From 739464fbc5dc328001fcc71e327938229c836204 Mon Sep 17 00:00:00 2001
From: Steven Hammerton <steven.hammerton@openmarket.com>
Date: Mon, 12 Oct 2015 16:02:17 +0100
Subject: [PATCH] Add a comment to clarify why we split on closing curly brace
 when reading CAS attribute tags

---
 synapse/rest/client/v1/login.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/synapse/rest/client/v1/login.py b/synapse/rest/client/v1/login.py
index c92dedcc0..2e3e4f39f 100644
--- a/synapse/rest/client/v1/login.py
+++ b/synapse/rest/client/v1/login.py
@@ -177,6 +177,11 @@ class LoginRestServlet(ClientV1RestServlet):
             if child.tag.endswith("attributes"):
                 attributes = {}
                 for attribute in child:
+                    # ElementTree library expands the namespace in attribute tags
+                    # to the full URL of the namespace.
+                    # See (https://docs.python.org/2/library/xml.etree.elementtree.html)
+                    # We don't care about namespace here and it will always be encased in
+                    # curly braces, so we remove them.
                     if "}" in attribute.tag:
                         attributes[attribute.tag.split("}")[1]] = attribute.text
                     else: