| 1 |
<?php |
| 2 |
// phpcs:disable Universal.NamingConventions.NoReservedKeywordParameterNames.parentFound |
| 3 |
namespace SubstackImporter; |
| 4 |
|
| 5 |
use WP_Error; |
| 6 |
use WXR_Generator\Generator; |
| 7 |
use ZipArchive; |
| 8 |
use DOMDocument; |
| 9 |
use DomComment; |
| 10 |
use DomElement; |
| 11 |
use DOMText; |
| 12 |
|
| 13 |
|
| 14 |
/** |
| 15 |
* The Substack Converter is responsible for taking in a Substack export and providing |
| 16 |
* data to the WXR generator. |
| 17 |
* |
| 18 |
* @package SubstackImporter |
| 19 |
*/ |
| 20 |
class Converter { |
| 21 |
|
| 22 |
/** |
| 23 |
* @var string $export_file_path Path of the export file. |
| 24 |
*/ |
| 25 |
protected $export_file_path; |
| 26 |
|
| 27 |
/** |
| 28 |
* Instance of the WXR generator |
| 29 |
* @var Generator $generator |
| 30 |
*/ |
| 31 |
protected $generator; |
| 32 |
|
| 33 |
/** |
| 34 |
* Authors. |
| 35 |
* @var array |
| 36 |
*/ |
| 37 |
protected $authors = array(); |
| 38 |
|
| 39 |
/** |
| 40 |
* Categories. |
| 41 |
* |
| 42 |
* @var array |
| 43 |
*/ |
| 44 |
protected $categories = array(); |
| 45 |
|
| 46 |
/** |
| 47 |
* URL of the Substack Newsletter. |
| 48 |
* |
| 49 |
* @var string |
| 50 |
*/ |
| 51 |
protected $substack_url; |
| 52 |
|
| 53 |
/** |
| 54 |
* The classnames of all possible embed nodes in the Substack HTML. |
| 55 |
* |
| 56 |
* @var string[] |
| 57 |
*/ |
| 58 |
protected $supported_embeds = array( |
| 59 |
'tweet', |
| 60 |
'instagram', // No longer supported. |
| 61 |
'youtube-wrap', |
| 62 |
'spotify-wrap', |
| 63 |
'soundcloud-wrap', |
| 64 |
'vimeo-wrap', |
| 65 |
'bandcamp-wrap', // Shortcode embed, |
| 66 |
'github-gist', // Not supported in core, using shortcode embed |
| 67 |
'tiktok-wrap', |
| 68 |
); |
| 69 |
|
| 70 |
|
| 71 |
/** |
| 72 |
* Converter constructor. |
| 73 |
* |
| 74 |
* @param Generator $generator Instance of the WXR Generator. |
| 75 |
* @param string $export_file_path Path to the Substack export zip file. |
| 76 |
* @param null $substack_url URL of the Substack newsletter. |
| 77 |
*/ |
| 78 |
public function __construct( Generator $generator, $export_file_path, $substack_url = null ) { |
| 79 |
$this->generator = $generator; |
| 80 |
$this->export_file_path = $export_file_path; |
| 81 |
$this->substack_url = $substack_url; |
| 82 |
} |
| 83 |
|
| 84 |
/** |
| 85 |
* Convert the Substack export to a WXR. |
| 86 |
* |
| 87 |
* @returns WP_Error|void |
| 88 |
* |
| 89 |
* @throws \OxymelException |
| 90 |
*/ |
| 91 |
public function convert() { |
| 92 |
if ( ! $this->export_file_path || ! file_exists( $this->export_file_path ) ) { |
| 93 |
return new WP_Error( 'export_file_not_exist', 'The export file does not exist' ); |
| 94 |
} |
| 95 |
|
| 96 |
$this->generator->initialize(); |
| 97 |
|
| 98 |
// Add posts. |
| 99 |
$out = $this->add_posts(); |
| 100 |
|
| 101 |
if ( is_wp_error( $out ) ) { |
| 102 |
return $out; |
| 103 |
} |
| 104 |
|
| 105 |
// Add Authors. |
| 106 |
foreach ( $this->authors as $author ) { |
| 107 |
$this->generator->add_author( $author ); |
| 108 |
} |
| 109 |
|
| 110 |
// Add categories. |
| 111 |
foreach ( $this->categories as $category ) { |
| 112 |
$this->generator->add_category( $category ); |
| 113 |
} |
| 114 |
|
| 115 |
$this->generator->finalize(); |
| 116 |
} |
| 117 |
|
| 118 |
/** |
| 119 |
* Lets make sure that the url always has a https protocol. |
| 120 |
* @param null $url |
| 121 |
*/ |
| 122 |
private function ensure_protocol( $url ) { |
| 123 |
if ( is_null( $url ) ) { |
| 124 |
return null; |
| 125 |
} |
| 126 |
|
| 127 |
// Ensure substack_url has a protocol (http or https) |
| 128 |
$parsed_url = parse_url( $url ); |
| 129 |
if ( ! isset( $parsed_url['scheme'] ) ) { |
| 130 |
$url = 'https://' . $url; |
| 131 |
} |
| 132 |
|
| 133 |
return preg_replace( '/^(https?:\/\/)?/', 'https://', $url ); |
| 134 |
} |
| 135 |
|
| 136 |
/** |
| 137 |
* Load additional information retrieved through the Substack API into the export zip file. |
| 138 |
* |
| 139 |
* @param int $offset 0-indexed starting offset for the post to start with. |
| 140 |
* @param int $limit Number of posts to process. |
| 141 |
* |
| 142 |
* @return array|WP_Error |
| 143 |
*/ |
| 144 |
public function load_meta_data( $offset = 0, $limit = 1 ) { |
| 145 |
|
| 146 |
$zip = $this->get_export_zip(); |
| 147 |
|
| 148 |
if ( is_wp_error( $zip ) ) { |
| 149 |
return $zip; |
| 150 |
} |
| 151 |
|
| 152 |
$total_count = 0; |
| 153 |
|
| 154 |
$this->substack_url = $this->ensure_protocol( $this->substack_url ); |
| 155 |
|
| 156 |
foreach ( $this->get_posts() as $idx => $post ) { |
| 157 |
++$total_count; |
| 158 |
|
| 159 |
if ( $idx < $offset || $idx >= $offset + $limit ) { |
| 160 |
continue; |
| 161 |
} |
| 162 |
|
| 163 |
list($id, $slug) = explode( '.', $post['post_id'], 2 ); |
| 164 |
$meta = $this->fetch_post_meta( $slug ); |
| 165 |
|
| 166 |
if ( $meta ) { |
| 167 |
$zip->addFromString( sprintf( 'meta/%s.json', $id ), $meta ); |
| 168 |
} |
| 169 |
} |
| 170 |
|
| 171 |
return array( |
| 172 |
'total' => $total_count, |
| 173 |
'processed' => min( $offset + $limit, $total_count ), |
| 174 |
); |
| 175 |
} |
| 176 |
|
| 177 |
/** |
| 178 |
* Convert each Substack post to a WordPress post and add it to the WXR. |
| 179 |
* |
| 180 |
* @return void|WP_Error |
| 181 |
* |
| 182 |
* @throws \OxymelException |
| 183 |
*/ |
| 184 |
protected function add_posts() { |
| 185 |
|
| 186 |
$posts_generator = $this->get_posts(); |
| 187 |
|
| 188 |
if ( is_wp_error( $posts_generator ) ) { |
| 189 |
return $posts_generator; |
| 190 |
} |
| 191 |
|
| 192 |
foreach ( $posts_generator as $post ) { |
| 193 |
|
| 194 |
$post_id = explode( '.', $post['post_id'], 2 ); |
| 195 |
$id = (int) $post_id[0]; |
| 196 |
$post_meta = $this->get_post_meta_from_export( $id ); |
| 197 |
|
| 198 |
/** |
| 199 |
* Filter the post metadata loaded from the Substack export. |
| 200 |
* |
| 201 |
* Allows modification of the metadata retrieved from the Substack API |
| 202 |
* before it is used for author, comments, and other post data. |
| 203 |
* |
| 204 |
* @since 1.2.0 |
| 205 |
* |
| 206 |
* @param array|null $post_meta The post metadata from the Substack API response. |
| 207 |
* @param array $post The raw Substack post data from the CSV. |
| 208 |
* @param int $id The Substack post ID. |
| 209 |
*/ |
| 210 |
$post_meta = apply_filters( 'substack_importer_post_meta', $post_meta, $post, $id ); |
| 211 |
|
| 212 |
/** |
| 213 |
* Fires before a single Substack post is processed and converted. |
| 214 |
* |
| 215 |
* Useful for setting up state or performing actions before conversion begins. |
| 216 |
* |
| 217 |
* @since 1.2.0 |
| 218 |
* |
| 219 |
* @param array $post The raw Substack post data from the CSV. |
| 220 |
* @param array|null $post_meta The post metadata from the Substack API response. |
| 221 |
* @param int $id The Substack post ID. |
| 222 |
*/ |
| 223 |
do_action( 'substack_importer_before_post', $post, $post_meta, $id ); |
| 224 |
|
| 225 |
if ( ! empty( $post['subtitle'] ) ) { |
| 226 |
$post['html_body'] = $this->add_subtitle( $post ); |
| 227 |
} |
| 228 |
|
| 229 |
/** |
| 230 |
* Filter the raw HTML content before Gutenberg conversion. |
| 231 |
* |
| 232 |
* This filter runs after the subtitle has been prepended (if present) |
| 233 |
* but before the HTML is parsed and converted to Gutenberg blocks. |
| 234 |
* Useful for cleaning up or transforming Substack-specific HTML, |
| 235 |
* adding custom elements, or stripping unwanted markup. |
| 236 |
* |
| 237 |
* @since 1.2.0 |
| 238 |
* |
| 239 |
* @param string $html_body The raw HTML content from the Substack export. |
| 240 |
* @param array $post The raw Substack post data from the CSV. |
| 241 |
* @param array|null $post_meta The post metadata from the Substack API response. |
| 242 |
*/ |
| 243 |
$post['html_body'] = apply_filters( |
| 244 |
'substack_importer_raw_content', |
| 245 |
$post['html_body'], |
| 246 |
$post, |
| 247 |
$post_meta |
| 248 |
); |
| 249 |
|
| 250 |
$post_content = $this->convert_html_to_gutenberg( $post['html_body'] ); |
| 251 |
|
| 252 |
/** |
| 253 |
* Filter the post content after Gutenberg conversion. |
| 254 |
* |
| 255 |
* This filter allows modification of the converted Gutenberg block content |
| 256 |
* before it is added to the WXR. Useful for wrapping paywalled content in |
| 257 |
* custom blocks (e.g., membership plugins). |
| 258 |
* |
| 259 |
* @since 1.2.0 |
| 260 |
* |
| 261 |
* @param string $post_content The converted Gutenberg block content. |
| 262 |
* @param array $post The original Substack post data. |
| 263 |
* @param array|null $post_meta Additional post metadata from Substack API. |
| 264 |
*/ |
| 265 |
$post_content = apply_filters( |
| 266 |
'substack_importer_post_content_after_conversion', |
| 267 |
$post_content, |
| 268 |
$post, |
| 269 |
$post_meta |
| 270 |
); |
| 271 |
|
| 272 |
$post_data = array( |
| 273 |
'id' => $id, |
| 274 |
'title' => $post['title'], |
| 275 |
'content' => $post_content, |
| 276 |
'date' => 'true' === $post['is_published'] ? $post['post_date'] : '', |
| 277 |
'status' => 'true' === $post['is_published'] ? 'publish' : 'draft', |
| 278 |
'post_date_gmt' => $post['post_date'], |
| 279 |
'post_date' => $post['post_date'], |
| 280 |
'post_taxonomies' => array(), |
| 281 |
'metas' => array(), |
| 282 |
); |
| 283 |
|
| 284 |
$first_image_url = $this->get_first_image_url_from_html( $post['html_body'] ); |
| 285 |
if ( ! empty( $first_image_url ) ) { |
| 286 |
$post_data['metas'][] = array( |
| 287 |
'key' => '_substack_first_image_url', |
| 288 |
'value' => $first_image_url, |
| 289 |
); |
| 290 |
} |
| 291 |
|
| 292 |
if ( isset( $post_id[1] ) ) { |
| 293 |
$post_data['post_name'] = $post_id[1]; |
| 294 |
} |
| 295 |
|
| 296 |
// If we were able to retrieve more information through the Substack API, we might have |
| 297 |
// author information and comments. |
| 298 |
$post_data['author'] = $post_meta ? $this->get_post_author( $post_meta, $post_data['status'] ) : $this->get_default_author( $post_data['status'] ); |
| 299 |
$post_data['comments'] = $post_meta ? $this->get_post_comments( $post_meta ) : array(); |
| 300 |
|
| 301 |
// Set the comment status. |
| 302 |
$post_data['comment_status'] = ! empty( $post_meta['write_comment_permissions'] ) && 'none' === $post_meta['write_comment_permissions'] |
| 303 |
? 'closed' |
| 304 |
: 'open'; |
| 305 |
|
| 306 |
// Handle podcast posts - prepend a Gutenberg audio block to the post content. |
| 307 |
if ( 'podcast' === $post['type'] && ! empty( $post['podcast_url'] ) ) { |
| 308 |
$post_data = $this->handle_podcast_post( $post_data, $post ); |
| 309 |
} |
| 310 |
|
| 311 |
/** |
| 312 |
* Allow for custom modifications to the post data. |
| 313 |
* |
| 314 |
* @param array $post_data The post data. |
| 315 |
* @param array $post The original post data. |
| 316 |
*/ |
| 317 |
$post_data = apply_filters( 'substack_importer_post_data', $post_data, $post ); |
| 318 |
|
| 319 |
$this->generator->add_post( $post_data ); |
| 320 |
|
| 321 |
/** |
| 322 |
* Fires after a single Substack post has been converted and added to the WXR. |
| 323 |
* |
| 324 |
* Useful for logging, progress tracking, or performing cleanup after each post. |
| 325 |
* |
| 326 |
* @since 1.2.0 |
| 327 |
* |
| 328 |
* @param array $post_data The final post data that was added to the WXR. |
| 329 |
* @param array $post The raw Substack post data from the CSV. |
| 330 |
* @param array|null $post_meta The post metadata from the Substack API response. |
| 331 |
* @param int $id The Substack post ID. |
| 332 |
*/ |
| 333 |
do_action( 'substack_importer_after_post', $post_data, $post, $post_meta, $id ); |
| 334 |
} |
| 335 |
} |
| 336 |
|
| 337 |
protected function handle_podcast_post( $post_data, $post ) { |
| 338 |
$post_data['content'] = $this->get_audio_block( $post['podcast_url'] ) . $post_data['content']; |
| 339 |
|
| 340 |
// Create a new attachment |
| 341 |
$this->generator->add_post( |
| 342 |
array( |
| 343 |
'title' => urldecode( basename( $post['podcast_url'] ) ), |
| 344 |
'link' => $post['podcast_url'], |
| 345 |
'post_date' => $post_data['post_date'], |
| 346 |
'type' => 'attachment', |
| 347 |
'attachment_url' => $post['podcast_url'], |
| 348 |
'metas' => array( |
| 349 |
array( |
| 350 |
'key' => '_wp_original_image_link', |
| 351 |
'value' => $post['podcast_url'], |
| 352 |
), |
| 353 |
), |
| 354 |
) |
| 355 |
); |
| 356 |
|
| 357 |
// Add this post to the podcast category. |
| 358 |
$this->categories['podcast'] = array( |
| 359 |
'slug' => 'podcast', |
| 360 |
'name' => 'Podcast', |
| 361 |
); |
| 362 |
$post_data['post_taxonomies'][] = array( |
| 363 |
'name' => 'Podcast', |
| 364 |
'slug' => 'podcast', |
| 365 |
'domain' => 'category', |
| 366 |
); |
| 367 |
|
| 368 |
return $post_data; |
| 369 |
} |
| 370 |
|
| 371 |
/** |
| 372 |
* Add the subtitle to the html_content by prepending a h2. |
| 373 |
* |
| 374 |
* @param array $post The post data containing subtitle and html_body. |
| 375 |
* |
| 376 |
* @return string html body content. |
| 377 |
*/ |
| 378 |
protected function add_subtitle( $post ) { |
| 379 |
$heading = sprintf( '<h2>%s</h2>', $post['subtitle'] ); |
| 380 |
|
| 381 |
/** |
| 382 |
* Filter the subtitle HTML before it is prepended to the post content. |
| 383 |
* |
| 384 |
* Return an empty string to skip the subtitle entirely. |
| 385 |
* Useful for changing the heading level, wrapping in custom markup, |
| 386 |
* or conditionally removing subtitles. |
| 387 |
* |
| 388 |
* @since 1.2.0 |
| 389 |
* |
| 390 |
* @param string $heading The subtitle HTML (default: an h2 element). |
| 391 |
* @param array $post The raw Substack post data containing 'subtitle' and 'html_body'. |
| 392 |
*/ |
| 393 |
$heading = apply_filters( 'substack_importer_subtitle', $heading, $post ); |
| 394 |
|
| 395 |
return $heading . $post['html_body']; |
| 396 |
} |
| 397 |
|
| 398 |
/** |
| 399 |
* Get a Gutenberg Audio block for the podcast. |
| 400 |
* |
| 401 |
* @param string $audio_url The URL of the podcast audio file. |
| 402 |
* |
| 403 |
* @return string The Gutenberg audio block HTML. |
| 404 |
*/ |
| 405 |
protected function get_audio_block( $audio_url ) { |
| 406 |
$code = '<!-- wp:audio --><figure class="wp-block-audio"><audio controls src="%s"></audio><figcaption>Podcast</figcaption></figure><!-- /wp:audio -->'; |
| 407 |
$block = sprintf( $code, $audio_url ); |
| 408 |
|
| 409 |
/** |
| 410 |
* Filter the Gutenberg audio block HTML for podcast posts. |
| 411 |
* |
| 412 |
* Allows modification or replacement of the audio block that is |
| 413 |
* prepended to podcast post content. Useful for using a custom |
| 414 |
* audio player block or adding additional markup. |
| 415 |
* |
| 416 |
* @since 1.2.0 |
| 417 |
* |
| 418 |
* @param string $block The Gutenberg audio block HTML. |
| 419 |
* @param string $audio_url The URL of the podcast audio file. |
| 420 |
*/ |
| 421 |
return apply_filters( 'substack_importer_audio_block', $block, $audio_url ); |
| 422 |
} |
| 423 |
|
| 424 |
protected function get_post_author( $post_meta, $post_status ) { |
| 425 |
// If we can't get the author information, return a default author. |
| 426 |
if ( empty( $post_meta['publishedBylines'] ) ) { |
| 427 |
return $this->get_default_author( $post_status ); |
| 428 |
} |
| 429 |
|
| 430 |
$byline = $post_meta['publishedBylines'][0]; |
| 431 |
$this->authors[ $byline['id'] ] = array( |
| 432 |
'login' => $byline['name'], |
| 433 |
'display_name' => $byline['name'], |
| 434 |
'id' => $byline['id'], |
| 435 |
); |
| 436 |
|
| 437 |
return $byline['name']; |
| 438 |
} |
| 439 |
|
| 440 |
protected function get_default_author( $post_status ) { |
| 441 |
$unknown_author_key = '_unknown'; |
| 442 |
$unknown_author_value = array( |
| 443 |
'login' => 'Unknown', |
| 444 |
'display_name' => 'Unknown', |
| 445 |
'id' => 1, |
| 446 |
); |
| 447 |
$draft_author_key = '_draft'; |
| 448 |
$draft_author_value = array( |
| 449 |
'login' => 'Draft', |
| 450 |
'display_name' => 'Draft Posts', |
| 451 |
'id' => 2, |
| 452 |
); |
| 453 |
|
| 454 |
if ( 'publish' === $post_status ) { |
| 455 |
$this->authors[ $unknown_author_key ] = $unknown_author_value; |
| 456 |
return $unknown_author_key; |
| 457 |
} else { |
| 458 |
$this->authors[ $draft_author_key ] = $draft_author_value; |
| 459 |
return $draft_author_key; |
| 460 |
} |
| 461 |
} |
| 462 |
|
| 463 |
/** |
| 464 |
* Get post comments retrieved through the Substack api. |
| 465 |
* |
| 466 |
* @param array $post_data Additional data about a post retrieved through the Substack Post API. |
| 467 |
* |
| 468 |
* @return mixed |
| 469 |
*/ |
| 470 |
protected function get_post_comments( $post_meta ) { |
| 471 |
if ( empty( $post_meta['comments'] ) ) { |
| 472 |
return array(); |
| 473 |
} |
| 474 |
|
| 475 |
return $this->parse_comments( $post_meta['comments'] ); |
| 476 |
} |
| 477 |
|
| 478 |
/** |
| 479 |
* Recursively parse the comments and prepare the data required for the WXR output. |
| 480 |
* |
| 481 |
* @param array $comments An array of comments provided by the Substack posts API endpoint. |
| 482 |
* @param array $out Output that is ready to be passed to the WXR generator. |
| 483 |
* @param null $parent If we are in a recursive call, the parent must be provided. |
| 484 |
* |
| 485 |
* @return array|mixed |
| 486 |
*/ |
| 487 |
protected function parse_comments( $comments, $out = array(), $parent = null ) { |
| 488 |
foreach ( $comments as $comment ) { |
| 489 |
$out[] = array( |
| 490 |
'id' => $comment['id'], |
| 491 |
'author' => $comment['name'], |
| 492 |
'date' => $comment['date'], |
| 493 |
'date_gmt' => $comment['date'], |
| 494 |
'content' => $comment['body'], |
| 495 |
'parent' => $parent, |
| 496 |
'metas' => array(), |
| 497 |
); |
| 498 |
|
| 499 |
if ( ! empty( $comment['children'] ) ) { |
| 500 |
$out = $this->parse_comments( $comment['children'], $out, (int) $comment['id'] ); |
| 501 |
} |
| 502 |
} |
| 503 |
|
| 504 |
return $out; |
| 505 |
} |
| 506 |
|
| 507 |
/** |
| 508 |
* Convert the content HTML to Gutenberg blocks and return the result. |
| 509 |
* |
| 510 |
* @param string $content The HTML provided by Substack. |
| 511 |
* |
| 512 |
* @return string|string[]|null |
| 513 |
* |
| 514 |
* @todo Load the content as XML to prevent errors from loadHTML. |
| 515 |
*/ |
| 516 |
protected function convert_html_to_gutenberg( $content ) { |
| 517 |
$dom = new DOMDocument(); |
| 518 |
|
| 519 |
// Suppress warnings and errors when loading HTML |
| 520 |
libxml_use_internal_errors( true ); |
| 521 |
|
| 522 |
// By inserting a meta tag with utf-8 encoding we make sure the content is converted to utf-8 |
| 523 |
$content = '<meta http-equiv="Content-Type" content="text/html; charset=utf-8">' . $content; |
| 524 |
$dom->loadHTML( $content ); |
| 525 |
|
| 526 |
// Clear any errors that were logged |
| 527 |
libxml_clear_errors(); |
| 528 |
|
| 529 |
$body = $dom->getElementsByTagName( 'body' )->item( 0 ); |
| 530 |
|
| 531 |
// We don't want to use the DomNodeList because it will change while we are iterating over the nodes. |
| 532 |
$nodes = array(); |
| 533 |
foreach ( $body->childNodes as $node ) { //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 534 |
if ( ! $node instanceof DomElement ) { |
| 535 |
continue; |
| 536 |
} |
| 537 |
$nodes[] = $node; |
| 538 |
} |
| 539 |
|
| 540 |
// We go through the top-level nodes and handle each of them. |
| 541 |
foreach ( $nodes as $idx => $node ) { |
| 542 |
$next_sibling = count( $nodes ) - 1 > $idx ? $nodes[ $idx + 1 ] : null; |
| 543 |
$this->convert_node( $node, $body, $next_sibling ); |
| 544 |
} |
| 545 |
|
| 546 |
// Save as XML otherwise we don't get HTMl5 elements correctly. |
| 547 |
$content = $dom->saveXML( $body ); |
| 548 |
|
| 549 |
// Strip the body tag. |
| 550 |
$content = preg_replace( '/<body>(.+)<\/body>/s', '$1', $content ); |
| 551 |
|
| 552 |
return $content; |
| 553 |
} |
| 554 |
|
| 555 |
|
| 556 |
/** |
| 557 |
* Convert a single node to a Gutenberg block. |
| 558 |
* |
| 559 |
* Tries to convert a given HTML node into a Gutenberg block. |
| 560 |
* |
| 561 |
* @param DomElement $node The node to be converted. |
| 562 |
* @param DomElement $parent The parent of the node to be converted. |
| 563 |
* @param DomElement|null $next_sibling The next sibling of the node to be converted, if it exists. |
| 564 |
* |
| 565 |
*/ |
| 566 |
protected function convert_node( DOMElement $node, DomElement $parent, ?DomElement $next_sibling = null ) { |
| 567 |
$block_name = null; |
| 568 |
$block_attributes = array(); |
| 569 |
|
| 570 |
$node_name = $node->nodeName; //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 571 |
switch ( $node_name ) { |
| 572 |
|
| 573 |
case 'p': |
| 574 |
$block_name = 'wp:paragraph'; |
| 575 |
$class = $node->getAttribute( 'class' ); |
| 576 |
|
| 577 |
// remove empty paragraphs. |
| 578 |
/** @todo Perhaps we can remove all empty nodes, not just paragraphs? */ |
| 579 |
if ( ! $node->childNodes->length ) { //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 580 |
$parent->removeChild( $node ); |
| 581 |
$node = null; |
| 582 |
} |
| 583 |
|
| 584 |
// Button |
| 585 |
if ( 'button-wrapper' === $class ) { |
| 586 |
$node = $this->convert_button_node( $node, $parent ); |
| 587 |
$block_name = 'wp:button'; |
| 588 |
} |
| 589 |
|
| 590 |
break; |
| 591 |
|
| 592 |
case 'blockquote': |
| 593 |
$block_name = 'wp:quote'; |
| 594 |
$node->setAttribute( 'class', 'wp-block-quote' ); |
| 595 |
break; |
| 596 |
|
| 597 |
case 'div': |
| 598 |
case 'iframe': |
| 599 |
$class = $node->getAttribute( 'class' ); |
| 600 |
|
| 601 |
// Preformatted text - these are Poetry blocks in Substack |
| 602 |
if ( 'preformatted-block' === $class ) { |
| 603 |
$node = $this->convert_preformatted_node( $node, $parent ); |
| 604 |
$block_name = 'wp:verse'; |
| 605 |
} |
| 606 |
|
| 607 |
// Pull Quotes |
| 608 |
if ( 'pullquote' === $class ) { |
| 609 |
$node = $this->convert_pullquote_node( $node, $parent ); |
| 610 |
$block_name = 'wp:pullquote'; |
| 611 |
} |
| 612 |
|
| 613 |
// Images |
| 614 |
if ( 'captioned-image-container' === $class ) { |
| 615 |
$result = $this->convert_image_node( $node, $parent ); |
| 616 |
$node = $result['node']; |
| 617 |
$block_attributes = $result['block_attributes']; |
| 618 |
$block_name = 'wp:image'; |
| 619 |
} |
| 620 |
|
| 621 |
// Horizontal separator |
| 622 |
if ( |
| 623 |
$node && |
| 624 |
$node->hasChildNodes() && //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 625 |
$node->childNodes->length && //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 626 |
'hr' === $node->childNodes[0]->nodeName //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 627 |
) { |
| 628 |
$node = $this->convert_separator_node( $node, $parent ); |
| 629 |
$block_name = 'wp:separator'; |
| 630 |
} |
| 631 |
|
| 632 |
// Embeds |
| 633 |
$first_class = explode( ' ', $class ); |
| 634 |
if ( ! empty( $first_class ) && in_array( $first_class[0], $this->supported_embeds, true ) ) { |
| 635 |
$result = $this->convert_embed_node( $node, $parent ); |
| 636 |
$node = $result['node']; |
| 637 |
$block_attributes = $result['block_attributes']; |
| 638 |
$block_name = $result['block_name']; |
| 639 |
} |
| 640 |
|
| 641 |
if ( 'paywall-jump' === $class ) { |
| 642 |
$result = $this->convert_paywall_node( $node, $parent ); |
| 643 |
$node = $result['node']; |
| 644 |
$block_attributes = $result['block_attributes']; |
| 645 |
$block_name = $result['block_name']; |
| 646 |
} |
| 647 |
|
| 648 |
if ( 'subscription-widget-wrap' === $class ) { |
| 649 |
$result = $this->convert_subscription_node( $node, $parent ); |
| 650 |
$node = $result['node']; |
| 651 |
$block_attributes = $result['block_attributes']; |
| 652 |
$block_name = $result['block_name']; |
| 653 |
} |
| 654 |
|
| 655 |
break; |
| 656 |
|
| 657 |
case 'ol': |
| 658 |
case 'ul': |
| 659 |
$block_name = 'wp:list'; |
| 660 |
|
| 661 |
if ( 'ol' === $node_name ) { |
| 662 |
$block_attributes['ordered'] = true; |
| 663 |
} |
| 664 |
|
| 665 |
break; |
| 666 |
|
| 667 |
case 'pre': |
| 668 |
$block_name = 'wp:code'; |
| 669 |
$node->setAttribute( 'class', 'wp-block-code' ); |
| 670 |
break; |
| 671 |
|
| 672 |
case 'h1': |
| 673 |
case 'h2': |
| 674 |
case 'h3': |
| 675 |
case 'h4': |
| 676 |
case 'h5': |
| 677 |
case 'h6': |
| 678 |
$block_name = 'wp:heading'; |
| 679 |
|
| 680 |
// Gutenberg defaults to h2 for heading blocks without a level attribute. |
| 681 |
if ( 'h1' === $node_name ) { |
| 682 |
$node = $this->replace_html_node_tag( $node, $parent, 'h2' ); |
| 683 |
} |
| 684 |
|
| 685 |
$node->setAttribute( 'class', 'wp-block-heading' ); |
| 686 |
|
| 687 |
if ( ! in_array( $node_name, array( 'h1', 'h2' ), true ) ) { |
| 688 |
$block_attributes['level'] = (int) substr( $node_name, 1, 1 ); |
| 689 |
} |
| 690 |
break; |
| 691 |
|
| 692 |
case 'a': |
| 693 |
$class = $node->getAttribute( 'class' ); |
| 694 |
if ( 'image-link image2' === trim( $class ) ) { |
| 695 |
$result = $this->convert_image_node( $node, $parent ); |
| 696 |
$node = $result['node']; |
| 697 |
$block_attributes = $result['block_attributes']; |
| 698 |
$block_name = 'wp:image'; |
| 699 |
} |
| 700 |
|
| 701 |
break; |
| 702 |
|
| 703 |
} |
| 704 |
|
| 705 |
if ( ! $block_name || ! $node ) { |
| 706 |
return; |
| 707 |
} |
| 708 |
|
| 709 |
/** |
| 710 |
* Filter the result of a single node conversion to a Gutenberg block. |
| 711 |
