for #167, deduplicate contacts by largest

4 files changed, 90 insertions, 1 deletions

diff --git a/service/pixelated/adapter/contacts.py b/service/pixelated/adapter/contacts.py
new file mode 100644
index 00000000..f286f220
--- /dev/null
+++ b/service/pixelated/adapter/contacts.py
@@ -0,0 +1,30 @@
+#
+# Copyright (c) 2014 ThoughtWorks, Inc.
+#
+# Pixelated is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# Pixelated is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with Pixelated. If not, see <http://www.gnu.org/licenses/>.
+from email.utils import parseaddr
+
+
+def address_duplication_filter(contacts):
+    contacts_by_mail = dict()
+
+    for contact in contacts:
+        mail_address = extract_mail_address(contact)
+        previous = contacts_by_mail.get(mail_address, '')
+        contacts_by_mail[mail_address] = contact if len(contact) > len(previous) else previous
+    return contacts_by_mail.values()
+
+
+def extract_mail_address(text):
+    return parseaddr(text)[1]
diff --git a/service/pixelated/adapter/search/__init__.py b/service/pixelated/adapter/search/__init__.py
index 5491632f..12829c65 100644
--- a/service/pixelated/adapter/search/__init__.py
+++ b/service/pixelated/adapter/search/__init__.py
@@ -18,6 +18,7 @@ from pixelated.support.encrypted_file_storage import EncryptedFileStorage
 
 import os
 from pixelated.adapter.model.status import Status
+from pixelated.adapter.contacts import address_duplication_filter
 from pixelated.support.functional import flatten
 from whoosh.index import FileIndex
 from whoosh.fields import *
@@ -211,6 +212,6 @@ class SearchEngine(object):
                                       groupedby=sorting.FieldFacet('bcc', allow_overlap=True)).groups()
                 sender = searcher.search(sender.parse("*%s*" % query), limit=None, mask=restrict_q,
                                          groupedby=sorting.FieldFacet('sender', allow_overlap=True)).groups()
-                return flatten([to, cc, bcc, sender])
+                return address_duplication_filter(flatten([to, cc, bcc, sender]))
 
         return []
diff --git a/service/test/integration/test_contacts.py b/service/test/integration/test_contacts.py
index 4974c471..c5baa094 100644
--- a/service/test/integration/test_contacts.py
+++ b/service/test/integration/test_contacts.py
@@ -57,3 +57,26 @@ class ContactsTest(SoledadTestBase):
             self.assertFalse('recipient@trash.com' in contacts)
         d.addCallback(_assert)
         return d
+
+    def test_deduplication_on_same_mail_address_using_largest(self):
+        input_mail = MailBuilder().with_tags(['important']).build_input_mail()
+
+        formatted_input_mail = MailBuilder().with_tags(['important'])
+        formatted_input_mail.with_to('Recipient Principal <recipient@to.com>')
+        formatted_input_mail.with_cc('Recipient Copied <recipient@cc.com>')
+        formatted_input_mail.with_bcc('Recipient Carbon <recipient@bcc.com>')
+        formatted_input_mail = formatted_input_mail.build_input_mail()
+
+        self.client.add_mail_to_inbox(input_mail)
+        self.client.add_mail_to_inbox(formatted_input_mail)
+
+        d = self.get_contacts(query='Recipient')
+
+        def _assert(contacts):
+            print contacts
+            self.assertEquals(3, len(contacts))
+            self.assertTrue('Recipient Principal <recipient@to.com>' in contacts)
+            self.assertTrue('Recipient Copied <recipient@cc.com>' in contacts)
+            self.assertTrue('Recipient Carbon <recipient@bcc.com>' in contacts)
+        d.addCallback(_assert)
+        return d
diff --git a/service/test/unit/adapter/test_contacts.py b/service/test/unit/adapter/test_contacts.py
new file mode 100644
index 00000000..ee9bf618
--- /dev/null
+++ b/service/test/unit/adapter/test_contacts.py
@@ -0,0 +1,35 @@
+#
+# Copyright (c) 2014 ThoughtWorks, Inc.
+#
+# Pixelated is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# Pixelated is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with Pixelated. If not, see <http://www.gnu.org/licenses/>.
+import unittest
+
+from pixelated.adapter.contacts import address_duplication_filter
+from pixelated.adapter.contacts import extract_mail_address
+
+
+class TestContacts(unittest.TestCase):
+
+    def test_contacts_filter_duplication_by_largest(self):
+        contacts = ['John Large Name <john@name.example.com>', 'john@name.example.com', 'dont.delete@example.com']
+        contacts_filtered = address_duplication_filter(contacts)
+        self.assertIn('dont.delete@example.com', contacts_filtered)
+        self.assertIn('John Large Name <john@name.example.com>', contacts_filtered)
+        self.assertNotIn('john@name.example.com', contacts_filtered)
+
+    def test_extract_mail_address_from_contact(self):
+        full_address = 'John Large Name <john@name.example.com>'
+        mail_address = 'john@name.example.com'
+
+        self.assertEquals(mail_address, extract_mail_address(full_address))