raindrop-ai 0.0.23__py3-none-any.whl → 0.0.25__py3-none-any.whl

raindrop/analytics.py

@@ -13,6 +13,7 @@ from threading import Timer
 from raindrop.version import VERSION
 from raindrop.models import TrackAIEvent, Attachment, SignalEvent, DefaultSignal, FeedbackSignal, EditSignal, PartialTrackAIEvent, PartialAIData
 from raindrop.interaction import Interaction
+from raindrop.redact import perform_pii_redaction
 
 
 
@@ -28,6 +29,7 @@ upload_interval = 1.0
 buffer = []
 flush_lock = threading.Lock()
 debug_logs = False
+redact_pii = False
 flush_thread = None
 shutdown_event = threading.Event()
 max_ingest_size_bytes = 1 * 1024 * 1024  # 1 MB
@@ -44,6 +46,14 @@ def set_debug_logs(value: bool):
     else:
         logger.setLevel(logging.INFO)
 
+def set_redact_pii(value: bool):
+    global redact_pii
+    redact_pii = value
+    if redact_pii:
+        logger.info("PII redaction enabled")
+    else:
+        logger.info("PII redaction disabled")
+
 def start_flush_thread():
     logger.debug("Opening flush thread")
     global flush_thread
@@ -177,6 +187,10 @@ def track_ai(
     payload.properties["$context"] = _get_context()
 
     data = payload.model_dump(mode="json")
+    
+    # Apply PII redaction if enabled
+    if redact_pii:
+        data = perform_pii_redaction(data)
 
     size = _get_size(data)
     if size > max_ingest_size_bytes:
@@ -399,6 +413,11 @@ def _flush_partial_event(event_id: str) -> None:
 
     # convert to ordinary TrackAIEvent-ish dict before send
     data = evt.model_dump(mode="json", exclude_none=True)
+    
+    # Apply PII redaction if enabled
+    if redact_pii:
+        data = perform_pii_redaction(data)
+    
     size = _get_size(data)
     if size > max_ingest_size_bytes:
         logger.warning(f"[raindrop] partial event {event_id} > 1 MB; skipping")

raindrop/redact.py

@@ -0,0 +1,134 @@
+import re
+import json
+import os
+from typing import Dict, Any
+
+
+class PIIRedactor:
+    """PII redactor that uses regex patterns to identify and replace PII."""
+    
+    def __init__(self):
+        
+        # Build regex patterns
+        self._build_patterns()
+    
+    def _build_patterns(self):
+        """Build regex patterns for PII detection."""
+        # Email pattern
+        self.email_pattern = re.compile(
+            r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b',
+            re.IGNORECASE
+        )
+        
+        # Credit card pattern (basic - matches 13-19 digits with optional spaces/dashes)
+        self.credit_card_pattern = re.compile(
+            r'\b(?:\d[ -]*?){13,19}\b'
+        )
+        
+        # Phone number pattern (US-style, but flexible)
+        self.phone_pattern = re.compile(
+            r'(\+?1[-.\s]?)?\(?\d{3}\)?[-.\s]?\d{3}[-.\s]?\d{4}\b'
+        )
+        
+        # SSN pattern
+        self.ssn_pattern = re.compile(
+            r'\b\d{3}[-\s]?\d{2}[-\s]?\d{4}\b'
+        )
+        
+        # Password/secret pattern (matches patterns like "password: xxx" or "secret: xxx")
+        self.password_pattern = re.compile(
+            r'\b(pass(word|phrase)?|secret|pwd|passwd)\s*[:=]\s*\S+',
+            re.IGNORECASE
+        )
+        
+        # Street address pattern (simplified)
+        self.address_pattern = re.compile(
+            r'\b\d+\s+[A-Za-z\s]+\s+(street|st|avenue|ave|road|rd|boulevard|blvd|lane|ln|drive|dr|court|ct|plaza|pl|terrace|ter|place|pl|way|parkway|pkwy)\b',
+            re.IGNORECASE
+        )
+        
+        # Greeting/closing patterns for name detection
+        self.greeting_pattern = re.compile(
+            r'(^|\.\s+)(dear|hi|hello|greetings|hey|hey there)[\s-]*',
+            re.IGNORECASE
+        )
+        
+        self.closing_pattern = re.compile(
+            r'(thx|thanks|thank you|regards|best|[a-z]+ly|[a-z]+ regards|all the best|happy [a-z]+ing|take care|have a [a-z]+ (weekend|night|day))\s*[,.!]*',
+            re.IGNORECASE
+        )
+        
+        # Generic name pattern (capitalized words)
+        self.generic_name_pattern = re.compile(
+            r'( ?(([A-Z][a-z]+)|([A-Z]\.))+)([,.]|[,.]?$)',
+            re.MULTILINE
+        )
+        
+        
+        # Credentials pattern (API keys, tokens, etc.)
+        self.credentials_pattern = re.compile(
+            r'\b(api[_-]?key|token|bearer|authorization|auth[_-]?token|access[_-]?token|secret[_-]?key)\s*[:=]\s*["\']?[\w-]+["\']?',
+            re.IGNORECASE
+        )
+
+    def redact(self, text: str) -> str:
+        """Redact PII from the given text using regex patterns."""
+        if not isinstance(text, str):
+            return text
+        
+        # Apply redactions in order
+        # Credentials
+        text = self.credentials_pattern.sub('<REDACTED_CREDENTIALS>', text)
+        
+        # Credit card numbers
+        text = self.credit_card_pattern.sub('<REDACTED_CREDIT_CARD>', text)
+        
+        # Email addresses
+        text = self.email_pattern.sub('<REDACTED_EMAIL>', text)
+        
+        # Phone numbers
+        text = self.phone_pattern.sub('<REDACTED_PHONE>', text)
+        
+        # SSN
+        text = self.ssn_pattern.sub('<REDACTED_SSN>', text)
+        
+        # Passwords/secrets
+        text = self.password_pattern.sub('<REDACTED_SECRET>', text)
+        
+        # Street addresses
+        text = self.address_pattern.sub('<REDACTED_ADDRESS>', text)
+        
+        # Note: IPs, URLs, usernames, and zipcodes are disabled by default
+        # to match JS SDK behavior
+        
+        return text
+
+
+def perform_pii_redaction(event_data: Dict[str, Any]) -> Dict[str, Any]:
+    """
+    Redact PII from event data, specifically targeting ai_data input and output fields.
+    
+    Args:
+        event_data: The event data dictionary to redact PII from
+        
+    Returns:
+        The event data with PII redacted
+    """
+    redactor = PIIRedactor()
+    
+    # Create a copy to avoid modifying the original
+    event_copy = event_data.copy()
+    
+    # Redact PII from ai_data fields if they exist
+    if 'ai_data' in event_copy and event_copy['ai_data']:
+        ai_data = event_copy['ai_data'].copy() if event_copy['ai_data'] else {}
+        
+        if 'input' in ai_data and ai_data['input']:
+            ai_data['input'] = redactor.redact(ai_data['input'])
+        
+        if 'output' in ai_data and ai_data['output']:
+            ai_data['output'] = redactor.redact(ai_data['output'])
+        
+        event_copy['ai_data'] = ai_data
+    
+    return event_copy 

