diff --git a/packages/playground/data-liberation/blueprints-library b/packages/playground/data-liberation/blueprints-library
index b52a93ce17..b1362cbe3c 160000
--- a/packages/playground/data-liberation/blueprints-library
+++ b/packages/playground/data-liberation/blueprints-library
@@ -1 +1 @@
-Subproject commit b52a93ce17562a1964fb27df770792fe165b217b
+Subproject commit b1362cbe3ca0956a36cc00dd769698bde0ba2ec7
diff --git a/packages/playground/data-liberation/bootstrap.php b/packages/playground/data-liberation/bootstrap.php
index 91038c1ae3..dcfab2623f 100644
--- a/packages/playground/data-liberation/bootstrap.php
+++ b/packages/playground/data-liberation/bootstrap.php
@@ -11,11 +11,19 @@
 require_once __DIR__ . '/blueprints-library/src/WordPress/AsyncHttp/Connection.php';
 require_once __DIR__ . '/blueprints-library/src/WordPress/AsyncHttp/Client.php';
 
-require_once __DIR__ . '/src/byte-readers/WP_Byte_Reader.php';
-require_once __DIR__ . '/src/byte-readers/WP_File_Reader.php';
-require_once __DIR__ . '/src/byte-readers/WP_GZ_File_Reader.php';
-require_once __DIR__ . '/src/byte-readers/WP_Remote_File_Reader.php';
-require_once __DIR__ . '/src/byte-readers/WP_Remote_File_Ranged_Reader.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/Filesystem/WP_Abstract_Filesystem.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/Filesystem/WP_Filesystem.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/Filesystem/WP_File_Visitor_Event.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/Filesystem/WP_Filesystem_Visitor.php';
+
+require_once __DIR__ . '/blueprints-library/src/WordPress/ByteReader/WP_Byte_Reader.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/ByteReader/WP_File_Reader.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/ByteReader/WP_GZ_File_Reader.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/ByteReader/WP_Remote_File_Reader.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/ByteReader/WP_Remote_File_Ranged_Reader.php';
+
+require_once __DIR__ . '/blueprints-library/src/WordPress/Zip/ZipStreamReader.php';
+require_once __DIR__ . '/blueprints-library/src/WordPress/Zip/WP_Zip_Filesystem.php';
 
 if (
 	! class_exists( 'WP_HTML_Tag_Processor' ) &&
@@ -48,17 +56,19 @@
 require_once __DIR__ . '/src/block-markup/WP_Block_Markup_Processor.php';
 require_once __DIR__ . '/src/block-markup/WP_Block_Markup_Url_Processor.php';
 require_once __DIR__ . '/src/block-markup/WP_URL_In_Text_Processor.php';
-require_once __DIR__ . '/src/block-markup/WP_HTML_To_Blocks.php';
 require_once __DIR__ . '/src/block-markup/WP_URL.php';
+require_once __DIR__ . '/src/block-markup/WP_HTML_To_Blocks.php';
+
+require_once __DIR__ . '/src/entity-readers/WP_Entity_Reader.php';
+require_once __DIR__ . '/src/entity-readers/WP_HTML_Entity_Reader.php';
+require_once __DIR__ . '/src/entity-readers/WP_EPub_Entity_Reader.php';
+require_once __DIR__ . '/src/entity-readers/WP_WXR_Entity_Reader.php';
+require_once __DIR__ . '/src/entity-readers/WP_Directory_Tree_Entity_Reader.php';
 
