<?php

/**
 * Work with post
 */
class Wpil_Post
{
    public static $advanced_custom_fields_list = null;
    public static $post_types_without_editors = array(
        'web-story'
    );
    public static $post_url_cache = array();
    public static $editor_insert_log = array();

    /**
     * Register services
     */
    public function register()
    {
        add_filter('wp_insert_post_data', [$this, 'addAutolinksToInsertRun'], 998, 3);
        add_filter('wp_insert_post_data', [$this, 'addLinksToContent'], 9999, 3);
        add_action('wp_ajax_wpil_editor_reload', [$this, 'editorReload']);
        add_action('wp_ajax_wpil_is_outbound_links_added', [$this, 'isOutboundLinksAdded']);
        add_action('wp_ajax_wpil_is_inbound_links_added', [$this, 'isInboundLinksAdded']);
        add_action('wp_ajax_wpil_ignore_orphaned_post', [$this, 'ajaxIgnoreOrphanedPost']);
        add_action('future_to_publish', [$this, 'saveAutolinksToTransitionedPosts'], 998, 1);
        add_action('save_post', [$this, 'saveAutolinksToPost'], 998);
        add_action('save_post', [$this, 'updateStatMark'], 99999);
        add_action('before_delete_post', [$this, 'delete_inbound_links_on_post_delete']);
        add_action('before_delete_post', [$this, 'deleteReferences'], 11);
        add_action('save_post', [$this, 'addLinkToMetaContent'], 9999, 1); // meta content is the last stop in the link inserting
        add_filter('wp_link_query_args', array(__CLASS__, 'filter_custom_link_post_types'), 10, 1);
        add_filter('wp_link_query', array(__CLASS__, 'custom_link_category_search'), 10, 2);
        add_filter('et_fb_ajax_save_verification_result', array(__CLASS__, 'verify_divi_save_status'));
    }

    /**
     * Add links to content before post update
     */
    public static function addLinksToContent($data, $post, $unsanitized_postarr = array(), $update = false, $direct_call = false, $meta = array())
    {
        //get links from DB
        $meta = (!empty($meta)) ? $meta: Wpil_Toolbox::get_encoded_post_meta($post['ID'], 'wpil_links', true);

        if (is_null($data) && isset($post['ID']) && !empty($post['ID'])) {
            $data = get_post($post['ID'], ARRAY_A);
            $data['post_content'] = addslashes($data['post_content']);
            $data_null = true;
        }

        // if this is a Google Web Story
        if(self::get_post_processing_type($data) === 'web-story' && !empty($unsanitized_postarr)){ // todo: implement a pre-insert content processor to handle this kind of thing! Also, create a post-insert handler.
            // filter the post data so we retain the full post as it's stored in the database
            $data = self::use_unsanitized_content($data, $unsanitized_postarr);
        }

        // if there are links to insert and we're dealing with a data object we can work with
        if (!empty($meta) && (is_array($meta) || is_object($meta)) && !empty($data)) { // should always be an array... but if the unserialize call fails, we'll get a string instead

            // if we have the last modified date and the user doesn't want the post modified date to change
            if( !empty($post) && 
                isset($post['post_modified']) && 
                isset($post['post_modified_gmt']) && 
                !Wpil_Settings::updatePostModifiedDate() &&
                !Wpil_Base::action_happened('doing_post_update')) // and we haven't detected this to be post of a content update
            {
                // set the current modified date for it
                $data['post_modified'] = $post['post_modified'];
                $data['post_modified_gmt'] = $post['post_modified_gmt'];
            }elseif(
                !empty($post) && 
                isset($post['ID']) && 
                !Wpil_Settings::updatePostModifiedDate() &&
                !Wpil_Base::action_happened('doing_post_update'))
            {
                $posty = get_post($post['ID']);
                // set the current modified date for it
                $data['post_modified'] = $posty->post_modified;
                $data['post_modified_gmt'] = $posty->post_modified_gmt;
            }

            //update post text
            foreach ($meta as $link) {
                $inserted = false;
                $changed_sentence = self::getSentenceWithAnchor($link);
//                $link['sentence'] = Wpil_Word::removeQuotes($link['sentence']);

                if (strpos($data['post_content'], $link['sentence']) === false) {
                    $sentence = addslashes($link['sentence']);
                } else {
                    $sentence = $link['sentence'];
                }

                // if the sentence still isn't here...
                if(false === strpos($data['post_content'], $sentence)){
                    // try seeing if the punchuation was stripped
                    $repunch = self::repunctuate_sentence($data['post_content'], $sentence, $changed_sentence); 
                    // if it was...
                    if(!empty($repunch['sentence']) && false !== strpos($data['post_content'], $repunch['sentence'])){
                        // update the sentence
                        $sentence = $repunch['sentence'];
                        // and the changed link
                        $changed_sentence = $repunch['changed_sentence'];
                    }
                }

                $force_insert = (isset($link['keyword_data']) && !empty($link['keyword_data']->force_insert)) ? true: false;

                Wpil_Editor_Kadence::insertLink($data['post_content'], $sentence, $changed_sentence);

                if (strpos($data['post_content'], $sentence) !== false) {
                    $changed_sentence = self::changeByACF($data['post_content'], $link['sentence'], $changed_sentence);
                    $inserted = self::insertLink($data['post_content'], $sentence, $changed_sentence, $force_insert);
                    Wpil_Base::track_action('link_inserted', $inserted);
                }

                if (isset($data['post_content_filtered']) && !empty($data['post_content_filtered']) && (strpos($data['post_content_filtered'], $sentence) !== false || strpos($data['post_content_filtered'], stripslashes($sentence)) !== false)) {
                    $changed_sentence = self::changeByACF($data['post_content_filtered'], $link['sentence'], $changed_sentence);
                    $inserted = self::insertLink($data['post_content_filtered'], $sentence, $changed_sentence, $force_insert);
                    Wpil_Base::track_action('link_inserted', $inserted);
                }

                // if this is a WooCommerce product, and the link is for the excerpt/short description & isn't possible for the main content
                if( strpos($data['post_content'], $sentence) === false && 
                    strpos($data['post_excerpt'], $sentence) !== false && 
                    defined('WC_PLUGIN_FILE') && 
                    'product' === $data['post_type'] && 
                    in_array('product', Wpil_Settings::getPostTypes()))
                {
                    // add the link to the excerpt
                    $inserted = self::insertLink($data['post_excerpt'], $sentence, $changed_sentence, $force_insert);
                    Wpil_Base::track_action('link_inserted', $inserted);
                }

                // if the Enfold Advanced editor is active
                if( isset($post['aviaLayoutBuilder_active']) && 'active' === $post['aviaLayoutBuilder_active'] && 
                    isset($post['_aviaLayoutBuilderCleanData']) && !empty($post['_aviaLayoutBuilderCleanData']))
                {
                    // add links to the submitted form content
                    if (strpos($_POST['_aviaLayoutBuilderCleanData'], $sentence) !== false) {
                        $inserted = self::insertLink($_POST['_aviaLayoutBuilderCleanData'], $sentence, $changed_sentence, $force_insert);
                        Wpil_Base::track_action('link_inserted', $inserted);
                    }
                }

                if($inserted){
                    Wpil_Keyword::update_auto_imported_link_count(1); // the function checcks to make sure that we're only counting during the link importing
                }
            }

            // if the post doesn't belong to a post type that an editor doesn't apply to

            if(!in_array(self::get_post_processing_type($data), self::$post_types_without_editors, true) && !in_array($post['ID'], self::$editor_insert_log)){
                self::$editor_insert_log[] = $post['ID'];
                self::editors('addLinks', [$meta, $post['ID'], &$data['post_content']]);
            }

            // if this is a Google Web Story
            if(self::get_post_processing_type($data) === 'web-story' && !empty($unsanitized_postarr)){
                // filter the post data so we retain the full post as it's stored in the database
                $data = self::filter_webstory_content($data, !empty($unsanitized_postarr));
            }

            // if we're not filtering an insert
            if (!empty($data_null)) {
                // go through our process to update the post from here

                Wpil_Word::addSlashesToNewLine($data['post_content']);

                $update = [
                    'ID' => $post['ID'],
                    'post_content' => $data['post_content'],
                    'post_excerpt' => $data['post_excerpt']
                ];

                if(isset($data['post_content_filtered'])){
                    $update['post_content_filtered'] = $data['post_content_filtered'];
                }

                // remove any kses rules that have given us trouble
                add_filter('wp_kses_allowed_html', [__CLASS__, 'allowed_tags_and_attributes'], 10, 2);

                // remove any post update hooks so we don't runn around in circles
                Wpil_Base::remove_hooked_function('wp_insert_post_data', 'Wpil_Post', 'addAutolinksToInsertRun', 998);
                Wpil_Base::remove_hooked_function('wp_insert_post_data', 'Wpil_Post', 'addLinksToContent', 9999);

                // and temporarily remove the metafield processing to avoid duplicate inserts
                Wpil_Base::remove_hooked_function('save_post', 'Wpil_Post', 'addLinkToMetaContent', 9999);

                // and remove any hooks that should have occurred before now
                Wpil_Base::remove_lower_priority_hooked_functions('wp_insert_post_data', 998); // remove 997 and down ↓↓↓

                // if we're not supposed to updated the modified date
                if(!Wpil_Settings::updatePostModifiedDate()){
                    $update['post_modified'] = $data['post_modified'];
                    $update['post_modified_gmt'] = $data['post_modified_gmt'];
                    add_filter('wp_insert_post_data', array(__CLASS__, 'prevent_date_modifying'), 10, 3);
                }

                // if the customer has opted to disable revisions...
                if(Wpil_Settings::disablePostRevisions()){
                    // remove the revision saver
                    remove_action('post_updated', 'wp_save_post_revision');
                }

                // update the post
                wp_update_post($update);

                // if the customer has opted to disable revisions...
                if(Wpil_Settings::disablePostRevisions()){
                    // put back revision saver
                    add_action('post_updated', 'wp_save_post_revision', 10, 1);
                }

                // remove our kses rule removing
                remove_filter('wp_kses_allowed_html', [__CLASS__, 'allowed_tags_and_attributes']);
                // and remove the date keeper
                remove_filter('wp_insert_post_data', [__CLASS__, 'prevent_date_modifying']);
                // and re-add the metafield linking if it's not already back // recursion man!
                if(!$direct_call && !Wpil_Base::has_hooked_function('save_post', 'Wpil_Post', 'addLinkToMetaContent', 9999)){
                    add_action('save_post', [__CLASS__, 'addLinkToMetaContent'], 9999, 1);
                }elseif($direct_call){
                    // if the link inserter has been called outside of the post saving routine, 
                    // such as by an ajax call, 
                    // add the meta links now because they aren't going to be added otherwise
                    self::addLinkToMetaContent($post['ID'], $direct_call);
                }

                // remove any ghost links
                $new_post = new Wpil_Model_Post($post['ID']);
                Wpil_Keyword::deleteGhostLinks($new_post);
            }

            if (WPIL_STATUS_LINK_TABLE_EXISTS){
                Wpil_Report::update_post_in_link_table($post['ID']);
            }
        }elseif(self::get_post_processing_type($data) === 'web-story' && !empty($unsanitized_postarr)){
            // if there are no links to insert and this is a Google Web Story
            // filter the post data so we retain the full post as it's stored in the database
            $data = self::filter_webstory_content($data, !empty($unsanitized_postarr));
        }

        //return updated post data
        return $data;
    }

    /**
     * Hooks to the "wp_insert_post_data" filter and prevents the post date from being modified if we're not doing a post update.
     * So if links are being inserted or changed, this will stop WP from auto updating the post modified date.
     **/
    public static function prevent_date_modifying($data, $post, $unsanitized_postarr){
        // if we have the last modified date and the user doesn't want the post modified date to change
        if( !empty($post) && 
            isset($post['post_modified']) && 
            isset($post['post_modified_gmt']) && 
            !Wpil_Settings::updatePostModifiedDate() &&
            !Wpil_Base::action_happened('doing_post_update')) // and we haven't detected this to be post of a content update
        {
            // set the current modified date for it
            $data['post_modified'] = $post['post_modified'];
            $data['post_modified_gmt'] = $post['post_modified_gmt'];
        }elseif(
            !empty($post) && 
            isset($post['ID']) && 
            !Wpil_Settings::updatePostModifiedDate() &&
            !Wpil_Base::action_happened('doing_post_update'))
        {
            $posty = get_post($post['ID']);
            if(!empty($posty) && is_a($posty, 'WP_Post') && isset($posty->post_modified) && isset($posty->post_modified_gmt)){
                // set the current modified date for it
                $data['post_modified'] = $posty->post_modified;
                $data['post_modified_gmt'] = $posty->post_modified_gmt;
            }
        }
        return $data;
    }

    /**
     * Replaces the content of the post object with the unsanitized version
     **/
    public static function use_unsanitized_content($data = array(), $unsanitized_postarr = array()){
		if(isset($unsanitized_postarr['post_content_filtered'])){
			$data['post_content_filtered'] = $unsanitized_postarr['post_content_filtered'];
		}

		if(isset($unsanitized_postarr['post_content'])){
			$data['post_content'] = $unsanitized_postarr['post_content'];
		}

		return $data;
    }

    public static function filter_webstory_content($data = array(), $using_unsanitized = false){
        if(empty($data)){
            return $data;
        }

        return self::filter_insert_post_data($data, $using_unsanitized);
    }

    /**
     * Google Web Story content filtering functionality borrowed from the Web Stories plugin version 1.22.1
     * Including license from the time of borrowing, and modifying as allowed by license.
     * As always, the license applies to the utilized code and is not a claim that Link Whisper is licensed with this license.
     **/

    /**
     * Copyright 2020 Google LLC
     *
     * Licensed under the Apache License, Version 2.0 (the "License");
     * you may not use this file except in compliance with the License.
     * You may obtain a copy of the License at
     *
     *     https://www.apache.org/licenses/LICENSE-2.0
     *
     * Unless required by applicable law or agreed to in writing, software
     * distributed under the License is distributed on an "AS IS" BASIS,
     * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     * See the License for the specific language governing permissions and
     * limitations under the License.
     */

	/**
	 * Filters slashed post data just before it is inserted into the database.
	 *
	 * Used to run story HTML markup through KSES on our own, but with some filters applied
	 * that should only affect the web-story post type.
	 *
	 * This allows storing full AMP HTML documents in post_content for stories, which require
	 * more allowed HTML tags and a patched version of {@see safecss_filter_attr}.
	 *
	 * @since 1.8.0
	 *
	 * @param array|mixed $data                An array of slashed, sanitized, and processed post data.
	 * @param array       $postarr             An array of sanitized (and slashed) but otherwise unmodified post data.
	 * @param array       $unsanitized_postarr An array of slashed yet *unsanitized* and unprocessed post data as
	 *                                         originally passed to wp_insert_post().
	 * @return array|mixed Filtered post data.
	 */
	public static function filter_insert_post_data($data, $using_unsanitized){
		if(!\is_array($data)){
			return $data;
		}

        // filter the post content filtered to apply formatting
		if(isset($data['post_content_filtered'])){
			$data['post_content_filtered'] = self::filter_story_data($data['post_content_filtered'], $using_unsanitized);
		}

        if(current_user_can('unfiltered_html')){
            return $data;
        }

		if(isset($data['post_content'])){
			add_filter('safe_style_css', [__CLASS__, 'filter_safe_style_css']);
			add_filter('wp_kses_allowed_html', [__CLASS__, 'filter_kses_allowed_html'], 10, 2);

			$data['post_content'] = self::filter_content_save_pre_before_kses($data['post_content']);

			$data['post_content'] = wp_filter_post_kses($data['post_content']);
			$data['post_content'] = self::filter_content_save_pre_after_kses($data['post_content']);

			remove_filter('safe_style_css', [__CLASS__, 'filter_safe_style_css']);
			remove_filter('wp_kses_allowed_html', [__CLASS__, 'filter_kses_allowed_html']);
		}

		return $data;
	}

	/**
	 * Filters story data.
	 *
	 * Provides simple sanity check to ensure story data is valid JSON.
	 *
	 * @since 1.22.0
	 *
	 * @param string $story_data JSON-encoded story data.
	 * @return string Sanitized & slashed story data.
	 */
	private static function filter_story_data(string $story_data, bool $using_unsanitized){
        // if we're using unsanitized data
        if($using_unsanitized){
            // handle the slashes that are added when a post is submitted
            $decoded = json_decode( (string) wp_unslash( $story_data ), true );
            return null === $decoded ? '' : wp_slash( (string) wp_json_encode( $decoded ) );
        }else{
            // if the data isn't unsanitized, it's from the database and doesn't need to be unslashed
            $decoded = json_decode( $story_data, true );
            return null === $decoded ? '' : (string) wp_slash(wp_json_encode( $decoded ));
        }
	}

	/**
	 * Temporarily renames the style attribute to data-temp-style in full story markup.
	 *
	 * @since 1.0.0
	 *
	 * @param string $post_content Post content.
	 * @return string Filtered post content.
	 */
	public static function filter_content_save_pre_before_kses($post_content){
		return (string) preg_replace_callback(
			'|(?P<before><\w+(?:-\w+)*\s[^>]*?)style=\\\"(?P<styles>[^"]*)\\\"(?P<after>([^>]+?)*>)|', // Extra slashes appear here because $post_content is pre-slashed..
			static function ( $matches ) {
				return $matches['before'] . sprintf( ' data-temp-style="%s" ', $matches['styles'] ) . $matches['after'];
			},
			$post_content
		);
	}

	/**
	 * Renames data-temp-style back to style in full story markup.
	 *
	 * @since 1.0.0
	 *
	 * @param string $post_content Post content.
	 * @return string Filtered post content.
	 */
	public static function filter_content_save_pre_after_kses($post_content){
		return (string) preg_replace_callback(
			'/ data-temp-style=\\\"(?P<styles>[^"]*)\\\"/',
			function($matches){
				$styles = str_replace('&quot;', '\"', $matches['styles']);
				return sprintf(' style="%s"', esc_attr(self::safecss_filter_attr(wp_kses_stripslashes($styles))));
			},
			$post_content
		);
	}

	/**
	 * Filters list of allowed CSS attributes.
	 *
	 * @since 1.0.0
	 *
	 * @param string[]|mixed $attr Array of allowed CSS attributes.
	 * @return array|mixed Filtered list of CSS attributes.
	 */
	public static function filter_safe_style_css( $attr ) {
		if ( ! \is_array( $attr ) ) {
			return $attr;
		}

		$additional = [
			'display',
			'opacity',
			'position',
			'top',
			'left',
			'transform',
			'white-space',
			'clip-path',
			'-webkit-clip-path',
			'pointer-events',
			'will-change',
			'--initial-opacity',
			'--initial-transform',
		];

		array_push( $attr, ...$additional );

		return $attr;
	}

	/**
	 * Filters an inline style attribute and removes disallowed rules.
	 *
	 * This is equivalent to the WordPress core function of the same name,
	 * except that this does not remove CSS with parentheses in it.
	 *
	 * A few more allowed attributes are added via the safe_style_css filter.
	 *
	 * @SuppressWarnings(PHPMD)
	 *
	 * @since 1.0.0
	 *
	 * @see safecss_filter_attr()
	 *
	 * @param string $css A string of CSS rules.
	 * @return string Filtered string of CSS rules.
	 */
	public static function safecss_filter_attr($css){
		$css = wp_kses_no_null( $css );
		$css = str_replace( [ "\n", "\r", "\t" ], '', $css );

		$allowed_protocols = wp_allowed_protocols();

		$css_array = explode( ';', trim( $css ) );

		/** This filter is documented in wp-includes/kses.php */
		$allowed_attr = apply_filters(
			'safe_style_css',
			[
				'background',
				'background-color',
				'background-image',
				'background-position',
				'background-size',
				'background-attachment',
				'background-blend-mode',

				'border',
				'border-radius',
				'border-width',
				'border-color',
				'border-style',
				'border-right',
				'border-right-color',
				'border-right-style',
				'border-right-width',
				'border-bottom',
				'border-bottom-color',
				'border-bottom-style',
				'border-bottom-width',
				'border-left',
				'border-left-color',
				'border-left-style',
				'border-left-width',
				'border-top',
				'border-top-color',
				'border-top-style',
				'border-top-width',

				'border-spacing',
				'border-collapse',
				'caption-side',

				'columns',
				'column-count',
				'column-fill',
				'column-gap',
				'column-rule',
				'column-span',
				'column-width',

				'color',
				'font',
				'font-family',
				'font-size',
				'font-style',
				'font-variant',
				'font-weight',
				'letter-spacing',
				'line-height',
				'text-align',
				'text-decoration',
				'text-indent',
				'text-transform',

				'height',
				'min-height',
				'max-height',

				'width',
				'min-width',
				'max-width',

				'margin',
				'margin-right',
				'margin-bottom',
				'margin-left',
				'margin-top',

				'padding',
				'padding-right',
				'padding-bottom',
				'padding-left',
				'padding-top',

				'flex',
				'flex-basis',
				'flex-direction',
				'flex-flow',
				'flex-grow',
				'flex-shrink',

				'grid-template-columns',
				'grid-auto-columns',
				'grid-column-start',
				'grid-column-end',
				'grid-column-gap',
				'grid-template-rows',
				'grid-auto-rows',
				'grid-row-start',
				'grid-row-end',
				'grid-row-gap',
				'grid-gap',

				'justify-content',
				'justify-items',
				'justify-self',
				'align-content',
				'align-items',
				'align-self',

				'clear',
				'cursor',
				'direction',
				'float',
				'overflow',
				'vertical-align',
				'list-style-type',

				'z-index',
			]
		);

		/*
		 * CSS attributes that accept URL data types.
		 *
		 * This is in accordance to the CSS spec and unrelated to
		 * the sub-set of supported attributes above.
		 *
		 * See: https://developer.mozilla.org/en-US/docs/Web/CSS/url
		 */
		$css_url_data_types = [
			'background',
			'background-image',

			'cursor',

			'list-style',
			'list-style-image',

			'clip-path',
			'-webkit-clip-path',
		];

		/*
		 * CSS attributes that accept gradient data types.
		 *
		 */
		$css_gradient_data_types = [
			'background',
			'background-image',
		];

		/*
		 * CSS attributes that accept color data types.
		 *
		 * This is in accordance to the CSS spec and unrelated to
		 * the sub-set of supported attributes above.
		 *
		 * See: https://developer.mozilla.org/en-US/docs/Web/CSS/color_value
		 */
		$css_color_data_types = [
			'color',
			'background',
			'background-color',
			'border-color',
			'box-shadow',
			'outline',
			'outline-color',
			'text-shadow',
		];

		if ( empty( $allowed_attr ) ) {
			return $css;
		}

		$css = '';
		foreach ( $css_array as $css_item ) {
			if ( '' === $css_item ) {
				continue;
			}

			$css_item        = trim( $css_item );
			$css_test_string = $css_item;
			$found           = false;
			$url_attr        = false;
			$gradient_attr   = false;
			$color_attr      = false;
			$transform_attr  = false;

			$parts = explode( ':', $css_item, 2 );

			if ( false === strpos( $css_item, ':' ) ) {
				$found = true;
			} else {
				$css_selector = trim( $parts[0] );

				if ( \in_array( $css_selector, $allowed_attr, true ) ) {
					$found         = true;
					$url_attr      = \in_array( $css_selector, $css_url_data_types, true );
					$gradient_attr = \in_array( $css_selector, $css_gradient_data_types, true );
					$color_attr    = \in_array( $css_selector, $css_color_data_types, true );

					// --initial-transform is a special custom property used by the story editor.
					$transform_attr = 'transform' === $css_selector || '--initial-transform' === $css_selector;
				}
			}

			if ( $found && $url_attr ) {
				$url_matches = [];

				// Simplified: matches the sequence `url(*)`.
				preg_match_all( '/url\([^)]+\)/', $parts[1], $url_matches );

				foreach ( $url_matches[0] as $url_match ) {
					$url_pieces = [];

					// Clean up the URL from each of the matches above.
					preg_match( '/^url\(\s*([\'\"]?)(.*)(\g1)\s*\)$/', $url_match, $url_pieces );

					if ( empty( $url_pieces[2] ) ) {
						$found = false;
						break;
					}

					$url = trim( $url_pieces[2] );

					if ( empty( $url ) || wp_kses_bad_protocol( $url, $allowed_protocols ) !== $url ) {
						$found = false;
						break;
					}

					// Remove the whole `url(*)` bit that was matched above from the CSS.
					$css_test_string = str_replace( $url_match, '', $css_test_string );
				}
			}

			if ( $found && $gradient_attr ) {
				$css_value = trim( $parts[1] );
				if ( preg_match( '/^(repeating-)?(linear|radial|conic)-gradient\(([^()]|rgb[a]?\([^()]*\))*\)$/', $css_value ) ) {
					// Remove the whole `gradient` bit that was matched above from the CSS.
					$css_test_string = str_replace( $css_value, '', $css_test_string );
				}
			}

			if ( $found && $color_attr ) {
				$color_matches = [];

				// Simplified: matches the sequence `rgb(*)` and `rgba(*)`.
				preg_match_all( '/rgba?\([^)]+\)/', $parts[1], $color_matches );

				foreach ( $color_matches[0] as $color_match ) {
					$color_pieces = [];

