Retrieves WordPress post and page content from the local installation using WP_Query. For media files, use the separate WordPress Media handler.

Architecture

Base Class: Extends FetchHandler (@since v0.2.1)

Inherited Functionality:

• Automatic deduplication via isItemProcessed() and markItemProcessed()
• Engine data storage via storeEngineData() for downstream handlers
• Standardized responses via successResponse(), emptyResponse(), errorResponse()
• Centralized logging and error handling

Implementation: Uses DataPacket class for consistent packet structure

Configuration Options

Basic Settings

Post Type

• Default: post
• Options: Any registered post type (post, page, custom post types)
• Purpose: Determines which content type to fetch

Post Status

• Default: publish
• Options: publish, draft, private, pending
• Purpose: Filter by publication status

Post ID

• Default: Empty (query-based selection)
• Purpose: Target specific post by ID (bypasses all other filters)

Selection Options

Randomize Selection

• Default: false (chronological by modified date)
• Purpose: Random post selection vs. most recently modified

Search Terms

• Default: Empty
• Purpose: Filter posts containing specific keywords

Timeframe Filtering

Timeframe Limit

• Default: all_time
• Options:
  • all_time – No time restriction
  • 24_hours – Last 24 hours
  • 72_hours – Last 72 hours
  • 7_days – Last 7 days
  • 30_days – Last 30 days

Taxonomy Filtering

Dynamic Taxonomy Filters

• Format: taxonomy_{slug}_filter
• Value: Term ID
• Purpose: Filter posts by taxonomy terms
• Example: taxonomy_category_filter = 5 (category term ID)

Data Output

Engine Data Filter Architecture

The WordPress Local handler generates clean data packets for AI processing while storing engine parameters in database for later retrieval via datamachine_engine_data filter.

Clean Data Packet (AI-Visible)

[
    'data' => [
        'content_string' => "Source: Site NamennTitle: Post TitlennPost Content...",
        'file_info' => null
    ],
    'metadata' => [
        'source_type' => 'wordpress_local',
        'item_identifier_to_log' => 123, // Post ID
        'original_id' => 123,
        'original_title' => 'Post Title',
        'original_date_gmt' => '2023-01-01 12:00:00'
        // Clean data packet for AI processing
    ]
]

Engine Parameters Storage

// Stored in database via centralized datamachine_engine_data filter (array storage)
if ($job_id) {
    apply_filters('datamachine_engine_data', null, $job_id, [
        'source_url' => $source_url,
        'image_url' => $image_url
    ]);
}

Return Structure

return [
    'processed_items' => [$clean_data_packet]
    // Engine parameters stored separately in database
];

Metadata Fields

Clean Data Packet:

• source_type: Always wordpress_local
• original_title: Post title as stored in database
• original_date_gmt: Post publication date in GMT
• URLs excluded to prevent AI content pollution

Engine Parameters:

• source_url: Full permalink for link attribution and post identification
• image_url: Featured image URL for media handling (null if none)

Processing Behavior

Deduplication

Uses processed items tracking to prevent duplicate processing:

$is_processed = apply_filters('datamachine_is_item_processed', false, $flow_step_id, 'wordpress_local', $post_id);

Tracking: Each processed post is marked by flow step and post ID
Scope: Per-flow deduplication (same post can be processed by different flows)

Query Optimization

Performance Settings:

• posts_per_page = 10 (finds first eligible item)
• no_found_rows = true
• update_post_meta_cache = false
• update_post_term_cache = false

Selection Strategy

1. Query Execution – Retrieves up to 10 posts matching criteria
2. Deduplication Check – Tests each post against processed items
3. First Match – Returns first unprocessed post immediately
4. Single Item – Always returns exactly one post or empty array

Implementation Examples

Basic WordPress Fetch

$handler_config = [
    'wordpress_posts' => [
        'post_type' => 'post',
        'post_status' => 'publish',
        'timeframe_limit' => '7_days'
    ]
];

$result = $wordpress_handler->get_fetch_data($pipeline_id, $handler_config, $job_id);
// Returns: ['processed_items' => [...]]
// Engine parameters (source_url, image_url) stored in database via centralized engine_data filter

Specific Post Targeting

$handler_config = [
    'wordpress_posts' => [
        'post_id' => 123 // Overrides all other criteria
    ]
];

Advanced Filtering

$handler_config = [
    'wordpress_posts' => [
        'post_type' => 'product',
        'search' => 'keyword',
        'taxonomy_category_filter' => 5,
        'taxonomy_product_tag_filter' => 12,
        'timeframe_limit' => '30_days',
        'randomize_selection' => true
    ]
];

Error Handling

Missing Configuration

Invalid Pipeline ID: Returns empty array
Missing Flow Step ID: Disables processed items tracking (logs debug message)

Database Errors

No Matching Posts: Returns empty array (normal operation)
Invalid Post ID: Returns empty array with warning log
Trashed Posts: Skipped with warning log

Content Processing

Missing Title: Uses ‘N/A’ as fallback
Empty Content: Passes empty string (not filtered)
Missing Featured Image: Sets image_source_url to null

Integration Notes

WordPress Core Dependency

• Uses WP_Query for database operations
• Requires get_post(), get_permalink(), get_post_thumbnail_id()
• Utilizes wp_get_attachment_image_url() for featured images
• Leverages get_bloginfo('name') for source identification

Performance Considerations

• Single Post Strategy – Returns immediately upon finding eligible post
• Optimized Query – Disables unnecessary WordPress features
• Targeted Selection – Uses specific criteria to reduce query scope
• Memory Efficient – Processes one post at a time

WordPress Compatibility

• Post Types – Works with any registered post type
• Taxonomies – Dynamic support for all taxonomies
• Custom Fields – Content includes shortcode-expanded custom fields
• Multisite – Operates within current site context