bento_search 1.4.4 → 1.5.0

data/test/search_engines/ebsco_host_engine_test.rb

@@ -310,6 +310,36 @@ class EbscoHostEngineTest < ActiveSupport::TestCase
     end
   end
 
+  test_with_cassette("multi-field author title", :ebscohost) do
+    results = @engine.search(:query => {:title => '"Reflections on the History of Debt Resistance"',
+                              :author => "Caffentzis"})
+
+    assert ! results.failed?
+
+    assert_equal 1, results.total_items
+
+    result = results.first
+
+    assert_equal "10.1215/00382876-2345315", result.doi
+  end
+
+  test_with_cassette("multi-field citation numbers", :ebscohost) do
+    results = @engine.search(:query => {
+      :issn       => "00382876",
+      :volume     => "112",
+      :issue      => "4",
+      :start_page => "824"
+    })
+
+    assert ! results.failed?
+    
+    assert_equal 1, results.total_items
+
+    result = results.first
+
+    assert_equal "10.1215/00382876-2345315", result.doi
+  end
+
 
   test("illegal arg for get with id with no colon") do
     assert_raise ArgumentError do
@@ -328,6 +358,14 @@ class EbscoHostEngineTest < ActiveSupport::TestCase
     assert_present first.issn
   end
 
+  test_with_cassette("affiliation search", :ebscohost) do
+    results = @engine.search('"johns hopkins"', :semantic_search_field => :author_affiliation)
+
+    assert ! results.failed?
+
+    assert_present results
+  end
+
 
 
 end

data/test/search_engines/google_books_engine_test.rb

@@ -122,8 +122,24 @@ class GoogleBooksEngineTest < ActiveSupport::TestCase
     
     query_params = CGI.parse( URI.parse(url).query )
     
-    assert_match /^|\sintitle\:cancer\s|$/, query_params["q"].first
-    assert_match /^|\sintitle\:"by radiation"\s|$/, query_params["q"].first
+    assert_match /(^|\s)intitle\:cancer(\s|$)/, query_params["q"].first
+    assert_match /(^|\s)intitle\:"by radiation"(\s|$)/, query_params["q"].first
+  end
+
+  def test_multi_search
+    # Have to use protected methods to get at what we want to test. 
+    # Is there a better way to factor this for testing?
+    norm_args = @engine.send(:parse_search_arguments, 
+      :query => {:title => "manufacturing consent", :author => "chomsky", "inpublisher" => "something"}
+    )
+    url = @engine.send(:args_to_search_url, norm_args)
+
+    query_params = CGI.parse( URI.parse(url).query )
+    
+    assert_match /(^|\s)intitle\:manufacturing(\s|$)/, query_params["q"].first
+    assert_match /(^|\s)intitle\:consent(\s|$)/, query_params["q"].first
+    assert_match /(^|\s)inauthor\:chomsky(\s|$)/, query_params["q"].first
+    assert_match /(^|\s)inpublisher\:something(\s|$)/, query_params["q"].first
   end
     
   

data/test/search_engines/scopus_engine_test.rb

@@ -32,6 +32,20 @@ class ScopusEngineTest < ActiveSupport::TestCase
     
     assert_equal "http://api.elsevier.com/content/search/index:SCOPUS?query=AUTH%28one+two%29&sort=refeid", url
   end
+
+  def test_construct_multi_field
+    url = @engine.send(:scopus_url, :query => {"TITLE" => "Terrible Diseases", "AUTH" => "John Smith"})
+
+    url           = URI.parse(url)
+    params        = CGI.parse(url.query)
+    scopus_query  = params["query"].first
+
+    query_components = scopus_query.split(" AND ")
+
+    assert_equal 2, query_components.length
+    assert_includes query_components, "AUTH(John Smith)"
+    assert_includes query_components, "TITLE(Terrible Diseases)"
+  end
   
   def test_construct_search_with_per_page
     url = @engine.send(:scopus_url, :query => "one two", :per_page => 30)
@@ -129,7 +143,37 @@ class ScopusEngineTest < ActiveSupport::TestCase
     assert results.first.title.downcase.include?("cancer"), "Title includes query term"    
   end
   
-  
+  test_with_cassette("multi-field search", :scopus) do
+    results = @engine.search(:query => 
+      { :title => "Protein measurement with the folin phenol reagent",
+        :author => "Lowry",
+        :issn => "0021-9258"
+      })
+
+    assert ! results.failed?
+    assert_present results
+
+    results.each do |result|
+      assert_equal "00219258", result.issn
+      assert_includes result.title.downcase, "Protein measurement with the folin phenol reagent".downcase
+    end
+  end
+
+  test_with_cassette("multi-fielded citation details search", :scopus) do
+    results = @engine.search(:query =>{
+      :issn       => "15385132",
+      :volume     => "12",
+      :issue      => "2",
+      :start_page => "179"
+    })
+
+    assert ! results.failed?
+    assert_equal 1, results.total_items 
+
+    item = results.first
+
+    assert item.title.start_with?("Architects and planners in the middle of a road war")
+  end
     
   
 end

data/test/search_engines/search_engine_base_test.rb

@@ -1,6 +1,8 @@
 require 'test_helper'
 
 class ParseSearchArgumentsTest < ActiveSupport::TestCase
+  MockEngine = BentoSearch::MockEngine
+
   class Dummy
     include BentoSearch::SearchEngine
     
@@ -153,6 +155,63 @@ class ParseSearchArgumentsTest < ActiveSupport::TestCase
     end
     
   end
+
+
+  describe "multi-field search" do
+    it "complains with multi-query and search_field" do
+      engine = MockEngine.new(:multi_field_search => true)
+      assert_raises(ArgumentError) { engine.search(:query => {:title => "foo"}, :semantic_search_field => :author)}
+      assert_raises(ArgumentError) { engine.search(:query => {:title => "foo"}, :search_field => "something")}
+    end
+
+    it "rejects if search engine does not support" do
+      engine = MockEngine.new(:multi_field_search => false)
+      assert_raises(ArgumentError) { engine.search(:query => {:title => "title", :author => "author"}) }
+    end
+
+    it "converts semantic search fields" do
+      engine = MockEngine.new(:multi_field_search => true, 
+        :search_field_definitions => {
+          "internal_title_field"  => {:semantic => :title},
+          "internal_author_field" => {:semantic => :author}  
+        })
+
+      engine.search(:query => {:title => "title query", :author => "author query"})
+
+      assert_equal(
+        { "internal_title_field"  => "title query", 
+          "internal_author_field" => "author query"},
+        engine.last_args[:query]
+      )
+    end
+
+    it "passes through other fields" do
+      engine = MockEngine.new(:multi_field_search => true, 
+        :search_field_definitions => {
+          "internal_title_field"  => {:semantic => :title},
+          "internal_author_field" => {:semantic => :author}  
+        })
+
+      engine.search(:query => {"internal_title_field" => "query", "other field" => "query"})
+
+      assert_equal(
+       {"internal_title_field" => "query", "other field" => "query"},
+        engine.last_args[:query]
+      )
+    end
+
+    it "complains on unrecognized field if configured" do
+      engine = MockEngine.new(:multi_field_search => true, 
+        :unrecognized_search_field => "raise",
+        :search_field_definitions => {
+          "internal_title_field"  => {:semantic => :title},
+          "internal_author_field" => {:semantic => :author}  
+        })      
+      assert_raises(ArgumentError) do
+        engine.search(:query => {"internal_title_field" => "query", "other field" => "query"})
+      end
+    end
+  end
   
   def test_semantic_blank_ignored
     d = Dummy.new

data/test/search_engines/worldcat_sru_dc_engine_test.rb

@@ -93,6 +93,18 @@ class WorldcatSruDcEngineTest < ActiveSupport::TestCase
       assert_includes components, clause
     end        
   end
+
+  def test_construt_cql_multi_field
+    cql = @engine.construct_cql_query(:query => {"srw.ti" => "manufacturing", "srw.au" => "chomsky"})
+
+    components = cql.split(" AND ")
+
+    assert_equal 2, components.length
+
+    ['srw.au = "chomsky"', 'srw.ti = "manufacturing"'].each do |clause|
+      assert_includes components, clause
+    end
+  end
   
   test_with_cassette("max_out_pagination", :worldcat_sru_dc) do
     #worldcat maxes out at start 9999, we silently correct
@@ -146,6 +158,11 @@ class WorldcatSruDcEngineTest < ActiveSupport::TestCase
       @engine.get("NOT EXISTING")
     end
   end
+
+  test_with_cassette("multi field search", :worldcat_sru_dc) do
+    results = @engine.search(:query => {:author => "Chomsky", :title => "Manufacturing"})
+    assert ! results.failed?
+  end
   
