datacatalog-importer 0.1.5 → 0.1.6

data/VERSION

@@ -1 +1 @@
-0.1.5
+0.1.6

data/datacatalog-importer.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{datacatalog-importer}
-  s.version = "0.1.5"
+  s.version = "0.1.6"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["David James"]
-  s.date = %q{2010-03-03}
+  s.date = %q{2010-03-05}
   s.description = %q{This framework makes it easier to write importers for the National Data Catalog.}
   s.email = %q{djames@sunlightfoundation.com}
   s.extra_rdoc_files = [
@@ -39,7 +39,7 @@ Gem::Specification.new do |s|
   s.homepage = %q{http://github.com/djsun/datacatalog-importer}
   s.rdoc_options = ["--charset=UTF-8"]
   s.require_paths = ["lib"]
-  s.rubygems_version = %q{1.3.5}
+  s.rubygems_version = %q{1.3.6}
   s.summary = %q{A framework to write National Data Catalog importers}
   s.test_files = [
     "spec/datacatalog-importer_spec.rb",

data/lib/pusher.rb

@@ -34,12 +34,14 @@ module DataCatalog
 
       def push_organizations
         read_data(:organization) do |data|
+          link_to_existing_organization!(data, :parent_id)
           create_or_update_organization(data)
         end
       end
       
       def push_sources
         read_data(:source) do |data|
+          link_to_existing_organization!(data, :organization_id)
           source = create_or_update_source(data)
           data[:downloads].each do |download_data|
             create_or_update_download(source, download_data)
@@ -68,25 +70,42 @@ module DataCatalog
       # ---
       
       def create_or_update_organization(data)
-        docs = DataCatalog::Organization.all(:url => data[:url])
+        url, name = data[:url], data[:name]
+        raise "#{name} has blank URL" if url.blank?
+        docs = DataCatalog::Organization.all(:url => url)
         n = docs.length
         case n
         when 0
-          puts "Creating Organization: #{data[:name]}"
+          puts "Creating Organization: #{name}"
           DataCatalog::Organization.create(data)
         when 1
-          puts "Updating Organization: #{data[:name]}"
+          puts "Updating Organization: #{name}"
           DataCatalog::Organization.update(docs[0].id, data)
         else
           multiple_matches("Organization", n, data)
         end
       end
       
+      def find_organization_by(field, name)
+        docs = DataCatalog::Organization.all(field => name)
+        n = docs.length
+        case n
+        when 0
+          nil
+        when 1
+          docs[0]
+        else
+          multiple_matches("Organization", n, { field => name })
+          nil
+        end
+      end
+      
       # Important: do not modify data
       def create_or_update_source(data)
         data = data.reject do |key, value|
           [:organization, :downloads].include?(key)
         end
+        puts "data[:url] : #{data[:url]}"
         docs = DataCatalog::Source.all(:url => data[:url])
         n = docs.length
         case n
@@ -121,6 +140,28 @@ module DataCatalog
         end
       end
       
+      # Try to link to an existing organization, first by
+      # using an URL, then by name.
+      #
+      # Note: modifies data (that is why I use the !)
+      def link_to_existing_organization!(data, organization_id_key)
+        organization_hash = data.delete(:organization)
+        raise "Could not find :organization key" unless organization_hash
+        name = organization_hash[:name]
+        url  = organization_hash[:url]
+        organization = if url
+          find_organization_by(:url, url)
+        elsif name
+          find_organization_by(:name, name)  
+        end
+        
+        if organization
+          data[organization_id_key] = organization.id
+        else
+          puts " - Could not find organization named: #{name}"
+        end
+      end
+      
       # ---
       
       def multiple_matches(model, n, data)

data/lib/utility.rb

@@ -28,7 +28,7 @@ module DataCatalog
 
     def self.headers
       {
-        "UserAgent" => "National Data Catalog Importer/0.1.5",
+        "UserAgent" => "National Data Catalog Importer/0.1.6",
       }
     end
     

metadata

@@ -1,7 +1,12 @@
 --- !ruby/object:Gem::Specification 
 name: datacatalog-importer
 version: !ruby/object:Gem::Version 
-  version: 0.1.5
+  prerelease: false
+  segments: 
+  - 0
+  - 1
+  - 6
+  version: 0.1.6
 platform: ruby
 authors: 
 - David James
@@ -9,29 +14,37 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-03-03 00:00:00 -05:00
+date: 2010-03-05 00:00:00 -05:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: nokogiri
-  type: :runtime
-  version_requirement: 
-  version_requirements: !ruby/object:Gem::Requirement 
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
+        segments: 
+        - 1
+        - 4
+        - 1
         version: 1.4.1
-    version: 
+  type: :runtime
+  version_requirements: *id001
 - !ruby/object:Gem::Dependency 
   name: rspec
-  type: :development
-  version_requirement: 
-  version_requirements: !ruby/object:Gem::Requirement 
+  prerelease: false
+  requirement: &id002 !ruby/object:Gem::Requirement 
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
+        segments: 
+        - 1
+        - 2
+        - 9
         version: 1.2.9
-    version: 
+  type: :development
+  version_requirements: *id002
 description: This framework makes it easier to write importers for the National Data Catalog.
 email: djames@sunlightfoundation.com
 executables: []
@@ -73,18 +86,20 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      segments: 
+      - 0
       version: "0"
-  version: 
 required_rubygems_version: !ruby/object:Gem::Requirement 
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      segments: 
+      - 0
       version: "0"
-  version: 
 requirements: []
 
 rubyforge_project: 
-rubygems_version: 1.3.5
+rubygems_version: 1.3.6
 signing_key: 
 specification_version: 3
 summary: A framework to write National Data Catalog importers