extract annotations: return anchor to annotation, not the word in text

[librarian.git] / librarian / dcparser.py

diff --git a/librarian/dcparser.py b/librarian/dcparser.py
index 07a90d7..9faffe8 100644 (file)
--- a/librarian/dcparser.py
+++ b/librarian/dcparser.py
@@ -38,9 +38,17 @@ class Person(object):
             names = [ name for name in parts[1].split() if len(name) ] # all non-whitespace tokens
         return cls(surname, *names)
 
+    def readable(self):
+        return u" ".join(self.first_names + (self.last_name,))
+
     def __eq__(self, right):
         return self.last_name == right.last_name and self.first_names == right.first_names
 
+    def __cmp__(self, other):
+        return cmp((self.last_name, self.first_names), (other.last_name, other.first_names))
+
+    def __hash__(self):
+        return hash((self.last_name, self.first_names))
 
     def __unicode__(self):
         if len(self.first_names) > 0:
@@ -260,7 +268,8 @@ class BookInfo(object):
                     if len(v) == 0: continue
                     for x in v:
                         e = etree.Element(field.uri)
-                        e.text = unicode(x)
+                        if x is not None:
+                            e.text = unicode(x)
                         description.append(e)
                 else:
                     e = etree.Element(field.uri)
@@ -280,7 +289,7 @@ class BookInfo(object):
             if v is not None:
                 if field.multiple:
                     if len(v) == 0: continue
-                    v = [ unicode(x) for x in v if v is not None ]
+                    v = [ unicode(x) for x in v if x is not None ]
                 else:
                     v = unicode(v)
 
@@ -296,7 +305,7 @@ class BookInfo(object):
             if v is not None:
                 if field.multiple:
                     if len(v) == 0: continue
-                    v = [ unicode(x) for x in v if v is not None ]
+                    v = [ unicode(x) for x in v if x is not None ]
                 else:
                     v = unicode(v)
                 result[field.name] = v