Add requirements for optional tests based on mf:requires. Specific requirements defined in the test vocabulary.

Fixes #365.

Author: gkellogg

Gemfile.lock

@@ -1,5 +1,5 @@
 GIT
-  remote: git://github.com/ruby-rdf/json-ld.git
+  remote: https://github.com/ruby-rdf/json-ld.git
   revision: 241ef57e096cefdff55ef6d39a917049eec04529
   branch: develop
   specs:
@@ -80,7 +80,7 @@ GEM
     nokogumbo (2.0.2)
       nokogiri (~> 1.8, >= 1.8.4)
     public_suffix (4.0.3)
-    rack (2.1.2)
+    rack (2.2.2)
     rake (13.0.1)
     rdf (3.1.1)
       hamster (~> 3.0)
@@ -176,4 +176,4 @@ DEPENDENCIES
   redcarpet
 
 BUNDLED WITH
-   1.17.3
+   2.1.4

tests/README.md

@@ -78,6 +78,7 @@ Implementations create their own infrastructure for running the test suite. In p
 * Some algorithms, particularly _fromRdf_, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of `@list`. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).
 * Some _toRdf_ tests require the use of [JSON Canonicalization Scheme](https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05) to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using [RDF Dataset Isomorphism](https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism). These tests are marked using the `useJCS` option.
 * When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from `_:b0` may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and _expected_ to datsets to extract a bijective mapping of blank node labels between the two datasets as described in [RDF Dataset Isomorphism](https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism)).
+* Some tests may have a `requires` property, indicating some optional behavior described by a test vocabulary term.
 
 # Contributing
 

tests/compact-manifest.html

@@ -73,6 +73,7 @@ <h1>Running tests</h1>
 <li>Some algorithms, particularly <em>fromRdf</em>, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of <code>@list</code>. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).</li>
 <li>Some <em>toRdf</em> tests require the use of <a href="https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05">JSON Canonicalization Scheme</a> to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>. These tests are marked using the <code>useJCS</code> option.</li>
 <li>When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from <code>_:b0</code> may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and <em>expected</em> to datsets to extract a bijective mapping of blank node labels between the two datasets as described in <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>).</li>
+<li>Some tests may have a <code>requires</code> property, indicating some optional behavior described by a test vocabulary term.</li>
 </ul>
 
 <h1>Contributing Tests</h1>

tests/expand-manifest.html

@@ -73,6 +73,7 @@ <h1>Running tests</h1>
 <li>Some algorithms, particularly <em>fromRdf</em>, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of <code>@list</code>. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).</li>
 <li>Some <em>toRdf</em> tests require the use of <a href="https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05">JSON Canonicalization Scheme</a> to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>. These tests are marked using the <code>useJCS</code> option.</li>
 <li>When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from <code>_:b0</code> may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and <em>expected</em> to datsets to extract a bijective mapping of blank node labels between the two datasets as described in <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>).</li>
+<li>Some tests may have a <code>requires</code> property, indicating some optional behavior described by a test vocabulary term.</li>
 </ul>
 
 <h1>Contributing Tests</h1>

tests/flatten-manifest.html

@@ -73,6 +73,7 @@ <h1>Running tests</h1>
 <li>Some algorithms, particularly <em>fromRdf</em>, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of <code>@list</code>. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).</li>
 <li>Some <em>toRdf</em> tests require the use of <a href="https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05">JSON Canonicalization Scheme</a> to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>. These tests are marked using the <code>useJCS</code> option.</li>
 <li>When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from <code>_:b0</code> may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and <em>expected</em> to datsets to extract a bijective mapping of blank node labels between the two datasets as described in <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>).</li>
+<li>Some tests may have a <code>requires</code> property, indicating some optional behavior described by a test vocabulary term.</li>
 </ul>
 
 <h1>Contributing Tests</h1>

tests/fromRdf-manifest.html

@@ -68,6 +68,7 @@ <h1>Running tests</h1>
 <li>Some algorithms, particularly <em>fromRdf</em>, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of <code>@list</code>. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).</li>
 <li>Some <em>toRdf</em> tests require the use of <a href="https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05">JSON Canonicalization Scheme</a> to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>. These tests are marked using the <code>useJCS</code> option.</li>
 <li>When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from <code>_:b0</code> may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and <em>expected</em> to datsets to extract a bijective mapping of blank node labels between the two datasets as described in <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>).</li>
+<li>Some tests may have a <code>requires</code> property, indicating some optional behavior described by a test vocabulary term.</li>
 </ul>
 
 <h1>Contributing Tests</h1>

tests/html-manifest.html

@@ -88,6 +88,7 @@ <h1>Running tests</h1>
 <li>Some algorithms, particularly <em>fromRdf</em>, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of <code>@list</code>. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).</li>
 <li>Some <em>toRdf</em> tests require the use of <a href="https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05">JSON Canonicalization Scheme</a> to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>. These tests are marked using the <code>useJCS</code> option.</li>
 <li>When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from <code>_:b0</code> may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and <em>expected</em> to datsets to extract a bijective mapping of blank node labels between the two datasets as described in <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>).</li>
+<li>Some tests may have a <code>requires</code> property, indicating some optional behavior described by a test vocabulary term.</li>
 </ul>
 
