extend authorization and rdf serialization for validation

9 files changed, 213 insertions, 28 deletions

diff --git a/lib/authorization.rb b/lib/authorization.rb
index f9499e6..c33f712 100644
--- a/lib/authorization.rb
+++ b/lib/authorization.rb
@@ -132,6 +132,20 @@ module OpenTox
       end
     end
     
+    # Lists policies alongside with affected uris
+    # @param [String] subjectid
+    # @return [Hash] keys: all policies of the subjectid owner, values: uris affected by those policies
+    def self.list_policy_uris( subjectid )
+      names = list_policies(subjectid)
+      policies = {}
+      names.each do |n|
+        p = OpenTox::Policies.new
+        p.load_xml( list_policy(n, subjectid) )
+        policies[n] = p.uris
+      end
+      policies
+    end
+    
     #Returns the owner (who created the first policy) of an URI
     # @param [String, String]uri,subjectid
     # return [String, nil]owner,nil returns owner of the URI
@@ -271,21 +285,37 @@ module OpenTox
       return true
     end
 
-    #Checks (if subjectid is valid) if a policy exist and create default policy if not    
+    # Checks (if subjectid is valid) if a policy exist and create default policy if not
+    # @param [String] uri
+    # @param [String] subjectid
+    # @return [Boolean] true if policy checked/created successfully (or no uri/subjectid given), false else 
     def self.check_policy(uri, subjectid)
+      return true unless uri and subjectid
       token_valid = OpenTox::Authorization.is_token_valid(subjectid)      
       LOGGER.debug "OpenTox::Authorization.check_policy with uri: #{uri}, subjectid: #{subjectid} is valid: #{token_valid}"
-      if uri and token_valid
-        if !uri_has_policy(uri, subjectid)     
-          return send_policy(uri, subjectid)
-        else
-          LOGGER.debug "OpenTox::Authorization.check_policy URI: #{uri} has already a Policy."
+      # check if subjectid is valid
+      unless token_valid
+        # abort if invalid
+        LOGGER.error "OpenTox::Authorization.check_policy, subjectid NOT valid: #{subjectid}" 
+        return false
+      end
+      
+      if !uri_has_policy(uri, subjectid)
+        # if no policy exists, create a policy, return result of send policy
+        send_policy(uri, subjectid)
+      else
+        LOGGER.debug "OpenTox::Authorization.check_policy URI: #{uri} has already a Policy."
+        # if policy exists check for POST rights 
+        if authorize(uri, "POST", subjectid)
+          true
+       else
+          LOGGER.error "OpenTox::Authorization.check_policy, already exists, but no POST-authorization with subjectid: #{subjectid}" 
+          false
         end
       end
-      true
-    end    
+    end
     
-  end 
+  end
 end
 
 
diff --git a/lib/dataset.rb b/lib/dataset.rb
index d45c821..ae86f5f 100644
--- a/lib/dataset.rb
+++ b/lib/dataset.rb
@@ -253,11 +253,12 @@ module OpenTox
     # @param [Array] compounds List of compound URIs
     # @param [Array] features List of feature URIs
     # @param [Hash] metadata Hash containing the metadata for the new dataset
+    # @param [String] subjectid
     # @return [OpenTox::Dataset] newly created dataset, already saved
-    def split( compounds, features, metadata)
+    def split( compounds, features, metadata, subjectid=nil)
       LOGGER.debug "split dataset using "+compounds.size.to_s+"/"+@compounds.size.to_s+" compounds"
       raise "no new compounds selected" unless compounds and compounds.size>0
-      dataset = OpenTox::Dataset.create
+      dataset = OpenTox::Dataset.create(CONFIG[:services]["opentox-dataset"],subjectid)
       if features.size==0
         compounds.each{ |c| dataset.add_compound(c) }
       else
@@ -270,7 +271,7 @@ module OpenTox
         end
       end
       dataset.add_metadata(metadata)
-      dataset.save
+      dataset.save(subjectid)
       dataset
     end
 