raindrop/version.py

@@ -1 +1 @@
-VERSION = "0.0.19"
+VERSION = "0.0.25"

{raindrop_ai-0.0.23.dist-info → raindrop_ai-0.0.25.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: raindrop-ai
-Version: 0.0.23
+Version: 0.0.25
 Summary: Raindrop AI (Python SDK)
 License: MIT
 Author: Raindrop AI
@@ -10,7 +10,7 @@ Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
-Requires-Dist: pydantic (>=2.11,<3)
+Requires-Dist: pydantic (>=2.09,<3)
 Requires-Dist: requests (>=2.32.3,<3.0.0)
 Description-Content-Type: text/markdown
 

raindrop_ai-0.0.25.dist-info/RECORD

@@ -0,0 +1,9 @@
+raindrop/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+raindrop/analytics.py,sha256=XlwCh0n9zuktHCbEjqCWpci5juZvHPAtV-M56jrfxac,14153
+raindrop/interaction.py,sha256=U8OapN5tZGSMVkn-dTLum5sIg_AgSYKX9lrtwgXJOuI,1613
+raindrop/models.py,sha256=8YKuyB34P7sfawPFL3jOt5RgYlNXzf9dEsj8_2rjZKU,5293
+raindrop/redact.py,sha256=9AkNQGT61NrRStFH0zu0Lj4PGB2ATSN2rRgQnX_fQjQ,4552
+raindrop/version.py,sha256=KLRDXPxiLFqIKIkRg3nvV0u3CDx1KKih1aBFxpsuTSE,18
+raindrop_ai-0.0.25.dist-info/METADATA,sha256=2_XMJLXqec95On6OFCvgmKYY_8b3RsDWhHGLf3KYkxg,774
+raindrop_ai-0.0.25.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+raindrop_ai-0.0.25.dist-info/RECORD,,

raindrop_ai-0.0.23.dist-info/RECORD

@@ -1,8 +0,0 @@
-raindrop/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-raindrop/analytics.py,sha256=kAAzxrQqRYKCVG7bk2wQrdVrpca5t2sJV-RMr5b08H0,13672
-raindrop/interaction.py,sha256=U8OapN5tZGSMVkn-dTLum5sIg_AgSYKX9lrtwgXJOuI,1613
-raindrop/models.py,sha256=8YKuyB34P7sfawPFL3jOt5RgYlNXzf9dEsj8_2rjZKU,5293
-raindrop/version.py,sha256=zKIDJIJluqVEGWC_VXIxFo3Hrk5Q7UGBBqP5uigDAN8,18
-raindrop_ai-0.0.23.dist-info/METADATA,sha256=l2QSZBwDesa-ANhaXXZVeexCBUU4f94JL6kePrVBdzw,774
-raindrop_ai-0.0.23.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-raindrop_ai-0.0.23.dist-info/RECORD,,