oai 0.0.1 → 1.3.0

data/test/provider/tc_simple_provider.rb

@@ -0,0 +1,138 @@
+require 'test_helper_provider'
+
+class TestSimpleProvider < Test::Unit::TestCase
+
+  def setup
+    @simple_provider = SimpleProvider.new
+    @model = @simple_provider.class.model
+  end
+
+  def test_identify
+    doc = REXML::Document.new(@simple_provider.identify)
+    assert_equal @simple_provider.class.name,
+      doc.elements["/OAI-PMH/Identify/repositoryName"].text
+    assert_equal SimpleModel.new.earliest.to_s,
+      doc.elements["/OAI-PMH/Identify/earliestDatestamp"].text
+
+      # PC
+  #  lambda { REXML::Document.new(@simple_provider.identify(:set => 'A')) }
+
+
+  end
+
+  def test_list_sets
+    doc = REXML::Document.new(@simple_provider.list_sets)
+    sets = doc.elements["/OAI-PMH/ListSets"]
+    assert_equal @model.sets.size, sets.size
+    assert_equal @model.sets[0].name, sets[0].elements["//setName"].text
+  end
+
+  def test_metadata_formats
+    assert_nothing_raised { REXML::Document.new(@simple_provider.list_metadata_formats) }
+    doc =  REXML::Document.new(@simple_provider.list_metadata_formats)
+    assert_equal "oai_dc",
+      doc.elements['/OAI-PMH/ListMetadataFormats/metadataFormat/metadataPrefix'].text
+  end
+
+  def test_metadata_formats_for_document
+    assert_nothing_raised { REXML::Document.new(@simple_provider.list_metadata_formats(:identifier => "oai:test:1")) }
+    doc =  REXML::Document.new(@simple_provider.list_metadata_formats)
+    assert_equal "oai_dc",
+      doc.elements['/OAI-PMH/ListMetadataFormats/metadataFormat/metadataPrefix'].text
+  end
+
+  def test_list_records_without_constraints
+    assert_nothing_raised { REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc')) }
+
+    total = @model.find(:all).size
+    doc = REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc'))
+    assert_equal total, doc.elements['OAI-PMH/ListRecords'].size
+  end
+
+  def test_list_records_with_from_constraints
+    assert_nothing_raised { REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc')) }
+
+    total = @model.find(:all).size
+    doc = REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc', from: "2002-10-05"))
+    assert_equal total, doc.elements['OAI-PMH/ListRecords'].size
+  end
+
+  def test_list_records_with_until_constraints
+    assert_nothing_raised { REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc')) }
+
+    total = @model.find(:all).size
+    doc = REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc', until: "2002-11-05"))
+    assert_equal total, doc.elements['OAI-PMH/ListRecords'].size
+  end
+
+  def test_list_records_with_set_equal_a
+    total = @model.find(:all, :set => 'A').size
+    doc = REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc', :set => 'A'))
+    assert_equal total, doc.elements['OAI-PMH/ListRecords'].size
+  end
+
+  def test_list_record_with_set_equal_ab
+    total = @model.find(:all, :set => 'A:B').size
+    doc = REXML::Document.new(@simple_provider.list_records(:metadata_prefix => 'oai_dc', :set => 'A:B'))
+    assert_equal total, doc.elements['OAI-PMH/ListRecords'].size
+  end
+
+  def test_list_identifiers_without_constraints
+    assert_nothing_raised { REXML::Document.new(@simple_provider.list_identifiers(:metadata_prefix => 'oai_dc')) }
+
+    total = @model.find(:all).size
+    doc = REXML::Document.new(@simple_provider.list_identifiers(:metadata_prefix => 'oai_dc'))
+    assert_equal total, doc.elements['OAI-PMH/ListIdentifiers'].to_a.size
+  end
+
+  def test_list_identifiers_with_set_equal_a
+    total = @model.find(:all, :set => 'A').size
+    doc = REXML::Document.new(@simple_provider.list_identifiers(:metadata_prefix => 'oai_dc', :set => 'A'))
+    assert_equal total, doc.elements['OAI-PMH/ListIdentifiers'].to_a.size
+  end
+
+  def test_list_indentifiers_with_set_equal_ab
+    total = @model.find(:all, :set => 'A:B').size
+    doc = REXML::Document.new(@simple_provider.list_identifiers(:metadata_prefix => 'oai_dc', :set => 'A:B'))
+    assert_equal total, doc.elements['OAI-PMH/ListIdentifiers'].to_a.size
+  end
+
+  def test_get_record
+    assert_nothing_raised do
+      REXML::Document.new(
+        @simple_provider.get_record(
+          :identifier => 'oai:test:1',
+          :metadataPrefix => 'oai_dc'
+        )
+      )
+    end
+    doc = REXML::Document.new(
+      @simple_provider.get_record(
+        :identifier => 'oai:test:1',
+        :metadataPrefix => 'oai_dc'
+      )
+    )
+    assert_equal 'oai:test:1',
+      doc.elements['OAI-PMH/GetRecord/record/header/identifier'].text
+  end
+
+  def test_deleted_record
+    assert_nothing_raised do
+      REXML::Document.new(
+        @simple_provider.get_record(
+          :identifier => 'oai:test:6',
+          :metadataPrefix => 'oai_dc'
+        )
+      )
+    end
+    doc = REXML::Document.new(
+      @simple_provider.get_record(
+        :identifier => 'oai:test:5',
+        :metadataPrefix => 'oai_dc'
+      )
+    )
+    assert_equal 'oai:test:5', doc.elements['OAI-PMH/GetRecord/record/header/identifier'].text
+    assert_equal 'deleted', doc.elements['OAI-PMH/GetRecord/record/header'].attributes["status"]
+  end
+
+end

