Update to latest RDF canonicalization test suite

test/acceptance/canonicalization_w3c_test.exs

@@ -13,11 +13,15 @@ defmodule RDF.Canonicalization.W3C.Test do
 
   @path RDF.TestData.path("rdf-canon-tests")
   @base "https://github.com/w3c/rdf-canon/tests/"
-  @manifest TestSuite.manifest_path(@path, "manifest-urdna2015.ttl")
+  @manifest TestSuite.manifest_path(@path, "manifest.ttl")
             |> TestSuite.manifest_graph(base: @base)
 
-  TestSuite.test_cases(@manifest, RDFC.Urdna2015EvalTest)
+  TestSuite.test_cases(@manifest, RDFC.RDFC10EvalTest)
   |> Enum.each(fn test_case ->
+    if RDFC.hashAlgorithm() in RDF.Description.predicates(test_case) do
+      @tag skip: "missing ability to change hash algorithm"
+    end
+
     @tag test_case: test_case
     test TestSuite.test_title(test_case), %{test_case: test_case} do
       file_url = to_string(TestSuite.test_input_file(test_case))

test/data/rdf-canon-tests/README.md

@@ -1,5 +1,29 @@
 ## General instructions for running the RDF Dataset Canonicalization Test suites
-FIXME
+
+### Tests for RDFC-1.0 take input files, specified as N-Quads, and generate Canonical N-Quads output as required by the RDFC-1.0 algorithm.
+
+The result file is in the N-Quads format.
+The test passes if the result compares identically as the expected result as text files.
+
+For a negative evaluation test, the test passes if the implementation generates an error due to excessive calls to
+[Hash N-Degree Quads](https://www.w3.org/TR/rdf-canon/#hash-nd-quads-algorithm).
+
+### Tests for RDFC-1.0 Issued Identifiers Map.
+
+The result file is in the JSON format with keys representing
+the blank node identifiers from the test input,
+and values representing the associated canonical identifier
+from the [issued identifiers map](https://www.w3.org/TR/rdf-canon/#dfn-issued-identifiers-map)
+created as an alternate result
+from [Step 7](https://www.w3.org/TR/rdf-canon/#ca.7) of the
+[RDFC1.0 Canonicalization Algorithm](https://www.w3.org/TR/rdf-canon/#canon-algo-algo).
+The test passes if the value of the resulting
+[issued identifiers map](https://www.w3.org/TR/rdf-canon/#dfn-issued-identifiers-map)
+matches the corresponding expected test result that can be loaded via the `result` field of the test.
+
+Additionally, the keys of the [issued identifiers map](https://www.w3.org/TR/rdf-canon/#dfn-issued-identifiers-map)
+must exactly match the values of the [input blank node identifier map](https://www.w3.org/TR/rdf-canon/#dfn-input-blank-node-identifier-map).
+Note that all blank nodes appearing in the test appear in input blank node identifier map represent blank nodes and the specific value is not considered for test purposes
 
 ## Contributing Tests
 The test manifests and entries are built automatically from
@@ -11,6 +35,10 @@ Tests may be contributed via pull request to
 with suitable changes to the
 [manifest.csv](manifest.csv) and necessary _action_ and _result_ files. 
 
+The [manifest-rdfc10.ttl](manifest-rdfc10.ttl), [manifest-rdfc10.jsonld](manifest-rdfc10.jsonld), and [index.html](index.html) files are built automatically via a GitHub Action when files change in this directory, and should not be edited directly.
+
+The [vocab.html](vocab.html) and [vocab.jsonld](vocab.jsonld) files are built using [mk_vocab.rb](mk_vocab.rb) and are not build automatically by GitHub PR, as they change infrequently.
+
 ## Distribution
 Distributed under both the
 [W3C Test Suite License](http://www.w3.org/Consortium/Legal/2008/04-testsuite-license)