Added in splitting up of emails when preprocessing content.

src/watson/backends.py

@@ -1,5 +1,7 @@
 """Search backends used by django-watson."""
 
+import re
+
 from django.conf import settings
 from django.core.exceptions import ImproperlyConfigured
 from django.contrib.contenttypes.models import ContentType
@@ -14,7 +16,7 @@ def regex_from_search_text(search_text):
     words = search_text.split()
     return u"|".join(
         u"(\s{word}\s)|(^{word}\s)|(\s{word}$)|(^{word}$)".format(
-            word = word,
+            word = re.escape(word),
         )
         for word in words
     )

src/watson/registration.py

@@ -1,6 +1,6 @@
 """Adapters for registering models with django-watson."""
 
-import operator
+import operator, re
 from threading import local
 from contextlib import contextmanager
 from functools import wraps
@@ -24,6 +24,10 @@ class SearchAdapterError(Exception):
     """Something went wrong with a search adapter."""
 
 
+# Used for splitting up email addresses.
+RE_EMAIL = re.compile(u"([a-z0-9][a-z0-9\.+]*)@([a-z0-9\.+]*[a-z])", re.IGNORECASE)
+
+
 class SearchAdapter(object):
 
     """An adapter for performing a full-text search on a model."""
@@ -66,6 +70,20 @@ class SearchAdapter(object):
         # Resolution complete!
         return value
     
+    def prepare_content(self, content):
+        """Sanitizes the given content string for better parsing by the search engine."""
+        # Strip out HTML tags.
+        content = strip_tags(content)
+        # Split up email addresess.
+        def split_email(match):
+            return u" ".join((
+                match.group(0),
+                match.group(1),
+                match.group(2),
+            ))
+        content = RE_EMAIL.sub(split_email, content)
+        return content
+    
     def get_title(self, obj):
         """Returns the search title for the given obj."""
         return unicode(obj)
@@ -84,10 +102,10 @@ class SearchAdapter(object):
         # Exclude named fields.
         field_names = (field_name for field_name in field_names if field_name not in self.exclude)
         # Create the text.
-        return u" ".join(
-            strip_tags(self._resolve_field(obj, field_name))
+        return self.prepare_content(u" ".join(
+            self._resolve_field(obj, field_name)
             for field_name in field_names
-        )        
+        ))
     
     def get_url(self, obj):
         """Return the URL of the given obj."""

src/watson/tests.py

@@ -168,7 +168,14 @@ class InternalsTest(SearchTestBase):
         call_command("buildwatson")
         # Make sure that we have four again (including duplicates).
         self.assertEqual(SearchEntry.objects.count(), 4)
-        
+    
+    def testSearchEmailParts(self):
+        with watson.context():
+            self.test11.content = "foo@bar.com"
+            self.test11.save()
+        self.assertEqual(watson.search("foo").count(), 1)
+        self.assertEqual(watson.search("bar.com").count(), 1)
+        self.assertEqual(watson.search("foo@bar.com").count(), 1)
         
 class SearchTest(SearchTestBase):