Add database migrations for public feed, metrics, and deduplication

- Add is_public column to episodes table - Add user_episodes
junction table for many-to-many relationship - Add episode_metrics
table for tracking engagement - Add original_url_hash column for
deduplication - Add Core.py functions for public episodes (mark_public,
get_public_episodes) - Add Core.py functions for user_episodes
(add_episode_to_user, user_has_episode, get_user_episodes) -
Add Core.py functions for metrics tracking (track_episode_event,
get_episode_metrics) - Add URL normalization and hashing utilities -
All tests passing

2 files changed, 320 insertions, 2 deletions

diff --git a/.tasks/tasks.jsonl b/.tasks/tasks.jsonl
index 0c1c36c..de997db 100644
--- a/.tasks/tasks.jsonl
+++ b/.tasks/tasks.jsonl
@@ -85,3 +85,28 @@
 {"taskCreatedAt":"2025-11-14T18:19:45.999699368Z","taskDependencies":[],"taskId":"t-1a1EwRH","taskNamespace":null,"taskParent":"t-19ZF6A8","taskStatus":"Done","taskTitle":"Test in emacs and narrow terminals","taskType":"WorkTask","taskUpdatedAt":"2025-11-14T19:02:18.766470937Z"}
 {"taskCreatedAt":"2025-11-14T18:19:46.028016768Z","taskDependencies":[],"taskId":"t-1a1EEer","taskNamespace":null,"taskParent":"t-19ZF6A8","taskStatus":"Done","taskTitle":"Handle edge cases and polish UX","taskType":"WorkTask","taskUpdatedAt":"2025-11-14T19:02:18.827147429Z"}
 {"taskCreatedAt":"2025-11-14T18:19:46.056655181Z","taskDependencies":[],"taskId":"t-1a1ELGl","taskNamespace":null,"taskParent":"t-19ZF6A8","taskStatus":"Done","taskTitle":"Update documentation","taskType":"WorkTask","taskUpdatedAt":"2025-11-14T19:02:47.319855049Z"}
+{"taskCreatedAt":"2025-11-16T04:06:48.014952363Z","taskDependencies":[],"taskId":"t-ga8V8O","taskNamespace":null,"taskParent":null,"taskStatus":"Open","taskTitle":"PodcastItLater: Public Feed, Metrics & Audio Improvements","taskType":"Epic","taskUpdatedAt":"2025-11-16T04:06:48.014952363Z"}
+{"taskCreatedAt":"2025-11-16T04:06:57.071621037Z","taskDependencies":[],"taskId":"t-gaKVc7","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Done","taskTitle":"Add database migrations for new columns (is_public, user_episodes table, episode_metrics table, original_url_hash)","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:09:27.336080901Z"}
+{"taskCreatedAt":"2025-11-16T04:06:57.609993104Z","taskDependencies":[],"taskId":"t-gaNbfx","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Done","taskTitle":"Implement URL hashing and normalization function for episode deduplication","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:09:27.896576613Z"}
+{"taskCreatedAt":"2025-11-16T04:06:58.132246645Z","taskDependencies":[],"taskId":"t-gaPn6Z","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Done","taskTitle":"Add Core.py database functions for public episodes (mark_public, unmark_public, get_public_episodes)","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:09:28.463907822Z"}
+{"taskCreatedAt":"2025-11-16T04:06:58.665794496Z","taskDependencies":[],"taskId":"t-gaRBUA","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Done","taskTitle":"Add Core.py database functions for user_episodes junction table","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:09:29.027348473Z"}
+{"taskCreatedAt":"2025-11-16T04:06:59.199139475Z","taskDependencies":[],"taskId":"t-gaTQEV","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Done","taskTitle":"Add Core.py database functions for episode metrics tracking","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:09:29.602931183Z"}
+{"taskCreatedAt":"2025-11-16T04:07:07.307576303Z","taskDependencies":[],"taskId":"t-gbrS2a","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Modify submission flow to check for existing episodes by URL hash","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:07.307576303Z"}
+{"taskCreatedAt":"2025-11-16T04:07:07.834181871Z","taskDependencies":[],"taskId":"t-gbu51O","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add /public route to display public feed","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:07.834181871Z"}
+{"taskCreatedAt":"2025-11-16T04:07:08.369657826Z","taskDependencies":[],"taskId":"t-gbwkkw","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add /public.rss route for public RSS feed generation","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:08.369657826Z"}
+{"taskCreatedAt":"2025-11-16T04:07:08.906237761Z","taskDependencies":[],"taskId":"t-gbyzV2","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Update home page to show public feed when user is logged out","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:08.906237761Z"}
+{"taskCreatedAt":"2025-11-16T04:07:09.433392796Z","taskDependencies":[],"taskId":"t-gbAN3x","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add admin toggle button to episode cards for public/private status","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:09.433392796Z"}
+{"taskCreatedAt":"2025-11-16T04:07:17.092115521Z","taskDependencies":[],"taskId":"t-gc6Vrk","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add POST /admin/episode/{id}/toggle-public endpoint","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:17.092115521Z"}
+{"taskCreatedAt":"2025-11-16T04:07:17.6266109Z","taskDependencies":[],"taskId":"t-gc9aud","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add '+ Add to your feed' button on episode pages for logged-in users","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:17.6266109Z"}
+{"taskCreatedAt":"2025-11-16T04:07:18.165342861Z","taskDependencies":[],"taskId":"t-gcbqDl","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add POST /episode/{id}/add-to-feed endpoint","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:18.165342861Z"}
+{"taskCreatedAt":"2025-11-16T04:07:18.700573408Z","taskDependencies":[],"taskId":"t-gcdFSb","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add POST /episode/{id}/track endpoint for metrics tracking","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:18.700573408Z"}
+{"taskCreatedAt":"2025-11-16T04:07:19.229153372Z","taskDependencies":[],"taskId":"t-gcfTnG","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add JavaScript to episode player for tracking play events","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:19.229153372Z"}
+{"taskCreatedAt":"2025-11-16T04:07:27.174644219Z","taskDependencies":[],"taskId":"t-gcNemK","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Enhance Worker.py to extract publication date and author metadata from articles","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:27.174644219Z"}
+{"taskCreatedAt":"2025-11-16T04:07:27.700527081Z","taskDependencies":[],"taskId":"t-gcPraJ","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add intro TTS generation with metadata (title, author, date)","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:27.700527081Z"}
+{"taskCreatedAt":"2025-11-16T04:07:28.221004581Z","taskDependencies":[],"taskId":"t-gcRCzw","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Add outro TTS generation with title and author","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:28.221004581Z"}
+{"taskCreatedAt":"2025-11-16T04:07:28.74867703Z","taskDependencies":[],"taskId":"t-gcTPQn","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Combine intro, pauses, article content, and outro in Worker.py","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:28.74867703Z"}
+{"taskCreatedAt":"2025-11-16T04:07:29.289653388Z","taskDependencies":[],"taskId":"t-gcW6zN","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Write tests for public feed functionality","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:29.289653388Z"}
+{"taskCreatedAt":"2025-11-16T04:07:35.447349966Z","taskDependencies":[],"taskId":"t-gdlWtu","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Write tests for episode deduplication","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:35.447349966Z"}
+{"taskCreatedAt":"2025-11-16T04:07:35.995113703Z","taskDependencies":[],"taskId":"t-gdoeYo","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Write tests for metrics tracking","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:35.995113703Z"}
+{"taskCreatedAt":"2025-11-16T04:07:36.52315156Z","taskDependencies":[],"taskId":"t-gdqsl7","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Write tests for audio intro/outro generation","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:36.52315156Z"}
+{"taskCreatedAt":"2025-11-16T04:07:37.059671738Z","taskDependencies":[],"taskId":"t-gdsHUA","taskNamespace":"Biz/PodcastItLater.hs","taskParent":"t-ga8V8O","taskStatus":"Open","taskTitle":"Create admin metrics dashboard view","taskType":"WorkTask","taskUpdatedAt":"2025-11-16T04:07:37.059671738Z"}
diff --git a/Biz/PodcastItLater/Core.py b/Biz/PodcastItLater/Core.py
index 9e3f830..7b48ac2 100644
--- a/Biz/PodcastItLater/Core.py
+++ b/Biz/PodcastItLater/Core.py
@@ -10,6 +10,7 @@ Includes:
 # : dep pytest
 # : dep pytest-asyncio
 # : dep pytest-mock
+import hashlib
 import logging
 import Omni.App as App
 import Omni.Test as Test
@@ -21,6 +22,7 @@ import sqlite3
 import sys
 import time
 import typing
+import urllib.parse
 from collections.abc import Iterator
 from contextlib import contextmanager
 from typing import Any
@@ -51,6 +53,56 @@ def is_admin(user: dict[str, typing.Any] | None) -> bool:
     ]
 
 
+def normalize_url(url: str) -> str:
+    """Normalize URL for comparison and hashing.
+
+    Normalizes:
+    - Protocol (http/https)
+    - Domain case (lowercase)
+    - www prefix (removed)
+    - Trailing slash (removed)
+    - Preserves query params and fragments as they may be meaningful
+
+    Args:
+        url: URL to normalize
+
+    Returns:
+        Normalized URL string
+    """
+    parsed = urllib.parse.urlparse(url.strip())
+
+    # Normalize domain to lowercase, remove www prefix
+    domain = parsed.netloc.lower()
+    domain = domain.removeprefix("www.")
+
+    # Normalize path - remove trailing slash unless it's the root
+    path = parsed.path.rstrip("/") if parsed.path != "/" else "/"
+
+    # Rebuild URL with normalized components
+    # Use https as the canonical protocol
+    return urllib.parse.urlunparse((
+        "https",  # Always use https
+        domain,
+        path,
+        parsed.params,
+        parsed.query,
+        parsed.fragment,
+    ))
+
+
+def hash_url(url: str) -> str:
+    """Generate a hash of a URL for deduplication.
+
+    Args:
+        url: URL to hash
+
+    Returns:
+        SHA256 hash of the normalized URL
+    """
+    normalized = normalize_url(url)
+    return hashlib.sha256(normalized.encode()).hexdigest()
+
+
 class Database:  # noqa: PLR0904
     """Data access layer for PodcastItLater database operations."""
 
@@ -147,6 +199,9 @@ class Database:  # noqa: PLR0904
         # Run migration to add stripe events table
         Database.migrate_add_stripe_events_table()
 
+        # Run migration to add public feed features
+        Database.migrate_add_public_feed()
+
     @staticmethod
     def add_to_queue(
         url: str,
@@ -240,6 +295,7 @@ class Database:  # noqa: PLR0904
         user_id: int | None = None,
         author: str | None = None,
         original_url: str | None = None,
+        original_url_hash: str | None = None,
     ) -> int:
         """Insert episode record, return episode ID.
 
@@ -250,8 +306,9 @@ class Database:  # noqa: PLR0904
             cursor = conn.cursor()
             cursor.execute(
                 "INSERT INTO episodes "
-                "(title, audio_url, duration, content_length, user_id, author, "
-                "original_url) VALUES (?, ?, ?, ?, ?, ?, ?)",
+                "(title, audio_url, duration, content_length, user_id, "
+                "author, original_url, original_url_hash) "
+                "VALUES (?, ?, ?, ?, ?, ?, ?, ?)",
                 (
                     title,
                     audio_url,
@@ -260,6 +317,7 @@ class Database:  # noqa: PLR0904
                     user_id,
                     author,
                     original_url,
+                    original_url_hash,
                 ),
             )
             conn.commit()
@@ -632,6 +690,87 @@ class Database:  # noqa: PLR0904
             logger.info("Created stripe_events table")
 
     @staticmethod
+    def migrate_add_public_feed() -> None:
+        """Add is_public column and related tables for public feed feature."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+
+            # Add is_public column to episodes
+            cursor.execute("PRAGMA table_info(episodes)")
+            episodes_info = cursor.fetchall()
+            episodes_columns = [col[1] for col in episodes_info]
+
+            if "is_public" not in episodes_columns:
+                cursor.execute(
+                    "ALTER TABLE episodes ADD COLUMN is_public INTEGER "
+                    "NOT NULL DEFAULT 0",
+                )
+                logger.info("Added is_public column to episodes")
+
+            # Create user_episodes junction table
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS user_episodes (
+                    user_id INTEGER NOT NULL,
+                    episode_id INTEGER NOT NULL,
+                    added_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+                    PRIMARY KEY (user_id, episode_id),
+                    FOREIGN KEY (user_id) REFERENCES users(id),
+                    FOREIGN KEY (episode_id) REFERENCES episodes(id)
+                )
+            """)
+            logger.info("Created user_episodes junction table")
+
+            # Create index on episode_id for reverse lookups
+            cursor.execute(
+                "CREATE INDEX IF NOT EXISTS idx_user_episodes_episode "
+                "ON user_episodes(episode_id)",
+            )
+
+            # Add original_url_hash column to episodes
+            if "original_url_hash" not in episodes_columns:
+                cursor.execute(
+                    "ALTER TABLE episodes ADD COLUMN original_url_hash TEXT",
+                )
+                cursor.execute(
+                    "CREATE INDEX IF NOT EXISTS idx_episodes_url_hash "
+                    "ON episodes(original_url_hash)",
+                )
+                logger.info("Added original_url_hash column to episodes")
+
+            # Create episode_metrics table
+            cursor.execute("""
+                CREATE TABLE IF NOT EXISTS episode_metrics (
+                    id INTEGER PRIMARY KEY AUTOINCREMENT,
+                    episode_id INTEGER NOT NULL,
+                    user_id INTEGER,
+                    event_type TEXT NOT NULL,
+                    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+                    FOREIGN KEY (episode_id) REFERENCES episodes(id),
+                    FOREIGN KEY (user_id) REFERENCES users(id)
+                )
+            """)
+            logger.info("Created episode_metrics table")
+
+            # Create indexes for metrics queries
+            cursor.execute(
+                "CREATE INDEX IF NOT EXISTS idx_episode_metrics_episode "
+                "ON episode_metrics(episode_id)",
+            )
+            cursor.execute(
+                "CREATE INDEX IF NOT EXISTS idx_episode_metrics_type "
+                "ON episode_metrics(event_type)",
+            )
+
+            # Create index on is_public for efficient public feed queries
+            cursor.execute(
+                "CREATE INDEX IF NOT EXISTS idx_episodes_public "
+                "ON episodes(is_public)",
+            )
+
+            conn.commit()
+            logger.info("Database migrated for public feed feature")
+
+    @staticmethod
     def migrate_add_default_titles() -> None:
         """Add default titles to queue items that have None titles."""
         with Database.get_connection() as conn:
@@ -809,6 +948,160 @@ class Database:  # noqa: PLR0904
             logger.info("Updated user %s status to %s", user_id, status)
 
