WIP - notes from review and tests

one failing bc Item.public is returning private items.
ualbertalib · Jan 24, 2018 · 7497c12 · 7497c12
1 parent dd8aaca
commit 7497c12
Show file tree

Hide file tree

Showing 13 changed files with 487 additions and 23 deletions.
diff --git a/app/controllers/sitemap_controller.rb b/app/controllers/sitemap_controller.rb
@@ -6,22 +6,23 @@ def index; end
 
   def communities
     @communities = Community.all
-    raise 'sitemap should contain less than 50,000 targets' if @communities.count > 50_000
+    # TODO: consider using Rollbar to catch this kind of thing
+    logger.warn 'communities sitemap should contain less than 50,000 targets' if @communities.count > 50_000
   end
 
   def collections
     @collections = Collection.all
-    raise 'sitemap should contain less than 50,000 targets' if @collections.count > 50_000
+    logger.warn 'collections sitemap should contain less than 50,000 targets' if @collections.count > 50_000
   end
 
   def items
     @items = Item.public
-    raise 'sitemap should contain less than 50,000 targets' if @items.count > 50_000
+    logger.warn 'items sitemap should contain less than 50,000 targets' if @items.count > 50_000
   end
 
   def theses
-    @theses = Thesis.all
-    raise 'sitemap should contain less than 50,000 targets' if @theses.count > 50_000
+    @theses = Thesis.public
+    logger.warn 'thesis sitemap should contain less than 50,000 targets' if @theses.count > 50_000
   end
 
 end
diff --git a/app/models/concerns/item_properties.rb b/app/models/concerns/item_properties.rb
@@ -150,7 +150,7 @@ def valid_visibilities
     end
 
     def public
-      Item.where(visibility: JupiterCore::VISIBILITY_PUBLIC)
+      self.where(visibility: JupiterCore::VISIBILITY_PUBLIC)
     end
   end
 

diff --git a/app/views/items/_google_scholar_metadata.html.erb b/app/views/items/_google_scholar_metadata.html.erb
@@ -11,11 +11,3 @@
 <% end %>
 <meta name="dc.identifier" content="<%= @item.doi %>"/>
 <meta name="citation_doi" content="<%= @item.doi %>"/>
-<!-- does not support these metadata -->
-<!--
-  <meta name="citation_journal_title" content=""/>
-  <meta name="citation_volume" content=""/>
-  <meta name="citation_issue" content=""/>
-  <meta name="citation_firstpage" content=""/>
-  <meta name="citation_lastpage" content=""/>
--->
diff --git a/app/views/sitemap/_object.xml.builder b/app/views/sitemap/_object.xml.builder
@@ -2,20 +2,21 @@ xml.instruct! :xml, version: '1.0'
 cache 'sitemap', expires_in: 24.hours do
   xml.urlset(xmlns: 'http://www.sitemaps.org/schemas/sitemap/0.9',
              'xmlns:rs' => 'http://www.openarchives.org/rs/terms/') do
-    xml.rs :md, capability: 'resourcelist', at: Time.current.utc.iso8601
+    xml.rs :md, capability: 'resourcelist', at: Time.current.iso8601
     xml.url do
       xml.loc        root_url
-      xml.lastmod    Time.current.utc.iso8601
+      xml.lastmod    Time.current.iso8601
       xml.changefreq 'weekly'
       xml.priority   1
+      xml.rs :md, type: 'text/html'
     end
 
     objects.each do |object|
       xml.url do
         xml.loc item_url(object)
+        xml.lastmod object.updated_at
         xml.changefreq 'weekly'
         xml.priority   1
-        xml.lastmod object.updated_at
         xml.rs :md, type: 'text/html'
         object.file_sets.each do |file_set|
           xml << "\t#{file_set.sitemap_link}\n"

diff --git a/app/views/sitemap/collections.xml.builder b/app/views/sitemap/collections.xml.builder
@@ -11,9 +11,9 @@ cache 'sitemap', expires_in: 24.hours do
     @collections.each do |collection|
       xml.url do
         xml.loc community_collection_url(collection.community, collection)
+        xml.lastmod collection.updated_at
         xml.changefreq 'weekly'
         xml.priority   1
-        xml.lastmod collection.updated_at
       end
     end
   end

diff --git a/app/views/sitemap/communities.xml.builder b/app/views/sitemap/communities.xml.builder
@@ -11,9 +11,9 @@ cache 'sitemap', expires_in: 24.hours do
     @communities.each do |community|
       xml.url do
         xml.loc community_url(community)
+        xml.lastmod community.updated_at
         xml.changefreq 'weekly'
         xml.priority   1
-        xml.lastmod community.updated_at
       end
     end
   end

diff --git a/config/routes.rb b/config/routes.rb
@@ -51,10 +51,10 @@
     mount Sidekiq::Web => '/sidekiq', constraints: AdminConstraint.new
   end
 
