adding string stripping for p- properties

I'm not really happy with this code: this sort of thing should be abstracted by the HTML/XML library. Kartik has suggested we switch to BeautifulSoup. it was a bit hacky last time I looked, but if it reduces this kind of thing, I'd be in favour.
microformats · Feb 21, 2014 · f31e3d9 · f31e3d9
1 parent 7b8de5a
commit f31e3d9
Show file tree

Hide file tree

Showing 3 changed files with 20 additions and 1 deletion.
diff --git a/mf2py/parser.py b/mf2py/parser.py
@@ -139,7 +139,10 @@ def parse_props(el, is_root_element=False):
                         # TODO: parse for value-class here
                         prop_name = prop[2:]
                         prop_value = props.get(prop_name, [])
-                        prop_value.append(el.firstChild.nodeValue)
+                        # TODO: this is a goddamn horror show right here
+                        text_value = " ".join(t.nodeValue for t in el.childNodes if t.nodeType == t.TEXT_NODE)
+                        text_value = text_value.strip()
+                        prop_value.append(text_value)
 
                         if prop_value is not []:
                             props[prop_name] = prop_value

diff --git a/test/examples/string_stripping.html b/test/examples/string_stripping.html
@@ -0,0 +1,11 @@
+<!DOCTYPE html>
+<html>
+  <head>
+    <title>String Stripping example</title>
+  </head>
+  <body>
+    <div class="h-card">
+      <span class="p-name">  Tom Morris </span>
+    </div>
+  </body>
+</html>
diff --git a/test/test_parser.py b/test/test_parser.py
@@ -115,6 +115,11 @@ def test_backcompat():
     result = parse_fixture("backcompat.html")
     assert set(result["items"][0]["type"]) == set(["h-card"])
 
+def test_string_strip():
+    result = parse_fixture("string_stripping.html")
+    print result 
+    assert result["items"][0]["properties"]["name"][0] == "Tom Morris"
+
 if __name__ == '__main__':
     result = parse_fixture("nested_multiple_classnames.html")
     pprint(result)