diff --git a/lib/error.rb b/lib/error.rb
new file mode 100644
index 0000000..87e1a5d
--- /dev/null
+++ b/lib/error.rb
@@ -0,0 +1,11 @@
+module OpenTox
+
+  class NotFoundError < RuntimeError
+    
+  end
+  
+  class BadRequestError < RuntimeError
+    
+  end
+  
+end
\ No newline at end of file
diff --git a/lib/helper.rb b/lib/helper.rb
index 965b4ad..bb0279e 100644
--- a/lib/helper.rb
+++ b/lib/helper.rb
@@ -3,22 +3,24 @@ helpers do
   # Authentification
   def protected!(subjectid)
     if env["session"]
-      flash[:notice] = "You don't have access to this section: " and \
-      redirect back and \
-      return unless authorized?(subjectid)
+      unless authorized?(subjectid)
+        flash[:notice] = "You don't have access to this section: "
+        redirect back
+      end
     elsif !env["session"] && subjectid
-      throw(:halt, [401, "Not authorized.\n"]) and \
-      redirect back and \
-      return unless authorized?(subjectid)
+      unless authorized?(subjectid)
+        throw(:halt, [401, "Not authorized.\n"])
+        redirect back
+      end
+    else
+      throw(:halt, [401, "Not authorized.\n"]) unless authorized?(subjectid)
     end
-    throw(:halt, [401, "Not authorized.\n"]) and \
-    return unless authorized?(subjectid)
   end
 
   def authorized?(subjectid)
     if CONFIG[:authorization][:authorize_request].include?(request.env['REQUEST_METHOD'])
       ret = OpenTox::Authorization.authorize("#{request.env['rack.url_scheme']}://#{request.env['HTTP_HOST']}#{request.env['REQUEST_URI']}", request.env['REQUEST_METHOD'], subjectid)
-      LOGGER.debug "OpenTox helpers OpenTox::Authorization authorized? method: #{request.env['REQUEST_METHOD']} , URI: #{request.env['rack.url_scheme']}://#{request.env['HTTP_HOST']}#{request.env['REQUEST_URI']}, subjectid: #{subjectid} with return #{ret}."
+      LOGGER.debug "OpenTox helpers OpenTox::Authorization authorized? method: #{request.env['REQUEST_METHOD']}, URI: #{request.env['rack.url_scheme']}://#{request.env['HTTP_HOST']}#{request.env['REQUEST_URI']}, subjectid: #{subjectid} with return #{ret}."
       return ret
     end
     if CONFIG[:authorization][:authenticate_request].include?(env['REQUEST_METHOD'])
@@ -49,7 +51,7 @@ helpers do
 end
 
 before do
-  unless unprotected_requests or CONFIG[:authorization][:free_request].include?(env['REQUEST_METHOD']) 
+  unless !AA_SERVER or unprotected_requests or CONFIG[:authorization][:free_request].include?(env['REQUEST_METHOD']) 
     begin
       subjectid = session[:subjectid] if session[:subjectid]
       subjectid = params[:subjectid]  if params[:subjectid]  and !check_subjectid(subjectid)
@@ -60,7 +62,8 @@ before do
       LOGGER.debug "OpenTox ruby api wrapper: helper before filter: NO subjectid for URI: #{request.env['rack.url_scheme']}://#{request.env['HTTP_HOST']}#{request.env['REQUEST_URI']}"
       subjectid = ""
     end
-    protected!(subjectid) if AA_SERVER
+    @subjectid = subjectid
+    protected!(subjectid)
   end
 end
 
diff --git a/lib/opentox-ruby.rb b/lib/opentox-ruby.rb
index fb3803b..fc1732d 100644
--- a/lib/opentox-ruby.rb
+++ b/lib/opentox-ruby.rb
@@ -8,6 +8,7 @@ rescue LoadError
 	puts "Please install Openbabel with 'rake openbabel:install' in the compound component"
 end
 