data/test/provider/test_helper_provider.rb

@@ -0,0 +1,62 @@
+require 'oai'
+require 'test/unit'
+
+require File.dirname(__FILE__) + '/models'
+include OAI
+
+class SimpleProvider < Provider::Base
+  repository_name 'Test Provider'
+  record_prefix 'oai:test'
+  source_model SimpleModel.new
+end
+
+class BigProvider < Provider::Base
+  repository_name 'Another Provider'
+  record_prefix 'oai:test'
+  source_model BigModel.new
+end
+
+class TokenProvider < Provider::Base
+  repository_name 'Token Provider'
+  record_prefix 'oai:test'
+  source_model BigModel.new(25)
+end
+
+class MappedProvider < Provider::Base
+  repository_name 'Mapped Provider'
+  record_prefix 'oai:test'
+  source_model MappedModel.new
+end
+
+class ComplexProvider < Provider::Base
+  repository_name 'Complex Provider'
+  repository_url 'http://localhost'
+  record_prefix 'oai:test'
+  source_model ComplexModel.new(100)
+end
+
+class DescribedProvider < Provider::Base
+  repository_name 'Described PRovider'
+  repository_url 'http://localhost'
+  record_prefix 'oai:test'
+  source_model SimpleModel.new
+  sample_id '13900'
+  extra_description "<my_custom_xml />"
+end
+
+class InstanceProvider < Provider::Base
+  repository_name 'Instance Provider'
+  record_prefix 'oai:test'
+  repository_url 'http://localhost'
+  source_model SimpleModel.new
+
+  def initialize(options = {})
+    super
+    @url_path = options.fetch(:url_path)
+  end
+  attr_reader :url_path
+
+  def url
+    File.join(super, url_path)
+  end
+end

data/test/test.xml

@@ -0,0 +1,22 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<OAI-PMH xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd">
+  <responseDate>2006-09-11T14:33:15Z</responseDate>
+  <request verb="Identify">http://www.pubmedcentral.gov/oai/oai.cgi</request>
+  <Identify>
+    <repositoryName>PubMed Central (PMC3 - NLM DTD)</repositoryName>
+    <baseURL>http://www.pubmedcentral.gov/oai/oai.cgi</baseURL>
+    <protocolVersion>2.0</protocolVersion>
+    <adminEmail>oai@ncbi.nlm.nih.gov</adminEmail>
+    <earliestDatestamp>1999-01-01</earliestDatestamp>
+    <deletedRecord>no</deletedRecord>
+    <granularity>YYYY-MM-DD</granularity>
+    <description>
+      <oai-identifier xmlns="http://www.openarchives.org/OAI/2.0/oai-identifier" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai-identifier http://www.openarchives.org/OAI/2.0/oai-identifier.xsd">
+        <scheme>oai</scheme>
+        <repositoryIdentifier>pubmedcentral.gov</repositoryIdentifier>
+        <delimiter>:</delimiter>
+        <sampleIdentifier>oai:pubmedcentral.gov:13900</sampleIdentifier>
+      </oai-identifier>
+    </description>
+  </Identify>
+</OAI-PMH>

metadata

@@ -1,61 +1,243 @@
---- !ruby/object:Gem::Specification 
-rubygems_version: 0.8.11
-specification_version: 1
+--- !ruby/object:Gem::Specification
 name: oai
-version: !ruby/object:Gem::Version 
-  version: 0.0.1
-date: 2006-04-20 00:00:00 -05:00
-summary: A ruby library for working with the Open Archive Initiative Protocol for Metadata Harvesting (OAI-PMH)
-require_paths: 
-  - lib
-email: ehs@pobox.com
-homepage: http://www.textualize.com/ruby-marc
-rubyforge_project: 
-description: 
+version: !ruby/object:Gem::Version
+  version: 1.3.0
+platform: ruby
+authors:
+- Ed Summers
 autorequire: oai
-default_executable: 
 bindir: bin
-has_rdoc: true
-required_ruby_version: !ruby/object:Gem::Version::Requirement 
-  requirements: 
-    - 
-      - ">"
-      - !ruby/object:Gem::Version 
-        version: 0.0.0
-  version: 
-platform: ruby
-signing_key: 
-cert_chain: 
-authors: 
-  - Ed Summers
-files: 
-  - lib/oai
-  - lib/oai.rb
-  - lib/oai/client.rb
-  - lib/oai/exception.rb
-  - lib/oai/get_record.rb
-  - lib/oai/header.rb
-  - lib/oai/identify.rb
-  - lib/oai/list_identifiers.rb
-  - lib/oai/list_metadata_formats.rb
-  - lib/oai/list_records.rb
-  - lib/oai/list_sets.rb
-  - lib/oai/metadata_format.rb
-  - lib/oai/record.rb
-  - lib/oai/response.rb
-  - lib/oai/set.rb
-  - lib/oai/xpath.rb
-  - test/tc_get_record.rb
-  - test/tc_identify.rb
-  - test/tc_list_identifiers.rb
-  - test/tc_list_metadata_formats.rb
-  - test/tc_list_records.rb
-  - test/tc_list_sets.rb
-test_files: 
-  - test.rb
-rdoc_options: []
-extra_rdoc_files: []
-executables: []
+cert_chain: []
+date: 2025-03-04 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: builder
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 3.1.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 3.1.0
+- !ruby/object:Gem::Dependency
+  name: faraday
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '3'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '3'
+- !ruby/object:Gem::Dependency
+  name: faraday-follow_redirects
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.3.0
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '2'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.3.0
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '2'
+- !ruby/object:Gem::Dependency
+  name: rexml
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: activerecord
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 5.2.0
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '8.1'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 5.2.0
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '8.1'
+- !ruby/object:Gem::Dependency
+  name: appraisal
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: webrick
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+description:
+email: ehs@pobox.com
+executables:
+- oai
 extensions: []
+extra_rdoc_files: []
+files:
+- README.md
+- Rakefile
+- bin/oai
+- examples/models/file_model.rb
+- examples/providers/dublin_core.rb
+- lib/oai.rb
+- lib/oai/client.rb
+- lib/oai/client/get_record.rb
+- lib/oai/client/header.rb
+- lib/oai/client/identify.rb
+- lib/oai/client/list_identifiers.rb
+- lib/oai/client/list_metadata_formats.rb
+- lib/oai/client/list_records.rb
+- lib/oai/client/list_sets.rb
+- lib/oai/client/metadata_format.rb
+- lib/oai/client/record.rb
+- lib/oai/client/response.rb
+- lib/oai/client/resumable.rb
+- lib/oai/constants.rb
+- lib/oai/exception.rb
+- lib/oai/harvester.rb
+- lib/oai/harvester/config.rb
+- lib/oai/harvester/harvest.rb
+- lib/oai/harvester/logging.rb
+- lib/oai/harvester/mailer.rb
+- lib/oai/harvester/shell.rb
+- lib/oai/provider.rb
+- lib/oai/provider/metadata_format.rb
+- lib/oai/provider/metadata_format/oai_dc.rb
+- lib/oai/provider/model.rb
+- lib/oai/provider/model/activerecord_caching_wrapper.rb
+- lib/oai/provider/model/activerecord_wrapper.rb
+- lib/oai/provider/partial_result.rb
+- lib/oai/provider/response.rb
+- lib/oai/provider/response/error.rb
+- lib/oai/provider/response/get_record.rb
+- lib/oai/provider/response/identify.rb
+- lib/oai/provider/response/list_identifiers.rb
+- lib/oai/provider/response/list_metadata_formats.rb
+- lib/oai/provider/response/list_records.rb
+- lib/oai/provider/response/list_sets.rb
+- lib/oai/provider/response/record_response.rb
+- lib/oai/provider/resumption_token.rb
+- lib/oai/set.rb
+- lib/oai/xpath.rb
+- lib/test.rb
+- test/activerecord_provider/config/connection.rb
+- test/activerecord_provider/database/0001_oaipmh_tables.rb
+- test/activerecord_provider/fixtures/dc.yml
+- test/activerecord_provider/helpers/providers.rb
+- test/activerecord_provider/helpers/set_provider.rb
+- test/activerecord_provider/helpers/transactional_test_case.rb
+- test/activerecord_provider/models/dc_field.rb
+- test/activerecord_provider/models/dc_lang.rb
+- test/activerecord_provider/models/dc_set.rb
+- test/activerecord_provider/models/exclusive_set_dc_field.rb
+- test/activerecord_provider/models/oai_token.rb
+- test/activerecord_provider/tc_activerecord_wrapper.rb
+- test/activerecord_provider/tc_ar_provider.rb
+- test/activerecord_provider/tc_ar_sets_provider.rb
+- test/activerecord_provider/tc_caching_paging_provider.rb
+- test/activerecord_provider/tc_simple_paging_provider.rb
+- test/activerecord_provider/test_helper_ar_provider.rb
+- test/client/helpers/provider.rb
+- test/client/helpers/test_wrapper.rb
+- test/client/tc_exception.rb
+- test/client/tc_get_record.rb
+- test/client/tc_http_client.rb
+- test/client/tc_identify.rb
+- test/client/tc_libxml.rb
+- test/client/tc_list_identifiers.rb
+- test/client/tc_list_metadata_formats.rb
+- test/client/tc_list_records.rb
+- test/client/tc_list_sets.rb
+- test/client/tc_low_resolution_dates.rb
+- test/client/tc_utf8_escaping.rb
+- test/client/tc_xpath.rb
+- test/client/test_helper_client.rb
+- test/harvester/tc_harvest.rb
+- test/harvester/test_helper_harvester.rb
+- test/provider/models.rb
+- test/provider/tc_exceptions.rb
+- test/provider/tc_functional_tokens.rb
+- test/provider/tc_instance_provider.rb
+- test/provider/tc_provider.rb
+- test/provider/tc_resumption_tokens.rb
+- test/provider/tc_simple_provider.rb
+- test/provider/test_helper_provider.rb
+- test/test.xml
+homepage: http://github.com/code4lib/ruby-oai
+licenses:
+- MIT
+metadata: {}
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
 requirements: []
-dependencies: []
+rubygems_version: 3.5.3
+signing_key:
+specification_version: 4
+summary: A ruby library for working with the Open Archive Initiative Protocol for
+  Metadata Harvesting (OAI-PMH)
+test_files: []

data/lib/oai/header.rb

@@ -1,12 +0,0 @@
-module OAI
-  class Header
-    include OAI::XPath
-    attr_accessor :identifier, :datestamp, :set_spec
-
-    def initialize(element)
-      @identifier = xpath(element, './/identifier')
-      @datestamp = xpath(element, './/datestamp')
-      @set_spec = xpath(element, './/setSpec')
-    end
-  end
-end

