intersperse class_set instead of using string join (resolves issue #188)

pylode/utils.py

@@ -656,11 +656,11 @@ def _setclass_html(ont__, obj__, back_onts__, ns__, fids__):
             """Makes lists of (union) 'ClassX or Class Y or ClassZ' or
             (intersection) 'ClassX and Class Y and ClassZ'"""
             if (obj__, OWL.unionOf, None) in ont__:
-                joining_word = ' <span class="cardinality">or</span> '
+                joining_word = span("or", _class="cardinality")
             elif (obj__, OWL.intersectionOf, None) in ont__:
-                joining_word = ' <span class="cardinality">and</span> '
+                joining_word = span("and", _class="cardinality")
             else:
-                joining_word = ' <span class="cardinality">,</span> '
+                joining_word = span(",", _class="cardinality")
 
             class_set = set()
             for o in ont__.objects(obj__, OWL.unionOf | OWL.intersectionOf):
@@ -671,7 +671,7 @@ def _setclass_html(ont__, obj__, back_onts__, ns__, fids__):
                         )
                     )
 
-            return raw(joining_word.join([mem.render() for mem in class_set]))
+            return intersperse(class_set, joining_word)
 
         def _bn_html(ont__, back_onts__, ns__, fids__, obj__: BNode):
             # TODO: remove back_onts and fids if not needed by subfunctions
@@ -871,6 +871,11 @@ def _element_html(
     return elems
 
 
+def intersperse(lst, sep):
+    result = [sep] * (len(lst) * 2 - 1)
+    result[0::2] = lst
+    return result
+
 def de_space_html(html):
     # s = "".join([l_.strip().replace("\n", " ") for l_ in html.split("\n")])
     # return re.sub(" +", " ", s)