feat: add a European WhatsApp date format and support ngrok domains (v…

…ana-com#5)

README.md

@@ -57,7 +57,7 @@ To launch Selfie, ensure that [python](https://www.python.org), [poetry](https:/
 
 This starts a local web server and should launch the UI in your browser at http://localhost:8181. API documentation is available at http://localhost:8181/docs. Now that the server is running, you can use the API to import your data and connect to your LLM.
 
-> Note: You can host selfie at a publicly-accessible URL with [ngrok](https://ngrok.com). Add your ngrok token in `selfie/.env` and run `poetry run python -m selfie --share`.
+> Note: You can host selfie at a publicly-accessible URL with [ngrok](https://ngrok.com). Add your ngrok token (and optionally, ngrok domain) in `selfie/.env` and run `poetry run python -m selfie --share`.
 
 ### Step 1:  Gather Messaging Data
 

selfie/.env.example

@@ -1,2 +1,3 @@
 PORT=8181
-NGROK_AUTHTOKEN=<YOUR_NGROK_AUTHTOKEN> # Only needed if running using "share" flag
+NGROK_AUTHTOKEN=<YOUR_NGROK_AUTHTOKEN> # Only needed if running using "share" flag.
+NGROK_DOMAIN=<YOUR_NGROK_DOMAIN> # Optional static domain. You can create one at https://dashboard.ngrok.com/cloud-edge/domains.

selfie/__main__.py

@@ -50,12 +50,13 @@ def get_configured_app(shareable=False):
         logging.getLogger("selfie").setLevel(level=logging.DEBUG)
 
     ngrok_auth_token = os.environ.get('NGROK_AUTHTOKEN', None)
+    ngrok_domain = os.environ.get('NGROK_DOMAIN', None)
 
     if shareable and args.share:
         if ngrok_auth_token is None:
             raise ValueError("NGROK_AUTHTOKEN environment variable is required to share the API. Visit https://dashboard.ngrok.com to get your token.")
 
-        listener = ngrok.forward(args.port, authtoken_from_env=True)
+        listener = ngrok.forward(args.port, authtoken_from_env=True, domain=ngrok_domain)
         logger.info(f"Application is available at {listener.url()}")
         # Update config directly as modifying args won't affect the env vars
         os.environ['SELFIE_HOST'] = listener.url()

selfie/parsers/chat/whatsapp.py

@@ -45,15 +45,19 @@ class WhatsAppParser(TextBasedChatParser):
             "regex": r"(?:\[)?(?P<timestamp>\d{1,2}-\d{1,2}-\d{2}, \d{1,2}:\d{2}:\d{2}\s(AM|PM))(?:\])? (?P<from>.+?): (?P<value>.+)",
             "timestamp_format": "%Y-%m-%d, %I:%M:%S %p",
         },
-        # Deprecated or unofficial formats
         {
             "regex": r"\[(?P<timestamp>\d{1,2}/\d{1,2}/\d{2,4}, \d{1,2}:\d{2}:\d{2}\s(AM|PM))\] (?P<from>.+?): (?P<value>.+)",
             "timestamp_format": "%m/%d/%y, %I:%M:%S %p",
         },
+        # Deprecated or unofficial format (?)
         {
             "regex": r"(?P<timestamp>\d{1,2}/\d{1,2}/\d{2}, \d{1,2}:\d{2}(AM|PM)) - (?P<from>.+?): (?P<value>.+)",
             "timestamp_format": "%m/%d/%y, %I:%M%p",
         },
+        {
+            "regex": r"\[(?P<timestamp>\d{2}\.\d{2}\.\d{2}, \d{2}:\d{2}:\d{2})\] (?P<from>.+?): (?P<value>.+)",
+            "timestamp_format": "%d%m%y, %H:%M:%S",
+        },
     ]
 
     DROP_LINES_LIKE = [
@@ -72,6 +76,9 @@ class WhatsAppParser(TextBasedChatParser):
         {
             "regex": r"(?:\[)?(?P<timestamp>\d{1,2}-\d{1,2}-\d{2}, \d{1,2}:\d{2}:\d{2}\s(AM|PM))(?:\])? (?P<value>.+)",
         },
+        {
+            "regex": r"(?:\[)?(?P<timestamp>\d{2}\.\d{2}\.\d{2}, \d{2}:\d{2}:\d{2})(?:\])? (?P<value>.+)",
+        }
     ]
 
     def _preprocess_hook(self, document: str) -> str: