Merge pull request #649 from tgbugs/master

namespace.py fix compute_qname missing namespaces

6 files changed, 121 insertions, 4 deletions

diff --git a/test/n3/n3-writer-test-30.n3 b/test/n3/n3-writer-test-30.n3
new file mode 100644
index 00000000..88cf2210
--- /dev/null
+++ b/test/n3/n3-writer-test-30.n3
@@ -0,0 +1,22 @@
+# Test full length qnames
+
+@prefix :       <http://example.org/here#> .
+@prefix full:   <http://example.org/full> .
+@prefix pref:     <http://example.org/prefix/> .
+@prefix more:   <http://example.org/prefix/more> .
+
+# Test namespace generation
+
+full: :x :y .
+:x full: :y .
+:x :y full: .
+
+full: full: full: . 
+
+# Test existing namespace
+
+more: :x :y .
+:x more: :y .
+:x :y more: .
+
+more: more: more: .
diff --git a/test/n3/n3-writer-test-31.n3 b/test/n3/n3-writer-test-31.n3
new file mode 100644
index 00000000..1c4494f1
--- /dev/null
+++ b/test/n3/n3-writer-test-31.n3
@@ -0,0 +1,15 @@
+# Test unshortenable strict qnames no predicates for xml sanity check
+
+@prefix :       <http://example.org/here#> .
+@prefix evil1:   <http://example.org/1> .
+@prefix evil2:   <http://example.org/prefix/1#> .
+
+# Test namespace generation
+
+evil1: :x :y .
+:x :y evil1: .
+
+# Test existing namespace
+
+evil2:1 :x :y .
+:x :y evil2:1 .
diff --git a/test/test_n3_suite.py b/test/test_n3_suite.py
index f6efe34a..21e6bcba 100644
--- a/test/test_n3_suite.py
+++ b/test/test_n3_suite.py
@@ -1,5 +1,8 @@
 import os
 import sys
+import logging
+
+log = logging.getLogger(__name__)
 
 try:
     from .testutils import check_serialize_parse
@@ -18,6 +21,15 @@ def _get_test_files_formats():
             elif f.endswith('.n3'):
                 yield fpath, 'n3'
 
+def all_n3_files():
+    skiptests = [
+        'test/n3/example-lots_of_graphs.n3',  # only n3 can serialize QuotedGraph, no point in testing roundtrip
+    ]
+    for fpath, fmt in _get_test_files_formats():
+        if fpath in skiptests:
+            log.debug("Skipping %s, known issue" % fpath)
+        else:
+            yield fpath, fmt
 
 def test_n3_writing():
     for fpath, fmt in _get_test_files_formats():
diff --git a/test/test_namespace.py b/test/test_namespace.py
index 9de7ff29..4041433e 100644
--- a/test/test_namespace.py
+++ b/test/test_namespace.py
@@ -22,6 +22,33 @@ class NamespacePrefixTest(unittest.TestCase):
         self.assertEqual(g.compute_qname(URIRef("http://blip/blop")),
                          ("ns4", URIRef("http://blip/"), "blop"))
 
+        # should return empty qnames correctly
+        self.assertEqual(g.compute_qname(URIRef("http://foo/bar/")),
+            ("ns1", URIRef("http://foo/bar/"), ""))
+
+    def test_reset(self):
+        data = ('@prefix a: <http://example.org/a> .\n'
+                'a: <http://example.org/b> <http://example.org/c> .')
+        graph = Graph().parse(data=data, format='turtle')
+        for p, n in tuple(graph.namespaces()):
+            graph.store._IOMemory__namespace.pop(p)
+            graph.store._IOMemory__prefix.pop(n)
+        graph.namespace_manager.reset()
+        self.assertFalse(tuple(graph.namespaces()))
+        u = URIRef('http://example.org/a')
+        prefix, namespace, name = graph.namespace_manager.compute_qname(u, generate=True)
+        self.assertNotEqual(namespace, u)
+
+    def test_reset_preserve_prefixes(self):
+        data = ('@prefix a: <http://example.org/a> .\n'
+                'a: <http://example.org/b> <http://example.org/c> .')
+        graph = Graph().parse(data=data, format='turtle')
+        graph.namespace_manager.reset()
+        self.assertTrue(tuple(graph.namespaces()))
+        u = URIRef('http://example.org/a')
+        prefix, namespace, name = graph.namespace_manager.compute_qname(u, generate=True)
+        self.assertEqual(namespace, u)
+
     def test_n3(self):
         g = Graph()
         g.add((URIRef("http://example.com/foo"),
diff --git a/test/test_roundtrip.py b/test/test_roundtrip.py
index 819c944a..9dfed952 100644
--- a/test/test_roundtrip.py
+++ b/test/test_roundtrip.py
@@ -5,8 +5,11 @@ import rdflib.compare
 try:
     from .test_nt_suite import all_nt_files
     assert all_nt_files
+    from .test_n3_suite import all_n3_files
+    assert all_n3_files
 except:
     from test.test_nt_suite import all_nt_files
+    from test.test_n3_suite import all_n3_files
 
