Merge pull request #10 from mcs07/refreader

Fix issues with reference link extraction using HTML/XML readers
mcs07 · Nov 1, 2016 · 608f719 · 608f719
2 parents be8ec3a + f59592b
commit 608f719
Show file tree

Hide file tree

Showing 3 changed files with 6,241 additions and 3 deletions.
diff --git a/chemdataextractor/reader/markup.py b/chemdataextractor/reader/markup.py
@@ -164,14 +164,14 @@ def _parse_table_footnotes(self, fns, refs, specials):
 
     def _parse_reference(self, el):
         """Return reference ID from href or text content."""
-        if el.get('href', '').startswith('#'):
-            return [el.get('href')[1:]]
+        if '#' in el.get('href', ''):
+            return [el.get('href').split('#', 1)[1]]
         elif 'rid' in el.attrib:
             return [el.attrib['rid']]
         elif 'idref' in el.attrib:
             return [el.attrib['idref']]
         else:
-            return [el.text.strip()]
+            return [''.join(el.itertext()).strip()]
 
     def _parse_table(self, el, refs, specials):
         caps = self._css(self.table_caption_css, el)