Codex/add podcast search and subscription feature

AGENTS.md

@@ -33,7 +33,7 @@ Podsumer is a self-hosted podcast aggregator (podcatcher) written in PHP with ze
 4. **Constants**: UPPERCASE with underscores (e.g., `VERSION`, `PODSUMER_PATH`)
 5. **Database Tables**: Plural snake_case (e.g., `feeds`, `items`, `file_contents`)
 6. **Database Columns**: snake_case (e.g., `url_hash`, `last_update`)
-7. **Configuration Keys**: snake_case (e.g., `store_media_on_disk`, `state_file`)
+7. **Configuration Keys**: snake_case (e.g., `media_dir`, `state_file`)
 
 ### File Organization
 
@@ -155,7 +155,7 @@ function feed(array $args): void
 
 ## Special Considerations
 
-1. **Media Storage**: Can be configured for database or disk storage, but cannot be changed after library is established
+1. **Media Storage**: All media files are stored on disk in the configured media_dir
 2. **Performance**: Database operations use transactions and optimized PRAGMA settings
 3. **Compatibility**: Requires SQLite 3.6.19+ with foreign key support
 4. **Memory**: No memory limit set (`memory_limit = -1`)
@@ -173,7 +173,6 @@ function feed(array $args): void
 2. Generate coverage: `composer coverage`
 3. View coverage locally: `composer coverage-dev`
 4. Ensure no regression in test coverage
-5. Test with both database and disk storage modes
 
 ## Contribution Behavior
 

Dockerfile

@@ -1,32 +1,52 @@
-FROM php:8.2.12-apache-bookworm
-RUN apt update && apt install -y git
-RUN php -r "copy('https://getcomposer.org/installer', 'composer-setup.php');" && \
-    php -r "if (hash_file('sha384', 'composer-setup.php') === 'dac665fdc30fdd8ec78b38b9800061b4150413ff2e3b6f88543c636f7cd84f6db9189d43a81e5503cda447da73c7e5b6') { echo 'Installer verified'; } else { echo 'Installer corrupt'; unlink('composer-setup.php'); } echo PHP_EOL;" && \
-    php composer-setup.php && \
-    php -r "unlink('composer-setup.php');" && \
-    mv composer.phar /usr/local/bin/composer
+FROM php:8.2-apache
+
+# Install system dependencies
 RUN apt-get update && apt-get install -y \
-		libzip-dev && \
-    docker-php-ext-configure zip && \
-	docker-php-ext-install -j$(nproc) zip
+    libsqlite3-dev \
+    sqlite3 \
+    ffmpeg \
+    cron \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install PHP extensions
+RUN docker-php-ext-install pdo pdo_sqlite
+
+# Enable Apache modules
 RUN a2enmod rewrite
-RUN pecl install pcov && docker-php-ext-enable pcov
-RUN mkdir -p /opt/podsumer/
-RUN mkdir -p /opt/podsumer/conf
-COPY ./apache.conf /etc/apache2/sites-available/000-default.conf
-COPY ./apache.conf /etc/apache2/sites-enabled/000-default.conf
-COPY ./conf/podsumer.conf /opt/podsumer/conf/podsumer.conf
-COPY ./sql /opt/podsumer/sql
-COPY ./src /opt/podsumer/src
-COPY ./templates /opt/podsumer/templates
-COPY ./www /opt/podsumer/www
-COPY ./composer.json /opt/podsumer/composer.json
-COPY ./composer.lock /opt/podsumer/composer.lock
-WORKDIR /opt/podsumer
-RUN chown -R www-data:www-data /opt/podsumer
-RUN chmod -R 755 /opt/podsumer
-RUN chown -R www-data:www-data /etc/apache2/sites-available/000-default.conf
-RUN chown -R www-data:www-data /etc/apache2/sites-enabled/000-default.conf
-RUN composer dump-autoload
-EXPOSE 3094
+
+# Set working directory
+WORKDIR /var/www/html
+
+# Copy application files
+COPY . .
+
+# Install Composer
+COPY --from=composer:latest /usr/bin/composer /usr/bin/composer
+
+# Install dependencies
+RUN composer install --no-dev --optimize-autoloader
+
+# Set permissions
+RUN chown -R www-data:www-data /var/www/html \
+    && chmod -R 755 /var/www/html \
+    && chmod +x /var/www/html/scripts/refresh_feeds.php
+
+# Create media directory
+RUN mkdir -p /opt/media && chown -R www-data:www-data /opt/media
+
+# Configure Apache - fix the path to use the correct apache.conf file
+COPY apache.conf /etc/apache2/sites-available/000-default.conf
+
+# Create a script to generate the crontab with the configured interval
+RUN echo '#!/bin/bash\n\
+REFRESH_INTERVAL=$(php -r "include \"/var/www/html/conf/podsumer.conf\"; echo \$feed_refresh_interval ?? 6;")\n\
+echo "0 */${REFRESH_INTERVAL} * * * www-data /usr/local/bin/php /var/www/html/scripts/refresh_feeds.php >> /var/log/cron.log 2>&1" > /etc/cron.d/podsumer-cron\n\
+chmod 0644 /etc/cron.d/podsumer-cron\n\
+crontab /etc/cron.d/podsumer-cron\n\
+service cron start\n\
+apache2-foreground' > /usr/local/bin/start.sh \
+    && chmod +x /usr/local/bin/start.sh
+
+# Start the container with our custom script
+CMD ["/usr/local/bin/start.sh"]
 

conf/podsumer.conf

@@ -11,20 +11,9 @@ per_item_art_download = 33
 items_per_page = 10
 ssl = false
 
-# File Storage Method
+# Media Directory
 #
-# By default all data is stored in the database. If you would prefer
-# to have your audio files and artwork stored on disk set this to
-# `true` and specify a path where you would like podsumer to save
-# files. If you have a very large libary this should improve the
-# performance and size of backups.
-#
-# NOTE: If this value is changed after a library is already established,
-# only _new_ items and feeds will use the new storage backend. Previously
-# downloaded files will remain where they are.
-
-store_media_on_disk = true
-
+# Directory where media files audio and images are stored.
 # Make sure podsumer has permission to read and write in this directory.
 # media_dir cannot be changed once library is established. If changed, some media
 # files may no longer be servable.
@@ -33,3 +22,64 @@ media_dir = /opt/media
 playback_interval = 5
 playback_rewind = 5
 
+# PodcastIndex API key and secret
+#
+# These are used to search for podcasts and episodes using the PodcastIndex API.
+#
+# You can get your own API key and secret by signing up at https://podcastindex.org/
+#
+# Once you have your API key and secret, you can add them here.
+podcastindex_key = "YOUR_PODCAST_INDEX_KEY_HERE"
+podcastindex_secret = "YOUR_PODCAST_INDEX_SECRET_HERE"
+
+# Ad Blocking Configuration
+#
+# Enable automatic ad detection and removal from podcasts.
+# This feature uses OpenAI Whisper for transcription and GPT-4o-mini for ad detection.
+# Default is off to save on API costs.
+#
+# Note: This requires an OpenAI API key and will incur costs based on usage.
+ad_blocking_enabled = true
+
+# OpenAI API key for transcription and ad detection
+openai_api_key = "YOUR_OPENAI_API_KEY_HERE"
+
+# OpenAI model for ad detection
+# Options: gpt-4o-mini cheaper, gpt-4o more accurate but expensive
+openai_ad_detection_model = "gpt-4o"
+
+# Whether to use ffmpeg to strip ads from audio files
+# If false, ads will be skipped in the web player instead
+use_ffmpeg_ad_removal = false
+
+# Ad segment merge buffer in seconds
+# Ad segments that are within this many seconds of each other will be merged
+# This reduces confusion and makes skipping easier by combining closely spaced ads
+# Default is 8 seconds - increase for more aggressive merging, decrease for more precision
+ad_merge_buffer_seconds = 8
+
+# OpenAI Pricing Configuration
+#
+# OpenAI pricing per 1000 tokens as of December 2024
+# Update these values as OpenAI changes their pricing
+# Current pricing can be found at: https://openai.com/api/pricing/
+#
+# Note: The system will use actual token counts from API responses for accurate cost calculation
+# rather than estimating based on character count.
+#
+# Whisper pricing per minute of audio
+openai_whisper_cost_per_minute = 0.006
+
+# GPT-4o-mini pricing per 1000 tokens - Current as of Dec 2024
+openai_gpt4o_mini_input_cost_per_1k_tokens = 0.00015
+openai_gpt4o_mini_output_cost_per_1k_tokens = 0.0006
+
+# GPT-4o pricing per 1000 tokens - in case you want to use a different model
+openai_gpt4o_input_cost_per_1k_tokens = 0.0025
+openai_gpt4o_output_cost_per_1k_tokens = 0.01
+
+# Feed Refresh Configuration
+#
+# How often to refresh feeds in hours
+# Default is 6 hours
+feed_refresh_interval = 1

conf/test.conf

@@ -14,3 +14,6 @@ ssl = false
 media_dir = /opt/media
 playback_interval = 5
 playback_rewind = 5
+
+podcastindex_key = ""
+podcastindex_secret = ""