-['opentox', 'compound','dataset', 'parser','serializer', 'algorithm','model','task','validation','feature', 'rest_client_wrapper', 'authorization', 'policy', 'helper', 'to-html'].each do |lib|
+['opentox', 'compound','dataset', 'parser','serializer', 'algorithm','model','task','validation','feature', 
+ 'rest_client_wrapper', 'authorization', 'policy', 'helper', 'to-html', 'error' ].each do |lib|
 	require lib
 end
diff --git a/lib/overwrite.rb b/lib/overwrite.rb
index ffeba21..720ed77 100644
--- a/lib/overwrite.rb
+++ b/lib/overwrite.rb
@@ -1,7 +1,11 @@
 # class overwrites aka monkey patches
-# hack: store sinatra in global var to make url_for and halt methods accessible
+# hack: store sinatra instance in global var $url_provider to make url_for and halt methods accessible
 before {
-  $sinatra = self unless $sinatra
+  raise "should not happen, url provider already differently initialized "+
+    $url_provider.request.host.to_s+" != "+self.request.host.to_s if
+    $url_provider and $url_provider.request.host!=self.request.host and 
+    $url_provider.request.script_name!=self.request.script_name
+  $url_provider = self
   # stupid internet explorer does not ask for text/html, add this manually 
   request.env['HTTP_ACCEPT'] += ";text/html" if request.env["HTTP_USER_AGENT"]=~/MSIE/
 }
@@ -91,7 +95,7 @@ class OTLogger < Logger
   end
   
   def format(msg)
-    pwd.ljust(18)+" :: "+msg.to_s+"           :: "+trace+" :: "+($sinatra ? $sinatra.request.env['REMOTE_ADDR'] : nil).to_s
+    pwd.ljust(18)+" :: "+msg.to_s+"           :: "+trace
   end
   
   def debug(msg)
diff --git a/lib/policy.rb b/lib/policy.rb
index 0ef8298..08bf6ed 100644
--- a/lib/policy.rb
+++ b/lib/policy.rb
@@ -32,6 +32,11 @@ module OpenTox
       end
       return true
     end
+    
+    # @return [Array] set of arrays affected by policies
+    def uris
+      @policies.collect{ |k,v| v.uris }.flatten.uniq
+    end
 
     #loads a default policy template in policies instance
     def load_default_policy(user, uri, group="member")    
@@ -190,6 +195,11 @@ module OpenTox
       @subjects[name] = Subject.new(name, type, value)
     end
     
+    # @return [Array] set of uris affected by policy
+    def uris
+      @rules.collect{ |k,v| v.uri }.uniq
+    end
+    
     #rule inside a policy
     class Rule
       
diff --git a/lib/serializer.rb b/lib/serializer.rb
index 495702a..03c2639 100644
--- a/lib/serializer.rb
+++ b/lib/serializer.rb
@@ -26,6 +26,15 @@ module OpenTox
           OT.Algorithm => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
           OT.Parameter => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
           OT.Task => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          #classes for validation
+          OT.Validation => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          OT.ClassificationStatistics => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          OT.ConfusionMatrix => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          OT.ConfusionMatrixCell => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          OT.ClassValueStatistics => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          OT.RegressionStatistics => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          OT.Crossvalidation => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
+          OT.CrossvalidationInfo => { RDF["type"] => [{ "type" => "uri", "value" => OWL['Class'] }] } ,
 
           OT.compound => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
           OT.feature => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
@@ -34,6 +43,22 @@ module OpenTox
           OT.values => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
           OT.algorithm => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
           OT.parameters => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          #object props for validation#           
+          OT.model => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.trainingDataset => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.predictionFeature => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.predictionDataset => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.crossvalidation => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.testTargetDataset => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.testDataset => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.classificationStatistics => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.confusionMatrix => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.confusionMatrixCell => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.classValueStatistics => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.regressionStatistics => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.validation => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.crossvalidationInfo => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
+          OT.dataset => { RDF["type"] => [{ "type" => "uri", "value" => OWL.ObjectProperty }] } ,
 
           DC.title => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
           DC.identifier => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
@@ -47,6 +72,47 @@ module OpenTox
           OT.hasStatus => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
           OT.resultURI => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
           OT.percentageCompleted => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          # annotation props for validation        
+          OT.numUnpredicted => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.crossvalidationFold => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numInstances => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numWithoutClass => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.percentWithoutClass => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.percentUnpredicted => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.confusionMatrixActual => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.confusionMatrixPredicted => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.confusionMatrixValue => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numIncorrect => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.percentCorrect => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numCorrect => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.accuracy => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.trueNegativeRate => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.truePositiveRate => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.falseNegativeRate => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.falsePositiveRate => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numTrueNegatives => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numTruePositives => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numFalseNegatives => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numFalsePositives => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.classValue => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.precision => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.areaUnderRoc => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.weightedAreaUnderRoc => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.fMeasure => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.percentIncorrect => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.validationType => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.realRuntime => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.sampleCorrelationCoefficient => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.targetVarianceActual => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.targetVariancePredicted => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.meanAbsoluteError => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.sumSquaredError => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.rootMeanSquaredError => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.rSquare => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.stratified => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.numFolds => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.randomSeed => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
+          OT.reportType => { RDF["type"] => [{ "type" => "uri", "value" => OWL.AnnotationProperty }] } ,
 
           OT.hasSource => { RDF["type"] => [{ "type" => "uri", "value" => OWL.DatatypeProperty }] } ,
           OT.value => { RDF["type"] => [{ "type" => "uri", "value" => OWL.DatatypeProperty }] } ,
@@ -121,6 +187,64 @@ module OpenTox
         @object[uri] = { RDF["type"] => [{ "type" => "uri", "value" => OT.Task }] }
         add_metadata uri, metadata
       end
+      
+      # Add a resource defined by resource_class and content
+      # (see documentation of add_content for example)
+      # @param [String] uri of resource
+      # @param [String] resource class, e.g. OT.Validation
+      # @param [Hash] content as hash
+      def add_resource(uri, resource_class, content)
+        @object[uri] = { RDF["type"] => [{ "type" => "uri", "value" => resource_class }] }
+        @@content_id = 1
+        add_content uri, content
+      end
+      
+      private
+      @@content_id = 1
+      
+      # Recursiv function to add content
+      # @example
+      # { DC.description => "bla",
+      #   OT.similar_resources => [ "http://uri1", "http://uri2" ],
+      #   OT.matrixCells => 
+      #     [ { RDF.type => OT.MatrixCell, OT.cellIndex=1 OT.cellValue => "xy" },
+      #       { RDF.type => OT.MatrixCell, OT.cellIndex=2 OT.cellValue => "z" } ],
+      #   OT.info => { RDF.type => OT.ImportantInfo, 
+      #                DC.description => "blub" }
+      # } 
+      # @param [String] uri
+      # @param [Hash] content as hash, uri must already have been added to @object
+      def add_content(uri, hash)
+        raise "content is no hash: "+hash.class.to_s unless hash.is_a?(Hash)
+        hash.each do |u,v|
+          if v.is_a? Hash
+            # value is again a hash, i.e. a new owl class is added
+            # first make sure type (==class) is set 
+            type = v[RDF.type]
+            raise "type missing for "+u.to_s+" content:\n"+v.inspect unless type
+            raise "class unknown "+type.to_s+" (for "+u.to_s+")" unless @object.has_key?(type)
+            # create new node and add to current uri
+            genid = "_:#{type.split('#')[-1]}#{@@content_id}"
+            @@content_id += 1
+            @object[uri] = {} unless @object[uri]
+            @object[uri][u] = [{ "type" => "bnode", "value" => genid }]
+            # add content to new class
+            add_hash(genid,v)
+          elsif v.is_a? Array
+            # value is an array, i.e. a list of values with property is added 
+            v.each{ |vv| add_hash( uri, { u => vv } ) }
+          else # v.is_a? String
+            # simple string value
+            @object[uri] = {} unless @object[uri]
+            @object[uri][u] = [] unless @object[uri][u]
+            raise "property unknown "+u.to_s if !@object.has_key?(u) and u!=RDF.type
+            # use << to allow different values for one property
+            @object[uri][u] << {"type" => type(v), "value" => v }
+          end
+        end
+      end
+      
+      public
 
       # Add metadata
       # @param [Hash] metadata
diff --git a/lib/to-html.rb b/lib/to-html.rb
index 1bc1496..e9764ef 100755
--- a/lib/to-html.rb
+++ b/lib/to-html.rb
@@ -27,7 +27,8 @@ module OpenTox
   # @return [String] html page
   def self.text_to_html( text, related_links=nil, description=nil, post_params=nil )
     
-    title = $sinatra.url_for($sinatra.request.env['PATH_INFO'], :full) if $sinatra
+    # TODO add title as parameter
+    title = nil #$sinatra.url_for($sinatra.request.env['PATH_INFO'], :full) if $sinatra
     
     html = <<EOF
 <html>