* |
| 712 |
* Allows modification of the block name and attributes after the default |
| 713 |
* conversion logic has run. Return a block_name of null to skip the node. |
| 714 |
* Useful for overriding how specific Substack elements are converted, |
| 715 |
* adding custom attributes, or changing block types. |
| 716 |
* |
| 717 |
* @since 1.2.0 |
| 718 |
* |
| 719 |
* @param array $block_data { |
| 720 |
* The block conversion result. |
| 721 |
* |
| 722 |
* @type string $block_name The Gutenberg block name (e.g. 'wp:paragraph'). |
| 723 |
* @type array $block_attributes The block attributes array. |
| 724 |
* } |
| 725 |
* @param DomElement $node The converted DOM node. |
| 726 |
* @param string $node_name The original HTML tag name (e.g. 'p', 'div', 'h2'). |
| 727 |
*/ |
| 728 |
$block_data = apply_filters( |
| 729 |
'substack_importer_converted_node', |
| 730 |
array( |
| 731 |
'block_name' => $block_name, |
| 732 |
'block_attributes' => $block_attributes, |
| 733 |
), |
| 734 |
$node, |
| 735 |
$node_name |
| 736 |
); |
| 737 |
|
| 738 |
$block_name = $block_data['block_name']; |
| 739 |
$block_attributes = $block_data['block_attributes']; |
| 740 |
|
| 741 |
if ( ! $block_name ) { |
| 742 |
return; |
| 743 |
} |
| 744 |
|
| 745 |
// Create the Gutenberg block code |
| 746 |
$attributes_part = ''; |
| 747 |
if ( is_countable( $block_attributes ) && count( $block_attributes ) ) { |
| 748 |
$attributes_part = ' ' . wp_json_encode( $block_attributes ); |
| 749 |
} |
| 750 |
$block_open = new DOMComment( ' ' . $block_name . $attributes_part . ' ' ); |
| 751 |
$block_close = new DOMComment( ' /' . $block_name . ' ' ); |
| 752 |
|
| 753 |
$parent->insertBefore( $block_open, $node ); |
| 754 |
|
| 755 |
$next_sibling |
| 756 |
? $parent->insertBefore( $block_close, $next_sibling ) |
| 757 |
: $parent->appendChild( $block_close ); |
| 758 |
} |
| 759 |
|
| 760 |
/** |
| 761 |
* Convert a preformatted text node to valid Gutenberg markup. |
| 762 |
* |
| 763 |
* @param DomElement $node The node to be converted. |
| 764 |
* @param DomElement $parent The parent of the node. |
| 765 |
* |
| 766 |
* @return DomElement The converted node. |
| 767 |
*/ |
| 768 |
protected function convert_preformatted_node( DomElement $node, DomElement $parent ) { |
| 769 |
$node_pres = $node->getElementsByTagName( 'pre' ); |
| 770 |
if ( 0 === $node_pres->length ) { |
| 771 |
return null; |
| 772 |
} |
| 773 |
|
| 774 |
$new_node = new DomElement( 'pre', $node_pres[0]->textContent ); |
| 775 |
$parent->replaceChild( $new_node, $node ); |
| 776 |
$new_node->setAttribute( 'class', 'wp-block-verse' ); |
| 777 |
|
| 778 |
return $new_node; |
| 779 |
} |
| 780 |
|
| 781 |
/** |
| 782 |
* Convert a pullquote node to valid Gutenberg markup. |
| 783 |
* |
| 784 |
* @param DomElement $node The node to be converted. |
| 785 |
* @param DomElement $parent The parent of the node. |
| 786 |
* |
| 787 |
* @return DomElement The converted node. |
| 788 |
*/ |
| 789 |
protected function convert_pullquote_node( DomElement $node, DomElement $parent ) { |
| 790 |
$node_paragraphs = $node->getElementsByTagName( 'p' ); |
| 791 |
if ( 0 === $node_paragraphs->length ) { |
| 792 |
return null; |
| 793 |
} |
| 794 |
|
| 795 |
// Create the figure and blockquote elements for pullquote block. |
| 796 |
$new_node = new DomElement( 'figure' ); |
| 797 |
$blockquote = new DomElement( 'blockquote' ); |
| 798 |
$paragraph = new DomElement( 'p' ); |
| 799 |
$quote_text = new DOMText( $node_paragraphs[0]->textContent ); |
| 800 |
|
| 801 |
// Assemble and replace node. |
| 802 |
$parent->replaceChild( $new_node, $node ); |
| 803 |
$new_node->appendChild( $blockquote ); |
| 804 |
$blockquote->appendChild( $paragraph ); |
| 805 |
$paragraph->appendChild( $quote_text ); |
| 806 |
$new_node->setAttribute( 'class', 'wp-block-pullquote' ); |
| 807 |
|
| 808 |
return $new_node; |
| 809 |
} |
| 810 |
|
| 811 |
/** |
| 812 |
* Handle a button node. |
| 813 |
* |
| 814 |
* @param DomElement $node The node to be converted. |
| 815 |
* @param DomElement $parent The parent of the node. |
| 816 |
* |
| 817 |
* @return DomElement |
| 818 |
* |
| 819 |
* @todo Support multiple types of buttons. For now buttons are removed. |
| 820 |
*/ |
| 821 |
protected function convert_button_node( DomElement $node, DomElement $parent ) { |
| 822 |
$parent->removeChild( $node ); |
| 823 |
return null; |
| 824 |
} |
| 825 |
|
| 826 |
/** |
| 827 |
* Convert an image node to a Gutenberg valid markup. |
| 828 |
* |
| 829 |
* @param DomElement $node The node to be converted. |
| 830 |
* @param DomElement $parent The parent of the node. |
| 831 |
* |
| 832 |
* @return array An array containing the Block attributes and the new node. |
| 833 |
* |
| 834 |
* @todo If the node is a (a) link we need to make this image a link as well. |
| 835 |
*/ |
| 836 |
protected function convert_image_node( DomElement $node, DomElement $parent ) { |
| 837 |
|
| 838 |
// Check if the image needs to be resized |
| 839 |
// Can we already upload the image here? |
| 840 |
/** @var DomElement $image */ |
| 841 |
$image = $node->getElementsByTagName( 'img' )[0]; |
| 842 |
|
| 843 |
// if there is no image we can't proceed. |
| 844 |
if ( ! $image ) { |
| 845 |
$parent->removeChild( $node ); |
| 846 |
return array( |
| 847 |
'block_attributes' => array(), |
| 848 |
'node' => null, |
| 849 |
); |
| 850 |
} |
| 851 |
|
| 852 |
$block_attributes = array(); |
| 853 |
|
| 854 |
$new_node = new DomElement( 'figure' ); |
| 855 |
|
| 856 |
$parent->replaceChild( $new_node, $node ); |
| 857 |
|
| 858 |
$classes = array( 'wp-block-image', 'size-large' ); |
| 859 |
|
| 860 |
// The data we need is set as json data attribute on the img node. |
| 861 |
$image_data = json_decode( $image->getAttribute( 'data-attrs' ), true ); |
| 862 |
|
| 863 |
// Add the image as an attachement post to the WXR. |
| 864 |
$this->generator->add_post( |
| 865 |
array( |
| 866 |
'title' => urldecode( basename( $image_data['src'] ) ), |
| 867 |
'link' => $image_data['src'], |
| 868 |
'type' => 'attachment', |
| 869 |
'attachment_url' => $image_data['src'], |
| 870 |
'metas' => array( |
| 871 |
array( |
| 872 |
'key' => '_wp_original_image_link', |
| 873 |
'value' => $image_data['src'], |
| 874 |
), |
| 875 |
), |
| 876 |
) |
| 877 |
); |
| 878 |
|
| 879 |
// Create the new image element. |
| 880 |
$new_image = new DomElement( 'img' ); |
| 881 |
$new_node->appendChild( $new_image ); |
| 882 |
$new_image->setAttribute( 'src', $image_data['src'] ); |
| 883 |
if ( ! is_null( $image_data['alt'] ) ) { |
| 884 |
$new_image->setAttribute( 'alt', $image_data['alt'] ); |
| 885 |
} |
| 886 |
|
| 887 |
// Deal with resizing. |
| 888 |
if ( $image_data['resizeWidth'] ) { |
| 889 |
$classes[] = 'is-resized'; |
| 890 |
$new_image->setAttribute( 'width', $image_data['resizeWidth'] ); |
| 891 |
$block_attributes['width'] = $image_data['resizeWidth']; |
| 892 |
} |
| 893 |
|
| 894 |
// Handle caption if it exists |
| 895 |
$caption = $node->getElementsByTagName( 'figcaption' ); |
| 896 |
if ( $caption->length > 0 ) { |
| 897 |
$caption_text = $caption->item( 0 )->textContent; |
| 898 |
$block_attributes['caption'] = $caption_text; |
| 899 |
|
| 900 |
// Create figcaption element |
| 901 |
$figcaption = new DomElement( 'figcaption' ); |
| 902 |
$new_node->appendChild( $figcaption ); |
| 903 |
// phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 904 |
$figcaption->textContent = $caption_text; |
| 905 |
} |
| 906 |
|
| 907 |
// Set the classes on the figure element. |
| 908 |
$new_node->setAttribute( 'class', implode( ' ', $classes ) ); |
| 909 |
|
| 910 |
$block_attributes['sizeSlug'] = 'large'; |
| 911 |
$block_attributes['linkDestination'] = 'none'; |
| 912 |
|
| 913 |
$result = array( |
| 914 |
'block_attributes' => $block_attributes, |
| 915 |
'node' => $new_node, |
| 916 |
); |
| 917 |
|
| 918 |
/** |
| 919 |
* Filter the image node conversion result. |
| 920 |
* |
| 921 |
* Allows modification of the image block attributes and node after |
| 922 |
* the default conversion. Useful for adjusting image sizes, adding |
| 923 |
* custom classes, modifying captions, or changing link destinations. |
| 924 |
* |
| 925 |
* @since 1.2.0 |
| 926 |
* |
| 927 |
* @param array $result { |
| 928 |
* The image conversion result. |
| 929 |
* |
| 930 |
* @type array $block_attributes The image block attributes (sizeSlug, linkDestination, width, caption). |
| 931 |
* @type DomElement $node The figure DOM element for the image block. |
| 932 |
* } |
| 933 |
* @param array|null $image_data The decoded image data from the Substack data-attrs attribute. |
| 934 |
*/ |
| 935 |
return apply_filters( 'substack_importer_image_result', $result, $image_data ); |
| 936 |
} |
| 937 |
|
| 938 |
/** |
| 939 |
* Convert the node to a valid Gutenberg separator block. |
| 940 |
* |
| 941 |
* @param DomElement $node The node to be converted. |
| 942 |
* @param DomElement $parent The parent of the node to be converted. |
| 943 |
* |
| 944 |
* @return DomElement The new node. |
| 945 |
*/ |
| 946 |
protected function convert_separator_node( DomElement $node, DomElement $parent ) { |
| 947 |
|
| 948 |
$new_node = new DomElement( 'hr' ); |
| 949 |
$parent->replaceChild( $new_node, $node ); |
| 950 |
$new_node->setAttribute( 'class', 'wp-block-separator' ); |
| 951 |
|
| 952 |
return $new_node; |
| 953 |
} |
| 954 |
|
| 955 |
/** |
| 956 |
* Convert a node that represents an embed to valid Gutenberg embed block markup. |
| 957 |
* |
| 958 |
* @param DomElement $node The node to be converted. |
| 959 |
* @param DomElement $parent The parent of the node to be coverted. |
| 960 |
* |
| 961 |
* @return array Containing the block_name, block_attributes and node. |
| 962 |
*/ |
| 963 |
protected function convert_embed_node( DomElement $node, DomElement $parent ) { |
| 964 |
|
| 965 |
$first_class = explode( ' ', $node->getAttribute( 'class' ) )[0]; |
| 966 |
|
| 967 |
/** |
| 968 |
* Short-circuit the embed node conversion before default handling. |
| 969 |
* |
| 970 |
* Return a non-null array to skip the built-in switch statement entirely. |
| 971 |
* The returned array must have keys: 'node', 'block_attributes', 'block_name'. |
| 972 |
* Useful for handling unsupported embed types, overriding the default |
| 973 |
* conversion for a specific provider, or adding entirely new providers. |
| 974 |
* |
| 975 |
* @since 1.2.0 |
| 976 |
* |
| 977 |
* @param array|null $pre_result Return non-null to short-circuit. Expected keys: |
| 978 |
* 'node' (DomElement|null), 'block_attributes' (array), |
| 979 |
* 'block_name' (string|null). |
| 980 |
* @param DomElement $node The embed DOM node before conversion. |
| 981 |
* @param DomElement $parent The parent DOM element. |
| 982 |
* @param string $first_class The CSS class identifying the embed type (e.g. 'youtube-wrap', 'tweet'). |
| 983 |
*/ |
| 984 |
$pre_result = apply_filters( |
| 985 |
'substack_importer_pre_embed_conversion', |
| 986 |
null, |
| 987 |
$node, |
| 988 |
$parent, |
| 989 |
$first_class |
| 990 |
); |
| 991 |
|
| 992 |
if ( null !== $pre_result ) { |
| 993 |
return $pre_result; |
| 994 |
} |
| 995 |
|
| 996 |
switch ( $first_class ) { |
| 997 |
|
| 998 |
case 'youtube-wrap': |
| 999 |
$output = $this->convert_youtube_embed( $node, $parent ); |
| 1000 |
break; |
| 1001 |
|
| 1002 |
case 'vimeo-wrap': |
| 1003 |
$output = $this->convert_vimeo_embed( $node, $parent ); |
| 1004 |
break; |
| 1005 |
|
| 1006 |
case 'soundcloud-wrap': |
| 1007 |
$output = $this->convert_soundcloud_embed( $node, $parent ); |
| 1008 |
break; |
| 1009 |
|
| 1010 |
case 'tweet': |
| 1011 |
$output = $this->convert_tweet_embed( $node, $parent ); |
| 1012 |
break; |
| 1013 |
|
| 1014 |
case 'spotify-wrap': |
| 1015 |
$output = $this->convert_spotify_embed( $node, $parent ); |
| 1016 |
break; |
| 1017 |
|
| 1018 |
case 'bandcamp-wrap': |
| 1019 |
$output = $this->convert_bandcamp_embed( $node, $parent ); |
| 1020 |
break; |
| 1021 |
|
| 1022 |
case 'github-gist': |
| 1023 |
$output = $this->convert_gist_embed( $node, $parent ); |
| 1024 |
break; |
| 1025 |
|
| 1026 |
case 'instagram': |
| 1027 |
$output = $this->convert_instagram_embed( $node, $parent ); |
| 1028 |
break; |
| 1029 |
|
| 1030 |
case 'tiktok-wrap': |
| 1031 |
$output = $this->convert_tiktok_embed( $node, $parent ); |
| 1032 |
break; |
| 1033 |
|
| 1034 |
default: |
| 1035 |
$parent->removeChild( $node ); |
| 1036 |
$output = array( |
| 1037 |
'node' => null, |
| 1038 |
'block_attributes' => array(), |
| 1039 |
'block_name' => null, |
| 1040 |
); |
| 1041 |
|
| 1042 |
} |
| 1043 |
|
| 1044 |
/** |
| 1045 |
* Filter the embed node conversion result. |
| 1046 |
* |
| 1047 |
* Allows modification of the embed block name, attributes, and node |
| 1048 |
* after the default conversion. Useful for adding support for |
| 1049 |
* additional embed providers, modifying embed URLs, or changing |
| 1050 |
* how specific embeds are represented. |
| 1051 |
* |
| 1052 |
* @since 1.2.0 |
| 1053 |
* |
| 1054 |
* @param array $output { |
| 1055 |
* The embed conversion result. |
| 1056 |
* |
| 1057 |
* @type string|null $block_name The Gutenberg block name (e.g. 'wp:embed'). |
| 1058 |
* @type array $block_attributes The block attributes (url, type, providerNameSlug, etc.). |
| 1059 |
* @type DomElement|null $node The converted DOM node. |
| 1060 |
* } |
| 1061 |
* @param string $first_class The CSS class identifying the embed type (e.g. 'youtube-wrap', 'tweet'). |
| 1062 |
*/ |
| 1063 |
return apply_filters( 'substack_importer_embed_result', $output, $first_class ); |
| 1064 |
} |
| 1065 |
|
| 1066 |
/** |
| 1067 |
* Convert the embed node into Gutenberg markup for a Youtube embed. |
| 1068 |
* |
| 1069 |
* @param DomElement $node The node to be converted. |
| 1070 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1071 |
* |
| 1072 |
* @return array Containing the block_name, block_attributes and node. |
| 1073 |
*/ |
| 1074 |
protected function convert_youtube_embed( DomElement $node, DomElement $parent ) { |
| 1075 |
|
| 1076 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1077 |
|
| 1078 |
$block_attributes = array( |
| 1079 |
'url' => 'https://youtu.be/' . $data_attributes['videoId'], |
| 1080 |
'type' => 'video', |
| 1081 |
'providerNameSlug' => 'youtube', |
| 1082 |
'responsive' => true, |
| 1083 |
'className' => 'wp-embed-aspect-16-9 wp-has-aspect-ratio', |
| 1084 |
); |
| 1085 |
|
| 1086 |
$node = $this->replace_embed_node( $node, $parent, $block_attributes['url'] ); |
| 1087 |
$classes = 'wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio'; |
| 1088 |
$node->setAttribute( 'class', $classes ); |
| 1089 |
|
| 1090 |
return array( |
| 1091 |
'block_name' => 'wp:embed', |
| 1092 |
'block_attributes' => $block_attributes, |
| 1093 |
'node' => $node, |
| 1094 |
); |
| 1095 |
} |
| 1096 |
|
| 1097 |
/** |
| 1098 |
* Convert the embed node into Gutenberg markup for a Vimeo embed. |
| 1099 |
* |
| 1100 |
* @param DomElement $node The node to be converted. |
| 1101 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1102 |
* |
| 1103 |
* @return array Containing the block_name, block_attributes and node. |
| 1104 |
*/ |
| 1105 |
protected function convert_vimeo_embed( DomElement $node, DomElement $parent ) { |
| 1106 |
|
| 1107 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1108 |
|
| 1109 |
$block_attributes = array( |
| 1110 |
'url' => 'https://vimeo.com/' . $data_attributes['videoId'], |
| 1111 |
'type' => 'video', |
| 1112 |
'providerNameSlug' => 'vimeo', |
| 1113 |
'responsive' => true, |
| 1114 |
'className' => 'wp-embed-aspect-16-9 wp-has-aspect-ratio', |
| 1115 |
); |
| 1116 |
|
| 1117 |
$node = $this->replace_embed_node( $node, $parent, $block_attributes['url'] ); |
| 1118 |
$classes = 'wp-block-embed is-type-video is-provider-vimeo wp-block-embed-vimeo wp-embed-aspect-16-9 wp-has-aspect-ratio'; |
| 1119 |
$node->setAttribute( 'class', $classes ); |
| 1120 |
|
| 1121 |
return array( |
| 1122 |
'block_name' => 'wp:embed', |
| 1123 |
'block_attributes' => $block_attributes, |
| 1124 |
'node' => $node, |
| 1125 |
); |
| 1126 |
} |
| 1127 |
|
| 1128 |
/** |
| 1129 |
* Convert the embed node into Gutenberg markup for a Soundcloud embed. |
| 1130 |
* |
| 1131 |
* @param DomElement $node The node to be converted. |
| 1132 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1133 |
* |
| 1134 |
* @return array Containing the block_name, block_attributes and node. |
| 1135 |
*/ |
| 1136 |
protected function convert_soundcloud_embed( DomElement $node, DomElement $parent ) { |
| 1137 |
|
| 1138 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1139 |
|
| 1140 |
// We construct the Soundcloud URL as a combination of Author URL and the |
| 1141 |
// Soundcloud Embed ID as this is recognized as a valid embed URL within |
| 1142 |
// WordPress. |
| 1143 |
$url_parts = explode( '/', $data_attributes['url'] ); |
| 1144 |
$id = array_pop( $url_parts ); |
| 1145 |
$url = $data_attributes['author_url'] . '/' . $id; |
| 1146 |
|
| 1147 |
$block_attributes = array( |
| 1148 |
'url' => $url, |
| 1149 |
'type' => 'rich', |
| 1150 |
'providerNameSlug' => 'soundcloud', |
| 1151 |
'responsive' => true, |
| 1152 |
); |
| 1153 |
|
| 1154 |
$node = $this->replace_embed_node( $node, $parent, $block_attributes['url'] ); |
| 1155 |
$classes = 'wp-block-embed is-type-rich is-provider-soundcloud wp-block-embed-soundcloud'; |
| 1156 |
$node->setAttribute( 'class', $classes ); |
| 1157 |
|
| 1158 |
return array( |
| 1159 |
'block_name' => 'wp:embed', |
| 1160 |
'block_attributes' => $block_attributes, |
| 1161 |
'node' => $node, |
| 1162 |
); |
| 1163 |
} |
| 1164 |
|
| 1165 |
/** |
| 1166 |
* Convert the embed node into Gutenberg markup for a Tweet embed. |
| 1167 |
* |
| 1168 |
* @param DomElement $node The node to be converted. |
| 1169 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1170 |
* |
| 1171 |
* @return array Containing the block_name, block_attributes and node. |
| 1172 |
*/ |
| 1173 |
protected function convert_tweet_embed( DomElement $node, DomElement $parent ) { |
| 1174 |
|
| 1175 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1176 |
|
| 1177 |
$block_attributes = array( |
| 1178 |
'url' => $data_attributes['url'], |
| 1179 |
'type' => 'rich', |
| 1180 |
'providerNameSlug' => 'twitter', |
| 1181 |
'responsive' => true, |
| 1182 |
); |
| 1183 |
|
| 1184 |
$node = $this->replace_embed_node( $node, $parent, $block_attributes['url'] ); |
| 1185 |
$classes = 'wp-block-embed is-type-rich is-provider-twitter wp-block-embed-twitter'; |
| 1186 |
$node->setAttribute( 'class', $classes ); |
| 1187 |
|
| 1188 |
return array( |
| 1189 |
'block_name' => 'wp:embed', |
| 1190 |
'block_attributes' => $block_attributes, |
| 1191 |
'node' => $node, |
| 1192 |
); |
| 1193 |
} |
| 1194 |
|
| 1195 |
/** |
| 1196 |
* Convert the embed node into Gutenberg markup for a Spotify embed. |
| 1197 |
* |
| 1198 |
* @param DomElement $node The node to be converted. |
| 1199 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1200 |
* |
| 1201 |
* @return array Containing the block_name, block_attributes and node. |
| 1202 |
*/ |
| 1203 |
protected function convert_spotify_embed( DomElement $node, DomElement $parent ) { |
| 1204 |
|
| 1205 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1206 |
|
| 1207 |
$block_attributes = array( |
| 1208 |
'url' => $data_attributes['url'], |
| 1209 |
'type' => 'rich', |
| 1210 |
'providerNameSlug' => 'spotify', |
| 1211 |
'responsive' => true, |
| 1212 |
'className' => 'wp-embed-aspect-21-9 wp-has-aspect-ratio', |
| 1213 |
); |
| 1214 |
|
| 1215 |
$node = $this->replace_embed_node( $node, $parent, $block_attributes['url'] ); |
| 1216 |
$classes = 'wp-block-embed is-type-rich is-provider-spotify wp-block-embed-spotify wp-embed-aspect-21-9 wp-has-aspect-ratio'; |
| 1217 |
$node->setAttribute( 'class', $classes ); |
| 1218 |
|
| 1219 |
return array( |
| 1220 |
'block_name' => 'wp:embed', |
| 1221 |
'block_attributes' => $block_attributes, |
| 1222 |
'node' => $node, |
| 1223 |
); |
| 1224 |
} |
| 1225 |
|
| 1226 |
/** |
| 1227 |
* Converts the node into a shortcode for Bandcamp. |
| 1228 |
* |
| 1229 |
* The shortcode is currently not supported in Core but is available by enabling the embeds module |
| 1230 |
* of the Jetpack plugin. |
| 1231 |
* |
| 1232 |
* @example [bandcamp width=350 height=470 album=473417827 size=large bgcol=ffffff linkcol=0687f5 tracklist=false] |
| 1233 |
* |
| 1234 |
* @param DomElement $node The node to be converted. |
| 1235 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1236 |
* |
| 1237 |
* @return array |
| 1238 |
*/ |
| 1239 |
protected function convert_bandcamp_embed( DomElement $node, DomElement $parent ) { |
| 1240 |
|
| 1241 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1242 |
|
| 1243 |
// The embed URL contains the attributes for the shortcode. Here we extract them and add them to the shortcode. |
| 1244 |
preg_match_all( '/[a-z]+=[a-z0-9]+/', $data_attributes['embed_url'], $matches ); |
| 1245 |
$shortcode = sprintf( '[bandcamp %s]', implode( ' ', $matches[0] ) ); |
| 1246 |
|
| 1247 |
$new_node = new DOMText( $shortcode ); |
| 1248 |
$parent->replaceChild( $new_node, $node ); |
| 1249 |
|
| 1250 |
return array( |
| 1251 |
'block_name' => 'wp:shortcode', |
| 1252 |
'block_attributes' => array(), |
| 1253 |
'node' => $new_node, |
| 1254 |
); |
| 1255 |
} |
| 1256 |
|
| 1257 |
/** |
| 1258 |
* Convert a Github Gist node into a shortcode. |
| 1259 |
* |
| 1260 |
* Tries to get the Gist id from the raw link or removes the entire Gist if the ID can not be determined. |
| 1261 |
* |
| 1262 |
* @param DomElement $node The node to be converted. |
| 1263 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1264 |
* |
| 1265 |
* @return array |
| 1266 |
*/ |
| 1267 |
protected function convert_gist_embed( DomElement $node, DomElement $parent ) { |
| 1268 |
|
| 1269 |
$a_elements = $node->getElementsByTagName( 'a' ); |
| 1270 |
|
| 1271 |
$url = $a_elements->length > 0 |
| 1272 |
? $a_elements[0]->getAttribute( 'href' ) |
| 1273 |
: null; |
| 1274 |
|
| 1275 |
if ( ! $url || ! preg_match( '/\/([a-z0-9]+)\/raw/', $a_elements[0]->getAttribute( 'href' ), $matches ) ) { |
| 1276 |
$parent->removeChild( $node ); |
| 1277 |
return array( |
| 1278 |
'node' => null, |
| 1279 |
'block_attributes' => array(), |
| 1280 |
'block_name' => null, |
| 1281 |
); |
| 1282 |
} |
| 1283 |
|
| 1284 |
$shortcode = sprintf( '[gist https://gist.github.com/%s]', $matches[1] ); |
| 1285 |
|
| 1286 |
$new_node = new DOMText( $shortcode ); |
| 1287 |
$parent->replaceChild( $new_node, $node ); |
| 1288 |
|
| 1289 |
return array( |
| 1290 |
'block_name' => 'wp:shortcode', |
| 1291 |
'block_attributes' => array(), |
| 1292 |
'node' => $new_node, |
| 1293 |
); |
| 1294 |
} |
| 1295 |
|
| 1296 |
/** |
| 1297 |
* Convert the embed node into Gutenberg markup for a TikTok embed. |
| 1298 |
* |
| 1299 |
* @param DomElement $node The node to be converted. |
| 1300 |
* @param DomElement $parent The parent of the node to be coverted. |
| 1301 |
* |
| 1302 |
* @return array Containing the block_name, block_attributes and node. |
| 1303 |
*/ |
| 1304 |
protected function convert_tiktok_embed( DomElement $node, DomElement $parent ) { |
| 1305 |
|
| 1306 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1307 |
|
| 1308 |
$block_attributes = array( |
| 1309 |
'url' => $data_attributes['url'], |
| 1310 |
'type' => 'video', |
| 1311 |
'providerNameSlug' => 'tiktok', |
| 1312 |
'responsive' => true, |
| 1313 |
'className' => 'wp-embed-aspect-9-16 wp-has-aspect-ratio', |
| 1314 |
); |
| 1315 |
|
| 1316 |
$node = $this->replace_embed_node( $node, $parent, $block_attributes['url'] ); |
| 1317 |
$classes = 'wp-block-embed is-type-video is-provider-tiktok wp-block-embed-tiktok'; |
| 1318 |
$node->setAttribute( 'class', $classes ); |
| 1319 |
|
| 1320 |
return array( |
| 1321 |
'block_name' => 'wp:embed', |
| 1322 |
'block_attributes' => $block_attributes, |
| 1323 |
'node' => $node, |
| 1324 |
); |
| 1325 |
} |
| 1326 |
|
| 1327 |
/** |
| 1328 |
* Convert Instagram embed to a link to the Instagram post. |
| 1329 |
* |
| 1330 |
* Currently, Instagram embeds are not supported without the installation |
| 1331 |
* of additional plugins. For this reason, the embed will be converted in |
| 1332 |
* a link to the post. |
| 1333 |
* |
| 1334 |
* @param DomElement $node |
| 1335 |
* @param DomElement $parent |
| 1336 |
* |
| 1337 |
* @return array |
| 1338 |
*/ |
| 1339 |
protected function convert_instagram_embed( DomElement $node, DomElement $parent ) { |
| 1340 |
|
| 1341 |
$data_attributes = json_decode( $node->getAttribute( 'data-attrs' ), true ); |
| 1342 |
|
| 1343 |
$new_node = new DomElement( 'p' ); |
| 1344 |
$link_node = new DomElement( 'a' ); |
| 1345 |
|
| 1346 |
$parent->replaceChild( $new_node, $node ); |
| 1347 |
|
| 1348 |
$new_node->appendChild( $link_node ); |
| 1349 |
|
| 1350 |
$instagram_link = sprintf( 'https://instagram.com/p/%s/', $data_attributes['instagram_id'] ); |
| 1351 |
$link_node->setAttribute( 'href', $instagram_link ); |
| 1352 |
$link_node->setAttribute( 'target', '_blank' ); |
| 1353 |
$link_node->setAttribute( 'rel', 'noreferrer noopener' ); |
| 1354 |
$link_node->textContent = $instagram_link; //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1355 |
|
| 1356 |
return array( |
| 1357 |
'block_name' => 'wp:paragraph', |
| 1358 |
'block_attributes' => array(), |
| 1359 |
'node' => $new_node, |
| 1360 |
); |
| 1361 |
} |
| 1362 |
|
| 1363 |
/** |
| 1364 |
* Convert the node to paragraph block indicating the content was paywalled. |
| 1365 |
* |
| 1366 |
* @param DomElement $node The node to be converted. |
| 1367 |
* @param DomElement $parent The parent of the node to be converted. |
| 1368 |
* |
| 1369 |
* @return array The converted node data. |
| 1370 |
*/ |
| 1371 |
protected function convert_paywall_node( DomElement $node, DomElement $parent ) { |
| 1372 |
/** |
| 1373 |
* Filter the paywall marker text. |
| 1374 |
* |
| 1375 |
* @since 1.2.0 |
| 1376 |
* |
| 1377 |
* @param string $marker_text The default paywall marker text. |
| 1378 |
* @param DomElement $node The paywall node being converted. |
| 1379 |
* @param DomElement $parent The parent element. |
| 1380 |
*/ |
| 1381 |
$marker_text = apply_filters( |
| 1382 |
'substack_importer_paywall_marker_text', |
| 1383 |
__( 'The content below was originally paywalled.', 'substack-importer' ), |
| 1384 |
$node, |
| 1385 |
$parent |
| 1386 |
); |
| 1387 |
|
| 1388 |
/** |
| 1389 |
* Filter the entire paywall conversion result. |
| 1390 |
* |
| 1391 |
* Return a non-null value to override the default conversion. |
| 1392 |
* The returned array should have keys: 'node', 'block_attributes', 'block_name'. |
| 1393 |
* |
| 1394 |
* @since 1.2.0 |
| 1395 |
* |
| 1396 |
* @param array|null $result The conversion result, null to use default. |
| 1397 |
* @param DomElement $node The paywall node being converted. |
| 1398 |
* @param DomElement $parent The parent element. |
| 1399 |
*/ |
| 1400 |
$filtered_result = apply_filters( |
| 1401 |
'substack_importer_paywall_content', |
| 1402 |
null, |
| 1403 |
$node, |
| 1404 |
$parent |
| 1405 |
); |
| 1406 |
|
| 1407 |
if ( null !== $filtered_result ) { |
| 1408 |
return $filtered_result; |
| 1409 |
} |
| 1410 |
|
| 1411 |
// Default behavior: create a paragraph with the marker text. |
| 1412 |
$new_node = new DomElement( 'p' ); |
| 1413 |
$text = new DOMText( $marker_text ); |
| 1414 |
|
| 1415 |
$parent->replaceChild( $new_node, $node ); |
| 1416 |
$new_node->appendChild( $text ); |
| 1417 |
|
| 1418 |
return array( |
| 1419 |
'node' => $new_node, |
| 1420 |
'block_attributes' => array(), |
| 1421 |
'block_name' => 'wp:paragraph', |
| 1422 |
); |
| 1423 |
} |
| 1424 |
|
| 1425 |
/** |
| 1426 |
* Removes the Subscription input field. |
| 1427 |
* |
| 1428 |
* @param DomElement $node The node to be converted. |
| 1429 |
* @param DomElement $parent The parent of the node to be converted. |
| 1430 |
* |
| 1431 |
* @return DomElement The new node. |
| 1432 |
*/ |
| 1433 |
protected function convert_subscription_node( DomElement $node, DomElement $parent ) { |
| 1434 |
$parent->removeChild( $node ); |
| 1435 |
|
| 1436 |
return array( |
| 1437 |
'node' => null, |
| 1438 |
'block_attributes' => array(), |
| 1439 |
'block_name' => null, |
| 1440 |
); |
| 1441 |
} |
| 1442 |
|
| 1443 |
/** |
| 1444 |
* Replace the Substack Embed node with embed markup that is valid for Gutenberg. |
| 1445 |
* |
| 1446 |
* Returns the replacement node. |
| 1447 |
* |
| 1448 |
* @param DomElement $node |
| 1449 |
* @param DomElement $parent |
| 1450 |
* |
| 1451 |
* @return DomElement |
| 1452 |
*/ |
| 1453 |
protected function replace_embed_node( DomElement $node, DomElement $parent, $content ) { |
| 1454 |
$new_node = new DomElement( 'figure' ); |
| 1455 |
$wrapper = new DomElement( 'div' ); |
| 1456 |
|
| 1457 |
$parent->replaceChild( $new_node, $node ); |
| 1458 |
$new_node->appendChild( $wrapper ); |
| 1459 |
$wrapper->setAttribute( 'class', 'wp-block-embed__wrapper' ); |
| 1460 |
|
| 1461 |
// URL needs to be on its own line, see: |
| 1462 |
// https://github.com/wordpress/gutenberg/blob/trunk/packages/block-library/src/embed/save.js#L27 |
| 1463 |
$content = new DOMText( "\n" . $content . "\n" ); |
| 1464 |
$new_node->getElementsByTagName( 'div' )[0]->appendChild( $content ); |
| 1465 |
|
| 1466 |
return $new_node; |
| 1467 |
} |
| 1468 |
|
| 1469 |
/** |
| 1470 |
* Replace an HTML node tag while preserving its attributes and child nodes. |
| 1471 |
* |
| 1472 |
* @param DomElement $node Existing node. |
| 1473 |
* @param DomElement $parent Parent node. |
| 1474 |
* @param string $new_tag_name New node tag. |
| 1475 |
* |
| 1476 |
* @return DomElement |
| 1477 |
*/ |
| 1478 |
protected function replace_html_node_tag( DomElement $node, DomElement $parent, $new_tag_name ) { |
| 1479 |
$document = $node->ownerDocument; //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1480 |
$new_node = $document->createElement( $new_tag_name ); |
| 1481 |
|
| 1482 |
if ( $node->hasAttributes() ) { //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1483 |
foreach ( $node->attributes as $attribute ) { //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1484 |
$new_node->setAttribute( $attribute->nodeName, $attribute->nodeValue ); //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1485 |
} |
| 1486 |
} |
| 1487 |
|
| 1488 |
while ( $node->hasChildNodes() ) { //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1489 |
$new_node->appendChild( $node->firstChild ); //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1490 |
} |
| 1491 |
|
| 1492 |
$parent->replaceChild( $new_node, $node ); |
| 1493 |
|
| 1494 |
return $new_node; |
| 1495 |
} |
| 1496 |
|
| 1497 |
/** |
| 1498 |
* Retrieve additional post information through the Substack Post API. |
| 1499 |
* |
| 1500 |
* The most important data we are after includes author information and comments as this currently is not provided |
| 1501 |
* in the export file. |
| 1502 |
* |
| 1503 |
* It is important to note that comments might not be included or might not contain any information |
| 1504 |
* if the comments are only visible to paid users or if post itself is only accessible to paid users. |
| 1505 |
* |
| 1506 |
* The completeness of information in the response depends on the type of the post (paid vs. public). |
| 1507 |
* |
| 1508 |
* @param string $slug The slug of the post. |
| 1509 |
* |
| 1510 |
* @return string|null Returns a JSON string with post information or null if it could not be retrieved. |
| 1511 |
*/ |
| 1512 |
protected function fetch_post_meta( $slug ) { |
| 1513 |
|
| 1514 |
// If the substack url is not set, we skip this step. |
| 1515 |
if ( ! $this->substack_url ) { |
| 1516 |
return null; |
| 1517 |
} |
| 1518 |
|
| 1519 |
$post_url = sprintf( '%s/api/v1/posts/%s?all_comments=true', $this->substack_url, $slug ); |
| 1520 |
|
| 1521 |
$response = wp_remote_get( $post_url, array( 'redirection' => 0 ) ); |
| 1522 |
|
| 1523 |
if ( is_wp_error( $response ) || 200 !== $response['response']['code'] ) { |
| 1524 |
return null; |
| 1525 |
} |
| 1526 |
|
| 1527 |
return wp_remote_retrieve_body( $response ); |
| 1528 |
} |
| 1529 |
|
| 1530 |
/** |
| 1531 |
* Get meta info from the substack export zip. Returns null if no meta was found. |
| 1532 |
* |
| 1533 |
* @param int $id Substack Post ID. |
| 1534 |
* |
| 1535 |
* @return array|null |
| 1536 |
*/ |
| 1537 |
protected function get_post_meta_from_export( $id ) { |
| 1538 |
$zip = $this->get_export_zip(); |
| 1539 |
|
| 1540 |
if ( is_wp_error( $zip ) ) { |
| 1541 |
return null; |
| 1542 |
} |
| 1543 |
|
| 1544 |
$meta = $zip->getFromName( sprintf( 'meta/%s.json', $id ) ); |
| 1545 |
|
| 1546 |
return $meta |
| 1547 |
? json_decode( $meta, true ) |
| 1548 |
: null; |
| 1549 |
} |
| 1550 |
|
| 1551 |
/** |
| 1552 |
* Returns a generator yielding posts retrieved from the Substack export. |
| 1553 |
* |
| 1554 |
* If a there was a problem retrieving the Zip file, a WP_Error will be returned. |
| 1555 |
* |
| 1556 |
* @return \Generator|WP_Error |
| 1557 |
*/ |
| 1558 |
public function get_posts() { |
| 1559 |
|
| 1560 |
$zip = $this->get_export_zip(); |
| 1561 |
|
| 1562 |
if ( is_wp_error( $zip ) ) { |
| 1563 |
return $zip; |
| 1564 |
} |
| 1565 |
|
| 1566 |
return $this->get_posts_generator( $zip ); |
| 1567 |
} |
| 1568 |
|
| 1569 |
public function get_subscribers() { |
| 1570 |
$zip = $this->get_export_zip(); |
| 1571 |
|
| 1572 |
if ( is_wp_error( $zip ) ) { |
| 1573 |
return $zip; |
| 1574 |
} |
| 1575 |
$email_list_csv_filename = null; |
| 1576 |
for ( $i = 0; $i < $zip->numFiles; $i++ ) { // phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1577 |
$stat = $zip->statIndex( $i ); // phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase |
| 1578 |
if ( str_starts_with( $stat['name'], 'email_list' ) ) { |
| 1579 |
$email_list_csv_filename = $stat['name']; |
| 1580 |
break; |
| 1581 |
} |
| 1582 |
} |
| 1583 |
|
| 1584 |
if ( is_null( $email_list_csv_filename ) ) { |
| 1585 |
return new WP_Error( |
| 1586 |
'no_email_list_in_export_file', |
| 1587 |
__( 'No email_list.*.csv file was found in the archive.' ), |
| 1588 |
$this->get_error_data( $zip, true ) |
| 1589 |
); |
| 1590 |
} |
| 1591 |
|
| 1592 |
$subscriber_csv = $zip->getFromName( $email_list_csv_filename ); |
| 1593 |
|
| 1594 |
$subscribers = explode( "\n", trim( $subscriber_csv ) ); |
| 1595 |
$map = str_getcsv( array_shift( $subscribers ) ); |
| 1596 |
|
| 1597 |
$mapped_subscribers = []; |
| 1598 |
foreach ( $subscribers as $subscriber ) { |
| 1599 |
$subscriber_data = str_getcsv( $subscriber ); |
| 1600 |
// Check if the number of columns matches the number of headers |
| 1601 |
if ( count( $subscriber_data ) === count( $map ) ) { |
| 1602 |
$mapped_subscribers[] = array_combine( $map, $subscriber_data ); |
| 1603 |
} else { |
| 1604 |
$extra = wp_json_encode( |
| 1605 |
[ |
| 1606 |
'headers' => $map, |
| 1607 |
'data' => $subscriber_data, |
| 1608 |
], |
| 1609 |
JSON_PRETTY_PRINT |
| 1610 |
); |
| 1611 |
return new WP_Error( |
| 1612 |
'csv_mismatch', |
| 1613 |
__( 'The number of columns in the email_list.*.csv file does not match the number of headers.' ), |
| 1614 |
$extra |
| 1615 |
); |
| 1616 |
} |
| 1617 |
} |
| 1618 |
|
| 1619 |
return [ $mapped_subscribers, $email_list_csv_filename ]; |
| 1620 |
} |
| 1621 |
|
| 1622 |
public function summarize_subscribers( $subscribers ) { |
| 1623 |
if ( is_wp_error( $subscribers ) ) { |
| 1624 |
return $subscribers; |
| 1625 |
} |
| 1626 |
|
| 1627 |
$num_subscribers = count( $subscribers ); |
| 1628 |
$active = 0; |
| 1629 |
$oldest = strtotime( 'now' ); |
| 1630 |
$newest = 0; |
| 1631 |
$plans = []; |
| 1632 |
foreach ( $subscribers as $subscriber ) { |
| 1633 |
if ( |
| 1634 |
isset( $subscriber['active_subscription'] ) |
| 1635 |
&& 'true' === $subscriber['active_subscription'] |
| 1636 |
) { |
| 1637 |
++$active; |
| 1638 |
} |
| 1639 |
if ( isset( $subscriber['created_at'] ) ) { |
| 1640 |
$created_at = strtotime( $subscriber['created_at'] ); |
| 1641 |
$oldest = min( $oldest, $created_at ); |
| 1642 |
$newest = max( $newest, $created_at ); |
| 1643 |
} |
| 1644 |
if ( isset( $subscriber['plan'] ) ) { |
| 1645 |
$plans[ $subscriber['plan'] ] = ( $plans[ $subscriber['plan'] ] ?? 0 ) + 1; |
| 1646 |
} |
| 1647 |
} |
| 1648 |
|
| 1649 |
return [ |
| 1650 |
'newest' => gmdate( 'F j, Y', $newest ), |
| 1651 |
'oldest' => gmdate( 'F j, Y', $oldest ), |
| 1652 |
'active' => $active, |
| 1653 |
'num_subscribers' => $num_subscribers, |
| 1654 |
'plans' => $plans, |
| 1655 |
]; |
| 1656 |
} |
| 1657 |
|
| 1658 |
protected function get_posts_generator( ZipArchive $zip ) { |
| 1659 |
$post_csv = $zip->getFromName( 'posts.csv' ); |
| 1660 |
|
| 1661 |
$posts = explode( "\n", trim( $post_csv ) ); |
| 1662 |
$map = str_getcsv( array_shift( $posts ) ); |
| 1663 |
|
| 1664 |
foreach ( $posts as $post ) { |
| 1665 |
$post = str_getcsv( $post, ',' ); |
| 1666 |
$post = array_combine( $map, $post ); |
| 1667 |
$post['html_body'] = $zip->getFromName( sprintf( 'posts/%s.html', $post['post_id'] ) ); |
| 1668 |
yield $post; |
| 1669 |
} |
| 1670 |
} |
| 1671 |
|
| 1672 |
/** |
| 1673 |
* Extract the first image URL from post HTML. |
| 1674 |
* |
| 1675 |
* @param string $html HTML body from the Substack export. |
| 1676 |
* |
| 1677 |
* @return string|null |
| 1678 |
*/ |
| 1679 |
protected function get_first_image_url_from_html( $html ) { |
| 1680 |
if ( empty( $html ) ) { |
| 1681 |
return null; |
| 1682 |
} |
| 1683 |
|
| 1684 |
$document = new DOMDocument(); |
| 1685 |
$previous = libxml_use_internal_errors( true ); |
| 1686 |
$loaded = $document->loadHTML( '<?xml encoding="utf-8" ?>' . $html ); |
| 1687 |
libxml_clear_errors(); |
| 1688 |
libxml_use_internal_errors( $previous ); |
| 1689 |
|
| 1690 |
if ( ! $loaded ) { |
| 1691 |
return null; |
| 1692 |
} |
| 1693 |
|
| 1694 |
$images = $document->getElementsByTagName( 'img' ); |
| 1695 |
if ( 0 === $images->count() ) { |
| 1696 |
return null; |
| 1697 |
} |
| 1698 |
|
| 1699 |
$image = $images->item( 0 ); |
| 1700 |
if ( ! $image instanceof DOMElement ) { |
| 1701 |
return null; |
| 1702 |
} |
| 1703 |
|
| 1704 |
$data_attrs = $image->getAttribute( 'data-attrs' ); |
| 1705 |
if ( ! empty( $data_attrs ) ) { |
| 1706 |
$decoded_data = json_decode( html_entity_decode( $data_attrs, ENT_QUOTES ), true ); |
| 1707 |
if ( is_array( $decoded_data ) && ! empty( $decoded_data['src'] ) ) { |
| 1708 |
return esc_url_raw( $decoded_data['src'] ); |
| 1709 |
} |
| 1710 |
} |
| 1711 |
|
| 1712 |
$src = $image->getAttribute( 'src' ); |
| 1713 |
if ( ! empty( $src ) ) { |
| 1714 |
return esc_url_raw( $src ); |
| 1715 |
} |
| 1716 |
|
| 1717 |
return null; |
| 1718 |
} |
| 1719 |
|
| 1720 |
/** |
| 1721 |
* Get error data for a zip archive. |
| 1722 |
* |
| 1723 |
* @param ZipArchive|null $zip The zip archive. |
| 1724 |
* @param int|bool|null $file_open_result The result of the file open operation. |
| 1725 |
* |
| 1726 |
* @return array The error data. |
| 1727 |
*/ |
| 1728 |
private function get_error_data( $zip = null, $file_open_result = null ) { |
| 1729 |
$files = []; |
| 1730 |
|
| 1731 |
if ( is_numeric( $file_open_result ) ) { |
| 1732 |
switch ( $file_open_result ) { |
| 1733 |
case ZipArchive::ER_INCONS: |
| 1734 |
$file_open_result = 'Zip archive inconsistent'; |
| 1735 |
break; |
| 1736 |
|
| 1737 |
case ZipArchive::ER_INVAL: |
| 1738 |
$file_open_result = 'Invalid argument'; |
| 1739 |
break; |
| 1740 |
|
| 1741 |
case ZipArchive::ER_MEMORY: |
| 1742 |
$file_open_result = 'Malloc failure'; |
| 1743 |
break; |
| 1744 |
|
| 1745 |
case ZipArchive::ER_INVAL: |
| 1746 |
$file_open_result = 'No such file'; |
| 1747 |
break; |
| 1748 |
|
| 1749 |
case ZipArchive::ER_NOZIP: |
| 1750 |
$file_open_result = 'Not a zip archive'; |
| 1751 |
break; |
| 1752 |
|
| 1753 |
case ZipArchive::ER_OPEN: |
| 1754 |
$file_open_result = 'Can\'t open file'; |
| 1755 |
break; |
| 1756 |
|
| 1757 |
case ZipArchive::ER_READ: |
| 1758 |
$file_open_result = 'Read error'; |
| 1759 |
break; |
| 1760 |
|
| 1761 |
case ZipArchive::ER_SEEK: |
| 1762 |
$file_open_result = 'Seek error'; |
| 1763 |
break; |
| 1764 |
|
| 1765 |
default: |
| 1766 |
$file_open_result = sprintf( 'Unknow error code (%s)', $file_open_result ); |
| 1767 |
break; |
| 1768 |
} |
| 1769 |
} |
| 1770 |
|
| 1771 |
if ( is_bool( $file_open_result ) ) { |
| 1772 |
$file_open_result = $file_open_result ? 'Success' : 'Failure, no error code'; |
| 1773 |
} |
| 1774 |
|
| 1775 |
if ( ! empty( $zip ) ) { |
| 1776 |
for ( $i = 0; $i < $zip->numFiles; $i++ ) { //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase -- ZipArchive property |
| 1777 |
$files[] = $zip->getNameIndex( $i ); |
| 1778 |
} |
| 1779 |
} |
| 1780 |
|
| 1781 |
return [ |
| 1782 |
'file-path' => $this->export_file_path, |
| 1783 |
'file-open-result' => $file_open_result, |
| 1784 |
'files' => $files, |
| 1785 |
'substack-url' => $this->substack_url, |
| 1786 |
]; |
| 1787 |
} |
| 1788 |
|
| 1789 |
/** |
| 1790 |
* Get a ZipArchive instance of the export file or return an error if it failed. |
| 1791 |
* |
| 1792 |
* @return WP_Error|ZipArchive The zip archive or a WP_error instance on failure. |
| 1793 |
*/ |
| 1794 |
protected function get_export_zip() { |
| 1795 |
if ( ! class_exists( 'ZipArchive' ) ) { |
| 1796 |
return new WP_Error( 'missing_zip_extension', __( 'Could not unzip the Substack export file.' ), $this->get_error_data() ); |
| 1797 |
} |
| 1798 |
|
| 1799 |
$zip = new ZipArchive(); |
| 1800 |
$result = $zip->open( $this->export_file_path ); |
| 1801 |
|
| 1802 |
if ( true !== $result || 0 === $zip->numFiles ) { //phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase -- ZipArchive property |
| 1803 |
return new WP_Error( 'invalid_export_file', __( 'Could not unzip the Substack export file.' ), $this->get_error_data( $zip, $result ) ); |
| 1804 |
} |
| 1805 |
|
| 1806 |
// If posts.csv was not found in the zip archive, the export is invalid. |
| 1807 |
if ( false === $zip->getFromName( 'posts.csv' ) ) { |
| 1808 |
return new WP_Error( 'no_posts_in_export_file', __( 'The export file is not a valid Substack export, no posts.csv was found in the archive. ' ), $this->get_error_data( $zip, $result ) ); |
| 1809 |
} |
| 1810 |
|
| 1811 |
return $zip; |
| 1812 |
} |
| 1813 |
} |
| 1814 |
|