					// Clean up the color from each of the matches above.
					preg_match( '/^rgba?\([^)]*\)$/', $color_match, $color_pieces );

					// Remove the whole `rgb(*)` / `rgba(*) bit that was matched above from the CSS.
					$css_test_string = str_replace( $color_match, '', $css_test_string );
				}
			}

			if ( $found && $transform_attr ) {
				$css_value = trim( $parts[1] );
				if ( preg_match( '/^((matrix|matrix3d|perspective|rotate|rotate3d|rotateX|rotateY|rotateZ|translate|translate3d|translateX|translatY|translatZ|scale|scale3d|scalX|scaleY|scaleZ|skew|skewX|skeY)\(([^()])*\) ?)+$/', $css_value ) ) {
					// Remove the whole `gradient` bit that was matched above from the CSS.
					$css_test_string = str_replace( $css_value, '', $css_test_string );
				}
			}

			if ( $found ) {
				// Allow CSS calc().
				$css_test_string = (string) preg_replace( '/calc\(((?:\([^()]*\)?|[^()])*)\)/', '', $css_test_string );
				// Allow CSS var().
				$css_test_string = (string) preg_replace( '/\(?var\(--[a-zA-Z0-9_-]*\)/', '', $css_test_string );

				// Check for any CSS containing \ ( & } = or comments,
				// except for url(), calc(), or var() usage checked above.
				$allow_css = ! preg_match( '%[\\\(&=}]|/\*%', $css_test_string );

				/** This filter is documented in wp-includes/kses.php */
				$allow_css = apply_filters( 'safecss_filter_attr_allow_css', $allow_css, $css_test_string );

				// Only add the CSS part if it passes the regex check.
				if ( $allow_css ) {
					if ( '' !== $css ) {
						$css .= ';';
					}

					$css .= $css_item;
				}
			}
		}

		return $css;
	}


	/**
	 * Filter the allowed tags for KSES to allow for complete amp-story document markup.
	 *
	 * @SuppressWarnings(PHPMD.ExcessiveMethodLength)
	 *
	 * @since 1.0.0
	 *
	 * @param array|mixed $allowed_tags Allowed tags.
	 * @return array|mixed Allowed tags.
	 */
	public static function filter_kses_allowed_html($allowed_tags){
		if(!\is_array($allowed_tags)){
			return $allowed_tags;
		}

		$story_components = [
			'html'                          => [
				'amp'  => true,
				'lang' => true,
			],
			'head'                          => [],
			'body'                          => [],
			'meta'                          => [
				'name'    => true,
				'content' => true,
				'charset' => true,
			],
			'script'                        => [
				'async'          => true,
				'src'            => true,
				'custom-element' => true,
				'type'           => true,
			],
			'noscript'                      => [],
			'link'                          => [
				'href' => true,
				'rel'  => true,
			],
			'style'                         => [
				'type'            => true,
				'amp-boilerplate' => true,
				'amp-custom'      => true,
			],
			'amp-story'                     => [
				'background-audio'     => true,
				'live-story'           => true,
				'live-story-disabled'  => true,
				'poster-landscape-src' => true,
				'poster-portrait-src'  => true,
				'poster-square-src'    => true,
				'publisher'            => true,
				'publisher-logo-src'   => true,
				'standalone'           => true,
				'supports-landscape'   => true,
				'title'                => true,
			],
			'amp-story-captions'            => [
				'height' => true,
			],
			'amp-story-shopping-attachment' => [
				'cta-text' => true,
				'theme'    => true,
				'src'      => true,
			],
			'amp-story-shopping-config'     => [
				'src' => true,
			],
			'amp-story-shopping-tag'        => [],
			'amp-story-page'                => [
				'auto-advance-after' => true,
				'background-audio'   => true,
				'id'                 => true,
			],
			'amp-story-page-attachment'     => [
				'href'  => true,
				'theme' => true,
			],
			'amp-story-page-outlink'        => [
				'cta-image'          => true,
				'theme'              => true,
				'cta-accent-color'   => true,
				'cta-accent-element' => true,
			],
			'amp-story-grid-layer'          => [
				'aspect-ratio' => true,
				'position'     => true,
				'template'     => true,
			],
			'amp-story-cta-layer'           => [],
			'amp-story-animation'           => [
				'trigger' => true,
			],
			'amp-img'                       => [
				'alt'                       => true,
				'attribution'               => true,
				'data-amp-bind-alt'         => true,
				'data-amp-bind-attribution' => true,
				'data-amp-bind-src'         => true,
				'data-amp-bind-srcset'      => true,
				'disable-inline-width'      => true,
				'lightbox'                  => true,
				'lightbox-thumbnail-id'     => true,
				'media'                     => true,
				'noloading'                 => true,
				'object-fit'                => true,
				'object-position'           => true,
				'placeholder'               => true,
				'sizes'                     => true,
				'src'                       => true,
				'srcset'                    => true,
			],
			'amp-video'                     => [
				'album'                      => true,
				'alt'                        => true,
				'artist'                     => true,
				'artwork'                    => true,
				'attribution'                => true,
				'autoplay'                   => true,
				'captions-id'                => true,
				'controls'                   => true,
				'controlslist'               => true,
				'crossorigin'                => true,
				'data-amp-bind-album'        => true,
				'data-amp-bind-alt'          => true,
				'data-amp-bind-artist'       => true,
				'data-amp-bind-artwork'      => true,
				'data-amp-bind-attribution'  => true,
				'data-amp-bind-controls'     => true,
				'data-amp-bind-controlslist' => true,
				'data-amp-bind-loop'         => true,
				'data-amp-bind-poster'       => true,
				'data-amp-bind-preload'      => true,
				'data-amp-bind-src'          => true,
				'data-amp-bind-title'        => true,
				'disableremoteplayback'      => true,
				'dock'                       => true,
				'lightbox'                   => true,
				'lightbox-thumbnail-id'      => true,
				'loop'                       => true,
				'media'                      => true,
				'muted'                      => true,
				'noaudio'                    => true,
				'noloading'                  => true,
				'object-fit'                 => true,
				'object-position'            => true,
				'placeholder'                => true,
				'poster'                     => true,
				'preload'                    => true,
				'rotate-to-fullscreen'       => true,
				'src'                        => true,
			],
			'source'                        => [
				'type' => true,
				'src'  => true,
			],
			'img'                           => [
				'alt'           => true,
				'attribution'   => true,
				'border'        => true,
				'decoding'      => true,
				'height'        => true,
				'importance'    => true,
				'intrinsicsize' => true,
				'ismap'         => true,
				'loading'       => true,
				'longdesc'      => true,
				'sizes'         => true,
				'src'           => true,
				'srcset'        => true,
				'srcwidth'      => true,
				'width'         => true,
			],
			'svg'                           => [
				'width'   => true,
				'height'  => true,
				'viewbox' => true,
				'fill'    => true,
				'xmlns'   => true,
			],
			'clippath'                      => [
				'transform'     => true,
				'clippathunits' => true,
				'path'          => true,
			],
			'defs'                          => [],
			'feblend'                       => [
				'in'     => true,
				'in2'    => true,
				'result' => true,
			],
			'fecolormatrix'                 => [
				'in'     => true,
				'values' => true,
			],
			'feflood'                       => [
				'flood-opacity' => true,
				'result'        => true,
			],
			'fegaussianblur'                => [
				'stddeviation' => true,
			],
			'feoffset'                      => [],
			'filter'                        => [
				'id'                          => true,
				'x'                           => true,
				'y'                           => true,
				'width'                       => true,
				'height'                      => true,
				'filterunits'                 => true,
				'color-interpolation-filters' => true,
			],
			'g'                             => [
				'filter'  => true,
				'opacity' => true,
			],
			'path'                          => [
				'd'         => true,
				'fill-rule' => true,
				'clip-rule' => true,
				'fill'      => true,
			],
		];

		$allowed_tags = self::array_merge_recursive_distinct($allowed_tags, $story_components);

		$allowed_tags = array_map([__CLASS__, 'add_global_attributes'], $allowed_tags);

		return $allowed_tags;
	}

	/**
	 * Recursively merge multiple arrays and ensure values are distinct.
	 *
	 * Based on information found in http://www.php.net/manual/en/function.array-merge-recursive.php
	 *
	 * @since 1.5.0
	 *
	 * @param array ...$arrays [optional] Variable list of arrays to recursively merge.
	 * @return array An array of values resulted from merging the arguments together.
	 */
	protected static function array_merge_recursive_distinct( array ...$arrays ){
		if ( \count( $arrays ) < 2 ) {
			if ( [] === $arrays ) {
				return $arrays;
			}

			return array_shift( $arrays );
		}

		$merged = array_shift( $arrays );

		foreach ( $arrays as $array ) {
			foreach ( $array as $key => $value ) {
				if ( \is_array( $value ) && ( isset( $merged[ $key ] ) && \is_array( $merged[ $key ] ) ) ) {
					$merged[ $key ] = self::array_merge_recursive_distinct( $merged[ $key ], $value );
				} else {
					$merged[ $key ] = $value;
				}
			}
		}

		return (array) $merged;
	}

	/**
	 * Helper function to add global attributes to a tag in the allowed HTML list.
	 *
	 * @since 1.0.0
	 *
	 * @see _wp_add_global_attributes
	 *
	 * @param array $value An array of attributes.
	 * @return array The array of attributes with global attributes added.
	 */
	protected static function add_global_attributes($value){
		$global_attributes = [
			'aria-describedby'    => true,
			'aria-details'        => true,
			'aria-label'          => true,
			'aria-labelledby'     => true,
			'aria-hidden'         => true,
			'class'               => true,
			'id'                  => true,
			'style'               => true,
			'title'               => true,
			'role'                => true,
			'data-*'              => true,
			'animate-in'          => true,
			'animate-in-duration' => true,
			'animate-in-delay'    => true,
			'animate-in-after'    => true,
			'animate-in-layout'   => true,
			'layout'              => true,
		];

		return (array) array_merge( $value, $global_attributes );
	}

    /**
     * End borrowed Google Web Story code.
     **/

    /**
     * Inserts links into Google Web Story "filtered_post_content".
     * Since the content is JSON encoded, we need a special updater for it.
     **/
    public static function insert_links_in_webstory_json(&$data, $sentence, $changed_sentence, $force_insert){
        if(empty($data)){
            return false;
        }

        // if there's no data or there's no pages
        if(empty($data) || !isset($data['pages']) || empty($data['pages'])){
            // exit
            return false;
        }

        foreach($data['pages'] as &$dat){
            if(!empty($dat) && isset($dat['elements']) && !empty($dat['elements'])){
                foreach($dat['elements'] as &$sub_dat){
                    if( isset($sub_dat['type']) && $sub_dat['type'] === 'text' && 
                        isset($sub_dat['content']) && !empty($sub_dat['content']))
                    {
                        self::insertLink($sub_dat['content'], $sentence, addslashes($changed_sentence), $force_insert);
                    }
                }
            }
        }
    }

    /**
     * 
     **/
    public static function allowed_tags_and_attributes($current_allowed, $context){
        $new_allowed = array();
        // if Ultimate Gutenberg Blocks (Now Spectra) is active
        if('post' === $context && defined('UAGB_VER')){
            $new_allowed['svg'] = array(
                'xmlns'   => true,
                'viewbox' => true,
            );
            $new_allowed['path'] = array(
                'd' => true,
            );
            $new_allowed['script'] = array(
                'async'          => true,
                'src'            => true,
                'custom-element' => true,
                'type'           => true,
            );
            $new_allowed['style'] = array(
                'type'          => true
            );
            $new_allowed['div'] = array(
                'tabindex'      => true
            );
        }
        $allowed_tags = self::array_merge_recursive_distinct($current_allowed, $new_allowed);

        return $allowed_tags;
    }

    /**
     * Check if it need to force page reload
     */
    function editorReload(){
        if (!empty($_POST['post_id'])) {
            $meta = get_post_meta((int)$_POST['post_id'], 'wpil_gutenberg_restart', true);
            if (!empty($meta)) {
                delete_post_meta((int)$_POST['post_id'], 'wpil_gutenberg_restart');
                echo 'reload';
            }
        }

        wp_die();
    }

    /**
     * Check if outbound links were added to show dialog box
     */
    function isOutboundLinksAdded(){
        if (!empty($_POST['id']) && !empty($_POST['type'])) {
            if ($_POST['type'] == 'term') {
                $meta = get_term_meta((int)$_POST['id'], 'wpil_is_outbound_links_added', true);
            } else {
                $meta = get_post_meta((int)$_POST['id'], 'wpil_is_outbound_links_added', true);
            }
            if (!empty($meta)) {
                if ($_POST['type'] == 'term') {
                    delete_term_meta((int)$_POST['id'], 'wpil_is_outbound_links_added');
                } else {
                    delete_post_meta((int)$_POST['id'], 'wpil_is_outbound_links_added');
                }
                echo 'success';
            }
        }

        wp_die();
    }

    /**
     * Check if inbound links were added to show dialog box
     */
    function isInboundLinksAdded(){
        if (!empty($_POST['id']) && !empty($_POST['type'])) {
            if ($_POST['type'] == 'term') {
                $meta = get_term_meta((int)$_POST['id'], 'wpil_is_inbound_links_added', true);
            } else {
                $meta = get_post_meta((int)$_POST['id'], 'wpil_is_inbound_links_added', true);
            }
            if (!empty($meta)) {
                if ($_POST['type'] == 'term') {
                    delete_term_meta((int)$_POST['id'], 'wpil_is_inbound_links_added');
                } else {
                    delete_post_meta((int)$_POST['id'], 'wpil_is_inbound_links_added');
                }
                echo 'success';
            }
        }

        wp_die();
    }

    /**
     * Ignores the selected orphaned post on the orphaned post view.
     **/
    function ajaxIgnoreOrphanedPost(){
        Wpil_Base::verify_nonce('ignore-orphaned-post-nonce');

        if(!isset($_POST['post_ids']) || empty($_POST['post_ids']) || !is_array($_POST['post_ids'])){
            wp_send_json(array('error' => array('title' => __('Post id empty', 'wpil'),'text' => __('The post id was missing from the ignore orphaned post request.', 'wpil'))));
        }

        // get all the ignored orphaned posts (including ignored by category)
        $ignored = Wpil_Settings::getIgnoreOrphanedPosts();

        // get any specifically ignored posts
        $ignored_posts = get_option('wpil_ignore_orphaned_posts', '');

        foreach($_POST['post_ids'] as $pid){
            // if the post is ignored, move on to the next one
            if(in_array($pid, $ignored, true)){
                continue;
            }

            $bits = explode('_', $pid);

            // get the post
            $post = new Wpil_Model_Post((int)$bits[1], sanitize_text_field($bits[0]));

            $post_link = $post->getViewLink();
            $ignore_post = self::getPostByLink($post_link); // try getting the post to ensure that there's no issues getting the post from url
            if(!empty($ignore_post) && $post->id === $ignore_post->id){
                $ignored_posts .= "\n" . $post_link;
            }else{
                $ignored_posts .= "\n" . $post->getViewLink(false, true); // if we can't turn the url into a viable post, go with the "Ugly" url instead.
            }
        }

        update_option('wpil_ignore_orphaned_posts', $ignored_posts, false);

        wp_send_json(array('success' => true));
    }

    /**
     * Filters the post types that the custom link search box will look for so the user is only shown selected post types
     **/
    public static function filter_custom_link_post_types($query_args){
        if(!empty($_POST) && isset($_POST['wpil_custom_link_search'])){
            $selected_post_types = Wpil_Settings::getPostTypes();
            if(!empty($selected_post_types)){
                $query_args['post_type'] = $selected_post_types;
            }
        }
        return $query_args;
    }

    /**
     * Queries for terms when the user does a custom link search for outbound suggestions.
     * The existing search only does posts, so we have to do the terms separately
     **/
    public static function custom_link_category_search($queried_items = array()){
        if(!empty($_POST) && isset($_POST['wpil_custom_link_search'])){

            $selected_terms = get_option('wpil_2_term_types', array());

            if(empty($selected_terms)){
                return $queried_items;
            }

            $args = array('taxonomy' => $selected_terms, 'search' => $_POST['search'], 'number' => 20);

            $term_query = new WP_Term_Query($args);
            $terms = $term_query->get_terms();

            if(empty($terms)){
                return $queried_items;
            }

            foreach($terms as $term){
                $queried_items[] = array(
                    'ID' => $term->term_id,
                    'title' => $term->name,
                    'permalink' => get_term_link($term->term_id),
                    'info' => ucfirst($term->taxonomy),
                );

            }
        }

        return $queried_items;
    }

    /**
     * Insert links into sentence
     *
     * @param $sentence
     * @param $anchor
     * @param $url
     * @param $to_post_id
     * @return string
     */
    public static function getSentenceWithAnchor($link) {
        if (!empty($link['custom_sentence'])) {
            $link['custom_sentence'] = mb_ereg_replace(preg_quote(',</a>'), '</a>,', $link['custom_sentence']);
            return $link['custom_sentence'];
        }

        //get URL
        preg_match('/<a href="([^\"]+)"[^>]*>(.*)<\/a>/i', wp_unslash($link['sentence_with_anchor']), $matches);
        if (empty($matches[1])) {
            return $link['sentence'];
        }

        // update the sentence's tags
        $link['sentence'] = self::update_sentence_tags($link['sentence'], $link['sentence_with_anchor']);

        $url = $matches[1];

        //get anchor from source sentence
        $words = [];
        $word_start = false;
        $word_end = 0;
        preg_match_all('/<span[^>]+>([^<]+)<\/span>/i', $matches[2], $matches);
        if (count($matches[1])) {
            // go over the matches
            foreach ($matches[1] as $word) {
                // and try to determine where the anchor text begins and ends
                // if the start hasn't been found
                if ($word_start === false) {
                    // see if the first word in the sentence is a free-standing word
                    $word_start = stripos($link['sentence'], $word . ' ');
                    // if it is not...
                    if(false === $word_start){
                        // look for the word more loosly and set the start for that
                        $word_start = stripos($link['sentence'], $word);
                    }
                    // and provisionally set the text ending for the end of the first word
                    $word_end = $word_start + strlen($word);
                } else {
                    // in all other cases, just keep moving the anchor text ending down the sentence until we're done.
                    $word_end = stripos($link['sentence'], $word, $word_end) + strlen($word);
                }

                // also add the word to a list of words
                $words[] = $word;
            }
        }

        $word_search = array();
        $string_length = mb_strlen($link['sentence']);
        foreach($words as $word){
            $offset = 0;
            for($i = 0; $i < mb_substr_count($link['sentence'], $word); $i++){
                if($offset > $string_length){
                    break; // make sure we don't commit the capital offense of measuring a string that's shorter than the offset
                }

                $position = Wpil_Word::mb_strpos($link['sentence'], $word, $offset);
                $word_search[] = array(
                    'pos' => $position,
                    'word' => $word
                );
                $offset = $position + 1;
            }
        }

        if(!empty($word_search)){
            usort($word_search, function($a, $b){
                if ($a['pos'] == $b['pos']) {
                    return 0;
                }
    
                return ($a['pos'] > $b['pos']) ? 1 : -1;
            });
        }

        // now search the found words for the ones that match the selected sentence
        $start_pos = null;
        $end_pos = 0;
        $sentence_pos = 0; // sentence word counter so we can keep track of what sequencial sentence word we're looking at
        foreach($word_search as $search){
            if(
                isset($words[$sentence_pos]) && 
                $search['word'] === $words[$sentence_pos]){
                if($start_pos === null){
                    $start_pos = $search['pos'];
                }
                $end_pos = ($search['pos'] + mb_strlen($search['word']));

                $sentence_pos++;
            }elseif($sentence_pos >= count($words)){
                break;
            }else{
                // if the next word in the sentence isn't the next one in the search
                // check if it's actually the next starting word
                if(isset($words[0]) && $search['word'] === $words[0]){
                    // if it is, start the sentence here
                    $start_pos = $search['pos'];
                    $end_pos = ($search['pos'] + mb_strlen($search['word']));
                    $sentence_pos = 1;
                }else{
                    // otherwise, reset the search variables
                    $start_pos = null;
                    $end_pos = 0;
                    $sentence_pos = 0;
                }
            }
        }

        $anchor = false;
        // if we have positions to work with
        if($end_pos > $start_pos){
            // get the anchor
            $anchor = mb_substr($link['sentence'], (int)$start_pos, ($end_pos - (int)$start_pos));
            // check if there's a styling tag inside it
            $tags = array('<b>' => '</b>', '<i>' => '</i>', '<u>' => '</u>', '<strong>' => '</strong>', '<em>' => '</em>', '<code>' => '</code>');
            foreach($tags as $opening => $closing){
                // if we have an opening one inside the anchor, but no closing one
                if(false !== strpos($anchor, $opening) && false === strpos($anchor, $closing)){
                    // check if the closer is just after the last word
                    if(false !== Wpil_Word::mb_strpos($link['sentence'], ($anchor . $closing))){
                        // if it is, add the closing tag to the anchor
                        $anchor = ($anchor . $closing);
                    }
                }
            }
        }

        // if that didn't work, use the old method of getting the anchor text
        if(empty($anchor)){
            //get start position by nearest whitespace
            $start = 0;
            $i = 0;
            while(strpos($link['sentence'], ' ', $start+1) < $word_start && $i < 100) {
                $start = strpos($link['sentence'], ' ', $start+1);
                $next_whitespace = strpos($link['sentence'], ' ', $start+1);
                $tag = strpos($link['sentence'], '>', $start +1);
                if ($tag && $tag < $next_whitespace) {
                    $start = $tag;
                }
                $tag = strpos($link['sentence'], '(', $start +1);
                if ($tag && $tag < $next_whitespace) {
                    $start = $tag;
                }
                $i++;

                // exit the loop if there's no further whitespace
                if(empty($next_whitespace)){
                    break;
                }
            }
            if ($start) {
                $start++;
            }

            $nbsp = urldecode('%C2%A0');

            //get end position by nearest whitespace
            $end = 0;
            $prev_end = 0;
            while($end < $word_end && $end !== false) {
                $prev_end = $end;
                $end = strpos($link['sentence'], ' ', $end + 1);
                $tag = strpos($link['sentence'], ')', $prev_end +1);

                if($end > $word_end){
                    $maybe_end = strpos($link['sentence'], $nbsp, $prev_end + 1);
                    if(!empty($maybe_end) && $maybe_end < $word_end){
                        $end = $maybe_end;
                    }
                }

                if ($tag && $tag < $end) {
                    $end = $tag;
                }
            }

            if (substr($link['sentence'], $end-1, 1) == ',') {
                $end -= 1;
            }

            if ($end === false) {
                $end = strlen($link['sentence']);
            }

            $anchor = substr($link['sentence'], $start, $end - $start);
        }

        $anchor2 = str_replace('$', '\\$', $anchor);
        $sentence = self::build_link_with_attrs($url, $anchor2, $link['sentence']);

        return $sentence;
    }

    /**
     * Updates the html style tags in the sentence with the results from sentence with anchor.
     **/
    public static function update_sentence_tags($sentence, $sentence_with_anchor){

        // find all the encoded style tags
        preg_match_all('/<span[^><]*?class=["\'][^"\']*?wpil_suggestion_tag[^"\']*?["\'][^>]*?>([^<]*?)<\/span>/', $sentence_with_anchor, $matches);

        if(empty($matches)){
            return $sentence;
        }

        foreach($matches[0] as $key => $match){
            $decoded = base64_decode($matches[1][$key]);
            if(preg_match('/' . preg_quote($match, '/') . '\s*/', $sentence_with_anchor)){
                $sentence_with_anchor = preg_replace('/' . preg_quote($match, '/') . '\s*/', $decoded, $sentence_with_anchor);
            }else{
                $sentence_with_anchor = str_replace($match, $decoded, $sentence_with_anchor);
            }
        }

        // find all the non word tags
        preg_match_all('/<span[^><]*?class=["\'][^"\']*?wpil-non-word[^"\']*?["\'][^>]*?>([^<]*?)<\/span>/', $sentence_with_anchor, $matches);

        // if there are non word tags, remove them so they don't throw off the formatting
        if(!empty($matches)){
            foreach($matches[0] as $key => $match){
                $sentence_with_anchor = str_replace($match, $matches[1][$key], $sentence_with_anchor);
            }
        }

        $new_sentence = strip_tags($sentence_with_anchor, '<b><i><u><strong><em><code>');

        // remove any tags that are opening and closing without content
        $new_sentence = str_replace(array('<b></b>', '<i></i>', '<u></u>', '<strong></strong>', '<em></em>', '<code></code>'), '', $new_sentence);
        $new_sentence = str_replace(array('<b> </b>', '<i> </i>', '<u> </u>', '<strong> </strong>', '<em> </em>', '<code> </code>'), '', $new_sentence);

        // if the sentences are the same after removing all tags
        if(trim(strip_tags($sentence)) === trim(strip_tags($sentence_with_anchor)) || trim(strip_tags($sentence)) === str_replace('  ', ' ', trim(strip_tags($sentence_with_anchor))) ){
            // update the sentence with the new tagged version
            $sentence = trim($new_sentence);
        }

        return $sentence;
    }

    /**
     * Adds attributes to links that are being created.
     **/
    public static function build_link_with_attrs($url = '', $anchor = '', $sentence = ''){
        $external = !Wpil_Link::isInternal($url);
        $open_new_tab = (int)get_option('wpil_2_links_open_new_tab', 0);
        $open_external_new_tab = false;
        if($external){
            $open_external_new_tab = get_option('wpil_external_links_open_new_tab', null);
        }