-  get 'sitemap.xml', to: 'sitemap#index', defaults: { format: :xml }
-  get 'sitemap-communities.xml', to: 'sitemap#communities', defaults: { format: :xml }
-  get 'sitemap-collections.xml', to: 'sitemap#collections', defaults: { format: :xml }
-  get 'sitemap-items.xml', to: 'sitemap#items', defaults: { format: :xml }
+  get 'sitemap.xml', to: 'sitemap#index', defaults: { format: :xml }, as: :sitemapindex
+  get 'sitemap-communities.xml', to: 'sitemap#communities', defaults: { format: :xml }, as: :communities_sitemap
+  get 'sitemap-collections.xml', to: 'sitemap#collections', defaults: { format: :xml }, as: :collections_sitemap
+  get 'sitemap-items.xml', to: 'sitemap#items', defaults: { format: :xml }, as: :items_sitemap
   get 'sitemap-theses.xml', to: 'sitemap#theses', defaults: { format: :xml }
   root to: 'welcome#index'
 end
diff --git a/test/controllers/resourcesync.xsd b/test/controllers/resourcesync.xsd
@@ -0,0 +1,147 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema" targetNamespace="http://www.openarchives.org/rs/terms/" xmlns="http://www.openarchives.org/rs/terms/" elementFormDefault="qualified">
+
+<xsd:annotation>
+<xsd:documentation>
+XML Schema for ResoureSync extensions to Sitemap files.
+See: http://www.openarchives.org/rs/ .
+2013-01-06 first stab [Simeon Warner],
+2013-08-15 for v0.9.1, v1.0 [Simeon Warner],
+2016-04-26 for v1.1 [Simeon Warner].
+</xsd:documentation>
+</xsd:annotation>
+
+  <xsd:element name="md">
+    <xsd:annotation>
+      <xsd:documentation>
+	Metadata about a resource
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:complexType>
+      <xsd:complexContent>
+        <xsd:restriction base="xsd:anyType">
+          <xsd:attribute name="at" type="xsd:dateTime"/>
+          <xsd:attribute name="capability" type="xsd:string"/>
+          <xsd:attribute name="change" type="changeType"/>
+          <xsd:attribute name="completed" type="xsd:dateTime"/>
+          <xsd:attribute name="datetime" type="xsd:dateTime"/>
+          <xsd:attribute name="encoding" type="xsd:string"/>
+          <xsd:attribute name="from" type="xsd:dateTime"/>
+          <xsd:attribute name="hash" type="xsd:string"/>
+          <xsd:attribute name="length" type="xsd:integer"/>
+          <xsd:attribute name="modified" type="xsd:dateTime"/>
+          <xsd:attribute name="path" type="pathType"/>
+          <xsd:attribute name="type" type="contentType"/>
+          <xsd:attribute name="until" type="xsd:dateTime"/>
+	  <xsd:anyAttribute namespace="##other" processContents="lax"/>
+        </xsd:restriction>
+      </xsd:complexContent>
+    </xsd:complexType>
+  </xsd:element>
+
+  <xsd:element name="ln">
+    <xsd:annotation>
+      <xsd:documentation>
+	Links to related resources
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:complexType>
+      <xsd:complexContent>
+        <xsd:restriction base="xsd:anyType">
+          <xsd:attribute name="encoding" type="xsd:string"/>
+          <xsd:attribute name="hash" type="xsd:string"/>
+          <xsd:attribute name="href" type="xsd:anyURI"/>
+          <xsd:attribute name="length" type="xsd:nonNegativeInteger"/>
+          <xsd:attribute name="modified" type="xsd:dateTime"/>
+          <xsd:attribute name="path" type="pathType"/>
+          <xsd:attribute name="pri" type="priType"/>
+          <xsd:attribute name="rel" type="relType"/>
+          <xsd:attribute name="type" type="contentType"/>
+	  <xsd:anyAttribute namespace="##other" processContents="lax"/>
+        </xsd:restriction>
+      </xsd:complexContent>
+    </xsd:complexType>
+  </xsd:element>
+
+  <xsd:simpleType name="changeType">
+    <xsd:annotation>
+      <xsd:documentation>
+        Allowed values of the change attribute are
+	created, updated, deleted
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:restriction base="xsd:string">
+      <xsd:enumeration value="created"/>
+      <xsd:enumeration value="updated"/>
+      <xsd:enumeration value="deleted"/>
+    </xsd:restriction>
+  </xsd:simpleType>
+
+  <xsd:simpleType name="pathType">
+    <xsd:annotation>
+      <xsd:documentation>
+        Path values must start with a slash, must not end with a slash
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:restriction base="xsd:string">
+      <xsd:pattern value="/.*[^/]"/>
+    </xsd:restriction>
+  </xsd:simpleType>
+
+  <xsd:simpleType name="contentType">
+    <xsd:annotation>
+      <xsd:documentation>
+        MIME Content Types are described in RFCs 2045,2046
