Issue 1003

rdflib/graph.py

@@ -293,8 +293,9 @@ class Graph(Node):
     For more on named graphs, see: http://www.w3.org/2004/03/trix/
     """
 
-    def __init__(self, store="default", identifier=None, namespace_manager=None):
+    def __init__(self, store="default", identifier=None, namespace_manager=None, base=None):
         super(Graph, self).__init__()
+        self.base = base
         self.__identifier = identifier or BNode()
 
         if not isinstance(self.__identifier, Node):
@@ -952,14 +953,19 @@ def serialize(
         Format support can be extended with plugins,
         but "xml", "n3", "turtle", "nt", "pretty-xml", "trix", "trig" and "nquads" are built in.
         """
+
+        # if a base is set here, use it, else if it was set at graph creation time, use that (self.base)
+        if base is not None:
+            self.base = base
+
         serializer = plugin.get(format, Serializer)(self)
         if destination is None:
             stream = BytesIO()
-            serializer.serialize(stream, base=base, encoding=encoding, **args)
+            serializer.serialize(stream, base=self.base, encoding=encoding, **args)
             return stream.getvalue()
         if hasattr(destination, "write"):
             stream = destination
-            serializer.serialize(stream, base=base, encoding=encoding, **args)
+            serializer.serialize(stream, base=self.base, encoding=encoding, **args)
         else:
             location = destination
             scheme, netloc, path, params, _query, fragment = urlparse(location)
@@ -970,7 +976,7 @@ def serialize(
                 return
             fd, name = tempfile.mkstemp()
             stream = os.fdopen(fd, "wb")
-            serializer.serialize(stream, base=base, encoding=encoding, **args)
+            serializer.serialize(stream, base=self.base, encoding=encoding, **args)
             stream.close()
             if hasattr(shutil, "move"):
                 shutil.move(name, path)

rdflib/plugins/serializers/turtle.py

@@ -224,7 +224,9 @@ def serialize(self, stream, base=None, encoding=None,
                   spacious=None, **args):
         self.reset()
         self.stream = stream
-        self.base = base
+        # if base is set here, override previously set base if set at graph init
+        if base is not None:
+            self.base = base
 
         if spacious is not None:
             self._spacious = spacious
@@ -291,6 +293,9 @@ def getQName(self, uri, gen_prefix=True):
     def startDocument(self):
         self._started = True
         ns_list = sorted(self.namespaces.items())
+
+        if self.base:
+            self.write(self.indent() + '@base <%s> .\n' % self.base)
         for prefix, uri in ns_list:
             self.write(self.indent() + '@prefix %s: <%s> .\n' % (prefix, uri))
         if ns_list and self._spacious:

test/test_issue1003.py

@@ -0,0 +1,58 @@
+from rdflib import Graph, Literal, Namespace, RDF, URIRef
+from rdflib.namespace import SKOS, DCTERMS
+
+"""
+Testing scenarios:
+    1. no base set
+    2. base set at graph creation
+    3. base set at serialization
+    4. base set at both graph creation & serialization, serialization overrides
+"""
+
+# variables
+base = Namespace("http://example.org/")
+title = Literal("Title", lang="en")
+description = Literal("Test Description", lang="en")
+creator = URIRef("https://creator.com")
+cs = URIRef("")
+
+# starting graph
+g = Graph()
+g.add((cs, RDF.type, SKOS.ConceptScheme))
+g.add((cs, DCTERMS.creator, creator))
+g.add((cs, DCTERMS.source, URIRef("nick")))
+g.bind("dct", DCTERMS)
+g.bind("skos", SKOS)
+
+
+# 1. no base set
+g1 = Graph()
+g1 += g
+# @base should not be in output
+assert "@base" not in g.serialize(format='turtle').decode("utf-8")
+
+
+# 2. base set at graph creation
+g2 = Graph(base=base)
+g2 += g
+# @base should be in output
+assert "@base" in g2.serialize(format='turtle').decode("utf-8")
+
+
+# 3. base set at serialization
+g3 = Graph()
+g3 += g
+# @base should be in output
+assert "@base" in g3.serialize(format='turtle', base=base).decode("utf-8")
+
+
+# 4. base set at both graph creation & serialization, serialization overrides
+g4 = Graph(base=Namespace("http://nothing.com/"))
+g4 += g
+
+# @base should be in output and it should be http://example.org/ (graph init copy)
+assert "@base <http://nothing.com/>" in g4.serialize(format='turtle').decode("utf-8")
+# @base should be in output and it should be http://example.org/, not http://nothing.com/ (serialization overwritten)
+assert "@base <http://example.org/>" in g4.serialize(format='turtle', base=base).decode("utf-8")
+# just checking that the graph init base isn't sneakily in output
+assert "@base <http://nothing.com/>" not in g4.serialize(format='turtle', base=base).decode("utf-8")