        //add target blank if needed
        $blank = '';
        $rel = '';
        if (($open_new_tab == 1 && empty($external)) || 
            ($external && $open_external_new_tab) ||
            ($open_new_tab == 1 && $open_external_new_tab === null)
        ) {
            $noreferrer = !empty(get_option('wpil_add_noreferrer', false)) ? ' noreferrer': '';
            $blank = 'target="_blank"';
            $rel = 'rel="noopener' . $noreferrer;
        }

        // if the user has set external links to be nofollow, this is an external link, and this isn't an interlinked site
        if(
            !empty(get_option('wpil_add_nofollow', false)) && 
            $external && 
            !empty(wp_parse_url($url, PHP_URL_HOST)) &&
            !in_array(wp_parse_url($url, PHP_URL_HOST), Wpil_SiteConnector::get_linked_site_domains(), true))
        {
            if(empty($rel)){
                $rel = 'rel="nofollow';
            }else{
                $rel .= ' nofollow';
            }
        }

        // if the user has set some domains to be listed as sponsored
        if(
            $external && 
            !empty(wp_parse_url($url, PHP_URL_HOST)) &&
            Wpil_Link::isSponsoredLink($url))
        {
            if(empty($rel)){
                $rel = 'rel="sponsored';
            }else{
                $rel .= ' sponsored';
            }
        }

        if(!empty($rel)){
            $rel .= '"';
        }

        $anchor2 = str_replace('$', '\\$', $anchor);

        /**
         * allow the users to add classes to the link
         * @param string The class list
         * @param bool $external Is the link going to an external site?
         * @param string The location of the filter
         **/
        $classes = apply_filters('wpil_link_classes', '', $external, 'suggestions');

        // if the user returned an array, stringify it
        if(is_array($classes)){
            $classes = implode(' ', $classes);
        }

        $classes = (!empty($classes)) ? 'class="' . sanitize_text_field($classes) . '"': '';

        $title = '';
        if(!empty(get_option('wpil_add_destination_title', false))){
            $dest_post = self::getPostByLink($url);

            if(!empty($dest_post)){
                if($dest_post->type === 'post'){
                    $post = get_post($dest_post->id);
                    $title = 'title="'. esc_attr(apply_filters('wpil_link_destination_title', str_replace(array('[', ']'), array('&#91;', '&#93;'), get_the_title($post)), $post)) .'"';
                }else{
                    $term = get_term($dest_post->id);
                    $title = 'title="'. esc_attr(apply_filters('wpil_link_destination_title', str_replace(array('[', ']'), array('&#91;', '&#93;'), $term->name), $term)) .'"';
                }
            }
        }

        // todo build into a separate attr function with the other checks
        $attrs = '';
        if(!empty($title)){
            $attrs .= ' ' . $title;
        }
        if(!empty($blank)){
            $attrs .= ' ' . $blank;
        }
        if(!empty($rel)){
            $attrs .= ' ' . $rel;
        }
        if(!empty($classes)){
            $attrs .= ' ' . $classes;
        }

        if(!Wpil_Settings::get_disable_link_insert_tracking()){
            $attrs .= ' data-wpil-replace=""';
        }

        // change the staging domain into the live domain if the user has opted to do so
        $url = Wpil_Link::filter_staging_to_live_domain($url);

        //add link to sentence
        $sentence = preg_replace('/'.preg_quote($anchor, '/').'/i', '<a href="'.$url.'"' . $attrs . '>'.$anchor2.'</a>', $sentence, 1);

        $sentence = str_replace('$', '\\$', $sentence);

        // format the tags inside the sentence to make sure there's no half-in half-out tags
        $sentence = self::format_sentence_tags($sentence);

        return $sentence;
    }

    /**
     * Makes sure there aren't any tags that are half-in/half-out of the anchor tag.
     * Moves any offending tags along the same lines as the JS mover:
     ** If just the closing tag is inside the anchor, move it left until it's outside the anchor.
     ** If just the opening tag is inside the anchor, move it right until it's outside the anchor.
     ** If opening and closing tags are next to each other, remove them.
     **/
    public static function format_sentence_tags($sentence){

        // return the sentence if there's no tags inside the anchor
        if(empty(preg_match('/<a.*?>.*?(<[A-Za-z\/]*?>).*?<\/a>/', $sentence, $check)) || !isset($check[1]) || empty($check[1])){
            return $sentence;
        }

        // get the anchor tag and it's position data
        $link_start = Wpil_Word::mb_strpos($sentence, '<a href="');
        $link_end = Wpil_Word::mb_strpos($sentence, '</a>', $link_start);
        $link_length = ($link_end + 4 - $link_start);
        $link = mb_substr($sentence, $link_start, $link_length);
        $link_copy = $link;

        $tags_before_anchor = array();
        $tags_after_anchor = array();

        // check the anchor to see what tags it contains
        $tags_to_check = array('(<b>|<\/b>)', '(<i>|<\/i>)', '(<u>|<\/u>)', '(<strong>|<\/strong>)', '(<em>|<\/em>)', '(<code>|<\/code>)');
        foreach($tags_to_check as $tag){
            // if it only contains one tag
            if(preg_match_all('/' . $tag . '/', $link, $matches, PREG_OFFSET_CAPTURE) === 1){
                // extract the tag
                $pulled_tag = $matches[0][0][0];
                // get the tag's position
                $position = $matches[0][0][1];
                // replace the tag in the copied link
                $link_copy = mb_ereg_replace(preg_quote($pulled_tag), '', $link_copy);
                // find out if the tag is the first thing after the opening link tag // allowing for space
                $at_start = preg_match('/<a.*?>[ ]*(' . preg_quote($pulled_tag, '/') . ').*?<\/a>/', $sentence);

                // if the tag is a closing tag
                if(strpos($pulled_tag, '/')){
                    // put it on the list of tags that come before the anchor
                    $tags_before_anchor[$position] = $pulled_tag;
                }else{
                    // if it's an opening tag, check to see if it it's immediately after the link's opening tag
                    if($at_start){
                        // if it does, put it on the list that comes before the link tag
                        $tags_before_anchor[$position] = $pulled_tag;
                    }else{
                        // if it doesn't come right after the link, put it on the list of tags that come after the anchor
                        $tags_after_anchor[$position] = $pulled_tag;
                    }
                }
            }
        }

        // if there are tags that should be moved in front of the anchor
        if(!empty($tags_before_anchor)){
            // sort them to make sure we don't make a mess
            ksort($tags_before_anchor);
            // and insert them before the anchor
            $link_copy = implode('', $tags_before_anchor) . $link_copy;
        }

        // if there are tags that should be moved past the end of the anchor
        if(!empty($tags_after_anchor)){
            // sort them to make sure we don't make a mess
            ksort($tags_after_anchor);
            // and add them after the anchor
            $link_copy = $link_copy . implode('', $tags_after_anchor);
        }

        // replace the old link with the new link
        $sentence = mb_ereg_replace(preg_quote($link), $link_copy, $sentence);

        // remove any double tags // it is possible that a user will have something like <strong><em><u></u></em></strong> that should be removed, but we'll cross that bridge when we get there
        $sentence = str_replace(array('<b></b>', '<i></i>', '<u></u>', '<strong></strong>', '<em></em>', '<code></code>'), '', $sentence);
        $sentence = str_replace(array('<b> </b>', '<i> </i>', '<u> </u>', '<strong> </strong>', '<em> </em>', '<code> </code>'), ' ', $sentence);

        return $sentence;
    }

    /**
     * Get post content
     *
     * @param $post_id integer
     * @return string
     */
    public static function getPostContent($post_id)
    {
        $post = get_post($post_id);

        return !empty($post->post_content) ? $post->post_content : '';
    }

    public static function addAutolinksToInsertRun($data, $post, $unsanitized_postarr = array()){
        if(empty($post) || !isset($post['ID']) || empty($data) || Wpil_Settings::disable_autolink_on_post_save()){
            return $data; // since we're on a filter, we want to remember to return the data if it doesn't meet our checks
        }

        if (!in_array($post['post_type'], Wpil_Settings::getAllTypes()) || !in_array($post['post_status'], Wpil_Settings::getPostStatuses())) {
            return $data;
        }

        if (in_array('post_' . $post['ID'], Wpil_Settings::getIgnoreKeywordsPosts())) {
            return $data;
        }

        // exit if we've just inserted selected links so we don't insert duplicates
        if(!empty($_POST) && isset($_POST['action']) && 'wpil_insert_selected_keyword_links' === $_POST['action']){
            return $data;
        }

        // make sure we're in the correct priority
        if(998 !== Wpil_Toolbox::get_current_action_priority()){
            return $data;
        }

        // check if this is a REST API post create run
        if( defined('REST_REQUEST') && !empty(REST_REQUEST) &&  // if this is a rest request
            isset($post['ID']) && empty($post['ID']) &&         // there's no post ID
            isset($post['post_type']) && in_array($post['post_type'], Wpil_Settings::getPostTypes())
        ){
            // if it is
            if(Wpil_Toolbox::is_doing_external_rest_api_action()){
                // we must be looking at a REST API created post, and we should setup an autolinking run after the insert
                // set the flag to force autolink running in saveAutolinksToPost
                Wpil_Base::track_action('doing_rest_api_post_create', true); // NOTE: if using the save_post hook isn't fast enough, look into creating a standalone rest api inserter that calls the link inserter during the intial creation
                // and exit
                return $data;
            }
        }

        // be sure to ignore any external object caches
        Wpil_Base::ignore_external_object_cache(true); // this might hurt...

        $wpil_post = new Wpil_Model_Post($post['ID']);

        // exit if we're already at the link limit
        if(Wpil_link::at_max_outbound_links($wpil_post)){
            return $data;
        }

        // if autolinks have been inserted from saveAutolinksToPost
        if(Wpil_Base::action_happened('did_active_autolink_insert')){
            return $data;
        }

        $content = wp_unslash($post['post_content']);
        $lowercased_content = mb_strtolower($content);

        if(empty($lowercased_content)){
            return $data;
        }

        $keywords = Wpil_Keyword::getKeywords();
        $url_index = array();
        $keyword_index = array();
        $post_language = self::getPostLanguageCode($wpil_post);
        foreach ($keywords as $key => $keyword) {
            $keyword->keyword = stripslashes($keyword->keyword);
            if(empty($keyword->keyword)){
                unset($keywords[$key]);
                continue;
            }

            if ( 
                (!empty($keyword->case_sensitive) && strpos($content, $keyword->keyword)) || // if this is a case sensitive keyword and it doesn't show up in the content
                strpos($lowercased_content, mb_strtolower($keyword->keyword)) === false || // OR if it's a non-case sensitive keyword
                empty(preg_match('/(?<![a-zA-Z][\'\-_]|[a-zA-Z]|="|=\\\"|\/)'.preg_quote(mb_strtolower($keyword->keyword), '/').'(?![a-zA-Z]|[\'\-_][a-zA-Z]|<\/a>|<\\\\\/a>)/', $lowercased_content, $matches)) || empty($matches)) // OR the keyword only shows up inside of existing lnks
            {
                unset($keywords[$key]);
                continue;
            }
            // if a link with the current link's url is slated to be installed and the current link doesn't have rules to insert more than once
            if(isset($url_index[$keyword->link]) && !empty($keyword->link_once) && empty($keyword->add_same_link)){
                // remove it from the list
                unset($keywords[$key]);
                continue;
            }

            // if there's a keyword with the same text already slated up
            if(isset($keyword_index[$keyword->keyword])){
                // remove it from the list
                unset($keywords[$key]);
                continue;
            }

            if(Wpil_Keyword::keywordAtLimit($keyword)){
                unset($keywords[$key]);
                continue;
            }

            // if the autolink is pointing to the target post
            $internal = Wpil_Link::isInternal($keyword->link);
            $link_post = ($internal || Wpil_Link::isAffiliateLink($keyword->link)) ? Wpil_Post::getPostByLink($keyword->link): null;
            
            if(apply_filters('wpil_force_include_autolink_post', false, $wpil_post, $keyword, $link_post)){
                $url_index[$keyword->link] = true;
                $keyword_index[$keyword->keyword] = true;
                continue;
            }
            
            if (!empty($link_post) && !empty($link_post->type) && $link_post->type == $wpil_post->type && $link_post->id == $wpil_post->id) {
                unset($keywords[$key]);
                continue;
            }

            // if the keyword is set to only link when the target is published and it hasn't been yet 
            if( (!empty($link_post) && !empty($keyword->restrict_to_live) && $link_post->type === 'post' && $link_post->getStatus() !== 'publish') ||
                (empty($link_post) && !empty($keyword->restrict_to_live) && $internal)
            ){
                // remove the keyword from the list
                unset($keywords[$key]);
                continue;
            }

            if( (!empty($post_language) &&      // if the post has a set language
                !empty($link_post) &&           // and the autolink is pointing to an internal post
                !empty($keyword->same_lang) &&  // and its restricted to a specific language
                $post_language !== self::getPostLanguageCode($link_post))   // and that language doesn't match the target post's
                ||  // ORRRR!!!
                (!empty($keyword->same_lang) && empty($post_language))      // the keyword is restricted by language, but we can't tell what the post's language is
            ){
                // remove it from the list
                unset($keywords[$key]);
                continue;
            }

            $url_index[$keyword->link] = true;
            $keyword_index[$keyword->keyword] = true;
        }

        // remove any existing possible links
        Wpil_Keyword::deletePossibleLinksByPost($wpil_post);

        if (!empty($keywords)) {
            // compile the keyword texts so we can ignore them when splitting the phrases
            $ignore_texts = array_map(function($keyword){ return $keyword->keyword; }, $keywords);
            // create a list for the insertable links
            $possible_links = array();

            $phrases = Wpil_Suggestion::getPhrases($content, true, array(), true, $ignore_texts);
            foreach ($keywords as $keyword) {
                $possible_links = array_merge($possible_links, Wpil_Keyword::makeLinks($phrases, $keyword, $wpil_post, true));
            }

            // if we have links
            if(!empty($possible_links)){
                // get any links that are slated to be inserted
                $existing_links = Wpil_Toolbox::get_encoded_post_meta($wpil_post->id, 'wpil_links', true);

                $known_sentences = array();
                if(!empty($existing_links)){
                    foreach($existing_links as $link){
                        $known_sentences[] = $link['sentence'];
                    }
                }else{
                    // if there are no existing links
                    // assume that the user has just done a post update, and set a flag so that we can act accordingly
                    Wpil_Base::track_action('doing_post_update', true);
                }

                // filter out duplicate sentence matches so we don't have a big stack of links for the same string
                $link_list = array();
                foreach($possible_links as $link){
                    if(!isset($link_list[$link['sentence']]) && !in_array($link['sentence'], $known_sentences, true)){
                        $link_list[$link['sentence']] = $link;
                    }
                }

                $possible_links = array_values($link_list);
                $possible_links = (!empty($existing_links) && is_array($existing_links)) ? array_merge($possible_links, $existing_links): $possible_links;

                // set a flag
                Wpil_Base::track_action('did_passive_autolink_queue', true);
                Wpil_Toolbox::update_encoded_post_meta($wpil_post->id, 'wpil_links', $possible_links);
            }
        }

        // return the unmodified data
        return $data;
    }

    /**
     * Saves autolinks to posts when they're saved or switched from draft to published.
     * Also fires the URLChanger once the autolinking is complete.
     * In the future, we may want to move it to it's own section to differentiate it from the autolinking.
     * ATM, it's here as a cleanup for "save_post" and to update any old autolinking rules that the user has on his site.
     * @param int $post_id The id of the post that we're saving the links to
     **/
    public static function saveAutolinksToPost($post_id){
        // don't save links for revisions
        if(wp_is_post_revision($post_id)){
            return;
        }

        // make sure the post isn't an auto-draft
        $post = get_post($post_id);
        if(!empty($post) && 'auto-draft' === $post->post_status){
            return;
        }

        // make sure this is for a post type that we track
        if(!in_array($post->post_type, Wpil_Settings::getPostTypes())){
            return;
        }

        // make sure we're in the correct priority
        if(998 !== Wpil_Toolbox::get_current_action_priority()){
            return;
        }

        // if the post processing flag isn't active, or hasn't been activated in the past 5 mins
        if (get_option('wpil_post_procession', 0) < (time() - 300)) {
            // create a new post instance to clear any previously edited data
            $post = new Wpil_Model_Post($post_id);
            // note that we're doing a post update
            Wpil_Base::track_action('doing_post_update', true);

            if( !Wpil_Settings::disable_autolink_on_post_save() && (
                    Wpil_Base::action_happened('doing_rest_api_post_create') || // run the autolink process if the user is creating a post via the REST API
                    Wpil_Toolbox::is_doing_external_rest_api_action() || // or thjis is an externally instagated REST request
                    !empty(Wpil_Post::get_active_editors()) || // or if there are any active page builders
                    class_exists('ACF') && empty(get_option('wpil_disable_acf', false)) || // or if ACF is active
                    (Wpil_Keyword::deleteGhostLinks($post) > 0) // or if we've removed links from the record
                )
            ){
                // and run the insert process
                Wpil_Keyword::addKeywordsToPost($post);
                // note that we've inserted links into the post
                Wpil_Base::track_action('did_active_autolink_insert', true);
            }

            Wpil_URLChanger::replacePostURLs($post);
        }
    }

    /**
     * Saves Autolinks to posts that have transitioned from one post status to another.
     * Currently, only intended for going from 'future' to 'publish'
     * @param object $post WP Post object
     **/
    public static function saveAutolinksToTransitionedPosts($post){
        // don't save links for revisions
        if(wp_is_post_revision($post->ID)){
            return;
        }

        // make sure the post isn't an auto-draft
        if(!empty($post) && 'auto-draft' === $post->post_status){
            return;
        }

        // make sure this is for a post type that we track
        if(!in_array($post->post_type, Wpil_Settings::getPostTypes())){
            return;
        }

        // make sure we're in the correct priority
        if(998 !== Wpil_Toolbox::get_current_action_priority()){
            return;
        }

        // make sure the user hasn't disabled autolinking on post update
        if(Wpil_Settings::disable_autolink_on_post_save()){
            return;
        }

        // set up our Link Whisper post object
        $wpil_post = new Wpil_Model_Post($post->ID);
        // and attempt keywording
        Wpil_Keyword::addKeywordsToPost($wpil_post);
    }

    /**
     * Updates the link stats when a post is saved.
     * Updates both the links report table and the link meta that belongs to the post.
     *
     * @param $post_id
     */
    public static function updateStatMark($post_id, $direct_call = false)
    {
        // don't save links for revisions
        if(wp_is_post_revision($post_id)){
            return;
        }

        // make sure the post isn't an auto-draft
        $post = get_post($post_id);
        if(!empty($post) && 'auto-draft' === $post->post_status){
            return;
        }

        // make sure we're checking the link stats at the end of the processing or that it's been called directly
        if(99999 !== Wpil_Toolbox::get_current_action_priority() && !$direct_call){
            return;
        }

        // if this is a reusable block
        if($post->post_type === 'wp_block'){
            // process it's links to see if we need to update posts that it links to
            Wpil_Report::update_reusable_block_links($post); // reusable blocks update separately of the main post, so we're able to check at this point in the process!
        }

        // make sure this is for a post type that we track
        if(!in_array($post->post_type, Wpil_Settings::getPostTypes())){
            return;
        }

        // clear the meta flag
        update_post_meta($post_id, 'wpil_sync_report3', 0);

        // check if the post is being ignored from Link Whisper
        $completely_ignored = Wpil_Settings::get_completely_ignored_pages();
        if(!empty($completely_ignored) && in_array('post_' . $post_id, $completely_ignored, true)){
            // exit now if it is
            return;
        }

        if (get_option('wpil_option_update_reporting_data_on_save', false)) {
            Wpil_Report::fillMeta();
            if(WPIL_STATUS_LINK_TABLE_EXISTS){
                Wpil_Report::remove_post_from_link_table(new Wpil_Model_Post($post_id));
                Wpil_Report::fillWpilLinkTable();
            }
            Wpil_Report::refreshAllStat();
        }else{
            if(WPIL_STATUS_LINK_TABLE_EXISTS){
                $post = new Wpil_Model_Post($post_id);
                // if the current post has the Thrive builder active, load the Thrive content
                $thrive_active = get_post_meta($post->id, 'tcb_editor_enabled', true);
                if(!empty($thrive_active)){
                    $thrive_content = Wpil_Editor_Thrive::getThriveContent($post->id);
                    if($thrive_content){
                        $post->setContent($thrive_content);
                    }
                }
                if(Wpil_Report::stored_link_content_changed($post)){
                    // get the fresh post content for the benefit of the descendent methods
                    $post->getFreshContent();
                    // find any inbound internal link references that are no longer valid
                    $removed_links = Wpil_Report::find_removed_report_inbound_links($post);
                    // update the links stored in the link table
                    Wpil_Report::update_post_in_link_table($post);
                    // if the user is not just using the link table
                    if(!Wpil_Settings::use_link_table_for_data()){
                        // update the meta data for the post
                        Wpil_Report::statUpdate($post, true);
                        // update the link counts for the posts that this one links to
                        Wpil_Report::updateReportInternallyLinkedPosts($post, $removed_links);
                    }
                    // remove any broken links that are no longer in the post
                    Wpil_Error::update_broken_link_post_listing($post);
                }

                // if the links haven't changed, reset the processing flag
                update_post_meta($post_id, 'wpil_sync_report3', 1);
            }
        }
    }

    /**
     * Delete all post meta on post delete
     *
     * @param $post_id
     */
    public static function deleteReferences($post_id)
    {
        // if this is a post revision
        if(wp_is_post_revision($post_id)){
            // don't delete the references since that will pull the data for the parent post!
            return;
        }

        $post = new Wpil_Model_Post($post_id);

        // if we're not only using the link table for data
        if(!Wpil_Settings::use_link_table_for_data()){

            // get the inbound links from the post meta
            $inbound = $post->getInboundInternalLinks();

            // if there are links
            if(!empty($inbound)){
                // remove each of the outbound links from the posts linking to this one
                foreach($inbound as $link){
                    if(!isset($link->post) || empty($link->post)){
                        continue;
                    }

                    $stored_link = array();
                    try {
                        $stored_links = $link->post->getOutboundInternalLinks();
                    } catch (Throwable $t) {
                    } catch (Exception $e) {
                    }

                    // if the current post does have links
                    if(!empty($stored_links)){
                        // count how many we're starting with
                        $link_count = count($stored_links);
                        // and go over all the links available
                        foreach($stored_links as $key => $stored_link){
                            if(!isset($stored_link->post) || empty($stored_link->post)){
                                continue;
                            }

                            // if the other post has a link pointing to this one
                            if(trailingslashit($stored_link->url) === trailingslashit($link->url)){
                                // remove the link from the stored data
                                unset($stored_links[$key]);
                            }
                        }

                        // re-count the links so we can tell if we removed any
                        $new_count = count($stored_links);

                        // if we have removed links
                        if($link_count > $new_count){
                            // rekey the link array just in case something is index sensitive
                            $stored_links = array_values($stored_links);
                            // update the stored data and the stored link count
                            if($link->post->type === 'post'){
                                $stored_links = Wpil_Toolbox::update_encoded_post_meta($link->post->id, 'wpil_links_outbound_internal_count_data', $stored_links);
                                $stored_links = Wpil_Toolbox::update_encoded_post_meta($link->post->id, 'wpil_links_outbound_internal_count', $new_count);
                            }else{
                                $stored_links = Wpil_Toolbox::update_encoded_post_meta($link->post->id, 'wpil_links_outbound_internal_count_data', $stored_links);
                                $stored_links = Wpil_Toolbox::update_encoded_term_meta($link->post->id, 'wpil_links_outbound_internal_count', $new_count);
                            }
                        }
                    }
                }
            }
        }

        // remove the meta-based link data for this posts
        foreach (array_merge(Wpil_Report::$meta_keys, ['wpil_sync_report3', 'wpil_sync_report2_time']) as $key) {
            delete_post_meta($post_id, $key);
        }
        if(WPIL_STATUS_LINK_TABLE_EXISTS){
            // remove the current post from the links table and the links that point to it
            Wpil_Report::remove_post_from_link_table(new Wpil_Model_Post($post_id), true);
        }
    }

    /**
     * Deletes the Inbound Internal links that are pointing to the post that is being deleted.
     *
     * @param $post_id
     */
    public static function delete_inbound_links_on_post_delete($post_id)
    {
        // if this is a post revision
        if(wp_is_post_revision($post_id)){
            // don't delete the references since that will pull the data for the parent post!
            return;
        }

        // if the user doesn't want to run the inbound link delete
        if(empty(Wpil_Settings::delete_inbound_internal_on_post_delete())){
            // exit without deleting the links
            return;
        }

        // create the post to work with
        $post = new Wpil_Model_Post($post_id);

        // get a list of the known Inbound Internal links
        $inbound = Wpil_Report::getReportInternalInboundLinks($post);
        // if there are links
        if(!empty($inbound)){
            // go over them and remove them from any posts that they are inserted in
            foreach($inbound as $link){
                // if there's no post available
                if(!isset($link->post) || empty($link->post)){
                    // skip to the next link
                    continue;
                }

