do
+    ids = ["25933345","25933144","23933344"]
+
+    hash = PubmedAPI::Interface.get_fulltext_links(ids)
+    paper = hash[ids[0]][0]
+    expect(paper.url).to eql('http://link.aps.org/abstract/PRL/v114/p158701')
+  end 
+
+
+
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pubmed_api
 version: !ruby/object:Gem::Version
-  version: 0.0.2
+  version: 0.0.3
 platform: ruby
 authors:
 - Kieran Higgins
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-05-03 00:00:00.000000000 Z
+date: 2015-05-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler