Optimismia.

wp_save_first_image_as_thumb.php

<?php
$dryRun = true;

$posts = get_posts(array("post_type" => (!empty($_GET['type']) ? $_GET['type'] : 'post'), "posts_per_page" => -1)); // post_type is kinda reserved.

foreach($posts as $post){
  $id = $post->ID;

  if(has_post_thumbnail($id)){
    continue;
    // it's already converted.
  }


  $image = '';
  $imageRegex = '/<img[^>]+src=[\'"]([^\'"]+)[\'"][^>]*>/i';

  $nmatches = preg_match_all($imageRegex, $post->post_content, $matches);
  if ($nmatches) {
    $image = $matches[1][0];
      // Additional check for spaces in filename
    $image = str_replace(' ', '%20', $image);
  }

  $dictionary = [
    "/modules/file/icons/application-pdf.png",
    "/modules/file/icons/x-office-document.png"
  ];
  
  if(in_array($image, $dictionary)){
    // Blacklist, if filename is this, don't use it. Skip to next one.
    continue;
  }


  if(empty($image)){
    // This post has no image. Skip to next one!
    continue;
  }

  // Still with me? Time to download the image!


  // But we need to make sure we can download the image first.

  $isExternal = (bool)parse_url($image); // check if it's a local image or an external image.

  if(!$isExternal){
    // It's not external, but we can't download it without url? Let's give it an url!

    if(strpos($image, "/") !== "0"){
      // additional check that it's not relative to our location.
      // If it is relative, we add a slash to make it absolute, but it's not quaranteed to work.

      $image = "/" . $image;
    }

    if(is_multisite()){
      // We need to check if we're running multisite, get_current_blog_id() isn't defined on single sites.
      $image = get_site_url(get_current_blog_id()) . $image;
    }

    else{
      $image = get_site_url() . $image;
    }
  }

  // Ok, now we can download the image!

  if(!$dryRun){
    $tempImage = file_get_contents($image);
    $uploadDir = wp_upload_dir();

    $filePath = $uploadDir['path'] . "/" . basename($image);

    file_put_contents($filePath, $tempImage);


    // straight copypaste from https://codex.wordpress.org/Function_Reference/wp_insert_attachment
    $filename = $filePath;

    // The ID of the post this attachment is for.
    $parent_post_id = $id;

    // Check the type of file. We'll use this as the 'post_mime_type'.
    $filetype = wp_check_filetype( basename( $filename ), null );

    // Get the path to the upload directory.
    $wp_upload_dir = $uploadDir;

    // Prepare an array of post data for the attachment.
    $attachment = array(
      'guid'           => $wp_upload_dir['url'] . '/' . basename( $filename ),
      'post_mime_type' => $filetype['type'],
      'post_title'     => preg_replace( '/\.[^.]+$/', '', basename( $filename ) ),
      'post_content'   => '',
      'post_status'    => 'inherit'
      );

    // Insert the attachment.
    $attach_id = wp_insert_attachment( $attachment, $filename, $parent_post_id );

    // Make sure that this file is included, as wp_generate_attachment_metadata() depends on it.
    require_once( ABSPATH . 'wp-admin/includes/image.php' );

    // Generate the metadata for the attachment, and update the database record.
    $attach_data = wp_generate_attachment_metadata( $attach_id, $filename );
    wp_update_attachment_metadata( $attach_id, $attach_data );

    set_post_thumbnail( $parent_post_id, $attach_id );

    // Tada. Everything *went* as expected, now delete the first image, and save the content.


    $post->post_content = preg_replace($imageRegex, "", $post->post_content, 1);

    wp_update_post($post);
  }

  else{
    // Dry run! Don't actually save anything, but instead show some output.

    $uploadDir = wp_upload_dir();
    $filePath = $uploadDir['path'] . "/" . basename($image);

    $post->post_content = preg_replace($imageRegex, "", $post->post_content, 1);

    echo "Image url: " . PHP_EOL;
    var_dump($image);

    echo PHP_EOL;

    echo "Save path: " . PHP_EOL;
    var_dump($filePath);

    echo PHP_EOL;

    echo "Post data: " . PHP_EOL;
    var_dump($post);


  }



}