HtmlPurifier for word/wordpress (assuming files are in a direct subdirectory of Wordpress root path)

cleandb.php

<?php

include('../wp-load.php');
include('vendor/autoload.php');

$config = \Symfony\Component\Yaml\Yaml::parse(file_get_contents('config.yml'));
$purifier_config = HTMLPurifier_Config::createDefault();

foreach ($config as $k => $v) {
    $purifier_config->set($k, $v);
}

$purifier = new HTMLPurifier($purifier_config);
$posts = get_posts(['posts_per_page' => -1]);

foreach ($posts as $post) {
    $content = $purifier->purify($post->post_content);
    $content = preg_replace( "/\r|\n/", " ", $content);
    $content = preg_replace("/\s{1,2}/", " ", $content);

    $post->post_content = $content;
    wp_insert_post($post);
}

composer.json

{
    "require": {
        "ezyang/htmlpurifier": "^4.7",
        "symfony/yaml": "^2.7"
    }
}

config.yml

HTML.AllowedElements: 'p, a, ul, ol, li, h1, h2, h3, h4, h5, h6, br, strong, em, b, i, table, tr, td, th, img'

HTML.AllowedAttributes: '*.attr, *.src, *.title, *.alt, table.cellspacing, table.cellpadding, table.align, *.href, *.width, *.height, *.border'

CSS.AllowedProperties: ''

AutoFormat.RemoveEmpty: true

AutoFormat.AutoParagraph: true

AutoFormat.RemoveEmpty.RemoveNbsp: true

AutoFormat.RemoveEmpty: false

AutoFormat.RemoveSpansWithoutAttributes: false