     @staticmethod
+    def mark_episode_public(episode_id: int) -> None:
+        """Mark an episode as public."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                "UPDATE episodes SET is_public = 1 WHERE id = ?",
+                (episode_id,),
+            )
+            conn.commit()
+            logger.info("Marked episode %s as public", episode_id)
+
+    @staticmethod
+    def unmark_episode_public(episode_id: int) -> None:
+        """Mark an episode as private (not public)."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                "UPDATE episodes SET is_public = 0 WHERE id = ?",
+                (episode_id,),
+            )
+            conn.commit()
+            logger.info("Unmarked episode %s as public", episode_id)
+
+    @staticmethod
+    def get_public_episodes(limit: int = 50) -> list[dict[str, Any]]:
+        """Get public episodes for public feed."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                """
+                SELECT id, title, audio_url, duration, created_at,
+                       content_length, author, original_url
+                FROM episodes
+                WHERE is_public = 1
+                ORDER BY created_at DESC
+                LIMIT ?
+            """,
+                (limit,),
+            )
+            rows = cursor.fetchall()
+            return [dict(row) for row in rows]
+
+    @staticmethod
+    def add_episode_to_user(user_id: int, episode_id: int) -> None:
+        """Add an episode to a user's feed."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            try:
+                cursor.execute(
+                    "INSERT INTO user_episodes (user_id, episode_id) "
+                    "VALUES (?, ?)",
+                    (user_id, episode_id),
+                )
+                conn.commit()
+                logger.info(
+                    "Added episode %s to user %s feed",
+                    episode_id,
+                    user_id,
+                )
+            except sqlite3.IntegrityError:
+                # Episode already in user's feed
+                logger.info(
+                    "Episode %s already in user %s feed",
+                    episode_id,
+                    user_id,
+                )
+
+    @staticmethod
+    def user_has_episode(user_id: int, episode_id: int) -> bool:
+        """Check if a user has an episode in their feed."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                "SELECT 1 FROM user_episodes "
+                "WHERE user_id = ? AND episode_id = ?",
+                (user_id, episode_id),
+            )
+            return cursor.fetchone() is not None
+
+    @staticmethod
+    def get_user_episodes(user_id: int) -> list[dict[str, Any]]:
+        """Get all episodes in a user's feed."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                """
+                SELECT e.id, e.title, e.audio_url, e.duration, e.created_at,
+                       e.content_length, e.author, e.original_url, e.is_public,
+                       ue.added_at
+                FROM episodes e
+                JOIN user_episodes ue ON e.id = ue.episode_id
+                WHERE ue.user_id = ?
+                ORDER BY ue.added_at DESC
+            """,
+                (user_id,),
+            )
+            rows = cursor.fetchall()
+            return [dict(row) for row in rows]
+
+    @staticmethod
+    def get_episode_by_url_hash(url_hash: str) -> dict[str, Any] | None:
+        """Get episode by original URL hash."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                "SELECT * FROM episodes WHERE original_url_hash = ?",
+                (url_hash,),
+            )
+            row = cursor.fetchone()
+            return dict(row) if row is not None else None
+
+    @staticmethod
+    def track_episode_event(
+        episode_id: int,
+        event_type: str,
+        user_id: int | None = None,
+    ) -> None:
+        """Track an episode event (added, played, downloaded)."""
+        if event_type not in {"added", "played", "downloaded"}:
+            msg = f"Invalid event type: {event_type}"
+            raise ValueError(msg)
+
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                "INSERT INTO episode_metrics "
+                "(episode_id, user_id, event_type) VALUES (?, ?, ?)",
+                (episode_id, user_id, event_type),
+            )
+            conn.commit()
+            logger.info(
+                "Tracked %s event for episode %s",
+                event_type,
+                episode_id,
+            )
+
+    @staticmethod
+    def get_episode_metrics(episode_id: int) -> dict[str, int]:
+        """Get aggregated metrics for an episode."""
+        with Database.get_connection() as conn:
+            cursor = conn.cursor()
+            cursor.execute(
+                """
+                SELECT event_type, COUNT(*) as count
+                FROM episode_metrics
+                WHERE episode_id = ?
+                GROUP BY event_type
+            """,
+                (episode_id,),
+            )
+            rows = cursor.fetchall()
+            return {row["event_type"]: row["count"] for row in rows}
+
+    @staticmethod
     def set_user_stripe_customer(user_id: int, customer_id: str) -> None:
         """Link Stripe customer ID to user."""
         with Database.get_connection() as conn: