umlaut 3.1.0 → 3.1.1

data/app/service_adaptors/tr_links.rb

@@ -1,278 +0,0 @@
-# encoding: UTF-8
-
-# Thomson Reuters Links Article Match Retrieval Service 
-#
-# Can generate links to "cited by" and "similar articles" pages for a specific
-# article in Web of Knowledge. 
-#
-# Can also generate links to the Journal Citation Reports (JCR) page for a given
-# journal title. 
-#
-# REQUIREMENTS: You must be an ISI customer if you want these links to actually
-# work for your users. Off-campus users should be sent through EZProxy, see
-# the EZProxy plug-in.
-# 
-# You need to register for the  the Thomson 'Links Article Match Retrieval'
-# (LAMR) service api, which is used here. To register, see:
-# http://wokinfo.com/products_tools/products/related/amr/
-#
-# You register by IP address, so no API key is needed once your registration
-# goes through. 
-#
-# If you later need to change the IP addresses entitled to use this API, use
-# http://scientific.thomson.com/scientific/techsupport/cpe/form.html.
-# to request a change.
-#
-# File support/outage tickets at http://ip-science.thomsonreuters.com/support/
-#
-# Note, as of 13 april 09, there's a bug in the API where journal titles including
-# ampersands cause an error. We will catch those errors and output a 'warning'
-# instead of an 'error', since it's a known problem. 
-class TrLinks < Service
-  require  'open-uri'
-  require 'nokogiri'
-  require 'net/http'
-  require 'builder'
-  
-  include MetadataHelper
-  
-  
-  def service_types_generated
-    types = []
-    types << ServiceTypeValue[:cited_by] if (@include_cited_by || @include_similar) 
-    types << ServiceTypeValue[:highlighted_link] if @include_jcr
-
-    return types
-  end
-
-  def initialize(config)
-    #defaults
-    @wos_app_name     = "Umlaut"
-    @wok_display_name = "Web of Knowledge\xc2\xae" # trademark symbol
-    @display_name     = @wok_display_name
-    @api_url          = "https://ws.isiknowledge.com/cps/xrpc"
-    @include_cited_by = true
-    @include_similar  = true
-    @include_jcr      = true
-    
-    @credits = {
-      @wok_display_name => "http://apps.webofknowledge.com"
-    }
-    
-    super(config)    
-  end
-
-  def handle(request)
-    
-    unless ( sufficient_metadata?(request.referent))
-       return request.dispatched(self, true)
-    end
-    
-    xml = gen_lamr_request(request)
-    
-    isi_response = do_lamr_request(xml)
-    
-    
-    begin
-      #raise if it's an error HTTP response
-      isi_response.value
-    
-      response_xml = Nokogiri::XML(isi_response.body)    
-      # Check for errors.
-      if (error = (response_xml.at('val[@name = "error"]') || response_xml.at("error") || response_xml.at('null[@name = "error"]')))
-        raise IsiResponseException.new("ISI service reported error: #{error.inner_text}")      
-      end
-
-      add_responses( request, response_xml )
-    rescue IsiResponseException => e
-      # Is this the known problem with ampersands?
-      # if so, output a warning, but report success not exception,
-      # because this is a known condition.
-      metadata = request.referent.metadata
-      if ( (metadata["title"] && metadata["title"].include?('&')) ||
-           (metadata["jtitle"] && metadata['jtitle'].include?('&')))
-        Rails.logger.warn("ISI LAMR still exhibiting ampersand problems: #{e.message} ; OpenURL: ?#{request.to_context_object.kev}")
-        return request.dispatched(self, true)
-      else    
-        # Log the error, return exception condition. 
-
-        error_parts = []
-        error_parts << "ISI URI: #{@api_url}"
-        error_parts << "Request XML: #{xml}"
-        error_parts << "ISI Response body: #{response_xml}"
-
-
-        Rails.logger.error("#{e.message} ; " + error_parts.join("\n     ") )
-        return request.dispatched(self, false, e)
-      end
-    end
-    
-    return request.dispatched(self, true)
-  end
-
-  # A DOI is always sufficient. Otherwise, it gets complicated because the ISI
-  # service is kind of picky in weird ways. ISSN alone is not enough, we need
-  # jtitle.  Once you have jtitle, Vol/issue/start page are often enough, but
-  # article title really helps, and jtitle+atitle+year is often enough too. 
-  def sufficient_metadata?(referent)
-    metadata = referent.metadata
-    return get_doi(referent) || get_pmid(referent) ||
-        (  ( metadata['jtitle'] || 
-             metadata['title'] )   &&           
-           (! (metadata['atitle'].blank? ||
-              metadata['date'].blank?
-              ) ||
-            ! ( metadata['volume'].blank? || metadata['issue'].blank? ||
-                metadata['spage'].blank?))
-        )    
-  end
-
-  # produces XML to be posted to Thomson 'Links Article Match Retrieval' service api. 
-  def gen_lamr_request(request)
-    output = ""
-    
-    builder = Builder::XmlMarkup.new(:target => output, :indent => 2)
-    builder.instruct!(:xml, :encoding => "UTF-8")    
-
-    builder.request(:xmlns => "http://www.isinet.com/xrpc41", :src => "app.id=Umlaut") do
-      builder.fn(:name => "LinksAMR.retrieve") do
-        builder.list do
-          # first map is authentication info. empty 'map' element since we are IP authenticated. 
-          builder.map
-          # specify what we're requesting
-          builder.map do
-            builder.list(:name=>"WOS") do
-              builder.val("timesCited")
-              builder.val("ut")
-              builder.val("doi")
-              builder.val("sourceURL")
-              builder.val("citingArticlesURL")
-              builder.val("relatedRecordsURL")
-            end
-          end
-          # specify our query
-          builder.map do
-            builder.map(:name => "cite_id") do
-              # Here's the tricky part, depends on what we've got.
-              metadata = request.referent.metadata
-
-              # DOI
-              if ( doi = get_doi(request.referent))
-                builder.val(doi, :name => "doi")
-              end
-
-              if ( pmid = get_pmid(request.referent))
-                builder.val(pmid, :name => "pmid")
-              end
-
-              # Journal title is crucial for ISI -- ISSN alone is
-              # not enough, weirdly!
-              if ( ! metadata['jtitle'].blank? )
-                builder.val(metadata['jtitle'], :name => "stitle" )
-              else
-                builder.val(metadata['title'], :name => 'stitle' )
-              end
-              
-              # ISSN, not actually used much by ISI, but can't hurt. 
-              if ( issn = request.referent.issn )
-                # ISSN _needs_ a hyphen for WoS, bah!
-                unless issn.match( /\-/ )
-                  issn = issn[0,4] + '-' + issn[4,7]
-                end
-                builder.val(issn, :name => "issn")
-              end
-
-              # article title often helpful. 
-              unless ( metadata['atitle'].blank?)
-                builder.val( metadata['atitle'], :name => "atitle")
-              end
-              # year
-              unless ( metadata['date'].blank?)
-                #first four digits are year
-                builder.val( metadata["date"][0,4], :name => "year" )
-              end
-
-              # Vol/issue/page.  Oddly, issue isn't used very strongly
-              # by ISI, but can't hurt. 
-              unless ( metadata['volume'].blank? )
-                builder.val(metadata['volume'], :name => 'vol')
-              end
-              unless ( metadata['issue'].blank? )
-                builder.val( metadata['issue'] , :name => 'issue')
-              end
-              unless ( metadata['spage'].blank? )
-                builder.val(metadata['spage'], :name => 'spage ')
-              end
-              
-            end
-          end          
-        end
-      end
-    end
-    return output
-  end
-
-  def do_lamr_request(xml)
-    uri = URI.parse(@api_url)
-    http = Net::HTTP.new(uri.host, uri.port)
-    http.use_ssl = true if (uri.scheme == 'https')
-
-    headers = {'Content-Type' => 'application/xml'}
-    
-    return http.post(uri.request_uri, xml, headers)
-  end
-
-  def add_responses(request, response_xml)
-
-    results = response_xml.at('map[@name ="cite_id"] map[@name="WOS"]')
-    unless (results)
-      error_message = "#{self.id}: "
-      error_message << 'Unexpected ISI response. The ISI response was not reported as an error, but did not contain a <map name="WOS"> inside a <map name="cite_id"> as we expected it to:'
-      error_message << "\n ISI XML request:\n#{gen_lamr_request(request)}\n"
-      error_message << "\n ISI http response status: #{isi_response.code}\n"
-      error_message << "\n ISI http response body:\n#{isi_response.body}\n"
-      Rails.logger.error(error_message)
-    end
-
-    
-    # cited by
-    count = results.at('val[@name="timesCited"]')
-    count = count ? count.inner_text.to_i : 0    
-    
-    cited_by_url = results.at('val[@name="citingArticlesURL"]')
-    cited_by_url = cited_by_url.inner_text if cited_by_url
-
-    if (@include_cited_by && count > 0 && cited_by_url )
-      label = ServiceTypeValue[:cited_by].display_name_pluralize.downcase.capitalize    
-      if count && count == 1
-        label = ServiceTypeValue[:cited_by].display_name.downcase.capitalize
-      end
-      
-      request.add_service_response(:service=>self, 
-        :display_text => "#{count} #{label}", 
-        :count=> count, 
-        :url => cited_by_url,
-        :debug_info => "url: " + cited_by_url,
-        :service_type_value => :cited_by)
-    end
-    
-    # similar
-    
-    similar_url = results.at('val[@name ="relatedRecordsURL"]')
-    similar_url = similar_url.inner_text if similar_url
-
-    if (@include_similar && similar_url )
-        request.add_service_response( :service=>self, 
-          :display_text => " #{ServiceTypeValue[:similar].display_name_pluralize.downcase.capitalize}", 
-          :url => similar_url,
-          :debug_info => "url: " + similar_url,
-          :service_type_value => :similar)
-    end
-    
-  end
-  
-end
-
-class IsiResponseException < StandardError
-  
-end

data/lib/tasks/Untitled-1

@@ -1,8 +0,0 @@
-
-
-file = open("Umlaut/log/production.log")
-
-file.each do |line|
-  puts line
-end
-

data/test/dummy/config/umlaut_services-JH.yml

@@ -1,223 +0,0 @@
-default:
-    services:
-      Pubmed:
-        priority: 0
-      
-      # Priority 3
-      
-      JH_SFX:
-        name: Find It @ JH 
-        display_name: Find It @ JH
-        type: Sfx
-        base_url: http://sfx.library.jhu.edu:8000/sfxlcl41 #http://sfx.library.jhu.edu:8000/sfxlcl3
-        priority: 3
-        status: active
-        # click_passthrough: true
-        click_passthrough: false
-        sfx_requests_expire_crontab: '00 00 * * 00' 
-        sfx_timeout: 9
-        # We do NOT include abstract and ToC from SFX.  
-        services_of_interest:
-          getFullTxt: fulltext
-          getSelectedFullTxt: fulltext
-          getDocumentDelivery: document_delivery                         
-          getDOI: highlighted_link
-        extra_targets_of_interest: 
-            LOCAL_FEEDBACK_MSE_LCL: help
-            LOCAL_FEEDBACK_SAIS_LCL: help
-            LOCAL_FEEDBACK_WELCH_LCL: help
-            LOCAL_FEEDBACK_AFL_LCL: help
-            LOCAL_FEEDBACK_APL_LCL: help
-      
-      # We have experimented with doing BL in the foreground wave 3 to
-      # get overall faster response, but for now leaving it at 'c' in bg
-      # where HIP was, to avoid slowing down the initial screen display. 