   
 end

data/test/vcr_cassettes/doaj_articles/basic_search.yml

@@ -0,0 +1,97 @@
+---
+http_interactions:
+- request:
+    method: get
+    uri: https://doaj.org/api/v1/search/articles/+Breast%20+cancer%20+patients%20+with%20+lobular%20+cancer%20+more%20+commonly%20+have%20+a%20+father%20+than%20+a%20+mother%20+diagnosed%20+with%20+cancer?pageSize=10
+    body:
+      encoding: UTF-8
+      string: ''
+    headers:
+      Cookie:
+      - session=UspzkET/3IS3Z8PIxKnLAhLnXDY=?_id=UycyNmYyZmZkMzhkMjA5MWE0NmNjZTU2YTI4OGJjNDZlNycKcDEKLg==
+  response:
+    status:
+      code: 200
+      message: OK
+    headers:
+      Server:
+      - nginx/1.4.6 (Ubuntu)
+      Date:
+      - Wed, 23 Sep 2015 15:51:24 GMT
+      Content-Type:
+      - application/json
+      Content-Length:
+      - '4515'
+      Connection:
+      - keep-alive
+      Vary:
+      - Accept-Encoding
+      - Accept-Encoding
+      Access-Control-Allow-Origin:
+      - "*"
+      Set-Cookie:
+      - session="UspzkET/3IS3Z8PIxKnLAhLnXDY=?_id=UycyNmYyZmZkMzhkMjA5MWE0NmNjZTU2YTI4OGJjNDZlNycKcDEKLg==";
+        Path=/; HttpOnly
+    body:
+      encoding: UTF-8
+      string: '{"pageSize": 10, "timestamp": "2015-0923T15:51:24Z", "results": [{"last_updated":
+        "2015-04-22T03:59:29Z", "id": "c7cbb0516de24de2bdcfb3d81a40ab1b", "bibjson":
+        {"start_page": "497", "title": "Breast cancer patients with lobular cancer
+        more commonly have a father than a mother diagnosed with cancer", "journal":
+        {"publisher": "BioMed Central", "language": ["English"], "license": [{"type":
+        "CC BY", "title": "CC BY"}], "title": "BMC Cancer", "country": "GB", "number":
+        "1", "volume": "11"}, "author": [{"name": "Ellberg Carolina"}, {"name": "Olsson
+        H\u00e5kan"}], "month": "11", "link": [{"url": "http://www.biomedcentral.com/1471-2407/11/497",
+        "type": "fulltext"}], "year": "2011", "identifier": [{"type": "doi", "id":
+        "10.1186/1471-2407-11-497"}, {"type": "pissn", "id": "1471-2407"}], "abstract":
+        "<p>Abstract</p> <p>Background</p> <p>The association between lobular breast
+        cancer and family history is not clear. The aim of the study was to possibly
+        identifying new hereditary patterns predisposing for cancer in the different
+        histopathologic subtypes of breast cancer, with focus on patients with lobular
+        breast cancer and cancer in their first degree relatives.</p> <p>Methods</p>
+        <p>In 1676 consecutive breast cancer patients detailed family history of cancer
+        was related to histopathologic subtype of breast cancer.</p> <p>Results</p>
+        <p>Patients with lobular breast cancer were found to be significantly positively
+        associated with having a father diagnosed with cancer, OR 2.17 (95% confidence
+        interval (CI) 1.37-3.46). The finding persisted after excluding breast cancer
+        in the family. Ductal breast cancer was associated with having a mother diagnosed
+        with cancer. There was a significant association between lobular breast cancer
+        and having a father with prostate cancer, OR 2.4 (CI 1.1-5.3). The occurrence
+        of having a father with prostate cancer for lobular breast cancer patients
+        was higher in the younger patient group, OR 2.9 (CI 1.1-7.8), and was still
+        high but lost statistical significance in the older patient group, OR 1.9
+        (CI 0.5-7.4). The association between lobular breast cancer and a father remained
+        significant after excluding fathers with prostate cancer, OR 1.94 (CI 1.20-3.14).
+        Other commonly occurring tumor types in the father included sarcoma and leukemia.</p>
+        <p>Conclusion</p> <p>We propose that lobular breast cancer is associated with
+        having a father diagnosed with cancer, most commonly prostate carcinoma. Since
+        the association remained after excluding family history of breast cancer,
+        the association seems independent of classical breast cancer heredity. The
+        association with a father diagnosed with cancer also remained after removing
+        prostate cancer, indicating an independence from prostate cancer as well.
+        The reason for this association is genetically unclear, but could involve
+        sex-specific imprinting.</p> ", "subject": [{"code": "RC254-282", "term":
+        "Neoplasms. Tumors. Oncology. Including cancer and carcinogens", "scheme":
+        "LCC"}, {"code": "RC31-1245", "term": "Internal medicine", "scheme": "LCC"},
+        {"code": "R", "term": "Medicine", "scheme": "LCC"}, {"term": "Oncology", "scheme":
+        "DOAJ"}, {"term": "Medicine (General)", "scheme": "DOAJ"}, {"term": "Health
+        Sciences", "scheme": "DOAJ"}, {"code": "RC254-282", "term": "Neoplasms. Tumors.
+        Oncology. Including cancer and carcinogens", "scheme": "LCC"}, {"code": "RC31-1245",
+        "term": "Internal medicine", "scheme": "LCC"}, {"code": "R", "term": "Medicine",
+        "scheme": "LCC"}, {"term": "Oncology", "scheme": "DOAJ"}, {"term": "Medicine
+        (General)", "scheme": "DOAJ"}, {"term": "Health Sciences", "scheme": "DOAJ"},
+        {"code": "RC254-282", "term": "Neoplasms. Tumors. Oncology. Including cancer
+        and carcinogens", "scheme": "LCC"}, {"code": "RC31-1245", "term": "Internal
+        medicine", "scheme": "LCC"}, {"code": "R", "term": "Medicine", "scheme": "LCC"},
+        {"code": "RC254-282", "term": "Neoplasms. Tumors. Oncology. Including cancer
+        and carcinogens", "scheme": "LCC"}, {"code": "RC31-1245", "term": "Internal
+        medicine", "scheme": "LCC"}, {"code": "R", "term": "Medicine", "scheme": "LCC"},
+        {"code": "RC254-282", "term": "Neoplasms. Tumors. Oncology. Including cancer
+        and carcinogens", "scheme": "LCC"}, {"code": "RC31-1245", "term": "Internal
+        medicine", "scheme": "LCC"}, {"code": "R", "term": "Medicine", "scheme": "LCC"}]},
+        "created_date": "2013-03-12T01:21:44Z"}], "query": "+Breast +cancer +patients
+        +with +lobular +cancer +more +commonly +have +a +father +than +a +mother +diagnosed
+        +with +cancer", "total": 1, "page": 1}'
+    http_version: 
+  recorded_at: Wed, 23 Sep 2015 15:51:24 GMT
+recorded_with: VCR 2.9.3

