talia_core 0.5.4 → 0.7.0

data/lib/talia_core/data_types/pdf_data.rb

@@ -1,16 +1,17 @@
 module TaliaCore
   module DataTypes
 
-    # Class to manage PDF data type
+    # FileRecord that contains a PDF document.
     class PdfData < FileRecord
      
-      # return the mime_type for a file
+      # The MIME type is always <tt>application/pdf</tt>
       def extract_mime_type(location)
         'application/pdf'
       end
 
-      # Create the PDF data from a PDF writer. This method needs a block
-      # which will be called with the writer object
+      # Create the PDF data using PDF::Writer. The writer will be passed
+      # to the block given to this method, and the resulting PDF will be
+      # saved as the record's file.
       def create_from_writer(writer_opts = {})
         activate_pdf
         writer = PDF::Writer.new(writer_opts) do |pdf|
@@ -24,14 +25,16 @@ module TaliaCore
 
       private
 
-      # Little helper method - no need to load the pdf classes unless needed,
-      # but if called every time round it will slow things down (in Rails it will).
+      # Little helper method to load the pdf writer only when needed. This will make sure
+      # the thing is loaded only once, otherwise it would be slow in Rails development
+      # (which uses #load for #require)
       def self.activate_pdf
         return if(@pdf_active)
         require 'pdf/writer'
         @pdf_active = true
       end
 
+      # See PdfData.activate_pdf
       def activate_pdf
         self.class.activate_pdf
       end

data/lib/talia_core/data_types/simple_text.rb

@@ -1,18 +1,19 @@
 require 'talia_core/data_types/file_store'
 
-# Class to manage data stored in a text file
+# File Record that contains a plain text file.
 module TaliaCore
   module DataTypes
   
     class SimpleText < FileRecord
       
-      # return the mime_type for a file
+      # The MIME type is always <tt>text/plain</tt>
       def extract_mime_type(location)
         'text/plain'
       end
 
-      # Get a line from a text file.
-      # At the end of file: close the file and return
+      # Read onle line from the text file. The file will
+      # be closed if the end of the file is hit, or
+      # if close_after_single_read is set to true
       def get_line(close_after_single_read=false)
         if !is_file_open?
           open_file

data/lib/talia_core/data_types/xml_data.rb

@@ -2,6 +2,8 @@ require 'talia_core/data_types/file_store'
 require 'rexml/document'
 # require 'xml/xslt'
 
+# Little helper to load the "tidy" library, to clean up
+# messy XML/HTML, if possible. 
 begin
   # if tidy is not present, disable it
   require 'tidy'
@@ -27,11 +29,14 @@ end
 module TaliaCore
   module DataTypes
   
-    # Class to manage XML and HTML data type
+    # FileRecord class to store XML (or XHTML) files.
     class XmlData < FileRecord
       
-      # return the mime_type for a file
+      # MIME type should be one of 'text/html' or 'text/xml' 
+      # ('text/hnml' is supported for legacy reasons)
       def extract_mime_type(location)
+        # TODO: Could probably use the Mime classes to get the
+        # type, or move to the superclass
         case File.extname(location).downcase
         when '.htm', '.html','.xhtml'
           'text/html'
@@ -42,24 +47,29 @@ module TaliaCore
         end
       end
       
-      # return the mime subtype for this specified class
+      # The mime subtype for this specified class
       def mime_subtype
         mime_type.split(/\//)[1]
       end    
 
-      # return contect of the object as REXML::Elements
-      # * options: Options for getting context. Default nil.
-      # * options[:xsl_file]: xsl file path for transformation.
-      def get_content(options = nil)
+      # The content of this document. This returns REXML elements
+      # for the document content. For plain XML files, this will
+      # return the children of the doucment root. For XHTML documents,
+      # this will return the children of the "body" tag.
+      #
+      # *Options*:
+      # 
+      # * [*xsl_file*] If given, the document will be transformed using this
+      #                XSL file before the document is extracted
+      def get_content(options = {})
+        # TODO: Maybe port this to hpricot/nokogiri too
         text_to_parse = all_text
       
-        if (!options.nil?)
-          # if xsl_file option is specified, execute transformation
-          if (!options[:xsl_file].nil?)
-            text_to_parse = xslt_transform(file_path, options[:xsl_file])
-          end
+        # if xsl_file option is specified, execute transformation
+        if (options[:xsl_file])
+          text_to_parse = xslt_transform(file_path, options[:xsl_file])
         end
-      
+
         # create document object
         document = REXML::Document.new text_to_parse
       
@@ -78,7 +88,7 @@ module TaliaCore
         return content
       end
     
-      # Returns an xml string of the elements returned by get_content
+      # Same as #get_content, but returns a string instead of the REXML documents
       def get_content_string(options = nil)
         xml_str = ''
         get_content(options).each do |element|
@@ -87,16 +97,26 @@ module TaliaCore
         xml_str
       end
     
-      # Returns an xml string that is escaped for HTML inclusing
+      # Same as #get_content_string, but with the XML escape for inclusion in 
+      # HTML documents
       def get_escaped_content_string(options = nil)
         get_content_string(options).gsub(/</, "&lt;").gsub(/>/, "&gt;")
       end
     
-      # Add data as string into file
-      # * location: location
-      # * data: data to write
-      # * options: options
-      # * options[:tidy]: enable or disable tidy (convert html into xhtml). Default value is true
+      # See the FileStore module for details on how creation of data file objects works.
+      # This version differs from the superclass version in that it will (optionally)
+      # clean the HTML using the "tidy" tool. Also see http://tidy.rubyforge.org/
+      #
+      # Tidy will be used under the following circumstances:
+      # 
+      # * The "tidy" option is given and
+      # * The library itself is available and
+      # * The file appears to be a (X)HTML file
+      #
+      # *Options:*:
+      #
+      # [*tidy*] Use the "tidy" tool to clean up (X)HTML. Defaults to true if no options
+      #          are given.
       def create_from_data(location, data, options = {:tidy => true})
         # check tidy option
         if (((options[:tidy] == true) and (Tidy_enable == true)) and 
@@ -118,9 +138,14 @@ module TaliaCore
       end
     
       private
-      # adjusted/replaced items path
-      # * item: REXML::Element to parse
+      
+      # Helper that updates the paths in an XML element. Takes a REXML::Element, and updates
+      # the paths for "img" and "a" tags to point to the Talia "source_data" controller.
+      #
+      # This is a quick hack to allow the rendering of HTML that contains those elements and 
+      # needs to be fixed to show linked files that are stored in Talia.
       def wrapItem item
+        # TODO: Quite hacky. Uses hardcoded paths, maybe should rather be a helper.
         if item.class == REXML::Element
           # recursive execution
           item.each_child { |subItem| wrapItem subItem}
@@ -150,9 +175,12 @@ module TaliaCore
         end
       end
     
-      # execute xslt transformation
-      # * document: xml document. Can be file path as string or REXML::Document
-      # * xsl_file: xsl file for transformation. Can be file path as string or REXML::Document
+      # Perform an XSLT transformation
+      #
+      # *Options*:
+      #
+      # [*document*] Xml document. Can be file path as string or REXML::Document
+      # [*xsl_file*] Xsl file for transformation. Can be file path as string or REXML::Document
       def xslt_transform(document, xsl_file)
         xslt = XML::XSLT.new()
         # get xml document

data/lib/talia_core/dummy_handler.rb

@@ -9,7 +9,8 @@ module TaliaCore
   # The problem is that invocations like source.namespace::name are
   # evaluated to (source.namespace).name 
   # This means that source.namespace must return an object on which
-  # "name" can be called with the desired effect. This is the "dummy handler"
+  # "name" can be called with the desired effect. This  object 
+  # is the "dummy handler"
   class DummyHandler
     
     # Create the new handler
@@ -24,7 +25,7 @@ module TaliaCore
     # Catch the invocations
     def method_missing(method, *args)
       # read value
-      raise(SemanticNamingError, "Wrong number of arguments") if(args.size != 0)
+      raise(ArgumentError, "Wrong number of arguments") if(args.size != 0)
       @subject[@namespace + method.to_s]
     end
     

data/lib/talia_core/errors.rb

@@ -1,29 +1,15 @@
-# This file contains all the errors that can be raised in the 
-# Talia core. Since these are not useful classes, they are grouped
-# together in one file.
+module TaliaCore
   
-# Indicates that it was tried to create an object with
-# an identifier (e.g. an URI) which already exists in the
-# system.
-class DuplicateIdentifierError < RuntimeError
-end
-
-# Indicates an error during initialization
-class SystemInitializationError < RuntimeError
-end
-
-# Indicates an error with the predicate naming
-class SemanticNamingError < RuntimeError
-end
-
-# Indicates an attempt to save attributes on an unsaved source
-class UnsavedSourceError < RuntimeError 
-end
-
-# Indicates an error in a query
-class QueryError < RuntimeError
-end
-
-# Indicates an error with import data
-class ImportError < RuntimeError
+  # Some Excptions/Errors that are specific to the Talia system
+  module Errors
+    
+    # Indicates an error during initialization
+    class SystemInitializationError < RuntimeError
+    end
+
+    # Indicates an error with import data
+    class ImportError < RuntimeError
+    end
+  
+  end
 end