-      Blacklight:
-         display_name: Blacklight JH Libraries Catalog
-         type: Blacklight
-         priority: c
-         base_url: https://catalyst.library.jhu.edu/catalog.atom
-         rft_id_bibnum_prefixes:
-            - http://catalog.library.jhu.edu/bib/
-            - http://blacklight.mse.jhu.edu/catalog/
-            - https://blacklight.mse.jhu.edu/catalog/
-            - http://blacklight.mse.jhu.edu/demo/catalog/
-            - https://blacklight.mse.jhu.edu/demo/catalog/
-            - https://catalyst.library.jhu.edu/catalog/
-            - http://catalyst.library.jhu.edu/catalog/
-         bl_fields:
-            issn: lsolr.issn
-            isbn: isbn
-            lccn: lccn
-            oclcnum: oclcnum
-            id: id
-            serials_limit_clause: "f[format][]=Serial"
-            title: title
-            author: author
-         exclude_holdings:
-            collection_str:
-              - Internet
-              - Internet Resource
-              - Welch Online Journals
-              - Welch Online Journal
-              - "Gibson-Electronic Journals & Indexes"
-              - "Gibson - Electronic Journals"
-      
-        
-      # First half of Amazon, run in foreground, get metadata and cover images.
-      Amazon:
-        display_name: Amazon.com
-        type: Amazon
-        url: http://webservices.amazon.com/onca/xml
-        api_key: "1S1R4KE79F07X8AS0JG2"
-        secret_key: "lgHW00st36oGlN6YJ4p6wUH1CEt+RygbwVeCWQYm"
-        associate_tag: "johnhopklibr-20"
-        priority: 3
-        status: active
-        service_types:
-          - abstract
-          - referent_enhance
-          - cover_image
-          
-      
-      # Priority c              
-      
-      # Second half of Amazon. It's Slow to lookup highlighted_link and search_inside
-      # availability, so we do it in a bg wave.  
-      Amazon_bg:
-        display_name: Amazon.com
-        type: Amazon
-        url: http://webservices.amazon.com/onca/xml
-        api_key: 1S1R4KE79F07X8AS0JG2
-        associate_tag: "johnhopklibr-20"
-        priority: c
-        status: active
-        make_aws_call: false
-        service_types:
-          - highlighted_link
-          - search_inside
-          - excerpts
-          
-
-      
-      CoverThing:
-        display_name: LibraryThing
-        type: CoverThing
-        developer_key: 17005f4154cc260c05164beb2d439dc4
-        priority: c
-        preempted_by:
-          existing_type: cover_image
-          
-      ElsevierCover:
-        type: ElsevierCover
-        priority: c
-        
-      UlrichsCover:
-        type: UlrichsCover
-        priority: c      
-          
-      GoogleBookSearch:
-        name: Google Book Search
-        type: GoogleBookSearch
-        priority: c
-        status: active
-        api_key: 'AIzaSyBxgykPTiYe0WQZQobFOEG9m-_NYDRyvTQ'
-                
-      Ulrichs:
-        display_name: Ulrich's Directory
-        type: UlrichsLink
-        priority: c
-        
-      IsbnDb:
-        display_name: isbndb.com
-        type: IsbnDb
-        priority: c
-        access_key: P8DP8QRC
-        
-      InternetArchive:
-        display_name: the Internet Archive
-        type: InternetArchive
-        priority: c
-        num_results_for_types:
-          audio: 0
-      
-      
-      # Priority d
-                
-      HathiTrust:
-        type: HathiTrust
-        priority: d
-        direct_link_base: 'https://babel.hathitrust.org/shcgi/'
-        preempted_by:
-          self_type: fulltext
-          existing_service: GoogleBookSearch
-          existing_type: fulltext
-          
-      EmailExport:
-        type: EmailExport
-        priority: d
-        
-      TxtHoldingExport:
-        type: TxtHoldingExport
-        priority: d
-      
-      # Worldcat scraper is quite slow, move it to wave 'd' to avoid
-      # holding up wave 'c'
-      Worldcat:
-        display_name: OCLC Worldcat.org
-        type: Worldcat
-        priority: d
-        display_text: "Find in other libraries"
-        suppress_precheck: true
-      
-      
-      WorldcatIdentities:
-        type: WorldcatIdentities
-        priority: d
-        #note_types: false
-        #wikipedia_link: false
-        #openurl_widely_held: false
-        #require_identifier: true  
-      
-      ScopusCitations:
-        type: Scopus
-        priority: c
-        json_api_key: 3Jek0AJKkIzcSUgArdeWIAeFo0pWj7
-        partner_id: E5wmcMWC
-        link_salt_key: "6c.aZT8TEQQO9ugCfGS3qfp(Z.6qCGnC"
-        
-      IsiCitations:
-        type: Isi
-        priority: c
-        
-      Jcr:
-        type: Jcr
-        priority: c
-        
-      Gpo:
-        type: Gpo
-        priority: e
-        #preempted_by:
-        #  - existing_type: fulltext
-        #  - existing_type: fulltext_title_level
-      
-      # Link out filter                
-      JH_EZProxy:
-        display_name: EZProxy
-        type: Ezproxy
-        task: link_out_filter  
-        proxy_server:  proxy.library.jhu.edu
-        # Don't bother prechecking with api, just send all clicks
-        # through EZProxy, EZProxy will redirect if not proxyable. 
-        precheck_with_api: false
-        priority: 5
-        proxy_password: d1NWbo
-        # Problems in our EZProxy setup have EZProxy proxying all
-        # jhu.edu links! We want to exclude most of them though. 
-        exclude: [ sfx.library.jhu.edu, catalog.library.jhu.edu, findit.library.jhu.edu , aplill.jhu.edu, ils.library.jhu.edu, saisill.jhu.edu, aplill.jhu.edu, aflill.peabody.jhu.edu, weldoc.welch.jhmi.edu ]
-        
-      SFX_backchannel_record:
-        display_name: SFX statistics
-        type: SfxBackchannelRecord
-        task: link_out_filter
-        priority: 6

data/test/dummy/config/umlaut_services-TEST.yml

@@ -1,37 +0,0 @@
-
-  # Configure what service plugins are used by Umlaut. This skeleton file
-  # has been generated into your app to help you get started. 
-  #
-  # If a service has "disabled:true", it's currently turned off. 
-  #
-  # Some services require local api key or connection details as config.
-  # Most services take other options for custom configuration too, not
-  # all options are neccesarily listed as examples here, see source
-  # or source-generated docs for more info.   
-
-default:
-    services:
-      JH_SFX:
-        name: Find It @ JH 
-        display_name: Find It @ JH
-        type: Sfx
-        base_url: http://sfx.library.jhu.edu:8000/sfxlcl41 #http://sfx.library.jhu.edu:8000/sfxlcl3
-        priority: 3
-        status: active
-        # click_passthrough: true
-        click_passthrough: false
-        sfx_requests_expire_crontab: '00 00 * * 00' 
-        sfx_timeout: 9
-        # We do NOT include abstract and ToC from SFX.  
-        services_of_interest:
-          getFullTxt: fulltext
-          getSelectedFullTxt: fulltext
-          getDocumentDelivery: document_delivery                         
-          getDOI: highlighted_link
-        extra_targets_of_interest: 
-            LOCAL_FEEDBACK_MSE_LCL: help
-            LOCAL_FEEDBACK_SAIS_LCL: help
-            LOCAL_FEEDBACK_WELCH_LCL: help
-            LOCAL_FEEDBACK_AFL_LCL: help
-            LOCAL_FEEDBACK_APL_LCL: help
-