Pimcore 11: Script to sanitize WYSIWYG fields data (DataObjects, Documents & Translations) ** IMPORTANT: Not tested properly so please backup your data before upgrade

pimcore11_sanitize_wysiwyg_data.php

<?php
// file should be located in project root

use Pimcore\Tool\Text;

ini_set('max_execution_time', 0);
ini_set("memory_limit", "-1");

include_once __DIR__ . "/vendor/autoload.php";
$bootstrap = new \Pimcore\Bootstrap();
$bootstrap->startup();
$pimcoreWysiwygSanitizer = \Pimcore::getContainer()->get(Text::PIMCORE_WYSIWYG_SANITIZER_ID);

// Sanitize DataObject WYSIWYG fields example
// Adapt as per your Data model to sanitize all persisted WYSIWYG data
echo "Sanitizing DataObject WYSIWYG fields....\n";

$offset = 0;
$limit = 100;

while (true) {
    $list = new \Pimcore\Model\DataObject\Test\Listing();
    $list->setOffset($offset);
    $list->setLimit($limit);
    $list = $list->load();
    if (!$list) {
        break;
    }

    foreach ($list as $object) {
        $save = false;

        //direct WYSIWYG field data
        if ($wysiwygData = $object->getDescription()) {
            $object->setDescription($wysiwygData);
            $save = true;
        }

        //WYSIWYG field in localized fields
        foreach (Pimcore\Tool::getValidLanguages() as $language) {
            if ($localizedWysiwygData = $object->getLocalizedDescription($language)) {
                $object->setLocalizedDescription($language, $localizedWysiwygData);
                $save = true;
            }
        }

        //WYSIWYG field in objectbricks field
        if ($bricksWysiwygData = $object->getBricks()->getTestBrick()?->getBrickDescription()) {
            $object->getBricks()->getTestBrick()->setBrickDescription($bricksWysiwygData);
            $save = true;
        }

        //WYSIWYG field in fieldcollections field
        $collectionArr = $object->getCollections() ?? [];
        /** @var \Pimcore\Model\DataObject\Fieldcollection\Data\TestCollection $collection */
        foreach ($collectionArr as $collection) {
            if ($collectionWysiwygData = $collection->getCollectionDescription()) {
                $collection->setCollectionDescription($collectionWysiwygData);
                $save = true;
            }
        }

        if ($save) {
            echo sprintf("WYSIWYG fields in Object %d\n", $object->getId());
            $object->save(['versionNote' => 'Sanitized WYSIWYG Data']);
        }
    }

    $offset += $limit;
    Pimcore::collectGarbage();
}

//Sanitize Document WYSIWYG fields
echo "\nSanitizing Document WYSIWYG editables....\n";

$offset = 0;
$limit = 100;

while (true) {
    $list = new \Pimcore\Model\Document\Listing();
    $list->setOffset($offset);
    $list->setLimit($limit);
    $list->setCondition('`id` = 46');
    $list = $list->load();
    if (!$list) {
        break;
    }

    /** @var Pimcore\Model\Document $document */
    foreach ($list as $document) {
        $save = false;

        if (!$document instanceof \Pimcore\Model\Document\PageSnippet) {
            continue;
        }

        $editables = $document->getEditables();
        foreach ($editables as $editable) {
            if (!$editable instanceof \Pimcore\Model\Document\Editable\Wysiwyg) {
                continue;
            }

            $editableData = $editable->getData();
            $editable->setDataFromEditmode($editableData);
            $document->setEditable($editable);
            $save = true;
        }

        if ($save) {
            echo sprintf("WYSIWYG data in Document %d\n", $document->getId());
            $document->save(['versionNote' => 'Sanitized WYSIWYG Data']);
        }
    }

    $offset += $limit;
    Pimcore::collectGarbage();
}

//Sanitize Translation WYSIWYG data
echo "\n\nSanitizing Translations WYSIWYG data.....\n";
$offset = 0;
$limit = 100;

while (true) {
    $list = new \Pimcore\Model\Translation\Listing();
    $list->setOffset($offset);
    $list->setLimit($limit);
    $list->setCondition("`key` = 'test'");
    $list = $list->load();
    if (!$list) {
        break;
    }

    foreach ($list as $transItem) {
        $save = false;
        foreach ($transItem->getTranslations() as $lang => $translation) {
            //process only translation containing html tags
            if ($translation != strip_tags(html_entity_decode($translation))) {
                $save = true;
                break;
            }
        }

        if ($save) {
            echo sprintf("WYSIWYG data in Translation: %d\n", $transItem->getKey());
            $transItem->save();
        }
    }

    $offset += $limit;
    Pimcore::collectGarbage();
}

echo "\n\nDone WYSIWYG sanitization.\n";