scheme-sdk 0.3.6__tar.gz → 0.3.7__tar.gz

{scheme_sdk-0.3.6 → scheme_sdk-0.3.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: scheme_sdk
-Version: 0.3.6
+Version: 0.3.7
 Summary: The Scheme SDK provides connectors for ingesting conversations, messages, and files across communication platforms.
 License:                                  Apache License
                                     Version 2.0, January 2004
@@ -203,6 +203,7 @@ License:                                  Apache License
             WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
             See the License for the specific language governing permissions and
             limitations under the License.
+Requires-Dist: html-sanitizer>=2.6.0
 Requires-Dist: requests>=2.32.5
 Requires-Python: >=3.11
 Project-URL: Homepage, https://www.schemebig.com/

{scheme_sdk-0.3.6 → scheme_sdk-0.3.7}/pyproject.toml

@@ -4,12 +4,13 @@ build-backend = "uv_build"
 
 [project]
 name = "scheme_sdk"
-version = "0.3.6"
+version = "0.3.7"
 description = "The Scheme SDK provides connectors for ingesting conversations, messages, and files across communication platforms."
 readme = "README.md"
 license = { file = "LICENSE" }
 requires-python = ">=3.11"
 dependencies = [
+    "html-sanitizer>=2.6.0",
     "requests>=2.32.5",
 ]
 

{scheme_sdk-0.3.6 → scheme_sdk-0.3.7}/src/scheme_sdk/connectors/base/message.py

@@ -2,6 +2,7 @@ from abc import ABC, abstractmethod
 from datetime import datetime
 from dataclasses import dataclass, field
 from typing import Any, Dict, Iterable, Optional
+from html_sanitizer import Sanitizer
 
 from .base import BaseConnector
 
@@ -11,6 +12,11 @@ class MessageConnector(BaseConnector, ABC):
     Abstract base class for all message connectors.
     """
 
+    _sanitizer: Sanitizer
+
+    def __init__(self):
+        self._sanitizer = Sanitizer()
+
     @abstractmethod
     def fetch_conversations(self) -> Iterable[Dict[str, Any]]:
         """
@@ -85,6 +91,7 @@ class MessageConnector(BaseConnector, ABC):
         """
         ...
 
+    @abstractmethod
     def normalize_message(self, raw_message: Dict[str, Any]) -> Dict[str, Any]:
         """
         Transform a platform-specific message into canonical format.
@@ -100,6 +107,7 @@ class MessageConnector(BaseConnector, ABC):
         """
         ...
 
+    @abstractmethod
     def normalize_conversation(
         self, raw_conversation: Dict[str, Any]
     ) -> Dict[str, Any]:
@@ -134,6 +142,18 @@ class MessageConnector(BaseConnector, ABC):
         self._logger.warning(f"search_messages not implemented for {self.platform}")
         return iter([])
 
+    def _sanitize_html(self, html: str) -> str:
+        """
+        Sanitize HTML content.
+
+        Args:
+            html: HTML content to sanitize
+
+        Returns:
+            Sanitized HTML content
+        """
+        return self._sanitizer.sanitize(html)
+
 
 @dataclass
 class Conversation:

{scheme_sdk-0.3.6 → scheme_sdk-0.3.7}/src/scheme_sdk/connectors/outlook.py

@@ -4,6 +4,12 @@ from typing import Any, Dict, List, Optional
 
 import requests
 
+from pprint import pprint as pp
+from dotenv import load_dotenv
+import os
+
+load_dotenv()
+
 from .base import MessageConnector
 
 
@@ -14,6 +20,7 @@ class OutlookConnector(MessageConnector):
     _backoff_cap_seconds = 30
 
     def __init__(self, token: str):
+        super().__init__()
         self.token = token
         self.base = "https://graph.microsoft.com/v1.0"
 
@@ -118,7 +125,7 @@ class OutlookConnector(MessageConnector):
         return {
             "title": message["subject"],
             "platform": self.platform,
-            "text": message["body"]["content"],
+            "text": self._sanitize_html(message["body"]["content"]),
             "direct_link": message["webLink"],
             "metadata": {
                 "platform_conversation_id": message["conversationId"],
@@ -212,3 +219,11 @@ class OutlookConnector(MessageConnector):
             next_params = None  # nextLink already includes the query string
 
         return items
+
+
+if __name__ == "__main__":
+    connector = OutlookConnector(os.getenv("OUTLOOK_TOKEN"))
+    conversations = connector.fetch_conversations(top=1)
+    pp(conversations)
+    messages = connector.fetch_messages(conversations[0]["id"])
+    pp(messages)