                // assemble the data needed to delete the link
                $parameters = array(
                    'post_id' => $link->post->id, 
                    'post_type' => $link->post->type, 
                    'url' => $link->url, 
                    'anchor' => $link->anchor
                );

                // and delete it from the post and any page builders that it may be inserted in
                $r = Wpil_Link::delete($parameters, true, true); // skip the stat updating since that will come later
            }
        }
    }

    /**
     * Get linked post Ids for current post
     *
     * @param $post
     * @param bool $return_ids Do we jsut return the linked post ids or the whole link object
     * @return array
     */
    public static function getLinkedPostIDs($post, $return_ids = true, $ignore_self = true)
    {
        $linked_post_ids = array();
        $prevent_twoway_linking = get_option('wpil_prevent_two_way_linking', false);

        // get the inbound post links
        if(WPIL_STATUS_LINK_TABLE_EXISTS){
            $links = Wpil_Report::getCachedReportInternalInboundLinks($post);
        }else{
            $links = Wpil_Report::getInternalInboundLinks($post);
        }

        // if we're to prevent twoway linking
        if($prevent_twoway_linking){
            // get the outbound links
            $outbound_links = Wpil_Report::getOutboundLinks($post);
            if(!empty($outbound_links['internal'])){
                $links = array_merge($links, $outbound_links['internal']);
            }
        }

        // if we're supposed to return just the ids
        if($return_ids){
            if($ignore_self){
                // process out the ids
                $linked_post_ids[] = $post->id;
            }

            foreach ($links as $link) {
                if (!empty($link->post->id)) {
                    $linked_post_ids[] = $link->post->id;
                }
            }
        }else{
            if($ignore_self){
                $url = $post->getLinks()->view;
                $host = parse_url($url, PHP_URL_HOST);


                $linked_post_ids[] = new Wpil_Model_Link([
                    'url' => $url,
                    'host' => str_replace('www.', '', $host),
                    'internal' => Wpil_Link::isInternal($url),
                    'post' => $post,
                    'anchor' => '',
                ]);
            }

            $linked_post_ids = array_merge($linked_post_ids, $links);
        }

        return $linked_post_ids;
    }

    /**
     * Get all Advanced Custom Fields names
     *
     * @return array
     */
    public static function getAdvancedCustomFieldsList($post_id)
    {
        global $wpdb;

        $fields = [];

        if(!class_exists('ACF') || get_option('wpil_disable_acf', false)){
            return $fields;
        }

        // get any ACF fields the user has ignored
        $ignored_fields = Wpil_Settings::getIgnoredACFFields();

        // get a list of ACF field rules to use for regular expressions
        $ignored_fields_wildcards = [];

        // get the content types that we'll be searching for
        $content_types = array('wysiwyg', 'textarea');
        if(!Wpil_Settings::get_ignore_acf_text_fields()){
            $content_types[] = 'text';
        }
        $content_types = " AND (`post_content` LIKE '%" . implode("%' OR `post_content` LIKE '%", $content_types) . "%')";

        if ( !empty($ignored_fields) ) {
            foreach($ignored_fields as $key => $rule) {
                if ( strpos($rule, '*') !== false ) {
                    $ignored_fields_wildcards[] = str_replace('*', '.*', $rule);
                    unset($ignored_fields[$key]);
                }
            }
            
            if ( !empty($ignored_fields_wildcards) ) {
                $ignored_fields_wildcards = implode('|', $ignored_fields_wildcards);
            }
        }

        // get any ACF fields that the user has chosen to focus on
        $acf_fields = Wpil_Query::querySpecifiedAcfFields();

        $fields_query = $wpdb->get_results("SELECT SUBSTR(meta_key, 2) as `name` FROM {$wpdb->postmeta} WHERE post_id = $post_id AND meta_value IN (SELECT DISTINCT post_name FROM {$wpdb->posts} WHERE post_name LIKE 'field_%' {$acf_fields} {$content_types}) AND SUBSTR(meta_key, 2) != ''");
       // print_r("SELECT SUBSTR(meta_key, 2) as `name` FROM {$wpdb->postmeta} WHERE post_id = $post_id AND meta_value IN (SELECT DISTINCT post_name FROM {$wpdb->posts} WHERE post_name LIKE 'field_%' {$acf_fields} {$content_types}) AND SUBSTR(meta_key, 2) != ''");
        foreach ($fields_query as $field) {
            $name = trim($field->name);
            if(in_array($name, $ignored_fields, true)){
                continue;
            }

            if ( !empty($ignored_fields_wildcards) && preg_match('/' . $ignored_fields_wildcards . '/', $name) ) {
                continue;
            }

            if ($name) {
                $fields[] = $field->name;
            }
        }

        // if there are any fields created with PHP/JSON
        $local_field_groups = (function_exists('acf_get_local_store')) ? acf_get_local_store('groups') : false;
        if(!empty($local_field_groups) && isset($local_field_groups->data)){
            $search_fields = array();
            $secondary_lookup_fields = array();
            foreach($local_field_groups->data as $group){
                // go to some pains to ignore options pages
                if( isset($group['location']) &&
                    isset($group['location'][0]) &&
                    isset($group['location'][0][0]) &&
                    isset($group['location'][0][0]['param']) &&
                    $group['location'][0][0]['param'] == 'options_page' &&
                    $group['location'][0][0]['operator'] == '==')
                {
                    continue;
                }

                if(isset($group['name'])){
                    $search_fields[$group['name']] = true;
                }elseif(isset($group['key']) && function_exists('acf_get_fields')){
                    $secondary_fields = acf_get_fields($group['key']);
                    if(!empty($secondary_fields)){
                        foreach($secondary_fields as $field){
                            if( isset($field['type']) && 
                                ($field['type'] === 'textarea' || $field['type'] === 'wysiwyg') &&
                                isset($field['key'])
                            ){
                                $secondary_lookup_fields[$field['key']] = true;
                            }elseif($field['type'] === 'text' && // if the field is a text AND
                                    !Wpil_Settings::get_ignore_acf_text_fields() && // we're not ignoring text fields AND
                                (
                                    isset($field['name']) && false !== strpos(strtolower($field['name']), 'url') // the text field contains "url" 
                                )
                            ){
                                // We're being extra cautious of text fields since they tend to be used for utility and title purposes.
                                // If there gets to be a lot of cases where we're missing oportunities because the search is limited, we'll see about widening the scope.
                                $secondary_lookup_fields[$field['key']] = true;
                            }elseif(isset($field['type']) && $field['type'] === 'flexible_content' && isset($field['layouts']) && !empty($field['layouts'])){
                                foreach($field['layouts'] as $layout){
                                    if(isset($layout['sub_fields']) && !empty($layout['sub_fields'])){
                                        $secondary_lookup_fields = array_merge($secondary_lookup_fields, self::getRecursiveACFSubFields($layout));
                                    }
                                }
                            }
                        }
                    }
                }
            }

            if(!empty($search_fields)){
                $search_fields = array_keys($search_fields);
                $search_fields = '`meta_key` LIKE \'' . implode('_%\' OR `meta_key` LIKE \'', $search_fields) . '_%\'';

                $fields_query = $wpdb->get_results("SELECT meta_key as 'name' FROM {$wpdb->postmeta} WHERE `post_id` = $post_id AND ({$search_fields})  AND `meta_value` != ''");

                if(!empty($fields_query)){
                    foreach ($fields_query as $field) {
                        $name = trim($field->name);
                        if(in_array($name, $ignored_fields, true)){
                            continue;
                        }
            
                        if ($name) {
                            $fields[] = $field->name;
                        }
                    }
                }
            }

            if(!empty($secondary_lookup_fields)){
                $secondary_lookup_fields = array_keys($secondary_lookup_fields);
                $search_fields = " AND `meta_value` IN ('" . implode("', '", $secondary_lookup_fields) . "')";
                $fields_query = $wpdb->get_col("SELECT meta_key FROM {$wpdb->postmeta} WHERE `post_id` = $post_id {$search_fields}");

                if(!empty($fields_query)){
                    foreach($fields_query as $field){
                        if(0 === strpos($field, '_')){
                            $name = trim(substr($field, 1));
                            if(in_array($name, $ignored_fields, true)){
                                continue;
                            }

                            $fields[] = $name;
                        }
                    }
                }
            }

            // remove any duplicate fields
            $fields = array_flip(array_flip($fields));
        }

        // if for some reason we couldn't find any fields
        if(empty($fields)){
            // try pulling and formatting fields from ACF using its own functions to make the field names
            $fields = self::flatten_get_fields_allowed_types($post_id);
        }

        return $fields;
    }

    /**
     * Recursively goes through the potential multitude of ACF subfields and pulls out all of the
     * textarea & WYSIWYG fields so we can search the database for them
     **/
    public static function getRecursiveACFSubFields($fields){
        $found_fields = array();
        if(isset($fields['sub_fields']) && !empty($fields['sub_fields'])){
            foreach($fields['sub_fields'] as $sub){
                // only get the fields that can reasonably be assumed to be linkable
                if( isset($sub['type']) &&
                    ($sub['type'] === 'textarea' || $sub['type'] === 'wysiwyg') &&
                    isset($sub['key'])
                ){
                    $found_fields[$sub['key']] = true;
                }elseif($sub['type'] === 'text' && // if the subfield is a text AND
                        !Wpil_Settings::get_ignore_acf_text_fields() && // we're not ignoring text fields AND
                    (
                        isset($sub['name']) && false !== strpos(strtolower($sub['name']), 'url') // the text field contains "url" 
                    )
                ){
                    $found_fields[$sub['key']] = true;
                }elseif(isset($sub['sub_fields']) && !empty($sub['sub_fields'])){
                    $found_fields = array_merge($found_fields, self::getRecursiveACFSubFields($sub));
                }
            }
        }

        return $found_fields;
    }

    /**
     * Flatten get_fields() and try to pull the fields we need!
     */
    public static function flatten_get_fields_allowed_types($post_id, $skip_empty = true){
        if(!function_exists('get_fields') || !class_exists('ACF')){
            return [];
        }

        $values = get_fields($post_id);
        if(empty($values) || !is_array($values)){
            return [];
        }

        // LW ignore rules
        $ignored_exact = Wpil_Settings::getIgnoredACFFields();
        $ignored_wildcards = [];
        if(!empty($ignored_exact)){
            foreach($ignored_exact as $i => $rule){
                if(strpos($rule, '*') !== false){
                    $ignored_wildcards[] = str_replace('*', '.*', $rule);
                    unset($ignored_exact[$i]);
                }
            }
        }
        $ignored_regex = !empty($ignored_wildcards) ? '/' . implode('|', $ignored_wildcards) . '/' : '';

        $schema = self::acf_get_allowed_schema_fields($post_id);
        $allowed_leaf = $schema['allowed_leaf'];

        $is_ignored = function($str) use ($ignored_exact, $ignored_regex){
            if(in_array($str, $ignored_exact, true)){
                return true;
            }
            if(!empty($ignored_regex) && preg_match($ignored_regex, $str)){
                return true;
            }
            return false;
        };

        $out = [];

        /**
         * @param mixed       $value   Current value
         * @param string      $prefix  Flattened LW-style key path
         * @param string|null $leafKey The field key at this node
         */
        $walk = function($value, $prefix, $leafKey = null) use (&$walk, &$out, $allowed_leaf, $skip_empty, $is_ignored){
            if($leafKey !== null){
                if($is_ignored($leafKey)){
                    return;
                }
            }

            if(!is_array($value)){
                if($leafKey === null || empty($allowed_leaf[$leafKey])){
                    return;
                }

                if(!is_string($value)){
                    return;
                }

                if($skip_empty && trim($value) === ''){
                    return;
                }

                $out[] = $prefix;
                return;
            }

            // Otherwise recurse through arrays (repeaters, flex, groups, etc.)
            foreach($value as $k => $v){
                $nextPrefix = ($prefix === '') ? (string)$k : ($prefix . '_' . $k);
                $nextLeafKey = is_string($k) ? $k : null;
                $walk($v, $nextPrefix, $nextLeafKey);
            }
        };

        foreach($values as $top_key => $top_value){
            if($is_ignored($top_key)){
                continue;
            }
            $walk($top_value, (string)$top_key, (string)$top_key);
        }

        return array_values(array_unique($out));
    }

    /**
     * Build a set of "allowed leaf field names" based on ACF schema:
     * - textarea, wysiwyg, and (optionally) text
     * - recursively through repeater/group/flexible_content/clone
     *
     * Returns: array{allowed_leaf: array<string,true>, allowed_roots: array<string,true>}
     */
    public static function acf_get_allowed_schema_fields($post_id){
        $allowed_leaf = [];
        $allowed_roots = [];

        if(!function_exists('get_field_objects') || !class_exists('ACF')){
            return ['allowed_leaf' => [], 'allowed_roots' => []];
        }

        $include_text = !Wpil_Settings::get_ignore_acf_text_fields();

        $objs = get_field_objects($post_id);
        if(empty($objs) || !is_array($objs)){
            return ['allowed_leaf' => [], 'allowed_roots' => []];
        }

        $walk_field = function($field) use (&$walk_field, &$allowed_leaf, $include_text){
            if(empty($field) || !is_array($field) || empty($field['name']) || empty($field['type'])){
                return;
            }

            $type = $field['type'];
            $name = $field['name'];

            // leaf types we care about
            if($type === 'textarea' || $type === 'wysiwyg' || ($include_text && $type === 'text')){
                $allowed_leaf[$name] = true;
                return;
            }

            // containers
            if(($type === 'group' || $type === 'repeater') && !empty($field['sub_fields'])){
                foreach($field['sub_fields'] as $sub){
                    $walk_field($sub);
                }
                return;
            }

            if($type === 'flexible_content' && !empty($field['layouts'])){
                foreach($field['layouts'] as $layout){
                    if(!empty($layout['sub_fields'])){
                        foreach($layout['sub_fields'] as $sub){
                            $walk_field($sub);
                        }
                    }
                }
                return;
            }

            if($type === 'clone'){
                // clone can inline sub_fields or reference "clone" keys depending on config
                if(!empty($field['sub_fields'])){
                    foreach($field['sub_fields'] as $sub){
                        $walk_field($sub);
                    }
                    return;
                }
            }
        };

        foreach($objs as $root){
            if(empty($root['name'])) continue;
            $allowed_roots[$root['name']] = true;

            $walk_field($root);
        }

        return ['allowed_leaf' => $allowed_leaf, 'allowed_roots' => $allowed_roots];
    }

    /**
     * Gets an array of all custom fields on the site.
     * @return array
     **/
    public static function getAllCustomFields()
    {
        global $wpdb;

        if(!class_exists('ACF') || get_option('wpil_disable_acf', false)){
            return array();
        }

        if (self::$advanced_custom_fields_list === null) {
            $ignored_fields = Wpil_Settings::getIgnoredACFFields();
            $only_search_fields = Wpil_Query::querySpecifiedAcfFields('pm');
            $content_types = array('wysiwyg', 'textarea');

            if(!Wpil_Settings::get_ignore_acf_text_fields()){
                $content_types[] = 'text';
            }

            $content_types = implode('|', $content_types);

            $fields = array();

            // try getting the main set of ACF fields
            //$post_names = $wpdb->get_col("SELECT DISTINCT pm.meta_key as `name` FROM {$wpdb->postmeta} pm INNER JOIN {$wpdb->posts} p ON pm.meta_value = p.post_name WHERE p.post_type = 'acf-field' AND p.post_name LIKE 'field_%'");
            $post_data = $wpdb->get_results("SELECT DISTINCT pm.meta_key as `name`, p.post_content as `content` FROM {$wpdb->postmeta} pm INNER JOIN {$wpdb->posts} p ON pm.meta_value = p.post_name WHERE p.post_type = 'acf-field' {$only_search_fields}");

            // if we found some
            if (!empty($post_data)) {
                // clean up their names and add them to the field list
                foreach ($post_data as $dat) {
                    if(!preg_match('/' . $content_types . '/', $dat->content)){
                        continue;
                    }

                    $name = trim(substr($dat->name, 1));
                    if (!empty($name)) {
                        $fields[] = $name;
                    }
                }
            }

            // if there are any fields created with PHP/JSON
            $local_field_groups = (function_exists('acf_get_local_store')) ? acf_get_local_store('groups') : false;
            if(!empty($local_field_groups) && isset($local_field_groups->data)){
                $search_fields = array();
                $secondary_lookup_fields = array();
                foreach($local_field_groups->data as $group){
                    // go to some pains to ignore options pages
                    if( isset($group['location']) &&
                        isset($group['location'][0]) &&
                        isset($group['location'][0][0]) &&
                        isset($group['location'][0][0]['param']) &&
                        $group['location'][0][0]['param'] == 'options_page' &&
                        $group['location'][0][0]['operator'] == '==')
                    {
                        continue;
                    }

                    if(isset($group['name'])){
                        $search_fields[] = $group['name'];
                    }elseif(isset($group['key']) && function_exists('acf_get_fields')){
                        $secondary_fields = acf_get_fields($group['key']);
                        if(!empty($secondary_fields)){
                            foreach($secondary_fields as $field){
                                if( isset($field['type']) && 
                                    ($field['type'] === 'textarea' || $field['type'] === 'wysiwyg') &&
                                    isset($field['key'])
                                ){
                                    $secondary_lookup_fields[$field['key']] = true;
                                }elseif(isset($field['type']) && $field['type'] === 'flexible_content' && isset($field['layouts']) && !empty($field['layouts'])){
                                    foreach($field['layouts'] as $layout){
                                        if(isset($layout['sub_fields']) && !empty($layout['sub_fields'])){
                                            $secondary_lookup_fields = array_merge($secondary_lookup_fields, self::getRecursiveACFSubFields($layout));
                                        }
                                    }
                                }
                            }
                        }
                    }
                }

                if(!empty($search_fields)){
                    $search_fields = '`meta_key` LIKE \'' . implode('_%\' OR `meta_key` LIKE \'', $search_fields) . '_%\'';

                    $fields_query = $wpdb->get_results("SELECT DISTINCT meta_key as `name` FROM {$wpdb->postmeta} WHERE ({$search_fields})");

                    if(!empty($fields_query)){
                        foreach ($fields_query as $field) {
                            $name = trim($field->name);
                            if ($name) {
                                $fields[] = $name;
                            }
                        }
                    }
                }

                if(!empty($secondary_lookup_fields)){
                    $secondary_lookup_fields = array_keys($secondary_lookup_fields);
                    $secondary_fields = "`meta_value` IN ('" . implode("', '", $secondary_lookup_fields) . "')";
                    $fields_query = $wpdb->get_col("SELECT DISTINCT meta_key FROM {$wpdb->postmeta} WHERE {$secondary_fields}");
                    if(!empty($fields_query)){

                        foreach($fields_query as $field){
                            if(0 === strpos($field, '_')){
                                $name = trim(substr($field, 1));
                                $fields[] = $name;
                            }
                        }
                    }
                }

                // if we've found some fields
                if(!empty($fields)){
                    // remove any duplicate fields
                    $fields = array_flip(array_flip($fields));

                    // get a list of ACF field rules to use for regular expressions
                    $ignored_fields_wildcards = [];

                    // remove any ignored fields that are defined
                    if(!empty($ignored_fields)){
                        foreach($ignored_fields as $key => $rule) {
                            if ( strpos($rule, '*') !== false ) {
                                $ignored_fields_wildcards[] = str_replace('*', '.*', $rule);
                                unset($ignored_fields[$key]);
                            }
                        }
                        
                        if ( !empty($ignored_fields_wildcards) ) {
                            $ignored_fields_wildcards = implode('|', $ignored_fields_wildcards);
                        }

                        foreach($fields as $ind => $field){
                            if(!empty($ignored_fields) && in_array($field, $ignored_fields, true)){
                                unset($fields[$ind]);
                            }

                            if ( !empty($ignored_fields_wildcards) && preg_match('/' . $ignored_fields_wildcards . '/', $field) ) {
                                unset($fields[$ind]);
                            }
                        }
                    }

                    // re-key the array in case something sensitive is listening
                    $fields = array_values($fields);
                }
            }

            self::$advanced_custom_fields_list = $fields;
        }

        return self::$advanced_custom_fields_list;
    }

    /**
     * Adds link to a specific metafield, updates the supplied content with the link
     **/
    public static function addLinkToField(&$content, $link = array(), $offset = 0){
        $force_insert = (isset($link['keyword_data']) && !empty($link['keyword_data']->force_insert)) ? true: false;
        $changed_sentence = self::getSentenceWithAnchor($link);
        return self::insertLink($content, $link['sentence'], $changed_sentence, $force_insert, $offset);
    }

    /**
     * Add link to the content in advanced custom fields
     *
     * @param $link
     * @param $post
     */
    public static function addLinkToAdvancedCustomFields($post_id)
    {
        // don't save the data if this is the result of using wp_update_post // there's no form submission, so $_POST will be empty
        if(empty($_POST)){
//            return;
        }

        $meta = Wpil_Toolbox::get_encoded_post_meta($post_id, 'wpil_links', true);

        // if there are links to insert
        if (!empty($meta)) {
            // see if the post has ACF fields
            $fields = self::getAdvancedCustomFieldsList($post_id);
            // if it does
            if (!empty($fields)) {
                // get the post's object
                $post = new Wpil_Model_Post($post_id);
                // set a flag for skipping the first sentences
                $found_first = false;

                // go over each field
                foreach ($fields as $field) {
                    // see if the field has string content
                    $content = get_post_meta($post_id, $field, true);
                    if(!empty($content) && is_string($content)){
                        // check to make sure the starting content isn't ACF content
                        $offset = 0;
                        // if we haven't found the first section
                        if(!$found_first){
                            // check if the ACF content is right at the beginning of the post's content
                            $position = Wpil_Word::mb_strpos($post->getContent(), $content);
                            // if it is
                            if($position !== false && $position < 5){
                                // parse the phrases out of the content
                                $phrases = Wpil_Suggestion::getPhrases($content);

                                // if doing so eliminated all of the content
                                if(empty($phrases)){
                                    // say that we've found the first paragraph and move on to the rest of the content
                                    $found_first = true;
                                    continue;
                                }

                                // if we do have phrases, find the position of the first one
                                $pos = Wpil_Word::mb_strpos($content, $phrases[0]->src);

                                // if the string was found in the content
                                if(false !== $pos){
                                    // set the offset for the start of the first clear sentence
                                    $offset = $pos;
                                }
                            }
                        }

                        // and see if we can put a link in it
                        foreach($meta as $key => $link){
                            if (strpos($content, $link['sentence']) !== false) {
                                // if the link was inserted
                                if(self::addLinkToField($content, $link, $offset)){
                                    // remove the link from the metadata
                                    unset($meta[$key]);
                                    // and update the field data
                                    update_post_meta($post_id, $field, $content);
                                }
                            }
                        }
                    }
                }

                // update the link meta with the remaining links
                $meta_values = (is_array($meta)) ? array_values($meta): $meta;
                Wpil_Toolbox::update_encoded_post_meta($post_id, 'wpil_links', $meta_values);
            }
        }
    }

    /**
     * Add link to the content in advanced custom fields
     *
     * @param $post Wpil_Model_Post
     * @param $old_link
     * @param $new_link
     * @param $anchor
     */
    public static function updateAdvancedCustomFieldLink($post, $old_link, $new_link, $anchor, $new_anchor = '', $sentence = '', $new_sentence = '')
    {
        if(empty($post) || $post->type !== 'post' || empty($old_link) || empty($new_link)){
            return;
        }

        // see if the post has ACF fields
        $fields = self::getAdvancedCustomFieldsList($post->id);
        // if it does
        if (!empty($fields)) {
            // go over each field
            foreach ($fields as $field) {
                // see if the field has string content
                $content = get_post_meta($post->id, $field, true);
                if(!empty($content) && is_string($content)){
                    // if it does, try updating the link URL
                    Wpil_Link::updateLinkUrl($content, $old_link, $new_link, $anchor, $new_anchor, $sentence, $new_sentence);
                    // if the URL was in the content and it has been changed
                    if(Wpil_Base::action_happened('link_url_updated')){
                        // update the field
                        update_post_meta($post->id, $field, $content);
                        // and exit while returning true for the benefit of anyone that's listening
                        return true;
                    }
                }
            }
        }

        // if we made it down here, the link hasn't been updated
        return false;
    }

    /**
     * Add link to the content in metafields
     *
     * @param $link
     * @param $post
     */
    public static function addLinkToMetaContent($post_id, $direct_call = false)
    {
        // don't save the data if this is the result of using wp_update_post // there's no form submission, so $_POST will be empty
        if(Wpil_Base::has_ancestor_function('update_item', 'WP_REST_Posts_Controller') && empty($_POST)){
            return;
        }

        // don't save on autosaves
        if(defined('DOING_AUTOSAVE') && DOING_AUTOSAVE){
            return;
        }

        // make sure this is for a post type that we track
        $post_type = get_post_type($post_id);
        if(empty($post_type) || !in_array($post_type, Wpil_Settings::getPostTypes())){
            return;
        }

        // make sure we're checking the link stats at the right stage of processing
        if(9999 !== Wpil_Toolbox::get_current_action_priority() && !$direct_call){
            return;
        }