data/lib/talia_core/initializer.rb

@@ -49,6 +49,46 @@ module TaliaCore
   #    # config file
   #    config['standalone_db'] = "true"
   #  end
+  #
+  # = Configuration options for <tt>talia_core.yml</tt>
+  #
+  # [*local_uri*] URL of the current installation. This should be set to
+  #               the URL where Talia is installed
+  # [*rdf_connection_file*] Name of the configuration file that contains the
+  #                         connection parameters for the RDF store. 
+  # [*rdf_connection*] Used to place the RDF options directly in the configuration
+  #                    file. the *rdf_connection_file* setting will take precedence,
+  #                    though.
+  # [*db_file*] File with the database configuration for ActiveRecord. When running
+  #             from within RAILS, Talia will *always* use the Rails configuration
+  # [*db_connection*] Used to place the DB options directly in the file. *db_file*
+  #                   will take precedence, if present
+  # [*standalone_db*] If set to true, Talia will be configured to use the ActiveRecord
+  #                   database connection without Rails. 
+  # [*ardf_log_level*] Used to indicate the log level for the ActiveRDF library (integer value)
+  # [*db_log*] Log file for the database/ActiveRecord. Does not take effect if
+  #            running from inside Rails
+  # [*data_directory_location*] Root directory for the FileRecord file storage
+  # [*namespaces*] Declares the namespaces that can be used within Talia 
+  #                (a number of "namespace": "url" pairs)
+  # [*iip_root_directory_location*] The directory that is used for the pyramidal files
+  #                                 that will be served by the IIP server
+  # [*iip_server_uri*] The URI to the IIP server. When using the default server, this will
+  #                    be the URL that calls the iipsrv.fcgi
+  # [*vips_command*] Path to the "vips" command, which is used to create the pyramidal images
+  # [*convert_command*] Path to the "convert" command from ImageMagick, which creates the thumbs
+  # [*thumb_options*] Options for thumbnail images. These can contain "height", "width" and the 
+  #                   "force" option. If the "force" option is set, the thumbnails will be the
+  #                   exact dimensions given (with a transparent background around the border).
+  #                   Otherwise the aspect ratio of the image will be preserved.
+  # [*environment*] The environment for the session, that is "development", "production" or "test". 
+  #                 If running from Rails, this will inherit Rails' setting
+  # [*standalone_log*] Log file to be used when in standalone mode (not using Rails)
+  # [*standalone_log_level*] Log level for the <tt>standalone_log</tt>
+  # [*assert*] If true, the <tt>assit</tt> assertions are enabled in the code. Defaults to true.
+  # [*auto_ontologies*] If set to a directory, Talia will automatically (re-)load the ontologies 
+  #                     from that directory on startup
+  # [*site_name*] Site name string. May (or may not) be used in the application.
   class Initializer
 
     # Is used to set the root directory manually. Must be written before
@@ -80,7 +120,7 @@ module TaliaCore
       # For now, the values are documented in the code below and in the default
       # configuration file.
       def run(config_file = nil, &initializer)
-        raise(SystemInitializationError, "System cannot be initialized twice") if(@@initialized || @@init_started)
+        raise(Errors::SystemInitializationError, "System cannot be initialized twice") if(@@initialized || @@init_started)
 
         @@init_started = true
 
@@ -352,10 +392,10 @@ module TaliaCore
       def load_ontologies
         return unless(@config['auto_ontologies'] && !['false', 'no'].include?(@config['auto_ontologies'].downcase))
         onto_dir = File.join(TALIA_ROOT, @config['auto_ontologies'])
-        raise(SystemInitializationError, "Cannot find configured ontology dir #{onto_dir}") unless(File.directory?(onto_dir))
+        raise(Errors::SystemInitializationError, "Cannot find configured ontology dir #{onto_dir}") unless(File.directory?(onto_dir))
         adapter = ActiveRDF::ConnectionPool.write_adapter
-        raise(SystemInitializationError, "Ontology autoloading without a context-aware adapter deletes all RDF data. This is only allowed in testing, please load the ontology manually.") unless(adapter.contexts? || (@environment == 'testing'))
-        raise(SystemInitializationError, "Ontology autoloading requires 'load' capability on the adapter.") unless(adapter.respond_to?(:load))
+        raise(Errors::SystemInitializationError, "Ontology autoloading without a context-aware adapter deletes all RDF data. This is only allowed in testing, please load the ontology manually.") unless(adapter.contexts? || (@environment == 'testing'))
+        raise(Errors::SystemInitializationError, "Ontology autoloading requires 'load' capability on the adapter.") unless(adapter.respond_to?(:load))
 