data/lib/oai/record.rb

@@ -1,11 +0,0 @@
-module OAI
-  class Record
-    include OAI::XPath
-    attr_accessor :header, :metadata
-
-    def initialize(element)
-      @header = OAI::Header.new xpath_first(element, './/header')
-      @metadata = xpath_first(element, './/metadata')
-    end
-  end
-end

data/lib/oai/response.rb

@@ -1,20 +0,0 @@
-module OAI
-  class Response
-    include OAI::XPath
-    attr_reader :doc, :resumption_token
-
-    def initialize(doc)
-      @doc = doc
-      @resumption_token = xpath(doc, './/resumptionToken')
-
-      # throw an exception if there was an error
-      error = xpath_first(doc, './/error')
-      if error
-        message = error.text
-        code = error.attributes['code']
-        raise OAI::Exception.new("#{message} [#{code}]")
-      end
-    end
-
-  end
-end

data/test/tc_get_record.rb

@@ -1,27 +0,0 @@
-class GetRecordTest < Test::Unit::TestCase
-  def test_get_one
-    client = OAI::Client.new 'http://www.pubmedcentral.gov/oai/oai.cgi'
-    response = client.get_record :identifier => 'oai:pubmedcentral.gov:13901'
-    assert_kind_of OAI::GetRecordResponse, response
-    assert_kind_of OAI::Record, response.record
-    assert_kind_of REXML::Element, response.record.metadata
-    assert_kind_of OAI::Header, response.record.header
-
-    # minimal check that the header is working
-    assert_equal 'oai:pubmedcentral.gov:13901', 
-      response.record.header.identifier
-
-    # minimal check that the metadata is working
-    assert 'en', response.record.metadata.elements['.//dc:language'].text
-  end
-
-  def test_missing_identifier
-    client = OAI::Client.new 'http://www.pubmedcentral.gov/oai/oai.cgi'
-    begin
-      client.get_record :metadata_prefix => 'oai_dc'
-      flunk 'invalid get_record did not throw OAI::Exception'
-    rescue OAI::Exception => e
-      assert_match /The request includes illegal arguments/, e.to_s
-    end
-  end
-end

data/test/tc_identify.rb

@@ -1,8 +0,0 @@
-class IdentifyTest < Test::Unit::TestCase
-  def test_ok
-    client = OAI::Client.new 'http://www.pubmedcentral.gov/oai/oai.cgi'
-    response = client.identify
-    assert_kind_of OAI::IdentifyResponse, response
-    assert_equal 'PubMed Central (PMC3 - NLM DTD) [http://www.pubmedcentral.nih.gov:80/oai/oai.cgi]', response.to_s
-  end
-end

data/test/tc_list_records.rb

@@ -1,9 +0,0 @@
-class GetRecordsTest < Test::Unit::TestCase
-  def test_get_records
-    client = OAI::Client.new 'http://www.pubmedcentral.gov/oai/oai.cgi'
-    response = client.list_records 
-    assert_kind_of OAI::ListRecordsResponse, response
-    assert response.entries.size > 0
-    assert_kind_of OAI::Record,  response.entries[0]
-  end
-end

data/test/tc_list_sets.rb

@@ -1,17 +0,0 @@
-class ListSetsTest < Test::Unit::TestCase
-
-  def test_list
-    client = OAI::Client.new 'http://www.pubmedcentral.gov/oai/oai.cgi'
-    response = client.list_sets
-    assert_kind_of OAI::ListSetsResponse, response
-    assert response.entries.size > 0
-    assert_kind_of OAI::Set,  response.entries[0]
-  
-    # test iterator
-    for set in response
-      assert_kind_of OAI::Set, set
-    end
-  end
-
-end
-

data/test.rb

@@ -1,12 +0,0 @@
-#!/usr/bin/env ruby
-
-$LOAD_PATH.unshift 'lib'
-
-require 'test/unit'
-require 'oai'
-require 'test/tc_list_identifiers'
-require 'test/tc_list_metadata_formats'
-require 'test/tc_identify'
-require 'test/tc_get_record'
-require 'test/tc_list_records'
-require 'test/tc_list_sets'