Fixes URL regex, updates nosetests set up, bumps up the version for release. (#9)

Author: cadmiumkitty

docs/source/conf.py

@@ -24,9 +24,9 @@
 author = 'Eugene Morozov'
 
 # The short X.Y version
-version = 'v1.1.0'
+version = 'v1.1.1'
 # The full version, including alpha/beta/rc tags
-release = 'v1.1.0'
+release = 'v1.1.1'
 
 
 # -- General configuration ---------------------------------------------------

rdfpandas/graph.py

@@ -39,8 +39,8 @@ def to_graph(df: pd.DataFrame, namespace_manager: NamespaceManager = None) -> Gr
 
     for (index, series) in df.iterrows():
         for (column, value) in series.iteritems():
-            match = re.search('([\w?:/.]*)(\{(\w*)\})?(\[(\d*)\])?(\(([\w?:/.]*)\))?(@(\w*))?', column)
-
+            # Matching unreserved, gen-delims and sub-delims with exception of "(", ")", "@", "[" and "]" from RFC 3986
+            match = re.search('([\w\-._~:/?#!$&\'*+,;=]*)(\{(\w*)\})?(\[(\d*)\])?(\(([\w?:/.]*)\))?(@(\w*))?', column)
             if pd.notna(value) and pd.notnull(value):
                 s = _get_identifier(prefixes, index)
                 p = _get_identifier(prefixes, match.group(1))

setup.py

@@ -11,7 +11,7 @@
 
 setup(
     name = 'rdfpandas',
-    version = '1.1.0',
+    version = '1.1.1',
     description = 'RDF support for Pandas',
     long_description = readme,
     author = 'Eugene Morozov',

tests/__init__.py

@@ -0,0 +1 @@
+from .test_graph import ConversionTestCase

tests/test_graph.py

@@ -177,10 +177,12 @@ def test_should_convert_data_frame_to_graph_uriref(self):
 
         ds1 = pd.Series(data=['https://google.com'], index = ['http://github.com/cadmiumkitty/rdfpandas/one'], dtype = np.unicode_)
         ds2 = pd.Series(data=['skos:broader'], index = ['http://github.com/cadmiumkitty/rdfpandas/one'], dtype = np.unicode_)
+        ds3 = pd.Series(data=['skos:Concept'], index = ['http://github.com/cadmiumkitty/rdfpandas/one'], dtype = np.unicode_)
 
         df = pd.DataFrame({
             'http://github.com/cadmiumkitty/rdfpandas/uri{URIRef}': ds1,
-            'http://github.com/cadmiumkitty/rdfpandas/curie{URIRef}': ds2
+            'http://github.com/cadmiumkitty/rdfpandas/curie{URIRef}': ds2,
+            'http://www.w3.org/1999/02/22-rdf-syntax-ns#type{URIRef}': ds3
             })
 
         g_expected = Graph()
@@ -191,6 +193,9 @@ def test_should_convert_data_frame_to_graph_uriref(self):
         g_expected.add((URIRef('http://github.com/cadmiumkitty/rdfpandas/one'),
                 URIRef('http://github.com/cadmiumkitty/rdfpandas/curie'), 
                 URIRef('skos:broader')))
+        g_expected.add((URIRef('http://github.com/cadmiumkitty/rdfpandas/one'),
+                URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#type'), 
+                URIRef('skos:Concept')))
 
         g_result = rdfpandas.to_graph(df)
 
@@ -333,7 +338,7 @@ def test_should_roundtrip_csv_to_graph_to_csv(self):
         """Should roundtrip DF -> Graph -> DF
         """
 
-        df = pd.read_csv('./csv/test.csv', index_col = '@id', keep_default_na = True)
+        df = pd.read_csv('./tests/csv/test.csv', index_col = '@id', keep_default_na = True)
         namespace_manager = NamespaceManager(Graph())
         namespace_manager.bind('skos', SKOS)
         namespace_manager.bind('rdfpandas', Namespace('http://github.com/cadmiumkitty/rdfpandas/'))
@@ -348,7 +353,7 @@ def test_should_roundtrip_graph_to_csv_to_graph(self):
         """
 
         g = rdflib.Graph()
-        g.parse('./rdf/test.ttl', format = 'ttl')
+        g.parse('./tests/rdf/test.ttl', format = 'ttl')
         df = rdfpandas.to_dataframe(g)
         print(df.T)
         g_result = rdfpandas.to_graph(df, g.namespace_manager)