 <h1>Contributing Tests</h1>

tests/manifest.html

@@ -38,7 +38,7 @@ <h3><a href="expand-manifest.jsonld">expand</a> tests have <em>input</em> and <e
 <p>The <em>expected</em> results can be compared using <a href="#json-ld-object-comparison">JSON-LD object comparison</a> with the processor output.</p>
 
 <p>For <em>NegativeEvaluationTests</em>, the result is a string associated with the expected error code.</p>
-<h3><a href="html-manifest.jsonld">html</a> tests have <em>input</em> and <em>expected</em> documents and an optional <em>context</em> document.</h3>
+<h3><a href="html.jsonld">html</a> tests have <em>input</em> and <em>expected</em> documents and an optional <em>context</em> document.</h3>
 
 <p>The <em>expected</em> results can be compared using <a href="#json-ld-object-comparison">JSON-LD object comparison</a> with the processor output after potentially remapping blank node identifiers (see below). Additionally, if the result is compacted and the <code>ordered</code> option is not set, result should be expanded and compared with the expanded <em>expected</em> document also using <a href="#json-ld-object-comparison">JSON-LD object comparison</a>.</p>
 
@@ -110,6 +110,7 @@ <h1>Running tests</h1>
 <li>Some algorithms, particularly <em>fromRdf</em>, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of <code>@list</code>. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).</li>
 <li>Some <em>toRdf</em> tests require the use of <a href="https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05">JSON Canonicalization Scheme</a> to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>. These tests are marked using the <code>useJCS</code> option.</li>
 <li>When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from <code>_:b0</code> may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and <em>expected</em> to datsets to extract a bijective mapping of blank node labels between the two datasets as described in <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>).</li>
+<li>Some tests may have a <code>requires</code> property, indicating some optional behavior described by a test vocabulary term.</li>
 </ul>
 
 <h1>Contributing Tests</h1>

tests/mk_vocab.rb

@@ -26,6 +26,7 @@
         ontology: compacted['@graph'].detect {|o| o['@id'] == "https://w3c.github.io/json-ld-api/tests/vocab#"},
         classes: compacted['@graph'].select {|o| o['@type'] == "rdfs:Class"}.sort_by {|o| o['rdfs:label']},
         properties: compacted['@graph'].select {|o| o['@type'] == "rdf:Property"}.sort_by {|o| o['rdfs:label']},
+        requirements: compacted['@graph'].select {|o| o['@type'] == "mf:Requirement"}.sort_by {|o| o['rdfs:label']},
         source: compacted.to_json(JSON::LD::JSON_STATE)
       )
       File.open("vocab.html", "w") {|fh| fh.write html}

tests/remote-doc-manifest.html

@@ -87,6 +87,7 @@ <h1>Running tests</h1>
 <li>Some algorithms, particularly <em>fromRdf</em>, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of <code>@list</code>. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).</li>
 <li>Some <em>toRdf</em> tests require the use of <a href="https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05">JSON Canonicalization Scheme</a> to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>. These tests are marked using the <code>useJCS</code> option.</li>
 <li>When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from <code>_:b0</code> may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and <em>expected</em> to datsets to extract a bijective mapping of blank node labels between the two datasets as described in <a href="https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism">RDF Dataset Isomorphism</a>).</li>
+<li>Some tests may have a <code>requires</code> property, indicating some optional behavior described by a test vocabulary term.</li>
 </ul>
 
 <h1>Contributing Tests</h1>

tests/template.haml

@@ -124,6 +124,7 @@
       * Some algorithms, particularly _fromRdf_, may not preserve the order of statements listed in the input document, and provision should be taken for performing unordered array comparison, for arrays other than values of `@list`. (This may be difficult for compacted results, where array value ordering is dependent on the associated term definition).
       * Some _toRdf_ tests require the use of [JSON Canonicalization Scheme](https://tools.ietf.org/html/draft-rundgren-json-canonicalization-scheme-05) to properly generate RDF Literals from JSON literal values. This algorithm is non-normative, but is assumed to be used to properly compare results using [RDF Dataset Isomorphism](https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism). These tests are marked using the `useJCS` option.
       * When comparing documents after flattening, framing or generating RDF, blank node identifiers may not be predictable. Implementations using the JSON-LD 1.0 algorithm, where output is always sorted and blank node identifiers are generated sequentially from `_:b0` may continue to use a simple object comparison. Otherwise, implementations should take this into consideration. (One way to do this may be to reduce both results and _expected_ to datsets to extract a bijective mapping of blank node labels between the two datasets as described in [RDF Dataset Isomorphism](https://www.w3.org/TR/rdf11-concepts/#dfn-dataset-isomorphism)).
+      * Some tests may have a `requires` property, indicating some optional behavior described by a test vocabulary term.
 
       # Contributing Tests
 
@@ -196,3 +197,6 @@
                         - entry['option'].each do |k, v|
                           %dt=k
                           %dd=v
+                  - if entry['requires']
+                    %dt="Requires"
+                    %dd= entry['requires']