Merge pull request #4475 from vigneshp826/master

Improve Slack Sanitization

CHANGELOG.rst

@@ -12,11 +12,14 @@ This project adheres to `Semantic Versioning`_ starting with version 1.0.
 
 Added
 -----
+
 - added ``DynamoTrackerStore`` to support persistence of agents running on AWS
 - added docstrings for ``TrackerStore`` classes
 - added buttons and images to mattermost.
 - `CRFEntityExtractor` updated to accept arbitrary token-level features like word vectors (issues/4214)
 - `SpacyFeaturizer` updated to add `ner_features` for `CRFEntityExtractor`
+- Sanitizing incoming messages from slack to remove slack formatting like <mailto:[email protected]|[email protected]> 
+  or <http://url.com|url.com> and substitute it with original content 
 
 Changed
 -------

rasa/core/channels/slack.py

@@ -180,7 +180,6 @@ def _is_user_message(slack_event):
     @staticmethod
     def _sanitize_user_message(text, uids_to_remove):
         """Remove superfluous/wrong/problematic tokens from a message.
-
         Probably a good starting point for pre-formatting of user-provided text
         to make NLU's life easier in case they go funky to the power of extreme
 
@@ -193,18 +192,32 @@ def _sanitize_user_message(text, uids_to_remove):
         Returns:
             str: parsed and cleaned version of the input text
         """
+
         for uid_to_remove in uids_to_remove:
             # heuristic to format majority cases OK
             # can be adjusted to taste later if needed,
             # but is a good first approximation
             for regex, replacement in [
                 (r"<@{}>\s".format(uid_to_remove), ""),
-                (r"\s<@{}>".format(uid_to_remove), ""),
-                # a bit arbitrary but probably OK
+                (
+                    r"\s<@{}>".format(uid_to_remove),
+                    "",
+                ),  # a bit arbitrary but probably OK
                 (r"<@{}>".format(uid_to_remove), " "),
             ]:
                 text = re.sub(regex, replacement, text)
 
+        """Find mailto or http links like <mailto:[email protected]|[email protected]> or '<http://url.com|url.com>in text and substitute it with original content
+        """
+
+        pattern = r"\<(mailto:|(http|https):\/\/).*\|.*\>"
+        match = re.search(pattern, text)
+
+        if match:
+            regex = match.group(0)
+            replacement = regex.split("|")[1]
+            replacement = replacement.replace(">", "")
+            text = text.replace(regex, replacement)
         return text.strip()
 
     @staticmethod