umlaut 3.3.1 → 4.0.0.beta1

data/test/dummy/config/umlaut_services.yml

@@ -24,6 +24,7 @@ default:
         type: Sfx
         display_name: Find It       
         base_url: http://sfx.library.edu/local
+        disabled: true
         priority: 3      
         roll_up_prefixes:
           - 'EBSCOHOST_'
@@ -248,3 +249,11 @@ default:
         type: SfxBackchannelRecord
         task: link_out_filter
         priority: 6
+
+      # Needs to be here for bx_test.rb, but this isn't a full definition
+      Bx:
+        disabled: true
+        type: Bx
+        priority: 3
+        token: fake
+

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

@@ -0,0 +1,289 @@
+
+  # 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:
+      
+      # 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
+        roll_up_prefixes:
+          - 'EBSCOHOST_'
+          - 'JSTOR_'
+          - 'PROQUEST_'
+          - 'SPRINGER_LINK_'
+          - 'ELSEVIER_SD_'
+          - 'HIGHWIRE_'
+          - 'AIP_'
+          #- 'WILEY_'
+        # 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
+
+      MSEL_Illiad:
+        type: Illiad
+        priority: 4
+        base_url: http://weldoc.welch.jhmi.edu/msel/illiad.dll/OpenURL
+        display_name: "MSE Borrowers DIRECT"
+        sid_suffix: " (via Find It)"
+        notes: This is a thing.
+      
+      #JournalTOCs:
+      #  display_name: JournalTocs
+      #  type: JournalTocs
+      #  priority: 3
+      #  service_type: highlighted_link
+        
+      # 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 4 : foreground
+      GoogleScholarLink:
+        priority: 4
+        type: GoogleScholarLink
+        #preempted_by:
+        #  existing_type: fulltext
+ 
+      # Priority c  : Background      
+            
+      
+      # 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: Catalyst 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][]=Journal%2FNewspaper"
+            title: title
+            author: author
+         exclude_holdings:
+            collection_str:
+              - Internet
+              - Internet Resource
+              - Welch Online Journals
+              - Welch Online Journal
+              - "Gibson-Electronic Journals & Indexes"
+              - "Gibson - Electronic Journals"
+
+      
+      # 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
+
+      HathiTrust:
+        type: HathiTrust
+        priority: c
+        # Uncomment to link through this 'internal' URL instead
+        # of following permalinks. jrochkind does this to use
+        # a WAYFless shibboleth login with EZProxy. 
+        #direct_link_base: 'https://babel.hathitrust.org/shcgi/'
+        #
+        # This next says:
+        # Don't add HathiTrust fulltext if there's already a
+        # GoogleBookSearch fulltext. Still add other HT response
+        # types. 
+        preempted_by:
+          self_type: fulltext
+          existing_service: GoogleBookSearch
+          existing_type: fulltext
+
+      
+      CoverThing:
+        type: CoverThing
+        developer_key: 17005f4154cc260c05164beb2d439dc4
+        priority: c
+        # Pre-empted by says if we already have a cover_image response
+        # from somewhere else that runs earlier, don't run this. 
+        preempted_by:
+          existing_type: cover_image
+          
+      ElsevierCover:
+        type: ElsevierCover
+        priority: c
+        
+      UlrichsCover:
+        type: UlrichsCover
+        priority: c      
+          
+      GoogleBookSearch:
+        type: GoogleBookSearch
+        priority: c        
+        api_key: 'AIzaSyBxgykPTiYe0WQZQobFOEG9m-_NYDRyvTQ'
+        
+        
+      InternetArchive:
+        display_name: the Internet Archive
+        type: InternetArchive
+        priority: c
+        # disable audio results
+        num_results_for_types:
+          audio: 0
+      
+
+
+      # Priority d
+                
+      Ulrichs:
+        display_name: "Ulrich's Directory"
+        type: UlrichsLink
+        priority: d
+        
+      #IsbnDb:
+      #  display_name: isbndb.com
+      #  type: IsbnDb
+      #  priority: d
+      #  access_key: P8DP8QRC
+
+      AllBooksDotCom:
+        type: AllBooksDotCom
+        priority: 3
+
+      ScopusCitations:
+        type: Scopus2
+        priority: d
+        api_key: "caa01d5a0db6377dd054005b625ac9ff"
+        partner_id: E5wmcMWC
+        link_salt_key: "6c.aZT8TEQQO9ugCfGS3qfp(Z.6qCGnC"
+          
+      # Web of Knowledge API has no api key, but your
+      # IP needs to be registered with them. See source docs
+      # for how. 
+      IsiCitations:
+        type: Isi
+        priority: d
+        
+      # Thomson JCR LAMR api has no api key, but your
+      # IP needs to be registered with them. Same registraiton
+      # as for ISI WoK generally. 
+      Jcr:
+        type: Jcr
+        priority: d
+
+          
+        
+      # Worldcat scraper can be quite slow, move it to wave 'd' to avoid
+      # holding up wave 'c'
+      Worldcat:
+        display_name: OCLC Worldcat.org
+        type: Worldcat
+        priority: d
+        # suppress_precheck says don't actually screen-scrape to
+        # see if the link is good, just link blindly. Mostly works
+        # for WorldCat, avoids the performance hit. 
+        suppress_precheck: true
+      
+      
+      WorldcatIdentities:
+        type: WorldcatIdentities
+        priority: d
+        note_types: false
+        wikipedia_link: false
+        openurl_widely_held: false
+        require_identifier: true 
+      
+            
+
+
+      ####
+      # Link out filters: Effect what happens when a user clicks
+      # on an Umlaut link to a third party source. Link out filters
+      # can operate to change where link goes or execute side effects
+      # on click. 
+      ####
+
+      # Redirect outgoing links through EZProxy                
+      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 ]
+
+        
+      # DOES NOT WORK WELL ENOUGH TO BOTHER.
+      #  
+      # Use of SFX api means that SFX can no longer keep statistics on
+      # clickthroughs. This link out filter will attempt to fake SFX into
+      # thinking a direct click happened when user clicks on an SFX link
+      # via Umlaut.  This is VERY fragile logic, required reverse engineering
+      # SFX and faking it out, still doesn't work all the time. But works
+      # much of the time. 
+      #SFX_backchannel_record:
+      #  disabled: true
+      #  type: SfxBackchannelRecord
+      #  task: link_out_filter
+      #  priority: 6

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

@@ -0,0 +1,250 @@
+
+  # 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:
+      
+      # SFX has pubmed lookup built in, so this service not generally
+      # needed with SFX, but included here as an example of it's availability
+      Pubmed:
+        disabled: true
+        priority: 0
+      
+      # Priority 3
+      
+      SFX:
+        type: Sfx
+        display_name: Find It       
+        base_url: http://sfx.library.edu/local
+        priority: 3      
+        roll_up_prefixes:
+          - 'EBSCOHOST_'
+          - 'JSTOR_'
+          - 'PROQUEST_'
+          - 'SPRINGER_LINK_'
+          - 'ELSEVIER_SD_'
+          - 'HIGHWIRE_'
+      
+      # First half of Amazon, run in foreground, get metadata and cover images.
+      Amazon:
+        disabled: true
+        display_name: Amazon.com
+        type: Amazon
+        url: http://webservices.amazon.com/onca/xml
+        api_key: "NEED_API_KEY"
+        secret_key: "NEED_SECRET_KEY"
+        associate_tag: "NEED_ASSOCIATE_ID"
+        priority: 3
+        service_types:
+          - abstract
+          - referent_enhance
+          - cover_image
+          
+      
+      # Priority c  : Background          
+      
+      # 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:
+        type: Amazon
+        disabled: true        
+        api_key: "NEED_API_KEY"
+        secret_key: "NEED_SECRET_KEY"
+        associate_tag: "NEED_ASSOCIATE_ID"
+        priority: c        
+        make_aws_call: false
+        service_types:
+          - highlighted_link
+          - search_inside
+          - excerpts
+          
+
+      
+      CoverThing:
+        type: CoverThing
+        disabled: true        
+        developer_key: YOUR_LT_KEY
+        priority: c
+        # Pre-empted by says if we already have a cover_image response
+        # from somewhere else that runs earlier, don't run this. 
+        preempted_by:
+          existing_type: cover_image
+          
+      ElsevierCover:
+        type: ElsevierCover
+        priority: c
+        
+      UlrichsCover:
+        type: UlrichsCover
+        priority: c      
+          
+      GoogleBookSearch:
+        type: GoogleBookSearch
+        disabled: true
+        priority: c        
+        #api_key: 'YOUR_GBS_API_KEY'
+                
+      DummyService:
+        type: DummyService
+        priority: c      
+        responses:
+          - service_type_value: holding
+            display_text: Library holding
+            url: http://holding.library.edu/DOCID
+
+      Ulrichs:
+        display_name: "Ulrich's Directory"
+        type: UlrichsLink
+        priority: c
+        
+      IsbnDb:
+        display_name: isbndb.com
+        disabled: true
+        type: IsbnDb
+        priority: c
+        access_key: YOUR_ISBNDB_ACCESS_KEY
+        
+      InternetArchive:
+        display_name: the Internet Archive
+        type: InternetArchive
+        priority: c
+        # disable audio results? uncomment:
+        #num_results_for_types:
+        #  audio: 0
+      
+      
+      # Priority d
+                
+      HathiTrust:
+        type: HathiTrust
+        priority: d
+        # Uncomment to link through this 'internal' URL instead
+        # of following permalinks. jrochkind does this to use
+        # a WAYFless shibboleth login with EZProxy. 
+        #direct_link_base: 'https://babel.hathitrust.org/shcgi/'
+        #
+        # This next says:
+        # Don't add HathiTrust fulltext if there's already a
+        # GoogleBookSearch fulltext. Still add other HT response
+        # types. 
+        preempted_by:
+          self_type: fulltext
+          existing_service: GoogleBookSearch
+          existing_type: fulltext
+          
+      EmailExport:
+        type: EmailExport
+        priority: d
+        
+      TxtHoldingExport:
+        type: TxtHoldingExport
+        priority: d
+      
+        
+      
+      ScopusCitations:
+        type: Scopus
+        disabled: true
+        priority: c
+        json_api_key: YOUR_SCOPUS_API_KEY
+        partner_id: YOUR_SCOPUS_PARTNER_ID
+        link_salt_key: "YOUR_SCOPUS_LINK_SALT_KEY"
+        
+      # Web of Knowledge API has no api key, but your
+      # IP needs to be registered with them. See source docs
+      # for how. 
+      IsiCitations:
+        type: Isi
+        disabled: true
+        priority: c
+        
+      # Thomson JCR LAMR api has no api key, but your
+      # IP needs to be registered with them. Same registraiton
+      # as for ISI WoK generally. 
+      Jcr:
+        type: Jcr
+        disabled: true
+        priority: c
+        
+      # Worldcat scraper can be quite slow, move it to wave 'd' to avoid
+      # holding up wave 'c'
+      Worldcat:
+        display_name: OCLC Worldcat.org
+        type: Worldcat
+        priority: d
+        # suppress_precheck says don't actually screen-scrape to
+        # see if the link is good, just link blindly. Mostly works
+        # for WorldCat, avoids the performance hit. 
+        suppress_precheck: true
+      
+      
+      WorldcatIdentities:
+        type: WorldcatIdentities
+        priority: d
+        # Turn off certain response types:
+        #note_types: false
+        #wikipedia_link: false
+        #openurl_widely_held: false
+        #require_identifier: true
+      
+      
+      # GPO is kind of experimental, it doesn't work great, but IF an 
+      # OpenURL includes a sudoc call number or other GPO identifiers,
+      # will try to link to full text via GPO catalog scrape. 
+      Gpo:
+        type: Gpo
+        disabled: true
+        priority: e
+        #preempted_by:
+        #  - existing_type: fulltext
+        #  - existing_type: fulltext_title_level
+
+
+      ####
+      # Link out filters: Effect what happens when a user clicks
+      # on an Umlaut link to a third party source. Link out filters
+      # can operate to change where link goes or execute side effects
+      # on click. 
+      ####
+
+      # Redirect outgoing links through EZProxy                
+      EZProxy:
+        type: Ezproxy
+        disabled: true
+        task: link_out_filter  
+        proxy_server:  HOSTNAME_OF_YOUR_EZPROXY
+        # By default, will pre-check with EZProxy api and
+        # only send links through EZProxy that are proxyable.
+        # requires proxy_password to be set. 
+        # optionally, set precheck_with_api false, and the EZProxy
+        # api won't be used, ALL links go through EZProxy. You may
+        # have EZProxy itself set to transparently redirect non-proxyable
+        # URLs back to non-proxied version.
+        #
+        #precheck_with_api: false
+        priority: 5
+        proxy_password: YOUR_EZPROXY_API_PWD
+        # Want to exclude certain hostnames from being directed through EZProxy?
+        # list them in array here:
+        #exclude: [ host.unversity.edu, otherhost.somewhere.com]
+        
+        
+      # Use of SFX api means that SFX can no longer keep statistics on
+      # clickthroughs. This link out filter will attempt to fake SFX into
+      # thinking a direct click happened when user clicks on an SFX link
+      # via Umlaut.  This is VERY fragile logic, required reverse engineering
+      # SFX and faking it out, still doesn't work all the time. But works
+      # much of the time. 
+      SFX_backchannel_record:
+        disabled: true
+        type: SfxBackchannelRecord
+        task: link_out_filter
+        priority: 6