data/test/vcr_cassettes/doaj_articles/catches_errors.yml

@@ -0,0 +1,42 @@
+---
+http_interactions:
+- request:
+    method: get
+    uri: https://doaj.org/api/v1/search/articles_bad_url/+something?pageSize=10
+    body:
+      encoding: UTF-8
+      string: ''
+    headers:
+      Cookie:
+      - session=UspzkET/3IS3Z8PIxKnLAhLnXDY=?_id=UycyNmYyZmZkMzhkMjA5MWE0NmNjZTU2YTI4OGJjNDZlNycKcDEKLg==
+  response:
+    status:
+      code: 404
+      message: NOT FOUND
+    headers:
+      Server:
+      - nginx/1.4.6 (Ubuntu)
+      Date:
+      - Wed, 23 Sep 2015 15:51:26 GMT
+      Content-Type:
+      - application/json
+      Content-Length:
+      - '246'
+      Connection:
+      - keep-alive
+      Vary:
+      - Accept-Encoding
+      - Accept-Encoding
+      Access-Control-Allow-Origin:
+      - "*"
+      Set-Cookie:
+      - session="UspzkET/3IS3Z8PIxKnLAhLnXDY=?_id=UycyNmYyZmZkMzhkMjA5MWE0NmNjZTU2YTI4OGJjNDZlNycKcDEKLg==";
+        Path=/; HttpOnly
+    body:
+      encoding: UTF-8
+      string: '{"status": "not_found", "error": "No endpoint at search/articles_bad_url/+something.
+        See https://doaj.org/api/v1/swagger.json for valid paths or read the documentation
+        at https://doaj.org/api/v1/docs. (ref: f2b2e3d2-620a-11e5-b4d0-04014a62b401)"}'
+    http_version: 
+  recorded_at: Wed, 23 Sep 2015 15:51:26 GMT
+recorded_with: VCR 2.9.3

data/test/vcr_cassettes/doaj_articles/complex_multi-field.yml

@@ -0,0 +1,67 @@
+---
+http_interactions:
+- request:
+    method: get
+    uri: https://doaj.org/api/v1/search/articles/+Anti%5C-war%20+bibjson.author.name:(+Caffentzis)%20+bibjson.title:(+%22Respect%20Your%20Enemies%22%20+first%20+rule%20+of%20+peace)%20+bibjson.journal.title:(+%22Revista%20Theomai%22)?pageSize=10
+    body:
+      encoding: UTF-8
+      string: ''
+    headers: {}
+  response:
+    status:
+      code: 200
+      message: OK
+    headers:
+      Server:
+      - nginx/1.4.6 (Ubuntu)
+      Date:
+      - Wed, 23 Sep 2015 15:58:02 GMT
+      Content-Type:
+      - application/json
+      Content-Length:
+      - '2314'
+      Connection:
+      - keep-alive
+      Vary:
+      - Accept-Encoding
+      - Accept-Encoding
+      Access-Control-Allow-Origin:
+      - "*"
+      Set-Cookie:
+      - session="UspzkET/3IS3Z8PIxKnLAhLnXDY=?_id=UycyNmYyZmZkMzhkMjA5MWE0NmNjZTU2YTI4OGJjNDZlNycKcDEKLg==";
+        Path=/; HttpOnly
+    body:
+      encoding: UTF-8
+      string: '{"pageSize": 10, "timestamp": "2015-0923T15:58:02Z", "results": [{"last_updated":
+        "2015-09-18T18:30:31Z", "id": "3de5b56aa21342f6adb2d313c16c6c63", "bibjson":
+        {"start_page": "0", "title": "Respect Your Enemies - The First Rule of Peace:
+        An Essay Addressed to the U. S. Anti-war Movement", "journal": {"country":
+        "AR", "publisher": "Red Internacional de Estudios sobre Sociedad, Naturaleza
+        y Desarrollo", "number": "7", "language": ["Spanish", "English", "French",
+        "Italian", "Portuguese"], "title": "Revista Theomai"}, "author": [{"name":
+        "George Caffentzis"}], "link": [{"url": "http://www.redalyc.org/articulo.oa?id=12400710",
+        "type": "fulltext", "content_type": "pdf"}], "year": "2003", "identifier":
+        [{"type": "pissn", "id": "1666-2830"}, {"type": "eissn", "id": "1515-6443"}],
+        "abstract": "There is now a fledgling anti-interventionist, anti-war movement
+        in the US. It will have a lot of work to do in the near future, although the
+        present threat of war on Iraq is the most pressing issue it faces. The question
+        is: can the antiwar movement do its work effectively and successfully? At
+        the moment it is not completely marginalized, if the votes in Congress are
+        any indication. On October 9, between one-quarter to one-third of the congressional
+        representatives voted against granting George W. Bush \"war powers.\" But
+        in order to show itself as expressing the majority perspective in this country,
+        it needs new arguments, a new respect (as in \"look again\") for its opponents,
+        a deeper understanding of the reasons for the actions of its opponents, and
+        a realistic assessment of their weaknesses. For its old arguments do not seem
+        convincing to the majority of US citizens, and its lack of curiosity about
+        its opponents and their reasoning is dulling its strategic sense.", "subject":
+        [{"code": "H1-99", "term": "Social sciences (General)", "scheme": "LCC"},
+        {"code": "H", "term": "Social Sciences", "scheme": "LCC"}, {"code": "GE1-350",
+        "term": "Environmental sciences", "scheme": "LCC"}, {"code": "G", "term":
+        "Geography. Anthropology. Recreation", "scheme": "LCC"}]}, "created_date":
+        "2012-03-28T00:03:02Z"}], "query": "+Anti\\-war +bibjson.author.name:(+Caffentzis)
+        +bibjson.title:(+\"Respect Your Enemies\" +first +rule +of +peace) +bibjson.journal.title:(+\"Revista
+        Theomai\")", "total": 1, "page": 1}'
+    http_version: 
+  recorded_at: Wed, 23 Sep 2015 15:58:02 GMT
+recorded_with: VCR 2.9.3