WordPress : Sort post titles in alphabetical order, disregarding initial definite and indefinite articles

wp-intart-sort.php

<?php

/**
 * File: Sort posts in alphabetical order, ignoring initial articles.
 * 
 * This condition is achieved using the `post_fields` and `posts_orderby`
 * filters. The implementer must provide their own list of definite and
 * indefinite articles for the hooks to ignore.
 *
 * To enable natural language sorting, pass the following Query parameter:
 * `"ignore_initial_articles" => true`
 * 
 * @link https://css-tricks.com/ignoring-the-in-wordpress-queries/
 */

namespace McAskill\WordPress\IgnoreInitialArticles;

/**
 * Retrieve definite and indefinite articles used when sorting titles.
 *
 * Based on {@see WP_Query::get_search_stopwords()}.
 *
 * @return array Initial articles.
 */

function get_initial_articles()
{
	static::$articles = null;

	if ( is_array( $articles ) ) {
		return $articles;
	}

	/**
	 * translators: This is a comma-separated list of very common initial words that
	 * should be excluded from sorting and filing processes, like "a", "an", and "the".
	 * These are usually called "articles". You should not simply translate these
	 * individual words into your language. Instead, look for and provide commonly
	 * accepted articles in your language.
	 *
	 * @link http://www.loc.gov/marc/bibliographic/bdapndxf.html
	 */
	$words = explode( ',', _x( 'a,an,the', 'Comma-separated list of initial articles in your language' ) );

	$articles = [];

	foreach ( $words as $word ) {
		$word = trim( $word, "\r\n\t " );
		if ( $word ) {
			$articles[] = $word;
		}
	}

	/**
	 * Filter articles used when parsing search terms.
	 *
	 * @param array $articles articles.
	 */

	$articles = apply_filters( 'wp_initial_articles', $articles );

	return $articles;
}

/**
 * Filter: Create a virtual table column named "title2" at the time
 * of the query. Populate the table column with the titles of the
 * queried posts. If the title has a matching initial article,
 * it will be removed.
 *
 * The list of English stopwords is the approximate search engines list,
 * and is translatable.
 *
 * @used-by Filter: "posts_fields"
 *
 * @param string   $fields    The SELECT clause of the query.
 * @param WP_Query &$wp_query The WP_Query instance (passed by reference).
 *
 * @return string $fields
 */

function create_temp_column( $fields, $wp_query )
{
	if ( $wp_query->get( 'ignore_initial_articles', false ) ) {
		global $wpdb;

		$matches = get_initial_articles();

		if ( count( $matches ) ) {
			$matches = implode( '|', $matches );

			$has_article = " CASE 
					WHEN $wpdb->posts.post_title regexp( '^($matches)[[:space:]]' )
						THEN trim( substr( $wpdb->posts.post_title from 4 ) )
					ELSE $wpdb->posts.post_title 
						END AS title2";

			if ( $has_article ) {
				$fields .= ( preg_match( '/^(\s+)?,/', $has_article ) ? $has_article : ", $has_article" );
			}

		}
	}

	return $fields;
}

/**
 * Filter: Create a new custom "orderby". It orders the posts by
 * the virtual table column "title2" using the ASC order.
 *
 * @used-by Filter: "posts_orderby"
 *
 * @param string   $orderby   The ORDER BY clause of the query.
 * @param WP_Query &$wp_query The WP_Query instance (passed by reference).
 *
 * @return string $orderby
 */

function sort_by_temp_column( $orderby, $wp_query )
{
	if ( $wp_query->get( 'ignore_initial_articles', false ) ) {
		$custom_orderby = ' UPPER( title2 ) ASC';

		if ( $custom_orderby ) {
			$orderby = $custom_orderby;
		}
	}

	return $orderby;
}

add_filter( 'posts_fields',  __NAMESPACE__ . '\\create_temp_column'  );
add_filter( 'posts_orderby', __NAMESPACE__ . '\\sort_by_temp_column' );