        $meta = Wpil_Toolbox::get_encoded_post_meta($post_id, 'wpil_links', true);

        if (!empty($meta)) {
            $fields = self::getMetaContentFieldList('post');
            if (!empty($fields)) {
                foreach ($fields as $field) {
                    if ($content = get_post_meta($post_id, $field, true)) {
                        if(!is_string($content) || empty($content)){
                            continue;
                        }
                        foreach ($meta as $key => $link) {
                            if (strpos($content, $link['sentence']) !== false) {
                                $force_insert = (isset($link['keyword_data']) && !empty($link['keyword_data']->force_insert)) ? true: false;
                                $changed_sentence = self::getSentenceWithAnchor($link);
                                $inserted = self::insertLink($content, $link['sentence'], $changed_sentence, $force_insert);

                                // if the link has been inserted
                                if($inserted){
                                    // remove it from the link meta
                                    unset($meta[$key]);
                                    // and update the field
                                    update_post_meta($post_id, $field, $content);
                                }
                            }
                        }
                    }
                }

                // update the link meta with the remaining links
                $meta_values = (is_array($meta)) ? array_values($meta): $meta;
                Wpil_Toolbox::update_encoded_post_meta($post_id, 'wpil_links', $meta_values);
            }

            // add links to any ACF fields
            self::addLinkToAdvancedCustomFields($post_id);
            // add links to Oxygen content
            $fake_content = false;
            Wpil_Editor_Oxygen::addLinks($meta, $post_id, $fake_content);
            // add links to Goodlayers content
            Wpil_Editor_Goodlayers::addLinks($meta, $post_id, $fake_content);
            // add links to any YooTheme content
            Wpil_Editor_YooTheme::addLinks($meta, $post_id, $fake_content);

            /**
             * Add the links to any custom data fields the customer may have
             * @param int $post_id
             * @param string $post_type (post|term)
             * @param array $meta
             **/
            do_action('wpil_meta_content_data_add_link', $post_id, 'post', $meta);

            //remove DB record with links
            delete_post_meta($post_id, 'wpil_links');
        }
    }

    /**
     * Gets a list of the possible meta content fields to add links to
     * @param string $type Is the content for a post or a term?
     * @return array $fields An array of the possible fields for the item
     **/
    public static function getMetaContentFieldList($type = 'post'){
        $fields = Wpil_Settings::getCustomFieldsToProcess();

        if(defined('RH_MAIN_THEME_VERSION') && $type === 'term'){
            $fields[] = 'brand_second_description';
        }

        return $fields;
    }

    /**
     * Get all posts with the same language
     *
     * @param $post_id
     * @return array
     */
    public static function getSameLanguagePosts($post_id)
    {
        global $wpdb;
        $ids = [];
        $posts = [];

        // if WPML is active and there's languages saved
        if(Wpil_Settings::wpml_enabled()) {
            $table = $wpdb->get_var("SHOW TABLES LIKE '{$wpdb->prefix}icl_languages'");
            if($table == $wpdb->prefix . 'icl_languages'){
                $post_types = self::getSelectedLanguagePostTypes();
                $language = $wpdb->get_var("SELECT language_code FROM {$wpdb->prefix}icl_translations WHERE element_id = $post_id AND `element_type` IN ({$post_types}) ");
                if (!empty($language)) {
                    $posts = $wpdb->get_results("SELECT element_id as id FROM {$wpdb->prefix}icl_translations WHERE element_id != $post_id AND language_code = '$language' AND `element_type` IN ({$post_types}) ");
                }
            }
        }

        // if Polylang is active
        if(Wpil_Settings::polylang_enabled()){
            $taxonomy_id = $wpdb->get_var("SELECT t.term_taxonomy_id FROM {$wpdb->term_taxonomy} t INNER JOIN {$wpdb->term_relationships} r ON t.term_taxonomy_id = r.term_taxonomy_id WHERE t.taxonomy = 'language' AND r.object_id = " . $post_id);
            if (!empty($taxonomy_id)) {
                $posts = $wpdb->get_results("SELECT object_id as id FROM {$wpdb->term_relationships} WHERE term_taxonomy_id = $taxonomy_id AND object_id != $post_id");
            }
        }

        if (!empty($posts)) {
            foreach ($posts as $post) {
                $ids[] = $post->id;
            }
        }

        return $ids;
    }

    /**
     * Get all posts from languages other than the current post's
     *
     * @param $post_id
     * @return array
     */
    public static function getNonSameLanguagePosts($post_id)
    {
        global $wpdb;
        $ids = [];
        $posts = [];

        // if WPML is active and there's languages saved
        if(Wpil_Settings::wpml_enabled()) {
            $table = $wpdb->get_var("SHOW TABLES LIKE '{$wpdb->prefix}icl_languages'");
            if($table == $wpdb->prefix . 'icl_languages'){
                $post_types = self::getSelectedLanguagePostTypes();
                $language = $wpdb->get_var("SELECT language_code FROM {$wpdb->prefix}icl_translations WHERE element_id = $post_id AND `element_type` IN ({$post_types}) ");
                if (!empty($language)) {
                    $other_languages = $wpdb->get_col("SELECT code FROM {$wpdb->prefix}icl_languages WHERE code != '{$language}' AND active = 1");
                    if(!empty($other_languages)){
                        
                        $other_languages = "('" . implode("', '", $other_languages) . "')";
                        $posts = $wpdb->get_results("SELECT element_id as id FROM {$wpdb->prefix}icl_translations WHERE element_id != $post_id AND language_code IN {$other_languages} AND `element_type` IN ({$post_types}) ");
                    }
                }
            }
        }

        // if Polylang is active
        if(Wpil_Settings::polylang_enabled()){
            $taxonomy_id = $wpdb->get_var("SELECT t.term_taxonomy_id FROM {$wpdb->term_taxonomy} t INNER JOIN {$wpdb->term_relationships} r ON t.term_taxonomy_id = r.term_taxonomy_id WHERE t.taxonomy = 'language' AND r.object_id = " . $post_id);
            if (!empty($taxonomy_id)) {
                $other_languages = $wpdb->get_col("SELECT DISTINCT t.term_taxonomy_id FROM {$wpdb->term_taxonomy} t INNER JOIN {$wpdb->term_relationships} r ON t.term_taxonomy_id = r.term_taxonomy_id WHERE t.taxonomy = 'language' AND t.term_taxonomy_id != $taxonomy_id");
                if(!empty($other_languages)){
                    $other_languages = implode(',', $other_languages);
                    $posts = $wpdb->get_results("SELECT object_id as id FROM {$wpdb->term_relationships} WHERE term_taxonomy_id IN ({$other_languages}) AND object_id != $post_id");
                }
            }
        }

        if (!empty($posts)) {
            foreach ($posts as $post) {
                $ids[] = $post->id;
            }
        }

        return $ids;
    }

    /**
     * Gets the selected post types formatted for WPML
     **/
    public static function getSelectedLanguagePostTypes(){
        $post_types = implode("', 'post_", Wpil_Suggestion::getSuggestionPostTypes());

        if(!empty($post_types)){
            $post_types = "'post_" . $post_types . "'";
        }

        return $post_types;
    }

    /**
     * Get all terms in the same language
     *
     * @param $term_id
     * @return array
     */
    public static function getSameLanguageTerms($term_id)
    {
        global $wpdb;
        $ids = [];

        // if WPML is active and there's languages saved
        if(defined('WPML_PLUGIN_BASENAME')) {
            $table = $wpdb->get_var("SHOW TABLES LIKE '{$wpdb->prefix}icl_languages'");
            if($table == $wpdb->prefix . 'icl_languages'){
                $term_types = self::getSelectedLanguageTermTypes();
                $language = $wpdb->get_var("SELECT language_code FROM {$wpdb->prefix}icl_translations WHERE element_id = $term_id AND `element_type` IN ({$term_types}) ");
                if (!empty($language)) {
                    $ids = $wpdb->get_col("SELECT element_id FROM {$wpdb->prefix}icl_translations WHERE element_id != $term_id AND language_code = '$language' AND `element_type` IN ({$term_types}) ");
                }
            }
        }

        // if Polylang is active
        if(defined('POLYLANG_VERSION')){
            // get the terms that have been translated... Eventually
            $taxonomy_description = $wpdb->get_var("SELECT `description` FROM {$wpdb->term_taxonomy} t INNER JOIN {$wpdb->term_relationships} r ON t.term_taxonomy_id = r.term_taxonomy_id WHERE t.taxonomy = 'term_translations' AND r.object_id = " . $term_id);
            if (!empty($taxonomy_description)) {
                $description_data = maybe_unserialize($taxonomy_description);
                $lang_code = array_search($term_id, $description_data);
                if(!empty($lang_code)){
                    $data = $wpdb->get_results("SELECT * FROM {$wpdb->term_taxonomy} WHERE `taxonomy` = 'term_translations' AND  `description` LIKE '%\"{$lang_code}\"%' AND term_id != $term_id");
                    if(!empty($data)){
                        foreach($data as $term){
                            $dat = maybe_unserialize($term->description);
                            if(!empty($dat) && isset($dat[$lang_code])){
                                $ids[] = $dat[$lang_code];
                            }
                        }
                    }
                }
            }
        }

        if (!empty($ids)) {
            $ids[] = array_flip(array_flip($ids));
        }

        return $ids;
    }

    /**
     * Gets the selected post types formatted for WPML
     **/
    public static function getSelectedLanguageTermTypes(){
        $term_types = implode("', 'tax_", Wpil_Settings::getTermTypes());

        if(!empty($term_types)){
            $term_types = "'tax_" . $term_types . "'";
        }

        return $term_types;
    }

    /**
     * Gets the language code for the supplied WPIL post.
     * Uses the functionality of the supported translation plugins to take advantage of their caching systems.
     * 
     * @param object $post A Wpil_Model_Post object to find the language for.
     * @return string|bool Returns the language code if it's set, and FALSE if no translation is detectable
     **/
    public static function getPostLanguageCode($post){
        global $sitepress;
        
        $code = false;

        if(empty($post)){
            return $code;
        }

        if(defined('WPML_PLUGIN_BASENAME') && !empty($sitepress) && method_exists($sitepress, 'get_language_for_element')){
            $type = ($post->type === 'term') ? 'tax_': 'post_';
            $code = $sitepress->get_language_for_element($post->id, $type . $post->getRealType());

            // if there's no language code, but we're pretty sure this is a post publish
            if( empty($code) && 
                !empty($_POST) && 
                isset($_POST['ID']) &&
                !empty($_POST['ID']) &&
                (int)$post->id === (int)$_POST['ID'] &&
                isset($_POST['icl_post_language']) && 
                !empty($_POST['icl_post_language']) && 
                (current_action() === 'wp_insert_post_data' || current_action() === 'save_post'))
            {
                // pull the code from the form data
                $current = filter_input( INPUT_POST, 'icl_post_language', FILTER_SANITIZE_FULL_SPECIAL_CHARS, FILTER_NULL_ON_FAILURE );
                if(!empty($current)){
                    $code = $current;
                }
            }
        }

        if(defined('POLYLANG_VERSION')){
            if($post->type === 'term' && function_exists('pll_get_term_language')){
                $code = pll_get_term_language($post->id);
            }elseif(function_exists('pll_get_post_language')){
                $code = pll_get_post_language($post->id);
            }
		}

        return $code;
    }

    public static function getAnchors($post)
    {
        preg_match_all('|<a [^>]+>([^<]+)</a>|i', $post->getContent(), $matches);

        if (!empty($matches[1])) {
            return $matches[1];
        }

        return [];
    }

    /**
     * Get URLs from post content
     *
     * @param $post
     * @return array|mixed
     */
    public static function getUrls($post)
    {
        preg_match_all('#<a\s.*?(?:href=[\'"](.*?)[\'"]).*?>#is', $post->getContent(), $matches);

        if (!empty($matches[1])) {
            return $matches[1];
        }

        return [];
    }

    public static function getSentencesWithUrls($post)
    {
        $data = [];
        $content = $post->getContent();

        // replace any base64ed image urls
        $content = preg_replace('`src="data:(?:image|text)\/(?:png|jpeg|svg\+xml|xml);base64,[\s]??[a-zA-Z0-9\/+=]+?"`', '', $content);
        $content = preg_replace('`alt="Source: data:image\/(?:png|jpeg|svg\+xml);base64,[\s]??[a-zA-Z0-9\/+=]+?"`', '', $content);

        preg_match_all('`(\!|\?|\.|^|)[^.!?\n]*<a\s[^>]*?(?:href=([\'"]|\\\")(.*?)([\'"]|\\\"))[^>]*?>(.*?)<\/a>((?!<a)[^.!?\n])*|<!-- wp:(?:core-embed\/wordpress|embed) {[\\\]*?"url[\\\]*?":[\\\]*?"([^"\\\]*?)[\\\]*?"[^}]*?[\\\]*?"} -->`is', $content, $matches);
        for ($i = 0; $i < count($matches[0]); $i++) {
            if (!empty($matches[0][$i]) && !empty($matches[3][$i])) {
                $sentence = $matches[0][$i];
                if (in_array(substr($sentence, 0, 1), ['.', '!', '?'])) {
                    $sentence = substr($sentence, 1);
                }

                $url = $matches[3][$i];

                // if the url is inside slashed quotes
                if( !empty($matches[2][$i]) && $matches[2][$i] === '\"' &&
                    !empty($matches[4][$i]) && $matches[4][$i] === '\"')
                {
                    // add the quotes to the url
                    $url = ($matches[2][$i] . $url . $matches[4][$i]);
                }

                // if there is an anchor
                if(!empty($matches[5][$i]) && $matches[5][$i]){
                    $anchor = $matches[5][$i];
                }else{
                    $anchor = '';
                }

                $data[] = [
                    'sentence' => trim(strip_tags($sentence)),
                    'anchor' => trim(strip_tags($anchor)),
                    'url' => $url
                ];
            }elseif(!empty($matches[7][$i])){
                $url = esc_attr($matches[7][$i]);
    
                $data[] = [
                    'sentence' => esc_attr__('Link is embedded, no sentence text detected', 'wpil'),
                    'anchor' => 'N/A',
                    'url' => $url
                ];
            }
        }

        // get the image tags too
        preg_match_all('#<img\s[^>]*?(?:(?:href|src)=([\'"]|\\\")(.*?)([\'"]|\\\"))[^>]*?>#is', $content, $matches);
        if(!empty($matches)){
            for ($i = 0; $i < count($matches[0]); $i++) {
                if (!empty($matches[0][$i]) && !empty($matches[1][$i])) {
                    $text = $matches[0][$i];

                    if(false !== strpos($text, 'title="') && false === strpos($text, 'title=""')){
                        $offset = (mb_strpos($text, 'title="') + 7);
                        $sentence = __('Broken Image. The title is: ', 'wpil') . '"' . mb_substr($text, $offset, (mb_strpos($text, '"', $offset) - $offset) ) . '"';
                    }elseif(false !== strpos($text, 'alt="') && false === strpos($text, 'alt=""')){
                        $offset = (mb_strpos($text, 'alt="') + 5);
                        $sentence = __('Broken Image. The alt text is: ', 'wpil') . '"' . mb_substr($text, $offset, (mb_strpos($text, '"', $offset) - $offset) ) . '"';
                    }else{
                        $sentence = __('Broken Image. The image doesn\'t have a title or alt text.', 'wpil');
                    }

                    $url = $matches[2][$i];

                    // if the url is inside slashed quotes
                    if( !empty($matches[1][$i]) && $matches[1][$i] === '\"' &&
                        !empty($matches[3][$i]) && $matches[3][$i] === '\"')
                    {
                        // add the quotes to the url
                        $url = ($matches[1][$i] . $url . $matches[3][$i]);
                    }

                    $data[] = [
                        'sentence' => trim(strip_tags($sentence)),
                        'anchor' => '',
                        'url' => $url
                    ];
                }
            }
        }

        // check to make sure that there aren't any empty anchors present
        if(strpos($content, '<a>') !== false){
            // if there are, pull those links too
            preg_match_all('`(\!|\?|\.|^|)[^.!?\n]*<a>(.*?)<\/a>((?!<a)[^.!?\n])*`is', $content, $matches);
            for ($i = 0; $i < count($matches[0]); $i++) {
                if (!empty($matches[0][$i])) {
                    $sentence = $matches[0][$i];
                    if (in_array(substr($sentence, 0, 1), ['.', '!', '?'])) {
                        $sentence = substr($sentence, 1);
                    }
    
                    $anchor = !empty($matches[2][$i]) ? $matches[2][$i]: '';

                    $data[] = [
                        'sentence' => trim(strip_tags($sentence)),
                        'url' => '{{wpil-empty-url}}',
                        'anchor' => $anchor
                    ];
                }
            }
        }

        return $data;
    }

    /**
     * Change sentence if it located inside embedded ACF blocks.
     * Changes the double qoutes in the link to insert's attributes into single quotes so we don't break the ACF blocks
     *
     * @param $content
     * @param $sentence
     * @param $changed_sentence
     * @return string
     */
    public static function changeByACF($content, $sentence, $changed_sentence){
        //find all blocks
        $blocks = [];
        $end = 0;
        while($end <= strlen($content) && strpos($content, '<!-- wp:acf', $end) !== false) {
            $begin = strpos($content, '<!-- wp:acf', $end);
            $end = strpos($content, '-->', $begin);
            $blocks[] = [$begin, $end];
        }

        //change sentence
        if (!empty($blocks)) {
            $pos = strpos($content, $sentence);
            foreach ($blocks as $block) {
                if ($block[0] < $pos && $block[1] > $pos) {
                    $changed_sentence = str_replace('"', "'", $changed_sentence);
                }
            }
        }

        return $changed_sentence;
    }

    /**
     * If quotes or apostrophes got shaved off before insert, try to borrow them back from the post content so we can actually insert the link.
     *
     * @param string $content
     * @param string $sentence
     * @param string $changed_sentence
     * @return array
     */
    public static function repunctuate_sentence($content, $sentence, $changed_sentence){
        $result = array(
            'sentence' => $sentence,
            'changed_sentence' => $changed_sentence,
        );

        if(empty($content) || empty($sentence) || empty($changed_sentence)){
            return $result;
        }

        // if the sentence is already in the content, we're already where we need to be
        if(
            false !== Wpil_Word::mb_strpos($content, $sentence) ||
            false !== Wpil_Word::mb_strpos(self::normalize_slashes($content), self::normalize_slashes($sentence))
        ){
            return $result;
        }

        $quote_chars = array("'", '"', '’', '‘', '“', '”');
        $split_chars = function($text){
            if($text === ''){
                return array();
            }

            $chars = preg_split('//u', (string) $text, -1, PREG_SPLIT_NO_EMPTY);
            return ($chars !== false) ? $chars : str_split((string) $text);
        };

        $is_quote = function($char) use ($quote_chars){
            return in_array($char, $quote_chars, true);
        };

        $build_match_data = function($text) use ($split_chars, $is_quote){
            $chars = $split_chars((string) $text);
            $clean = '';
            $map = array();

            foreach($chars as $index => $char){
                if($is_quote($char)){
                    continue;
                }

                $clean .= $char;
                $map[] = $index;
            }

            return array(
                'chars' => $chars,
                'clean' => $clean,
                'map' => $map,
            );
        };

        $content_data = $build_match_data($content);
        $sentence_data = $build_match_data($sentence);
        if(empty($content_data['clean']) || empty($sentence_data['clean'])){
            return $result;
        }

        $match_position = Wpil_Word::mb_strpos($content_data['clean'], $sentence_data['clean']);
        if(false === $match_position){
            return $result;
        }

        $clean_length = count($sentence_data['map']);
        if(
            empty($clean_length) || 
            !isset($content_data['map'][$match_position]) || 
            !isset($content_data['map'][$match_position + $clean_length - 1])
        ){
            return $result;
        }

        $match_start = $content_data['map'][$match_position];
        $match_end = $content_data['map'][$match_position + $clean_length - 1];
        $matched_sentence = implode('', array_slice($content_data['chars'], $match_start, (($match_end - $match_start) + 1)));
        if(empty($matched_sentence)){
            return $result;
        }

        $matched_data = $build_match_data($matched_sentence);
        if($matched_data['clean'] !== $sentence_data['clean']){
            return $result;
        }

        $plain_changed_sentence = trim(strip_tags(html_entity_decode($changed_sentence, ENT_QUOTES, 'UTF-8')));
        $changed_data = $build_match_data($plain_changed_sentence);
        if(!empty($changed_data['clean']) && $changed_data['clean'] === $sentence_data['clean']){
            $tokens = preg_split('/(<[^>]+>)/u', $changed_sentence, -1, PREG_SPLIT_DELIM_CAPTURE);

            if($tokens !== false){
                $rebuilt_changed_sentence = '';
                $source_index = 0;

                foreach($tokens as $token){
                    if($token === ''){
                        continue;
                    }

                    if(substr($token, 0, 1) === '<'){
                        $rebuilt_changed_sentence .= $token;
                        continue;
                    }

                    foreach($split_chars($token) as $char){
                        while(isset($matched_data['chars'][$source_index]) && $is_quote($matched_data['chars'][$source_index]) && $matched_data['chars'][$source_index] !== $char){
                            $rebuilt_changed_sentence .= $matched_data['chars'][$source_index];
                            $source_index++;
                        }

                        $rebuilt_changed_sentence .= $char;

                        if(isset($matched_data['chars'][$source_index]) && $matched_data['chars'][$source_index] === $char){
                            $source_index++;
                        }
                    }

                    while(isset($matched_data['chars'][$source_index]) && $is_quote($matched_data['chars'][$source_index])){
                        $rebuilt_changed_sentence .= $matched_data['chars'][$source_index];
                        $source_index++;
                    }
                }

                if(!empty($rebuilt_changed_sentence)){
                    $result['changed_sentence'] = $rebuilt_changed_sentence;
                }
            }
        }

        $result['sentence'] = $matched_sentence;

        return $result;
    }

    /**
     * Get post ID from any URL
     *
     * @param string $url
     * @return int|false
     */
    public static function get_post_id_from_any_url($url) {
        $url = Wpil_Settings::makeLinkAbsolute($url);

        $url_parts = parse_url($url);

        if(!isset($url_parts['path']) || empty($url_parts['path'])){
            return false;
        }

        $path = trim($url_parts['path'], '/');
        $path_parts = explode('/', $path);
        $slug = end($path_parts);

        // Get all public post types
        $post_types = get_post_types(array('public' => true));

        // First try exact path match
        $args = array(
            'post_type'      => $post_types,
            'post_status'    => 'publish',
            'posts_per_page' => 1,
            'fields'         => 'ids',
            'no_found_rows'  => true,
            'update_post_meta_cache' => false,
            'update_post_term_cache' => false
        );

        // Try matching the full path first
        $args['name'] = $path;
        $query = new WP_Query($args);

        // If no match, try with just the slug
        if (!$query->have_posts() && !empty($slug)) {
            $args['name'] = $slug;
            $query = new WP_Query($args);
        }

        if ($query->have_posts()) {
            foreach ($query->posts as $post_id) {
                $post = get_post($post_id);

                // Verify the slug matches
                if ($post->post_name !== $slug) {
                    continue;
                }

                $post_url = parse_url(get_permalink($post_id));
                if (trim($post_url['path'], '/') !== $path) {
                    continue;
                }

                return $post_id;
            }
        }

        return false;
    }

    /**
     * Get post model by view link.
     * URLtoPost
     * IDFROMLINK
     * IDFROMURL
     *
     * @param $link
     * @return Wpil_Model_Post|null
     */
    public static function getPostByLink($link)
    {
        global $wpdb;
        $post = null;
        $link = trim($link);
    //    $link = Wpil_Link::get_url_redirection($link) ?: $link; //todo: make work with
        $starting_link = $link;

        // check to see if we've already come across this link
        $cached = self::get_cached_url_post($link);
        // if we have
        if(!empty($cached)){
            //return the cached version
            return $cached;
        }

        // check to make sure that we are reasonably sure we can trace the link
        if(!Wpil_Link::is_traceable($link)){
            // if we're not, return null
            return $post;
        }

        // check to see if the link isn't a pretty link
        if(preg_match('#[?&](p|page_id|attachment_id)=(\d+)#', $link, $values)){
            // if it's not, get the id
            $id = absint($values[2]);
            // if there is an id
            if($id){
                // get the post so we can make sure it exists
                $wp_post = get_post($id);
                // if it does exist, set the id. Else, set it to null
                $post_id = (!empty($wp_post)) ? $wp_post->ID: null;
            }
        }elseif(preg_match('#[?&](tag_ID)=(\d+)#', $link, $values)){ // if it looks to be a tag id
            // if it's not, get the id
            $id = absint($values[2]);
            // if there is an id
            if($id){
                // get the term so we can make sure it exists
                $wp_term = get_term($id);
                // if it does exist, set the id. Else, set it to null
                $term_id = (!empty($wp_term)) ? $wp_term->term_id: null;
            }
        }else{
            // make sure the link isn't double slashed anywhere that it's not supposed to be
            if(!empty(preg_match('/(?<!http:|https:)\/\//', $link, $m)) || !empty($m)){
                $link = preg_replace('/(?<!http:|https:)(?:\/\/\/|\/\/)/', '/', $link);
                $link = preg_replace('/(?<!http:|https:)(?:\/\/\/|\/\/)/', '/', $link);
            }

            $link = explode('$', $link); // TODO: Review and confirm that no users report that links aren't being traced || the number of Inbound Internal links is lower after update 2.5.7
            $link = $link[0];

            // change the live domain into the staging domain if the user has opted to do so
            $link = Wpil_Link::filter_live_to_staging_domain($link);
            // clean up any translations if it's a relative link
            $link = Wpil_Link::clean_translated_relative_links($link);

            // if the user isn't using permalinks that override the post name
            if(!defined('CUSTOM_PERMALINKS_FILE')){
                $post_id = url_to_postid($link); // use this
            }
        }

        if (!empty($post_id)) {
            $post = new Wpil_Model_Post($post_id);
        }elseif(!empty($term_id)){
            $post = new Wpil_Model_Post($term_id, 'term');
        }

        // if we couldn't find the post and custom permalinks is active
        if(empty($post) && defined('CUSTOM_PERMALINKS_FILE')){
            // consult it's database listings to see if we can find the post the link belongs to
            $search_url = $link;

            // get the home url and clean it up
            $site_url = get_home_url();
            $site_url = preg_replace('/http:\/\/|https:\/\/|www\./', '', $site_url);
            // make sure the supplied link is similarly clean
            $search_url = preg_replace('/http:\/\/|https:\/\/|www\./', '', $search_url);

            // and replace the home portion of the link to make it relative
            $search_url = trim(str_replace($site_url, '', $search_url), '/'); // Don't add slashes around the url
            
            // get the stati and types to search
            $status = Wpil_Query::postStatuses('p');
            $type = Wpil_Query::postTypes('p');

            // now search the db
            $search = $wpdb->get_col(
                $wpdb->prepare(
                    'SELECT p.ID ' .
                    " FROM $wpdb->posts AS p INNER JOIN $wpdb->postmeta AS pm ON (pm.post_id = p.ID) " .
                    " WHERE pm.meta_key = 'custom_permalink' " .
                    ' AND (pm.meta_value = %s OR pm.meta_value = %s) ' .
                    " {$status} {$type} " .
                    " LIMIT 1",
                    $search_url,
                    $search_url . '/'
                )
            );
            // if we found a post
            if(!empty($search)){
                // that is our new post object
                $post = new Wpil_Model_Post($search[0]);
            }
        }

        // if all that didn't work, the post might be draft or Polylang Pro might be active and we'll have to check for multiple posts with the same name
        // so we'll try pulling the post name from the URL and seeing if that will get us an id
        if((empty($post) || Wpil_Settings::polylang_enabled()) && is_string($link) && !empty($link) && Wpil_Link::isInternal($link)){
            // get the permalink structure
            $link_structure = get_option('permalink_structure', '');
            if(!empty($link_structure)){
                // see if the post name is in it
                if(false !== strpos($link_structure, '%postname%')){
                    // if it is, blow up the link structure
                    $exploded_structure = explode('/', '/' . trim($link_structure, '/') . '/'); // frame the permalink with "/" so that we're consistently comparing it to the link
                    // make the supplied link relative, and blow it up too
                    if(!Wpil_Link::isRelativeLink($link)){
                        // get the home url and clean it up
                        $site_url = get_home_url();
                        $site_url = preg_replace('/http:\/\/|https:\/\/|www\./', '', $site_url);
                        // make sure the supplied link is similarly clean
                        $link = preg_replace('/http:\/\/|https:\/\/|www\./', '', $link);

                        // and replace the home portion of the link to make it relative
                        $link = '/'. trim(str_replace($site_url, '', $link), '/') . '/'; // we're going to assume that the user isn't using a draft post as the home url... That would give us just "/" at this point, and "///" isn't a valid url
                    }

                    // if polylang is active
                    if(Wpil_Settings::translation_enabled() && Wpil_Settings::polylang_enabled()){
                        global $polylang;

                        if(!empty($polylang)){
                            // get the link's language
                            $lang = $polylang->links_model->get_language_from_url($link);

                            // if we got the language, try getting it's term
                            if(!empty($lang)){
                                $language_term = get_term_by('slug', $lang, 'language');
                            }

                            // and remove any translation effect from the url
                            $link = $polylang->links_model->remove_language_from_link($link);
                        }
                    }

                    // now blow up the link
                    $exploded_link = explode('/', $link);

                    // check to see if we're looking at a child page link
                    $offset = 0;
                    if(count($exploded_link) > count($exploded_structure)){
                        // if we are, account for the parent slugs
                        $offset = (count($exploded_link) - count($exploded_structure));
                    }

                    // and see if the link has a postname in the same position as the permalink structure
                    $name = '';
                    foreach($exploded_structure as $key => $piece){
                        $ind = $key + $offset;
                        if( $piece === '%postname%' &&          // if we're focussed on the postname
                            isset($exploded_link[$ind]) &&      // and there's a corresponding piece in the link
                            !empty($exploded_link[$ind]) &&     // and there's something in the corresponding piece
                            is_string($exploded_link[$ind]) &&  // and the corresponding is a string
                            strlen($exploded_link[$ind]) > 0)   // and it's at least 1 char long
                        {
                            // extract the piece as the post name and exit the loop
                            $name = $exploded_link[$ind];
                            break;
                        }
                    }

                    // if we've found something
                    if(!empty($name)){
                        $post_types = Wpil_Query::postTypes();

                        if(Wpil_Settings::translation_enabled() && !empty($language_term)){
                            $query = $wpdb->prepare("SELECT a.ID FROM {$wpdb->posts} a LEFT JOIN {$wpdb->term_relationships} b ON a.ID = b.object_id WHERE a.post_name = %s && b.term_taxonomy_id = %d {$post_types} LIMIT 1", $name, $language_term->term_id);
                        }else{
                            $query = $wpdb->prepare("SELECT `ID` FROM {$wpdb->posts} WHERE `post_name` = %s {$post_types} LIMIT 1", $name);
                        }

                        // see if there's a post in the database with the same name from among the post types that the user has selected
                        $dat = $wpdb->get_col($query);

                        // if there isn't one, check across all the post types
                        if(empty($dat)){
                            $all_post_types = Wpil_Query::postTypes('', true);
                            $dat = $wpdb->get_col($wpdb->prepare("SELECT `ID` FROM {$wpdb->posts} WHERE `post_name` = %s {$all_post_types} LIMIT 1", $name));
                        }

                        // if that didn't work either, try looking for the title
                        if(empty($dat)){ // TODO: set up some kind of a post title lookup table. The post_title column isn't indexed, and searching it for many results can take forever
                            // replace any hyphens with spaces
                            $name = str_replace('-', ' ', $name);
                            // and search through our post types
                            $dat = $wpdb->get_col($wpdb->prepare("SELECT `ID` FROM {$wpdb->posts} WHERE `post_title` = %s {$post_types} LIMIT 1", $name)); // for exceedingly long titles, I might consider re-adding the LIKE check. But we'll cross that bridge when we get there

                            // if that still didn't work, check the title across all the post types that we are reasonably sure are active
                            if(empty($dat)){
                                $all_post_types = Wpil_Query::postTypes('', true);
                                $dat = $wpdb->get_col($wpdb->prepare("SELECT `ID` FROM {$wpdb->posts} WHERE `post_title` = %s {$all_post_types} LIMIT 1", $name));
                            }
                        }

                        // if we've found a post id
                        if(!empty($dat) && isset($dat[0]) && !empty($dat[0])){
                            // create the post object we've been striving for
                            $post = new Wpil_Model_Post($dat[0]);
                        }
                    }
                }
            }
        }

        if(empty($post)){
            $post_id = self::get_post_id_from_any_url($starting_link);

            // if we've found the post that the link belongs to
            if(!empty($post_id)){
                // setup the post object with it
                $post = new Wpil_Model_Post($post_id);
            }
        }

        // if we _still_ haven't found a post
        if (empty($post)) {
            // see if the URL is actually for a term instead of a post
            $slug = array_filter(explode('/', $starting_link));
            $term = Wpil_Term::getTermBySlug(end($slug), $starting_link);
            if(!empty($term)){
                $post = new Wpil_Model_Post($term->term_id, 'term');
            }
        }

        // if we've gone this far and haven't 

        // cache the results of our efforts in case we come across this link again
        self::update_cached_url_post($starting_link, $post);

        return $post;
    }

    /**
     * Checks to see if the url was previously processed into a post object.
     * If it is in the cache, it returns the cached post so we don't have to run through the process again.
     * Returns false if the url hasn't been processed yet, or it doesn't go to a known post
     **/
    public static function get_cached_url_post($url = ''){
        if(empty($url) || !is_string($url)){
            return false;
        }

        // clean up the url a little so we have consistency between slightly different links
        // filter the live url to staging if needed
        $url = Wpil_Link::filter_live_to_staging_domain($url);
        // clean up any translations if it's a relative link
        $url = Wpil_Link::clean_translated_relative_links($url);
        // remove www & protocol bits
        $url = str_replace(['http', 'https'], '', str_replace('www.', '', $url));

        if(empty($url) || !isset(self::$post_url_cache[$url])){
            return false;
        }

        return self::$post_url_cache[$url];
    }

    /**
     * Updates the url cache when we come across a url + post that we haven't stored yet.
     * Also does some housekeeping to make sure the cache doesn't grow too big
     **/
    public static function update_cached_url_post($url, $post){
        if(empty($url) || empty($post) || isset(self::$post_url_cache[$url]) || !is_string($url)){
            return false;
        }

        // clean up the url a little so we have consistency between slightly different links
        // filter the live url to staging if needed
        $url = Wpil_Link::filter_live_to_staging_domain($url);
        // clean up any translations if it's a relative link
        $url = Wpil_Link::clean_translated_relative_links($url);
        // remove www & protocol bits
        $url = str_replace(['http', 'https'], '', str_replace('www.', '', $url));

        if(empty($url)){
            return false;
        }

        self::$post_url_cache[$url] = $post;

        if(count(self::$post_url_cache) > 5000){
            $ind = key(self::$post_url_cache);
            unset(self::$post_url_cache[$ind]);
        }
    }

    /**
     * Insert link into content
     *
     * @param $content The post content or content segment to update
     * @param $sentence The sentence in the content that will have a link inserted
     * @param $changed_sentence The sentence with a link inserted
     * @param $ignore_links Should we overlook links that are inside the phrase? Currently applies to autolinks that are set to "force_insert" links.
     * @param $offset How much of the content should we skip over? Uses mb_str positioning
     * @return bool Returns true if the link was inserted, and false if it could not be inserted or data was missing
     */
    public static function insertLink(&$content, $sentence, $changed_sentence, $ignore_links = false, $offset = 0)
    {
        if(empty($sentence)){
            return false;
        }

        // trim the sentences to avoid inconsequential whitespace from preventing the insert!
        $sentence = trim(trim($sentence, " ")); // trimming non-breaking spaces in the inner trims (U+00a0)
        $changed_sentence = trim(trim($changed_sentence, " "));

        // if there's no content to work with
        if(empty($sentence) || empty($changed_sentence) || empty($content) || empty(self::normalize_slashes($content)) || empty(self::normalize_slashes($changed_sentence))){
            // say that the link wasn't inserted...
            return false;
        }

        // if the content already has the link, exit
        if( false !== Wpil_Word::mb_strpos($content, $changed_sentence) ||
            false !== Wpil_Word::mb_strpos(self::normalize_slashes($content), self::normalize_slashes($changed_sentence)) || // do a double check to make sure the quotes are the same for content and sentence
            false !== Wpil_Word::mb_strpos(self::normalize_slashes($content), self::normalize_slashes(str_replace(['<', '>'], ['\u003c', '\u003e'], $changed_sentence))) ||  // do a tripl check to make sure there isn't an ACF gutneberg block link AND quotes are the same for content and sentence
            self::check_for_url_anchor($content, $changed_sentence) // Ok fine, also make sure that the url + anchor aren't appearing in the content
            ){
            // check one more thing... make sure it's not an insertable autolink
            if(!self::check_if_insertable_autolink($changed_sentence)){
                // if it's not, exit
                return false;
            }
        }

        $position_start = Wpil_Word::mb_strpos($content, $sentence, $offset);
        if(false === $position_start){
            $position_start = Wpil_Word::mb_strpos(self::normalize_slashes($content), self::normalize_slashes($sentence), $offset);

            // if we have a start point now that the slashes have been normalized
            if(false !== $position_start){
                // find out if normalizing the slashes has changed the start position
                $letter1 = mb_substr(self::normalize_slashes($content), $position_start, 1);
                $letter2 = mb_substr($content, $position_start, 1);
                // if the letters don't match
                if($letter1 !== $letter2){
                    // figure out how far the string has changed to calculate the correct start point
                    $search = mb_substr(self::normalize_slashes($content), $position_start, mb_strlen($sentence));

                    // clean up the string to hopefully get a good search term
                    $search = explode('{wpil-explode-token}', str_replace(array('\'', '"', '\\'), '{wpil-explode-token}', $search));

                    $term = '';
                    foreach($search as $part){
                        if(strlen($part) > strlen($term)){
                            $term = $part;
                        }
                    }

                    // if we've found a term
                    if(!empty($term)){
                        $start1 = Wpil_Word::mb_strpos($content, $term);
                        $start2 = Wpil_Word::mb_strpos(self::normalize_slashes($content), $term);

                        if(false !== $start1 && false !== $start2){
                            if($start1 < $start2){ // 29 < 30; 30 - 29 = 1; 
                                $position_start = $position_start - ($start2 - $start1);
                            }else{ // 30 > 29; 30 - 29 = 1;
                                $position_start = $position_start + ($start1 - $start2);
                            }
                        }else{
                            $position_start = abs(intval($start1) - intval($start2));
                        }

                    }else{
                        $position_start = 0;
                    }
                }
            }
        }

        $position_end = 0;
        $old_end = 0;
        $endings = array_diff(Wpil_Word::$endings, array('\'', '"', ','));
        $sent_len = mb_strlen($sentence);

        // while we have words
        while($position_start !== false && $position_start < mb_strlen($content)){

            // go over all the endings and find out which one is the actual end to the current string
            $shortest = false;
            foreach($endings as $ending){
                // the shortest string will have the ending punctuation
                $current_end = Wpil_Word::mb_strpos($content, $ending, ($position_start + $sent_len));
                if(false === $shortest){
                    $shortest = $current_end;
                }elseif($current_end < $shortest && $current_end !== false){
                    $shortest = $current_end;
                }
            }

            $position_end = (false !== $shortest) ? $shortest: mb_strlen($content); // if no ending was found, give the end of the content

            // now find the ending of the string that comes before the current one.
            $old_shortest = false;
            foreach($endings as $ending){
                // the longest string will have the ending punctuation since it's closest to the end of the current string.
                $current_end = mb_strrpos($content, $ending, (1 + $position_start - mb_strlen($content)));

                if(false === $current_end){
                    continue;
                }

                // if there's a closing html tag that comes after the current old ending
                $closing = Wpil_Word::mb_strpos($content, '>', $current_end);
                if(false !== $closing && $closing < $position_end){
                    // find the opening tag so we can tell what kind of tag this is
                    $current_end = mb_strrpos($content, '<', (1 + $current_end - mb_strlen($content)));
                }

                if(false === $old_shortest){
                    $old_shortest = $current_end;
                }elseif($current_end > $old_shortest && $current_end !== false){
                    $old_shortest = $current_end;
                }
            }

            $old_end = (false !== $old_shortest) ? $old_shortest: 0;
            $length = ($position_end - $position_start);
            $replace = mb_substr($content, $position_start, $length);

            // get the slice of text that we'll be checking for links
            $examine_text = mb_substr($content, $old_end, ($position_end - $old_end));

            // if there isn't a link in the text AND the text isn't inside of a tag
            if( mb_strlen($replace) > 0 &&
                !Wpil_Link::checkForForbiddenTags($examine_text, $replace, $sentence, $ignore_links) && 
                !self::check_if_inside_tag($content, $position_start, $position_end) &&
                !self::check_if_inside_json($content, $position_start, $position_end) && // or JSON
                !self::check_if_inside_attribute($content, $position_start, $position_end)) // or an attribute
            {
                // get the text that comes before and after the sentence
                $front = mb_substr($content, 0, $position_start);
                $back = mb_substr($content, ($position_start + $length));

                // remove any quotes from the sentence to change
                $changed_sentence = Wpil_Word::removeQuotes($changed_sentence);

                // check if the user only wants to insert relative links
                if(!empty(get_option('wpil_insert_links_as_relative', false))){
                    // if he does, extract the url
                    preg_match('/<a href="([^\"]+)"[^>]*?>(.*)<\/a>/i', $changed_sentence, $matches);

                    // if we've got the url
                    if(!empty($matches) && isset($matches[1])){
                        // check the url to make sure it's internal
                        if(Wpil_Link::isInternal($matches[1])){
                            // if it is, make it relative
                            $url = wp_make_link_relative($matches[1]);
                            // and replace the existing url with the new one
                            $changed_sentence = mb_ereg_replace(preg_quote($matches[1]), $url, $changed_sentence);
                        }
                    }
                }

                $trackingA = '';
                $trackingB = '';
                if(false !== strpos($changed_sentence, 'data-wpil-replace=')){
                    if(!Wpil_Settings::get_disable_link_insert_tracking()){
                        $id = Wpil_Link::create_next_tracked_link_id();
                        if(!empty($id)){
                            $trackingA = ' data-wpil-monitor-id="' . $id . '"';
                            $trackingB = " data-wpil-monitor-id='" . $id . "'";
                        }
                    }

                    $changed_sentence = mb_eregi_replace(preg_quote('data-wpil-replace=""'), $trackingA, $changed_sentence);
                    $changed_sentence = mb_eregi_replace(preg_quote("data-wpil-replace=''"), $trackingB, $changed_sentence);
                }

                // escape the link if it's in json content
                $changed_sentence = self::processLinkForJsonContent($content, $changed_sentence, $position_start, $position_end);
                
                $changed_sentence = self::adjust_sentence_slashes($changed_sentence, $sentence, $position_start, $position_end, $content);
                $sentence = self::adjust_sentence_slashes($sentence, $changed_sentence, $position_start, $position_end, $content);

                $changed_text = mb_eregi_replace('(?<!=[\"\'\\\"\\\'])(' . preg_quote($sentence) . ')(?![\"\'\\\"\\\'].*?>)', $changed_sentence, $replace);

                // if the link has been inserted multiple times
                if(substr_count($changed_text, '</a>') > 1){
                    // remove all but the first version of the link
                    global $wpil_link_insert_count, $wpil_link_insert_sentence;
                    $wpil_link_insert_count = 0;
                    $wpil_link_insert_sentence = $sentence;

                    $changed_text = mb_ereg_replace_callback(preg_quote($changed_sentence), function($matches){global $wpil_link_insert_count, $wpil_link_insert_sentence; $wpil_link_insert_count++; return ($wpil_link_insert_count === 1) ? $matches[0] : $wpil_link_insert_sentence; }, $changed_text);
                }

                // if the link has been inserted
                if($changed_text !== $replace){
                    // add the link to the text
                    $content = ($front . $changed_text . $back);
                    // and exit the loop since we only add one link at a time.
                    return true;
                }else{
                    // if the link couldn't be inserted, continue the loop so hopefully we find the place to insert the link
                    $position_start = Wpil_Word::mb_strpos($content, $sentence, $position_end + 1);
                }
            }else{
                // if the keyword text is in a link, move to the next instance of the keyword
                try {
                    $position_start = Wpil_Word::mb_strpos($content, $sentence, $position_end + 1);
                    if(false === $position_start){
                        $position_start = Wpil_Word::mb_strpos(self::normalize_slashes($content), self::normalize_slashes($sentence), $position_end + 1);
                        if(!empty($position_start)){
                            // find out if normalizing the slashes has changed the start position
                            $letters1 = mb_substr(self::normalize_slashes($content), $position_start, 5);
                            $letters2 = mb_substr($content, $position_start, 5);

                            // if the letters don't match
                            if($letters1 !== $letters2){
                                // figure out how far the string has changed to calculate the correct start point
                                $search = mb_substr(self::normalize_slashes($content), $position_start, mb_strlen($sentence));
            
                                // clean up the string to hopefully get a good search term
                                $search = explode('{wpil-explode-token}', str_replace(array('\'', '"', '\\'), '{wpil-explode-token}', $search));

                                $term = '';
                                foreach($search as $part){
                                    // find the longest piece of text to get a search term
                                    if(strlen($part) > strlen($term)){
                                        $term = $part;
                                    }
                                }
            
                                // if we've found a term
                                if(!empty($term)){
                                    $start1 = Wpil_Word::mb_strpos($content, $term, $position_end + 1);
                                    $start2 = Wpil_Word::mb_strpos(self::normalize_slashes($content), $term, $position_end + 1);
            
                                    if(false !== $start1 && false !== $start2){
                                        if($start1 < $start2){ // 29 < 30; 30 - 29 = 1; 
                                            $position_start = $position_start - ($start2 - $start1);
                                        }else{ // 30 > 29; 30 - 29 = 1;
                                            $position_start = $position_start + ($start1 - $start2);
                                        }
                                    }else{
                                        $position_start = abs(intval($start1) - intval($start2));
                                    }
            
                                }
                            }
                        }
                    }
                } catch (Throwable $t) {
                    $position_start = false;
                } catch (Exception $e) {
                    $position_start = false;
                }
            }
        }

        return false;
    }

    /**
     * Checks if the current link is an autolink that should be checked if it can be inserted
     **/
    public static function check_if_insertable_autolink($changed_sentence = ''){
        if(empty($changed_sentence)){
            return false;
        }

        // if
        if( false !== strpos($changed_sentence, 'wpil_keyword_link') && // this is an autolink
            0 === strpos($changed_sentence, '<a class="wpil_keyword_link"') &&  // there's no sentence outside of the keyword
            strrpos($changed_sentence, '</a>') === mb_strlen($changed_sentence) - 4)
        {
            // extract the url
            $found = preg_match('`<a class="wpil_keyword_link"[^>]*?href=[\'"]([^\'"]*?)[\'"][^>]*?>`', $changed_sentence, $matches);

            // if we've got something
            if($found && isset($matches[1]) && !empty($matches[1])){
                // pull up the autolink
                $keyword = Wpil_Keyword::getKeywordByURL($matches[1]);
                // if we've got an autolink and it's supposed to be inserted more than once
                if(!empty($keyword) && isset($keyword->add_same_link) && !empty($keyword->add_same_link) && (!isset($keyword->link_once) || empty($keyword->link_once))){
                    // say that it's ok to insert the link
                    return true;
                }
            }
        }

        // failing the complex check for worthiness, return false
        return false;
    }

    /**
     * Checks to see if the current slice of text is in the middle of a tag.
     * Currently, scans the text backwards from the start position to see if it can find an opening HTML tag before coming across a closer.
     * If it can't find the tag, it assumes that the link isn't inside of a tag
     * @param string $content The body content that the link will be inserted in.
     * @param int $position_start The start of the "replace" text
     * @param int $position_end The ending of the "replace" text
     * @return bool Returns true if the text is inside of a tag, and false if it's not || there are no tags || it can't tell
     **/
    public static function check_if_inside_tag($content, $position_start, $position_end){
        // if we're missing data, we can't be inside a tag... Or at least we can't know it
        if(empty($content) || false === $position_start || false === $position_end){
            return false;
        }

        // check to make sure the content has HTML tags
        $has_normal = (false !== strpos($content, '<') && false !== strpos($content, '>'));
        $has_encoded = (false !== strpos($content, '&gt;') && false !== strpos($content, '&lt;'));

        // if we don't appear to have tags, say that we're not in a tag
        if(!$has_normal && !$has_encoded){
            return false;
        }

        $base64_img_positions = array();
        if(false !== strpos($content, '<img')){
            if(preg_match_all('/<img [^>]*?data:image\/(?:jpeg|png|gif|bmp|svg\+xml);base64,[^>]*?>/', $content, $m) && isset($m[0]) && !empty($m[0])){
                $last_pos = 0;
                foreach($m[0] as $img){
                    $str_pos = mb_strpos($content, $img, $last_pos);
                    if(false !== $str_pos){
                        $base64_img_positions[$str_pos] = ($str_pos + mb_strlen($img));
                        $last_pos = $str_pos;
                    }
                }
            }
        }

        $inside_tag = null;
        $tag_name_chars = '[a-zA-Z-]+'; // the list of chars that can be in a tag name
        $self_closing_tags = array('img', 'button', 'input'); // a list of tags that are self closing so we can skip searching for a closing tag

        // if the start position is past the beginning of the content
        if($position_start > 0){
            // start counting letters backwards to see if we can find an opening HTML tag
            for($i = ($position_start - 1); $i >= 0; $i--){
                $char = mb_substr($content, $i, 1);

                // if we find what looks like a closing tag (before finding what might be an opening tag)
                if($char === '>' || $char === '&gt;'){
                    // check to make sure that we're not looking through an img with base64 data
                    if(!empty($base64_img_positions)){
                        foreach($base64_img_positions as $start => $end){
                            if($end === ($i + 1)){
                                $i = $start;
                                continue 2;
                            }
                        }
                    }

                    // check to make sure it's a closing tag
                    for($j = ($i - 1); $j >= 0; $j--){
                        // find the opening char "<" for the possible closing tag
                        $tag_char = mb_substr($content, $j, 1);
                        if($tag_char === '<' || $tag_char === '&lt;'){
                            // extract the tag
                            $tag = mb_substr($content, $j, ($i - $j) + 1);
                            // check to see if it's a closing tag
                            $closing = preg_match('/'.$tag_char . '\/' . $tag_name_chars . $char.'/', $tag);
                            // if it is
                            if(!empty($closing)){
                                // we _shouldn't_ be inside of an HTML tag, so set the flag for it
                                $inside_tag = false;
                                // and exit the loops
                                break 2;
                            }else{
                                // if it's not a closing tag, see if it's an opening one
                                /*$opening = preg_match('/'.$tag_char . $tag_name_chars . $char.'/', $tag);
                                // if it's an opening tag
                                if(!empty($opening)){
                                    // see if there's any signs of opening HTML tags between here and where we started

                                }*/

                                // decide that we've done enough looking and say that we're not in a tag
                                $inside_tag = false;
                                // and exit both loops
                                break 2;
                                // exit the tag loop
                                //break;
                            }
                        }
                    }

                }elseif($char === '<' || $char === '&lt;'){
                    // if the current char is the opening sign for a tag

                    // track back down the letters to put together a potential tag
                    $tag_name = '';
                    for($k = ($i + 1); $k <= $position_end; $k++){
                        $tag_char = mb_substr($content, $k, 1);
                        $valid_char = preg_match('/'.$tag_name_chars.'/', $tag_char);

                        // if the char can be in a tag name
                        if($valid_char){
                            // add it to our provisional tag name
                            $tag_name .= $tag_char;
                        }elseif(!empty($tag_name) && empty($valid_char)){
                            // if we have a tag name, and the current char isn't recognized as an html tag
                            // check the content for a closing version of it
                            $closing_tag = preg_match('/'. $char . '\/' .$tag_name.'(?:>|&gt;)/', $content);
                        
                            // if there is a closing version of it
                            if(!empty($closing_tag) || in_array($tag_name, $self_closing_tags, true)){
                                // we're inside a tag!
                                $inside_tag = true;
                                // exit the loops
                                break 2;
                            }
                        }else{
                            // if the first letter isn't part of a tag, break out of this loop
                            break;
                        }
                    }
                }
            }
        }

        // TODO: Check from the ending part of the string if the need ever arises...
        // Since it takes 2 ends to make a tag, just checking the beginning so do it

        return (!empty($inside_tag));
    }

    /**
     * Checks to see if the current link candidate is inside a json string
     * @param string $content The body content that the link will be inserted in.
     * @param int $position_start The start of the "replace" text
     * @param int $position_end The ending of the "replace" text
     * @return bool Returns if the link target is inside a JSON string
     **/
    public static function check_if_inside_json($content, $position_start, $position_end){
        // try to find all JSON in the content
        preg_match_all('/(\{(?:[^{}]|(?R))*?\})/', $content, $matches);

        // if no matches are found, we _shouldn't_ be in JSON
        if(empty($matches)){
            return false;
        }

        $acf_json_blocks = array();
        if( class_exists('ACF') && 
            empty(get_option('wpil_disable_acf', false)) && 
            Wpil_Settings::getContentFormattingLevel() > 1) // make sure we can see the links that we're about to create!
        {
            preg_match_all('/<!--\s*wp:(acf(?:\/[A-Za-z0-9_-]+)?)\s+(\{(?:[^{}]|(?2))*\})\s*(?:\/)?-->/s', $content, $acf_json_blocks);
        }

        // loop over all the "JSON"s to check them for our content
        $pos = 0;
        foreach($matches[0] as $match){
            $start = Wpil_Word::mb_strpos($content, $match, $pos); // offset the search so we can keep moving down the string if there's repeating elements
            if(false === $start){
                continue;
            }
            $pos = $start;
            $end = $start + mb_strlen($match);
            // if the text we're examining falls inside the "JSON"
            if($start < $position_start && $position_end < $end){
                // try decoding the "JSON" to see if it really is JSON. Try both with slashes and with slashes removed
                if( (!empty(json_decode($match)) || !empty(json_decode(wp_unslash($match)))) ||
                    (false !== strpos(wp_unslash($match), '"https://schema.org"')) // if the JSON didn't parse, check if the match contains the schema declaration
                ){
                    // if we've got acf jsons
                    if(!empty($acf_json_blocks)){
                        // check to make sure this isn't one
                        foreach($acf_json_blocks[0] as $block){
                            // if it is
                            if(false !== strpos($block, $match)){
                                // skip this check
                                continue 2;
                            }
                        }
                    }

                    // if we've got something, this is JSON!
                    return true;
                }
            }elseif($start > $position_end){
                // exit the loop if the start of the current "JSON" is past the end of the search string
                break;
            }
        }

        // if we made it here, the text shouldn't be in json
        return false;
    }

    /**
     * Does a simple check to see if the current piece of text under examination is likely to be inside an attribute
     * @param string $content The body content that the link will be inserted in.
     * @param int $position_start The start of the "replace" text
     * @param int $position_end The ending of the "replace" text
     * @return bool Returns if the link target is inside an attribute
     **/
    public static function check_if_inside_attribute($content, $position_start, $position_end){
        // first try to find all of the attributes
        preg_match_all('/\s[a-zA-Z0-9-_]*?=(?:\\\)*?(\'|")([^><]*?)(?:\\\)*?\1(?!<|\[|&lt;)/u', $content, $attributes);

        // if there are none
        if(empty($attributes) || !isset($attributes[0]) || empty($attributes[0])){
            // we're clearly not inside one
            return false;
        }

        // pull up the text under consideration
        $text = mb_substr($content, $position_start, ($position_end - $position_start));

        // if that's empty for some reason
        if(empty($text)){
            // return false
            return false;
        }

        // remove all of the attributes that we know the examination text isn't inside so we don't have to search them
        $attributes = array_filter(array_map(function($att) use ($text){ return (false !== strpos($att, $text)) ? $att: false; }, $attributes[0]));

        // if that removed all of the attribuets
        if(empty($attributes)){
            // we're in the clear!
            return false;
        }

        // look over all of the remaining attributes
        $offset = 0;
        foreach($attributes as $attribute){
            // find the start of the attribute
            $att_start = Wpil_Word::mb_strpos($content, $attribute, $offset);

            // if that's not possible
            if(empty($att_start)){
                // skip to the next one
                continue;
            }

            // now pull up the end of the attribute
            $att_end = $att_start + mb_strlen($attribute);

            // if the text to examine is somewhere after the end of the
            if($att_start < $position_start && $att_end > $position_end){
                return true;
            }else{
                $offset = $att_end;
            }
        }

        return false;
    }

    /**
     * Get post IDs from certain category
     *
     * @param $category_id
     * @return array
     */
    public static function getCategoryPosts($category_id)
    {
        global $wpdb;

        $posts = [];
        $categories = $wpdb->get_results("SELECT r.object_id as `id` FROM {$wpdb->term_relationships} r INNER JOIN {$wpdb->term_taxonomy} tt ON tt.term_taxonomy_id = r.term_taxonomy_id WHERE tt.term_id = " . $category_id);
        foreach ($categories as $post) {
            $posts[] = $post->id;
        }

        return $posts;
    }

    /**
     * Run function for all editors
     *
     * @param $action
     * @param $params
     */
    public static function editors($action, $params)
    {
        $editors = [
            'Beaver',
            'Elementor',
            'Origin',
            'Oxygen',
            'Thrive',
            'Themify',
            'Muffin',
            'Enfold',
            'Cornerstone',
            'WPRecipe',
            'Goodlayers',
            'Divi'
        ];

        foreach ($editors as $editor) {
            $class = 'Wpil_Editor_' . $editor;
            call_user_func_array([$class, $action], $params);
        }
    }

    /**
     * TODO: Fill out so that we can pull the editors that are actually active and run through them.
     */
    public static function get_active_editors(){
        $editors = array();
        // check for active editors by looking for major constants or classes
        if(defined('FL_BUILDER_VERSION')){
            $editors[] = 'Beaver';
        }
        if(defined('ELEMENTOR_VERSION')){
            $editors[] = 'Elementor';
        }
        if(defined('SITEORIGIN_PANELS_VERSION')){
            $editors[] = 'Origin';
        }
        if(defined('CT_VERSION')){
            $editors[] = 'Oxygen';
        }
        if(defined('TVE_PLUGIN_FILE') || defined('TVE_EDITOR_URL')){
            $editors[] = 'Thrive';
        }
        if(class_exists('ThemifyBuilder_Data_Manager')){
            $editors[] = 'Themify';
        }
        if(defined('MFN_THEME_VERSION')){
            $editors[] = 'Muffin';
        }
        if(defined('AV_FRAMEWORK_VERSION')){
            $editors[] = 'Enfold';
        }
        if(class_exists('Cornerstone_Plugin')){
            $editors[] = 'Cornerstone';
        }
        if(defined('WPRM_POST_TYPE') && in_array('wprm_recipe', Wpil_Settings::getPostTypes())){
            $editors[] = 'WPRecipe';
        }
        if(defined('GDLR_CORE_LOCAL')){
            $editors[] = 'Goodlayers';
        }
        
        return $editors;
    }

    /**
     * Gets the meta keys for content areas created with page builders so we can search the database for content.
     * @return array
     **/
    public static function get_builder_meta_keys(){
        $builder_meta = array();
        // if Goodlayers is active
        if(defined('GDLR_CORE_LOCAL')){
            $builder_meta[] = 'gdlr-core-page-builder';
        }
        // if Themify builder is active
        if(class_exists('ThemifyBuilder_Data_Manager')){
            $builder_meta[] = '_themify_builder_settings_json';
        }
        // if Oxygen is active
        if(defined('CT_VERSION')){
            $builder_meta[] = 'ct_builder_shortcodes';
        }
        // if Muffin is active
        if(defined('MFN_THEME_VERSION')){
            $builder_meta[] = 'mfn-page-items-seo';
        }
        // if "Thrive" is active
        if(defined('TVE_PLUGIN_FILE') || defined('TVE_EDITOR_URL')){
            $builder_meta[] = 'tve_updated_post';
        }
        // if Elementor is active
        if(defined('ELEMENTOR_VERSION')){
            $builder_meta[] = '_elementor_data';
        }

        return $builder_meta;
    }

    /**
     * Makes sure all single and double qoutes are excaped once in the supplied text.
     * @param string $text The text that needs to have it's quotes escaped
     * @return string $text The updated text with the single and double qoutes escaped
     **/
    public static function normalize_slashes($text){
        $old_text = $text;
        // add slashes to the single qoutes
        $text = mb_eregi_replace("(?<!\\\\)'", "\'", $text);
        // add slashes to the double qoutes
        $text = mb_eregi_replace('(?<!\\\\)"', '\"', $text);
        // and return the text if it's not empty
        return !empty($text) ? $text: $old_text; // if it is empty, return the original text
    }

    /**
     * Checks to make sure that the supplied links anchor + url don't already show up in the post content.
     * We need this when inserting links from Gutenberg since they effectively shoot two processing runs, and any post insert link changes will cause the other cheks to fail
     * @param string $content The content we're about to stick a link in
     * @return string $changed_sentence The link we're about to stick in the content
     **/
    public static function check_for_url_anchor($content, $changed_sentence){
        if(empty($content) || empty($changed_sentence)){
            return false; // because why not!
        }

        // first, rip the url && anchor text from the link we are about to insert
        if(!preg_match('~<a\b[^>]*\bhref\s*=\s*(["\'])(.*?)\1[^>]*>(.*?)</a>~is', $changed_sentence, $m)){
            return false;
        }

        $target_url = rtrim(html_entity_decode($m[2], ENT_QUOTES, 'UTF-8'), '/');
        $target_anchor = preg_replace('/\s+/u', ' ', trim(strip_tags(html_entity_decode($m[3], ENT_QUOTES, 'UTF-8'))));

        // next, all the links that are currently in the content (in all their happy fun flavors!)
        if(!preg_match_all('~(?:<a\b[\s\S]*?</a>|&lt;a\b[\s\S]*?&lt;/a&gt;|\\\\u003ca\b[\s\S]*?(?:\\\\u003c\/a\\\\u003e|\\\\u003c/a\\\\u003e))~is', $content, $all) || empty($all[0])){
            return false;
        }

        // now go over each link, and checkc to see if any match our link!
        foreach($all[0] as $frag){

            // normalize for our sanity
            // turn everything into normal brackets
            $decoded = str_replace(array('\\u003c', '\\u003e'), array('<', '>'), html_entity_decode($frag, ENT_QUOTES, 'UTF-8'));
            // clean up json escapes
            $decoded = str_replace(array('\\"', "\\'", '\\/'), array('"', "'", '/'), $decoded);

            // make sure that we actually havea link
            if(!preg_match('~<a\b[^>]*\b(?:href|url)\s*=\s*(["\'])(.*?)\1[^>]*>(.*?)</a>~is', $decoded, $lm)){
                continue;
            }

            // normalize again like we did abouve
            $url  = rtrim(html_entity_decode($lm[2], ENT_QUOTES, 'UTF-8'), '/');
            $anchor = preg_replace('/\s+/u', ' ', trim(strip_tags(html_entity_decode($lm[3], ENT_QUOTES, 'UTF-8'))));

            // and see if we're the same
            if($url === $target_url && $anchor === $target_anchor){
                // if we are, yay!
                return true;
            }
        }

        // if we've made it this far, the link doens't existing in the content
        return false;
    }


    /**
     * Makes sure that the supplied sentence uses the same slashing as the rest of the post content.
     * Does a check to see if the sentence to insert is custom sentence since we're not able to normalize there without guessing
     * @param string $text The text that needs to have it's quotes escaped
     * @param string $compare_sentence A string (sentence or changed_sentence) to compare with so we can tell if this is a custom sentence
     * @param int $start The beginning of the replace_text where the link is supposed to be inserted
     * @param int $end The end of the replace_text where the link is supposed to be inserted
     * @param string $content The content that we're looking to insert the link into
     * @return string $text The updated text with the single and double qoutes escaped
     **/
    public static function adjust_sentence_slashes($sentence = '', $compare_sentence = '', $start = 0, $end = 0, $content = ''){
        if(empty($sentence) || empty($compare_sentence) || empty($content)){
            return $sentence;
        }

        // get the slightly cleaned up versions of the sentence and changed sentence
        $cleaned_sentence = trim(strip_tags($sentence));
        $cleaned_compare_sentence = trim(strip_tags($compare_sentence));
        $insert_text = mb_substr($content, $start, ($end - $start));

        // first,  
        if( (self::normalize_slashes($cleaned_sentence) !== self::normalize_slashes($cleaned_compare_sentence)) || // or if this is a modified sentence
            empty($insert_text) || // or there is no insert text
            false === Wpil_Word::mb_strpos(self::normalize_slashes($insert_text), self::normalize_slashes($sentence))) // or if the sentence is not in the search range
        {
            // if any of these are true, return the search sentence now
            return $sentence;
        }

        // search the insert text for a basis to work with
        preg_match_all('/\\\*[\'"]/', $insert_text, $matches);

        $q1_count = false;
        $q2_count = false;
        if(!empty($matches)){
            foreach($matches[0] as $match){
                if($q1_count === false && false !== strpos($match, "'")){
                    $q1_count = substr_count($match, '\\');
                }elseif($q2_count === false && false !== strpos($match, '"')){
                    $q2_count = substr_count($match, '\\');
                }
            }
        }

        // search the content to see what the html tags are usually slashed as
        preg_match_all('/(?:(?:href|class)=([\\\]*["\'])([^"]*?)[\\\]*[\'"])[^<>]*?/', $content, $matches);
        $html_quote = false;
        $quote_avg = array();
        if(!empty($matches)){

            foreach($matches[1] as $match){
                $q_count = substr_count($match, '\\');
                if(!isset($quote_avg[$q_count])){
                    $quote_avg[$q_count] = 1;
                }else{
                    $quote_avg[$q_count]++;
                }
            }

            if(!empty($quote_avg)){
                arsort($quote_avg);
                $html_quote = key($quote_avg);
            }
        }

        // base64 encode any tags so we don't break them by adding slashes
        $sentence = preg_replace_callback('|(<([a-zA-Z]*).*?>)(.*?)(<\/\2>)|i', function($i){ 
            $i[0] = str_replace($i[1], 'wpil-html-replace_' . base64_encode($i[1]) . '_wpil-html-replace', $i[0]);
            $i[0] = str_replace($i[4], 'wpil-html-replace_' . base64_encode($i[4]) . '_wpil-html-replace', $i[0]);
            return $i[0]; 
        }, $sentence);

        // reslash the changed sentence using the slash data that we've pulled about the post
        if($q1_count !== false){
            $replace = "'";
            for($i = 0; $i < $q1_count; $i++){
                $replace = ('\\' . $replace);
            }
            $sentence = preg_replace("/\\\*'/", $replace, $sentence);
        }
        if($q2_count !== false){
            $replace = '"';
            for($i = 0; $i < $q2_count; $i++){
                $replace = ('\\' . $replace);
            }
            $sentence = preg_replace('/\\\*"/', $replace, $sentence);
        }

        // decode all the encoded elements and do any slashing that's required
        if(false !== strpos($sentence, 'wpil-html-replace_')){
            $sentence = preg_replace_callback('`wpil-html-replace_(([A-Za-z0-9+\/]{4})*([A-Za-z0-9+\/]{3}=|[A-Za-z0-9+\/]{2}==)?)_wpil-html-replace`i', function($i) use ($html_quote){
                $html_content = base64_decode($i[1]);

                if(false !== $html_quote){
                    $replace = '"';
                    for($j = 0; $j < $html_quote; $j++){
                        $replace = ('\\' . $replace);
                    }
                    $html_content = preg_replace('/\\\*"/', $replace, $html_content);
                }

                return str_replace($i[0], $html_content, $i[0]);
            }, $sentence);
        }

        return $sentence;
    }

    /**
     * Checks to see if the link is inside a JSON element and adds appropriate escaping for it.
     * @param string $content The post content that we're processing.
     * @param string $changed_sentence The sentence that contains the link that we might be escaping.
     * @param int $position_start The starting character position of the SENTENCE that we'll be replacing. (Sentence contains the link)
     * @param int $position_start The final character position of the SENTENCE that we'll be replacing. (Sentence contains the link)
     **/
    public static function processLinkForJsonContent($content, $changed_sentence, $position_start = 0, $position_end = 0){
        // if Block Lab is active
        if(class_exists('Block_Lab\\Component_Abstract')){
            // check if the content is inside a block
            $len = strlen($content);
            $block_start = strrpos($content, '<!-- wp:block-lab', ($position_start - $len));
            $block_end = strrpos($content, '/-->', ($position_start - $len));

            // if the replace is inside a block, the opening block tag will be closer than the closing tag of whatever block came before.
            if($block_start > $block_end){
                // return double slashed content for the block labs block
                return addslashes(addslashes($changed_sentence));
            }
        }

        // todo continue work on a later date
        if(false){
            preg_match_all('#(?:<!-- wp:[a-zA-Z\/_\-1-9]*? )({(?:.*?)})(?: \/-->)#u', $content, $matches, PREG_OFFSET_CAPTURE);

            if(isset($matches[1]) && !empty($matches[1])){
                foreach($matches[1] as $match){

                    // if the changed sentence is within a json data object
                    if($match[1] < $position_start && ($match[1] + mb_strlen($match[0]) > $position_end)){
                        // try to detect any pre-existing html tags so we can get an idea of the formatting used

                        // regex to pull out html tags. Check over any tags and see how their content is encoded for storage so we can tell how to handle it
                        /***
                         * (<([a-z])*? ).*?>(.*?)<\/\2>|(\\u003c([a-z])*? ).*?\\u003e(.*?)\\u003c\/\5\\u003e 
                         *
                         ****/



                        // return true
                        return true;
                    }elseif($match[1] > $position_end){
                        // if we're searching objects beyond the sentence, exit
                        break;
                    }
                }
            }
        }

        // by default, content is not json.
        return $changed_sentence;
    }

    /**
     * Gets the most recent revision id for the current post.
     * @param int|string|$post The id or post object that we want to get the revsions for
     * @return int|false The id of the most recent revsion or false if we couldn't find it.
     **/
    public static function get_most_recent_revision_id($post_id = 0){
        if(empty($post_id)){
            return false;
        }

        $revisions = wp_get_post_revisions($post_id);

        if(empty($revisions) || !is_array($revisions)){
            return false;
        }

        $latest = 0;
        foreach($revisions as $revision){
            if($latest < $revision->ID){
                $latest = $revision->ID;
            }
        }

        return $latest;
    }

    /**
     * Checks to see if the recently saved Divi content without links is the same as the Divi content submitted with the edit form.
     * What happens normally is we save the links before Divi has a chance to save it's content.
     * Then after Divi saves it's content, it checks to see if the 
     **/
    public static function verify_divi_save_status($content_saved){
        global $wpdb;

        if($content_saved){
            return $content_saved;
        }

        $post_id = absint( $_POST['post_id'] );
        $saved_post = get_post( $post_id );
        $current_gmt = current_time('mysql', true);

        // if it's been longer than 5 minutes since the post was last updated
        if( empty($saved_post) ||
            !isset($saved_post->post_modified_gmt) ||
            abs(strtotime($saved_post->post_modified_gmt) - strtotime($current_gmt)) > 300)
        {
            // return the current state of content savedness because it's unlikely that the post has been updated
            return $content_saved;
        }

        $layout_type = isset( $_POST['layout_type'] ) ? sanitize_text_field( $_POST['layout_type'] ) : '';
        $shortcode_data = json_decode( stripslashes( $_POST['modules'] ), true );
        $post_content = et_fb_process_to_shortcode( $shortcode_data, $_POST['options'], $layout_type );
        $sanitized_content = sanitize_post_field( 'post_content', $post_content, $post_id, 'db' );

		$saved_post_content   = $saved_post->post_content;
		$builder_post_content = stripslashes( $sanitized_content );

		if ( 'utf8' === $wpdb->get_col_charset( $wpdb->posts, 'post_content' ) ) {
			$builder_post_content = wp_encode_emoji( $builder_post_content );
		}

        $unlinked_saved_post_content = Wpil_Link::remove_all_links_from_text($saved_post_content);
        $unlinked_builder_post_content = Wpil_Link::remove_all_links_from_text($builder_post_content);

        $saved_verification = $unlinked_saved_post_content === $unlinked_builder_post_content;

        return $saved_verification;
    }

    /**
     * Checks to make sure that the given post's type is one that we've created special support for.
     * Returns the post's type if it is, and false if it's not
     * @param WP_Post|Wpil_Model_Post|Array of WP post data
     * @return string|bool Post name if it's one that we have special support for. False if we don't
     **/
    public static function get_post_processing_type($post){
        if(empty($post)){
            return false;
        }
        $type = '';

        if(is_a($post, 'WP_Post') && isset($post->post_type)){
            $type = $post->post_type;
        }elseif(is_a($post, 'Wpil_Model_Post') && $post->type === 'post'){
            $type = get_post_type($post->id);
        }elseif(is_array($post) && isset($post['post_type']) && !empty($post['post_type'])){
            $type = $post['post_type'];
        }

        if(empty($type)){
            return false;
        }

        // now go over a list of the post types that we've built support for.
        // If one of their identifying constants or classes is active, we know we're not dealing with a duplicate named PT
        switch($type){
            case 'web-story':
                if(defined('WEBSTORIES_VERSION')){
                    return $type;
                }
            break;
            default:

            break;
        }

        return false;
    }

    /**
     * Helper function for checking if the supplied string is json
     * @return bool
     **/
    public static function is_json($str) {
        if(empty($str) || !is_string($str)){
            return false;
        }
        $json = json_decode($str);
        return $json && $str != $json;
    }

    /**
     * Function to fetch the primary term for the main hierarchical taxonomy of a post
     * @param $post_id
     * @param $post_type
     * @return mixed|null
     */
    public static function get_primary_term_for_main_taxonomy($post_id, $post_type) {
        // First, get the main hierarchical taxonomy
        $taxonomy = self::get_main_hierarchical_taxonomy($post_type, $post_id);

        // Check if Yoast SEO's Primary Term functionality exists
        if (class_exists('WPSEO_Primary_Term')) {
            $primary_term = new WPSEO_Primary_Term($taxonomy, $post_id);
            $primary_term_id = $primary_term->get_primary_term();
            if ($primary_term_id && !is_wp_error($primary_term_id)) {
                $term = get_term($primary_term_id, $taxonomy);
                if ($term && !is_wp_error($term)) {
                    return $term;
                }
            }
        }

        // Use the first term if no primary term is set
        $terms = wp_get_post_terms($post_id, $taxonomy);

        return !empty($terms) ? $terms[0] : null;
    }

    /**
     * Function to get the main hierarchical taxonomy for a post type
     * @param $post_type
     * @param $post_id
     * @return string
     */
    public static function get_main_hierarchical_taxonomy($post_type, $post_id) {
        // check for 'post' post type
        if ($post_type === 'post') {
            $categories = get_the_terms($post_id, 'category');
            if ($categories && !is_wp_error($categories)) {
                // Check if any term is not 'uncategorized'
                foreach ($categories as $category) {
                    if ($category->slug !== 'uncategorized') {
                        return 'category';
                    }
                }
            }
            // If only 'uncategorized' exists, continue to other taxonomies
        }

        // Check all taxonomies associated with the post type
        $taxonomies = get_object_taxonomies($post_type, 'objects');
        foreach ($taxonomies as $taxonomy) {
            // Check for hierarchical taxonomies
            if ($taxonomy->hierarchical) {
                // check for the 'category' taxonomy
                if ($taxonomy->name === 'category') {
                    $terms = get_the_terms($post_id, $taxonomy->name);
                    if ($terms && !is_wp_error($terms)) {
                        $has_valid_term = false;
                        foreach ($terms as $term) {
                            if ($term->slug !== 'uncategorized') {
                                $has_valid_term = true;
                                break;
                            }
                        }
                        if ($has_valid_term) {
                            return $taxonomy->name;
                        }
                    }
                    // If only 'uncategorized' exists, proceed to the next taxonomy
                    continue;
                }

                // For other hierarchical taxonomies, simply return the first one found
                return $taxonomy->name;
            }
        }

        return 'category';
    }

    /**
     * Gets a detailed list of ids...
     * Focussing on posts sincce 97% of people will be using those
     **/
    public static function get_money_pages(){
        $list = array(); // yay! our list of page ids!

        // first, go for the easy ones
        // get the ones that the user has set!
        $list = array_merge($list, get_option('wpil_pillar_content_post_ids', []));

        // quit now if the user has defined money pages
        if(!empty($list)){
            return $list;
        }

        // pull cornerstone from Yoast if it's set
        $list = array_merge($list, Wpil_Toolbox::get_cornerstone_ids());

        // and pillar from Rank Math // because they're totally different!
        $list = array_merge($list, Wpil_Toolbox::get_pillar_content_ids());

        // todo: pull in other seo plugins

        // next, pull what we can from the menus
        $menu_links = self::get_nav_menu_items();

        // if we've got something
        if(!empty($menu_links)){
            // add it to the list
            foreach($menu_links as $link){
                $list[] = $link['object_id'];
            }
        }
        
        // next, lets analyse linking trends to see if there are any posts that stand out...
        $list = array_merge($list, Wpil_Report::get_link_targetted_pages());
        // if there are, add them to the list

        // next, check for forms... // later...
        // add the non help forms to the list

        // try pulling in any e-com pages // yeah, later too...

        // sift out the garbage
        $ignored = Wpil_Settings::get_completely_ignored_pages();
        if(!empty($ignored)){
            $ignore_posts = array();
            foreach($ignored as $pid){
                $bits = explode('_', $pid);
                if(!empty($bits) && $bits[1] === 'post'){
                    $ignore_posts[] = $bits[0];
                }
            }

            if(!empty($ignore_posts)){
                $list = array_diff($list, $ignore_posts); 
            }
        }

        // if we've got posts
        if(!empty($list)){
            // filter to get the uniques
            $list = array_keys(array_flip($list)); // and we're left with profit!
        }

        return $list; // profit!
    }

    /**
     * Returns all nav menu items for the site.
     * Makes sure to only give us post objects that can be traced.
     *
     * @return array
     */
    public static function get_nav_menu_items(){
        $results = [];
        $seen    = [];

        // location => term_id
        $locations = get_nav_menu_locations();
        $location_by_term_id = [];
        foreach ($locations as $loc => $term_id) {
            $location_by_term_id[(int) $term_id] = (string) $loc;
        }

        // Pull menus that exist on the site.
        $menus = wp_get_nav_menus(); // array of WP_Term objects

        foreach ($menus as $menu_term) {
            $menu_id   = (int) $menu_term->term_id;
            $menu_name = (string) $menu_term->name;
            $location  = isset($location_by_term_id[$menu_id]) ? $location_by_term_id[$menu_id]: null;

            // Get items for this menu.
            $items = wp_get_nav_menu_items($menu_id);
            if (empty($items) || is_wp_error($items)) {
                continue;
            }

            foreach ($items as $item) {
                // $item is WP_Post with extra properties
                $url = isset($item->url) ? trim((string) $item->url) : '';
                if($url === '' || $url === '#' || !Wpil_Link::isInternal($url)){ 
                    continue;
                }

                $key = strtolower($url);
                if(isset($seen[$key])){
                    continue;
                }
                $seen[$key] = true;

                if($item->type === 'custom'){
                    $post = self::getPostByLink($url);
                    if(!empty($post) && $post->type === 'post'){
                        $object_id = $post->id;
                    }else{
                        continue;
                    }
                }else{
                    $object_id = $item->object_id;
                }

                $results[] = [
                    'menu'      => $menu_name,
                    'location'  => $location,               // null if not assigned to a theme location
                    'item_id'   => (int) $item->ID,
                    'title'     => (string) $item->title,
                    'url'       => $url,                    // already the resolved URL WP outputs in menus
                    'type'      => (string) $item->type,    // 'post_type', 'taxonomy', 'custom', etc.
                    'object'    => (string) $item->object,  // 'page', 'category', etc.
                    'object_id' => (int) $object_id,
                ];
            }
        }

        return $results;
    }

    /**
     * Main entry. Returns best about page candidate or null.
     *
     * @return array|null { post_id, url, score, reasons }
     */
    public static function find_about_page(){
        $candidates = self::find_about_page_candidates();
        if(empty($candidates)){
            return null;
        }

        $scored = [];
        foreach($candidates as $candidate){
            $scored[] = self::score_about_page_candidate($candidate);
        }

        usort($scored, function($a, $b){
            $a_score = isset($a['score']) ? $a['score'] : 0;
            $b_score = isset($b['score']) ? $b['score'] : 0;

            if($a_score == $b_score){
                return 0;
            }

            return ($a_score < $b_score) ? 1 : -1;
        });

        $best = (isset($scored[0]) && !empty($scored[0])) ? $scored[0]: null;

        // Threshold to avoid returning nonsense.
        if (!$best || !isset($best['score']) || $best['score'] < 8) {
            return null;
        }

        return $best;
    }

    /**
     * Collect candidates from menus, homepage header/footer scrape, and fallback published pages.
     *
     * @return array[] Each item: { url, anchor_text, source, post_id }
     */
    public static function find_about_page_candidates() {
        $home_url = home_url('/');
        $home_host = parse_url($home_url, PHP_URL_HOST);

        $raw_candidates = [];

        // 1) Menus
        $raw_candidates = array_merge($raw_candidates, self::pull_nav_menu_links());

        // 2) Homepage scrape for header and footer links
        $raw_candidates = array_merge($raw_candidates, self::pull_links_from_homepage());

        // 3) Fallback: pages list (helps if homepage is locked down or minimal)
        $pages = get_posts([
            'post_type' => 'page',
            'post_status' => 'publish',
            'posts_per_page' => 200,
            'orderby' => 'menu_order',
            'order' => 'ASC',
            'fields' => 'ids',
            'no_found_rows' => true,
        ]);

        if(!empty($pages)){
            foreach ($pages as $page_id) {
                $url = get_permalink($page_id);
                if (!$url) {
                    continue;
                }
                $raw_candidates[] = [
                    'url' => $url,
                    'anchor_text' => get_the_title($page_id),
                    'source' => 'published_pages',
                    'post_id' => (int) $page_id,
                ];
            }
        }

        // Normalize and dedupe
        $dedup = [];
        foreach ($raw_candidates as $cand) {
            $url = isset($cand['url']) ? trim($cand['url']) : '';
            if ($url === '') {
                continue;
            }

            // Only internal links
            $host = parse_url($url, PHP_URL_HOST);
            if ($host && $home_host && strcasecmp($host, $home_host) !== 0) {
                continue;
            }

            // Normalize to absolute
            if (strpos($url, '//') === 0) {
                $url = (is_ssl() ? 'https:' : 'http:') . $url;
            } elseif (strpos($url, 'http') !== 0) {
                $url = home_url($url);
            }

            // Remove fragments
            $url = preg_replace('/#.*$/', '', $url);

            // Skip empty or homepage
            if ($url === $home_url) {
                continue;
            }

            // Resolve to post id when possible
            if(!isset($cand['post_id']) || empty($cand['post_id'])){
                $cand['post_id'] = self::getPostByLink($url);
            }

            $cand['url'] = $url; // make sure the url is fully normalized

            $key = strtolower($url);
            if (!isset($dedup[$key])) {
                $dedup[$key] = $cand;
            } else {
                // Prefer candidates that have anchor_text or better source
                if (empty($dedup[$key]['anchor_text']) && !empty($cand['anchor_text'])) {
                    $dedup[$key]['anchor_text'] = $cand['anchor_text'];
                }
                if (($dedup[$key]['source'] ?? '') !== 'homepage_header_footer' && ($cand['source'] ?? '') === 'homepage_header_footer') {
                    $dedup[$key]['source'] = $cand['source'];
                }
            }
        }

        return array_values($dedup);
    }

    private static function pull_nav_menu_links() {
        $out = [];

        $pages = self::get_nav_menu_items();
        if (!is_array($pages) || empty($pages)) {
            return $out;
        }

        foreach($pages as $page){
            $out[] = array_merge([
                'anchor_text' => $page['title'],
                'source' => 'menu_' . $page['location'],
                'post_id' => $page['object_id']
            ], $page);
        }

        return $out;
    }

    private static function pull_links_from_homepage() {
        $out = [];

        if(!class_exists('DOMDocument')){
            return $out;
        }

        // grabe the homepage directly to ensure we get all the nav links
        $response = wp_remote_get(home_url('/'), [
            'timeout' => 10,
            'redirection' => 5,
            'user-agent' => WPIL_DATA_USER_AGENT,
            'headers' => [
                'Accept' => 'text/html,application/xhtml+xml',
            ],
        ]);

        if(is_wp_error($response)){
            return $out;
        }

        $html = (string) wp_remote_retrieve_body($response);
        if($html === ''){
            return $out;
        }

        // iuf our page is really long
        if(strlen($html) > 1500000){
            // trime it
            $html = substr($html, 0, 1500000);
        }

        // since we're only checking one page, we'll cheat with domdoc
        libxml_use_internal_errors(true);
        $dom = new DOMDocument();
        $dom->loadHTML($html);
        libxml_clear_errors();

        $xpath = new DOMXPath($dom);

        // look for the normal nave areas as well as any page builder sections standing in for them
        $contexts = [
            '//header//a[@href]',
            '//footer//a[@href]',
            '//nav//a[@href]',
            "//*[contains(translate(@id,'ABCDEFGHIJKLMNOPQRSTUVWXYZ','abcdefghijklmnopqrstuvwxyz'),'header') or contains(translate(@class,'ABCDEFGHIJKLMNOPQRSTUVWXYZ','abcdefghijklmnopqrstuvwxyz'),'header')]//a[@href]",
            "//*[contains(translate(@id,'ABCDEFGHIJKLMNOPQRSTUVWXYZ','abcdefghijklmnopqrstuvwxyz'),'footer') or contains(translate(@class,'ABCDEFGHIJKLMNOPQRSTUVWXYZ','abcdefghijklmnopqrstuvwxyz'),'footer')]//a[@href]",
            "//*[contains(translate(@id,'ABCDEFGHIJKLMNOPQRSTUVWXYZ','abcdefghijklmnopqrstuvwxyz'),'menu') or contains(translate(@class,'ABCDEFGHIJKLMNOPQRSTUVWXYZ','abcdefghijklmnopqrstuvwxyz'),'menu')]//a[@href]",
        ];

        $seen = [];
        foreach($contexts as $query){
            $nodes = $xpath->query($query);
            if(!$nodes){
                continue;
            }
            foreach($nodes as $a){
                $href = $a->getAttribute('href');
                if (!$href) {
                    continue;
                }
                $text = trim($a->textContent ?? '');
                $key = strtolower($href . '|' . $text);
                if (isset($seen[$key])) {
                    continue;
                }
                $seen[$key] = true;

                $out[] = [
                    'url' => $href,
                    'anchor_text' => $text,
                    'source' => 'homepage_header_footer',
                ];
            }
        }

        return $out;
    }

    function score_about_page_candidate(array $candidate) {
        $url = (string) ($candidate['url'] ?? '');
        $anchor_text = (string) ($candidate['anchor_text'] ?? '');
        $source = (string) ($candidate['source'] ?? '');
        $post_id = (int) ($candidate['post_id'] ?? 0);

        $reasons = [];
        $score = 0;

        $about_keywords = self::get_standard_about_page_keywords();
        $negative_keywords = self::get_standard_non_about_page_keywords();

        $url_path = strtolower((string) parse_url($url, PHP_URL_PATH));
        $url_host = strtolower((string) parse_url($url, PHP_URL_HOST));

        // Skip obvious non content pages
        foreach ($negative_keywords as $bad) {
            if ($bad !== '' && (strpos($url_path, $bad) !== false || strpos(strtolower($anchor_text), $bad) !== false)) {
                $score -= 20;
                $reasons[] = 'negative_keyword:' . $bad;
                break;
            }
        }

        // Keyword match in URL path
        foreach ($about_keywords as $kw) {
            if ($kw !== '' && strpos($url_path, $kw) !== false) {
                $score += 10;
                $reasons[] = 'url_keyword:' . $kw;
                break;
            }
        }

        // Keyword match in anchor text
        $normalized_anchor = self::about_page_normalize_text($anchor_text);
        foreach ($about_keywords as $kw) {
            if ($kw !== '' && strpos($normalized_anchor, $kw) !== false) {
                $score += 8;
                $reasons[] = 'anchor_keyword:' . $kw;
                break;
            }
        }

        // Source weighting
        if (strpos($source, 'menu_') === 0) {
            $score += 2;
            $reasons[] = 'source:menu';
        }
        if ($source === 'homepage_header_footer') {
            $score += 3;
            $reasons[] = 'source:homepage_layout';
        }

        // If we can resolve to a page, use title and template signals
        if ($post_id > 0) {
            $post = get_post($post_id);
            if ($post && $post->post_type === 'page' && $post->post_status === 'publish') {
                $score += 2;
                $reasons[] = 'is_page';

                $title_norm = self::about_page_normalize_text(get_the_title($post_id));
                foreach ($about_keywords as $kw) {
                    if ($kw !== '' && strpos($title_norm, $kw) !== false) {
                        $score += 8;
                        $reasons[] = 'title_keyword:' . $kw;
                        break;
                    }
                }

                $template = (string) get_page_template_slug($post_id);
                $template_norm = self::about_page_normalize_text($template);
                if ($template_norm && strpos($template_norm, 'about') !== false) {
                    $score += 6;
                    $reasons[] = 'template_mentions_about';
                }

                // Many about pages are top level pages
                if ((int) $post->post_parent === 0) {
                    $score += 1;
                    $reasons[] = 'top_level_page';
                }
            }
        }

        // Prefer shorter paths that look like a page slug
        $segments = array_values(array_filter(explode('/', trim($url_path, '/'))));
        if (count($segments) === 1) {
            $score += 1;
            $reasons[] = 'single_segment_path';
        }

        // Penalize media files and feeds
        if (preg_match('/\.(jpg|jpeg|png|gif|webp|pdf|zip|xml)$/i', $url_path)) {
            $score -= 15;
            $reasons[] = 'file_like_url';
        }
        if (strpos($url_path, '/feed') !== false) {
            $score -= 10;
            $reasons[] = 'feed_url';
        }

        return [
            'post_id' => $post_id,
            'url' => $url,
            'score' => $score,
            'reasons' => $reasons,
            'source' => $source,
            'anchor_text' => $anchor_text,
        ];
    }

    private static function about_page_normalize_text($text) {
        $text = (string) $text;
        $text = html_entity_decode($text, ENT_QUOTES | ENT_HTML5, 'UTF-8');
        $text = strtolower($text);
        $text = preg_replace('/\s+/', ' ', $text);
        $text = trim($text);

        // Remove punctuation but keep unicode letters and numbers
        $text = preg_replace('/[^\p{L}\p{N}\s]+/u', ' ', $text);
        $text = preg_replace('/\s+/', ' ', $text);
        return trim($text);
    }

    private static function get_standard_about_page_keywords() {
        return [
            // English
            'about', 'about-us', 'aboutus', 'about-me', 'aboutme',
            'about-the-author', 'about-the-company', 'about-the-site',
            'our-story', 'our team', 'our-team', 'meet-the-team', 'meet our team',
            'who-we-are', 'who we are', 'who-i-am', 'who i am', 'mission',

            // Spanish (Español)
            'acerca', 'acerca-de', 'acerca de',
            'acerca-de-nosotros', 'acerca de nosotros',
            'acerca-de-mi', 'acerca de mi', 'acerca-de-mí', 'acerca de mí',
            'sobre', 'sobre-mi', 'sobre mi', 'sobre-mí', 'sobre mí',
            'sobre-nosotros', 'sobre nosotros', 'sobre-nosotras', 'sobre nosotras',
            'quienes-somos', 'quienes somos',
            'quiénes-somos', 'quiénes somos',
            'nuestra-historia', 'nuestra historia',
            'nuestro-equipo', 'nuestro equipo',
            'conocenos', 'conócenos',

            // French (Français)
            'a-propos', 'a propos',
            'a-propos-de-nous', 'a propos de nous',
            'a-propos-de-moi', 'a propos de moi',
            'apropos',
            'qui-sommes-nous', 'qui sommes nous',
            'notre-histoire', 'notre histoire',
            'notre-equipe', 'notre équipe',
            'l-equipe', 'l equipe',

            // German (Deutsch)
            'uber-uns', 'ueber-uns', 'über-uns',
            'uber uns', 'ueber uns', 'über uns',
            'uber-mich', 'ueber-mich', 'über-mich',
            'uber mich', 'ueber mich', 'über mich',
            'uber-das-unternehmen', 'ueber-das-unternehmen', 'über-das-unternehmen',
            'unsere-geschichte', 'unsere geschichte',
            'unser-team', 'unser team',

            // Russian (Русский)
            'о нас', 'обо мне', 'о компании', 'о проекте', 'о фирме', 'о магазине',
            'o-nas', 'o nas',
            'o-kompanii', 'o kompanii',
            'o-proekte', 'o proekte',

            // Portuguese (Português)
            'sobre-nos', 'sobre nos', 'sobre-nós', 'sobre nós',
            'sobre-mim', 'sobre mim',
            'quem-somos', 'quem somos',
            'nossa-historia', 'nossa história',
            'nossa-equipe', 'nossa equipe',

            // Dutch (Dutch)
            'over-ons', 'over ons',
            'over-mij', 'over mij',
            'ons-verhaal', 'ons verhaal',
            'ons-team', 'ons team',
            'wie-zijn-wij', 'wie zijn wij',

            // Danish / Norwegian / Swedish (Dansk / Norsk bokmål / Svenska)
            'om-oss', 'om oss',
            'om-os', 'om os',
            'om-meg', 'om meg',
            'om-mig', 'om mig',
            'om-foretaget', 'om foretaget',
            'om-foretag', 'om foretag',
            'om-bedriften', 'om bedriften',
            'om-selskapet', 'om selskapet',
            'om-virksomheden', 'om virksomheden',
            'om-virksomheten', 'om virksomheten',
            'var-historia', 'vår-historie', 'vår historik', 'var historie',
            'mot-teamet', 'möt teamet',
            'møte-teamet', 'mote teamet',

            // Italian (Italiano)
            'chi-siamo', 'chi siamo',
            'su-di-noi', 'su di noi',
            'su-di-me', 'su di me',
            'la-nostra-storia', 'la nostra storia',
            'il-nostro-team', 'il nostro team',

            // Polish (Polskie)
            'o-nas', 'o nas',
            'o-mnie', 'o mnie',
            'o-firmie', 'o firmie',
            'nasz-zespol', 'nasz zespol', 'nasz-zespół', 'nasz zespół',
            'kim-jestesmy', 'kim jestesmy', 'kim jesteśmy',

            // Slovak (Slovenčina)
            'o-nas', 'o nás', 'o-nás',
            'o-mne', 'o mne',
            'o-firme', 'o firme',
            'o-spolocnosti', 'o spoločnosti', 'o-spoločnosti',
            'o-projekte', 'o projekte',

            // Arabic (عربي)
            'من نحن', 'عن الشركة', 'عن الموقع', 'نبذة عنا',
            'من-نحن', 'عن-الشركة', 'عن-الموقع', 'نبذة-عنا',

            // Serbian (Српски / srpski)
            'о нама', 'о мени', 'о компанији', 'о фирми', 'о пројекту',
            'o-nama', 'o nama',
            'o-meni', 'o meni',
            'o-kompaniji', 'o kompaniji',
            'o-firmi', 'o firmi',
            'o-projektu', 'o projektu',

            // Finnish (Suomi)
            'meista', 'meistä', 'meistämme',
            'minusta',
            'tietoa',
            'tietoa-meista', 'tietoa meistä',
            'yrityksesta', 'yrityksestä',
            'tietoa-yrityksesta', 'tietoa yrityksestä',

            // Hebrew (עִבְרִית)
            'עלינו', 'על החברה', 'עליי', 'מי אנחנו',
            'על-ינו', 'על-החברה', 'מי-אנחנו',

            // Hindi (हिन्दी)
            'हमारे बारे में', 'मेरे बारे में', 'कंपनी के बारे में',
            'hamare-bare-mein', 'hamare bare mein',
            'mere-bare-mein', 'mere bare mein',
            'company-ke-bare-mein', 'company ke bare mein',

            // Hungarian (Magyar)
            'rólunk', 'rolunk',
            'rólam', 'rolam',
            'csapatunk',
            'kuldetesunk', 'küldetésünk',
            'cégünkről', 'cegunkrol',

            // Romanian (Română)
            'despre', 'despre-noi', 'despre noi',
            'despre-mine', 'despre mine',
            'cine-suntem', 'cine suntem',
            'echipa-noastra', 'echipa noastră',
            'povestea-noastra', 'povestea noastră',

            // Ukrainian (Українська)
            'про нас', 'про мене', 'про компанію', 'про проект', 'про проєкт', 'хто ми',
            'pro-nas', 'pro nas',
            'pro-mene', 'pro mene',
            'pro-kompaniyu', 'pro kompaniyu',
            'pro-kompaniia', 'pro kompaniia',
            'khto-my', 'khto my',

            // Indonesian (Bahasa Indonesia)
            'tentang-kami', 'tentang kami',
            'tentang-saya', 'tentang saya',
            'tentang-perusahaan', 'tentang perusahaan',
            'profil-perusahaan', 'profil perusahaan',
            'kisah-kami', 'kisah kami',

            // Czech (Čeština)
            'o nás', 'o-nas', 'o nas',
            'o mně', 'o-mne', 'o mne',
            'o-spolecnosti', 'o společnosti', 'o spolecnosti',
            'o-projektu', 'o projektu',

            // Bulgarian (български)
            'за нас', 'за мен', 'за компанията', 'за фирмата', 'за проекта',
            'za-nas', 'za nas',
            'za-men', 'za men',
            'za-kompaniyata', 'za kompaniyata',
            'za-firmata', 'za firmata',
            'za-proekta', 'za proekta',

            // Lithuanian (Lietuvių)
            'apie-mus', 'apie mus',
            'apie-mane', 'apie mane',
            'apie-imone', 'apie įmonę', 'apie įmone',
            'apie-projekta', 'apie projektą', 'apie projekta',
            'kas-mes-esame', 'kas mes esame',
            'musu-istorija', 'mūsų istorija', 'musu istorija',
            'musu-komanda', 'mūsų komanda', 'musu komanda',

            // Latvian (Latviešu)
            'par-mums', 'par mums',
            'par-mani', 'par mani',
            'par-uznemumu', 'par uzņēmumu', 'par uznemumu',
            'musu-komanda', 'mūsu komanda', 'musu komanda',
            'musu-stasts', 'mūsu stāsts', 'musu stasts',
            'kas-mes-esam', 'kas mēs esam', 'kas mes esam',

            // Estonian (Eesti)
            'meist',
            'minust',
            'ettevottest', 'ettevõttest', 'ettevoittest',
            'meie-lugu', 'meie lugu',
            'meie-meeskond', 'meie meeskond',
            'kes-me-oleme', 'kes me oleme',
            
            // Greek (someday!)
            'sxetika', 'σχετικα', 'σχετικά',

            // Turkish
            'hakkimizda', 'hakkımızda',

            // Vietnamese (Tiếng Việt)
            'gioi-thieu', 'giới thiệu', 'gioi thieu',
            've-chung-toi', 'về chúng tôi', 've chung toi',
            've-toi', 'về tôi', 've toi',
            've-cong-ty', 'về công ty', 've cong ty',
            'cau-chuyen-cua-chung-toi', 'câu chuyện của chúng tôi', 'cau chuyen cua chung toi',
            'doi-ngu-cua-chung-toi', 'đội ngũ của chúng tôi', 'doi ngu cua chung toi',
            'ai-chung-toi-la', 'ai chúng tôi là', 'ai chung toi la',

            // Japanese (someday!)
            'about', 'プロフィール', '私について', '運営者情報',

            // Chinese (someday!)
            '关于', '关于我们', '关于我',
        ];
    }

    private static function get_standard_non_about_page_keywords() {
        return [
            'privacy', 'privacy policy', 'terms', 'terms of use', 'cookies', 'cookie policy',
            'login', 'sign in', 'signup', 'register', 'account',
            'cart', 'checkout', 'my account',
            'contact', 'support', 'help',
            'refund', 'returns', 'shipping',
            'wp admin', 'wp-login',
        ];
    }




}