+        http://tools.ietf.org/html/rfc2045
+        http://tools.ietf.org/html/rfc2046
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:restriction base="xsd:string">
+      <xsd:pattern value="[\w-]+/[\w-]+"/>
+    </xsd:restriction>
+  </xsd:simpleType>
+
+  <xsd:simpleType name="priType">
+    <xsd:annotation>
+      <xsd:documentation>
+        The priority attribute may have values 1 through 999999
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:restriction base="xsd:integer">
+      <xsd:minInclusive value="1"/>
+      <xsd:maxInclusive value="999999"/>
+    </xsd:restriction>
+  </xsd:simpleType>
+
+  <xsd:simpleType name="relType">
+    <xsd:annotation>
+      <xsd:documentation>
+	Syntax for link relation values is specified by
+	http://tools.ietf.org/html/rfc5988 and the relevant
+	portion is:
+
+	# link-param     = ( ( "rel" "=" relation-types )
+	#	           ... 
+	# relation-type  = reg-rel-type | ext-rel-type 
+	# reg-rel-type   = LOALPHA *( LOALPHA | DIGIT | "." | "-" ) 
+	# ext-rel-type   = URI
+	#
+
+	Registered link relations (reg-rel-type) are listed at
+	http://www.iana.org/assignments/link-relations/link-relations.xhtml
+	and URIs are allowed for extension (ext-rel-type)
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:union>
+      <xsd:simpleType>
+	<xsd:restriction base="xsd:string">
+          <xsd:pattern value="[a-z][a-z0-9\.\-]*"/>
+        </xsd:restriction>
+      </xsd:simpleType>
+      <xsd:simpleType>
+	<xsd:restriction base="xsd:anyURI"/>
+      </xsd:simpleType>
+    </xsd:union>
+  </xsd:simpleType>
+
+</xsd:schema>
diff --git a/test/controllers/siteindex.xsd b/test/controllers/siteindex.xsd
@@ -0,0 +1,72 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema" targetNamespace="http://www.sitemaps.org/schemas/sitemap/0.9" xmlns="http://www.sitemaps.org/schemas/sitemap/0.9" elementFormDefault="qualified">
+  <xsd:annotation>
+    <xsd:documentation>
+      XML Schema for Sitemap index files.
+      Last Modifed 2009-04-08
+    </xsd:documentation>
+  </xsd:annotation>
+
+  <xsd:element name="sitemapindex">
+    <xsd:annotation>
+      <xsd:documentation>
+        Container for a set of up to 50,000 sitemap URLs.
+        This is the root element of the XML file.
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:complexType>
+      <xsd:sequence>
+        <xsd:any namespace="##other" minOccurs="0" maxOccurs="unbounded" processContents="strict"/>
+        <xsd:element name="sitemap" type="tSitemap" maxOccurs="unbounded"/>
+      </xsd:sequence>
+    </xsd:complexType>
+  </xsd:element>
+
+  <xsd:complexType name="tSitemap">
+    <xsd:annotation>
+      <xsd:documentation>
+        Container for the data needed to describe a sitemap.
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:sequence>
+      <xsd:element name="loc" type="tLocSitemap"/>
+      <xsd:element name="lastmod" type="tLastmodSitemap" minOccurs="0"/>
+      <xsd:any namespace="##other" minOccurs="0" maxOccurs="unbounded" processContents="strict"/>
+    </xsd:sequence>
+  </xsd:complexType>
+
+  <xsd:simpleType name="tLocSitemap">
+    <xsd:annotation>
+      <xsd:documentation>
+        REQUIRED: The location URI of a sitemap.
+        The URI must conform to RFC 2396 (http://www.ietf.org/rfc/rfc2396.txt).
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:restriction base="xsd:anyURI">
+      <xsd:minLength value="12"/>
+      <xsd:maxLength value="2048"/>
+    </xsd:restriction>
+  </xsd:simpleType>
+
+  <xsd:simpleType name="tLastmodSitemap">
+    <xsd:annotation>
+      <xsd:documentation>
+        OPTIONAL: The date the document was last modified. The date must conform
+        to the W3C DATETIME format (http://www.w3.org/TR/NOTE-datetime).
+        Example: 2005-05-10
+        Lastmod may also contain a timestamp.
+        Example: 2005-05-10T17:33:30+08:00
+      </xsd:documentation>
+    </xsd:annotation>
+    <xsd:union>
+      <xsd:simpleType>
+        <xsd:restriction base="xsd:date"/>
+      </xsd:simpleType>
+      <xsd:simpleType>
+        <xsd:restriction base="xsd:dateTime"/>
+      </xsd:simpleType>
+    </xsd:union>
+  </xsd:simpleType>
+
+
+</xsd:schema>
-Original file line number
+Diff line change
@@ Expand Up / @@ -150,7 +150,7 @@ def valid_visibilities @@
         end
         def public
-          Item.where(visibility: JupiterCore::VISIBILITY_PUBLIC)
+          self.where(visibility: JupiterCore::VISIBILITY_PUBLIC)
         end
       end
@@ Expand Down @@