 """
 Test round-tripping by all serializers/parser that are registerd.
@@ -25,9 +28,10 @@ tests roundtripping through rdf/xml with only the literals-02 file
 
 
 SKIP = [
+    ('xml', 'test/n3/n3-writer-test-29.n3'),  # has predicates that cannot be shortened to strict qnames
     ('xml', 'test/nt/qname-02.nt'),  # uses a property that cannot be qname'd
-    # uses a property that cannot be qname'd
-    ('application/rdf+xml', 'test/nt/qname-02.nt'),
+    ('trix', 'test/n3/strquot.n3'),  # contains charachters forbidden by the xml spec
+    ('xml', 'test/n3/strquot.n3'),  # contains charachters forbidden by the xml spec
 ]
 
 
@@ -43,6 +47,7 @@ def roundtrip(e, verbose=False):
     if verbose:
         print("S:")
         print(s)
+        print(s.decode())
 
     g2 = rdflib.ConjunctiveGraph()
     g2.parse(data=s, format=testfmt)
@@ -52,12 +57,12 @@ def roundtrip(e, verbose=False):
         print("Diff:")
         print("%d triples in both" % len(both))
         print("G1 Only:")
-        for t in first:
+        for t in sorted(first):
             print(t)
 
         print("--------------------")
         print("G2 Only")
-        for t in second:
+        for t in sorted(second):
             print(t)
 
     assert rdflib.compare.isomorphic(g1, g2)
@@ -88,6 +93,24 @@ def test_cases():
                 yield roundtrip, (infmt, testfmt, f)
 
 
+def test_n3():
+    global formats
+    if not formats:
+        serializers = set(
+            x.name for x in rdflib.plugin.plugins(
+                None, rdflib.plugin.Serializer))
+        parsers = set(
+            x.name for x in rdflib.plugin.plugins(
+                None, rdflib.plugin.Parser))
+        formats = parsers.intersection(serializers)
+
+    for testfmt in formats:
+        if "/" in testfmt: continue # skip double testing
+        for f, infmt in all_n3_files():
+            if (testfmt, f) not in SKIP:
+                yield roundtrip, (infmt, testfmt, f)
+
+
 if __name__ == "__main__":
     import nose
     if len(sys.argv) == 1:
diff --git a/test/test_turtle_serialize.py b/test/test_turtle_serialize.py
index 0b602b34..81f57847 100644
--- a/test/test_turtle_serialize.py
+++ b/test/test_turtle_serialize.py
@@ -70,6 +70,24 @@ def test_turtle_valid_list():
         assert turtle_serializer.isValidList(o)
 
 
+def test_turtle_namespace():
+   graph = Graph()
+   graph.bind('OBO', 'http://purl.obolibrary.org/obo/')
+   graph.bind('GENO', 'http://purl.obolibrary.org/obo/GENO_')
+   graph.bind('RO', 'http://purl.obolibrary.org/obo/RO_')
+   graph.bind('RO_has_phenotype',
+                   'http://purl.obolibrary.org/obo/RO_0002200')
+   graph.add((URIRef('http://example.org'),
+              URIRef('http://purl.obolibrary.org/obo/RO_0002200'),
+              URIRef('http://purl.obolibrary.org/obo/GENO_0000385')))
+   output = [val for val in
+             graph.serialize(format='turtle').decode().splitlines()
+             if not val.startswith('@prefix')]
+   output = ' '.join(output)
+   assert 'RO_has_phenotype:' in output
+   assert 'GENO:0000385' in output
+
+
 if __name__ == "__main__":
     import nose
     import sys