         # Clear out the RDF
         if(adapter.contexts?)

data/lib/talia_core/oai/active_source_model.rb

@@ -40,13 +40,15 @@ module TaliaCore
         return select_partial(options[:resumption_token]) if(options[:resumption_token])
         
         conditions = sql_conditions(options)
-        
+       
         if(selector == :first)
-          model_adapter.get_wrapper_for(ActiveSource.find(selector, :prefetch_relations => true, :conditions => conditions))
+          #model_adapter.get_wrapper_for(ActiveSource.find(selector, :prefetch_relations => true, :conditions => conditions))
+          model_adapter.get_wrapper_for(Source.find(selector, :prefetch_relations => true, :conditions => conditions))
         elsif(selector == :all)
           select_partial(OAI::Provider::ResumptionToken.new(last_id(conditions), options.merge(:last => 0)))
         else
-          model_adapter.get_wrapper_for(ActiveSource.find(selector, :prefetch_relations => true, :conditions => conditions))
+          #model_adapter.get_wrapper_for(ActiveSource.find(selector, :prefetch_relations => true, :conditions => conditions))
+          model_adapter.get_wrapper_for(Source.find(selector, :prefetch_relations => true, :conditions => conditions))
         end
       rescue ActiveRecord::RecordNotFound
         nil
@@ -59,11 +61,13 @@ module TaliaCore
         token = OAI::Provider::ResumptionToken.parse(token) if(token.is_a?(String))
         
         conditions = token_conditions(token)
-        total = ActiveSource.count(:id, :conditions => conditions)
+        #total = ActiveSource.count(:id, :conditions => conditions)
+        total = Source.count(:id, :conditions => conditions)
         
         return [] if(total == 0)
         
-        records = ActiveSource.find(:all, :conditions => token_conditions(token),
+        #records = ActiveSource.find(:all, :conditions => token_conditions(token),
+        records = Source.find(:all, :conditions => token_conditions(token),
           :limit => @limit,
           :order => 'id asc',
           :prefetch_relations => true
@@ -77,7 +81,10 @@ module TaliaCore
       def select_first_or_last(order, conditions = nil)
         select_options = { :select => "id, #{timestamp_field}", :order => "#{timestamp_field} #{order}" }
         select_options[:conditions] = conditions if(conditions)
-        TaliaCore::ActiveSource.find(:first, select_options)
+        #result = TaliaCore::ActiveSource.find(:first, select_options)
+        result = TaliaCore::Source.find(:first, select_options)
+        raise OAI::NoMatchException if result.nil? 
+        result
       end
       
       # Sql condition for the given token

data/lib/talia_core/oai/active_source_oai_adapter.rb

@@ -56,17 +56,18 @@ module TaliaCore
       # Get the author/creator
       def creator
         @record[N::DCNS.creator].collect do |creator|
-          if(creator.is_a?(TaliaCore::ActiveSource))
-            author = ''
-            author_name = creator.hyper::author_name.first
-            author << author_name << ' ' if(author_name)
-            author_surname = creator.hyper::author_surname.first || ''
-            author << author_surname if(author_surname)
-            author = "No lookup. Author should be at #{creator.uri}" if(author == '')
-            author
-          else
-            creator
-          end
+          creator
+          # if(creator.is_a?(TaliaCore::ActiveSource))
+          #   author = ''
+          #   author_name = creator.hyper::author_name.first
+          #   author << author_name << ' ' if(author_name)
+          #   author_surname = creator.hyper::author_surname.first || ''
+          #   author << author_surname if(author_surname)
+          #   author = "No lookup. Author should be at #{creator.uri}" if(author == '')
+          #   author
+          # else
+          #   creator
+          # end
         end
       end
       
@@ -89,7 +90,7 @@ module TaliaCore
       # wrapper class that corresponds to the given record's class
       def self.get_wrapper_for(records)
         if(records.is_a?(ActiveSource))
-          self.new(record)
+          self.new(records)
         elsif(records.respond_to?(:collect))
           records.collect { |rec| self.new(rec) }
         elsif(records.nil?)

data/lib/talia_core/rdf_import.rb

@@ -27,7 +27,7 @@ module TaliaCore
           import_file(file, rdf_syntax, context)
         end
 
-        adapter.save if(adapter.respond_to?(:save))
+        adapter.try_call.save
 
         puts "\n--> Importing rdf/rdfs file: complete!\n\n"
       end

data/lib/talia_core/rdf_resource.rb

@@ -29,6 +29,7 @@ module TaliaCore
     
     # Initialize a new resource with the given URI
     def initialize(uri)
+      raise ArgumentError, "No blank URI allowed here" if(uri.to_s.blank?)
       @uri = N::URI.new(uri)
     end
 
@@ -73,7 +74,7 @@ module TaliaCore
       inverseobj.instance_variable_set(:@obj_uri, self)
       inverseobj.instance_variable_set(:@obj_class, object_class)
       
-      class <<inverseobj     
+      class << inverseobj     
         
         def [](property)
           property = N::URI.new(property) unless(property.kind_of?(N::URI))