 require_once __DIR__ . '/src/xml-api/WP_XML_Decoder.php';
 require_once __DIR__ . '/src/xml-api/WP_XML_Processor.php';
-require_once __DIR__ . '/src/wxr/WP_WXR_Reader.php';
-require_once __DIR__ . '/src/import/WP_Import_Utils.php';
 require_once __DIR__ . '/src/import/WP_Block_Object.php';
 require_once __DIR__ . '/src/import/WP_Entity_Importer.php';
-require_once __DIR__ . '/src/import/WP_File_Visitor.php';
-require_once __DIR__ . '/src/import/WP_File_Visitor_Event.php';
 require_once __DIR__ . '/src/import/WP_Imported_Entity.php';
 require_once __DIR__ . '/src/import/WP_Attachment_Downloader.php';
 require_once __DIR__ . '/src/import/WP_Attachment_Downloader_Event.php';
@@ -66,9 +76,8 @@
 require_once __DIR__ . '/src/import/WP_Stream_Importer.php';
 require_once __DIR__ . '/src/import/WP_Entity_Iterator_Chain.php';
 require_once __DIR__ . '/src/import/WP_Retry_Frontloading_Iterator.php';
-require_once __DIR__ . '/src/entity-readers/WP_Entity_Reader.php';
-require_once __DIR__ . '/src/entity-readers/WP_HTML_Entity_Reader.php';
-
+require_once __DIR__ . '/src/import/WP_Import_HTML_Processor.php';
+require_once __DIR__ . '/src/import/WP_Import_Utils.php';
 require_once __DIR__ . '/src/utf8_decoder.php';
 
 // When running in Playground, the composer autoloader script sees CLI SAPI and
diff --git a/packages/playground/data-liberation/dist/data-liberation-core.phar.gz b/packages/playground/data-liberation/dist/data-liberation-core.phar.gz
index ed86d01db1..5762f9c3fb 100755
Binary files a/packages/playground/data-liberation/dist/data-liberation-core.phar.gz and b/packages/playground/data-liberation/dist/data-liberation-core.phar.gz differ
diff --git a/packages/playground/data-liberation/phpunit.xml b/packages/playground/data-liberation/phpunit.xml
index 9646f33205..cd97c0ec90 100644
--- a/packages/playground/data-liberation/phpunit.xml
+++ b/packages/playground/data-liberation/phpunit.xml
@@ -2,10 +2,11 @@
 <phpunit xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" bootstrap="bootstrap.php" colors="true" xsi:noNamespaceSchemaLocation="https://schema.phpunit.de/10.0/phpunit.xsd" cacheDirectory=".phpunit.cache">
   <testsuites>
     <testsuite name="Application Test Suite">
-      <file>tests/WPHTMLEntityReaderTests.php</file>
-      <file>tests/WPHTMLToBlocksTests.php</file>
       <file>tests/WPWXRReaderTests.php</file>
       <file>tests/WPRewriteUrlsTests.php</file>
+      <file>tests/WPHTMLToBlocksTests.php</file>
+      <file>tests/WPHTMLEntityReaderTests.php</file>
+      <file>tests/WPEPubEntityReaderTests.php</file>
       <file>tests/WPURLInTextProcessorTests.php</file>
       <file>tests/WPBlockMarkupProcessorTests.php</file>
       <file>tests/WPBlockMarkupUrlProcessorTests.php</file>
diff --git a/packages/playground/data-liberation/project.json b/packages/playground/data-liberation/project.json
index f2712fef71..c2e51df4ac 100644
--- a/packages/playground/data-liberation/project.json
+++ b/packages/playground/data-liberation/project.json
@@ -33,7 +33,7 @@
 			"options": {
 				"cwd": "packages/playground/data-liberation",
 				"commands": [
-					"./vendor/bin/phpcs --standard=./phpcs.xml -s ./src ./*.php"
+					"./vendor/bin/phpcs --standard=./phpcs.xml -s ./src ./*.php "
 				],
 				"parallel": false
 			}
@@ -43,7 +43,7 @@
 			"options": {
 				"cwd": "packages/playground/data-liberation",
 				"commands": [
-					"./vendor/bin/phpcbf --standard=./phpcs.xml ./src ./*.php"
+					"./vendor/bin/phpcbf --standard=./phpcs.xml ./src ./*.php "
 				],
 				"parallel": false
 			}
diff --git a/packages/playground/data-liberation/src/byte-readers/WP_Byte_Reader.php b/packages/playground/data-liberation/src/byte-readers/WP_Byte_Reader.php
deleted file mode 100644
index 4671eca2cb..0000000000
--- a/packages/playground/data-liberation/src/byte-readers/WP_Byte_Reader.php
+++ /dev/null
@@ -1,10 +0,0 @@
-<?php
-
-interface WP_Byte_Reader {
-	public function tell(): int;
-	public function seek( int $offset ): bool;
-	public function is_finished(): bool;
-	public function next_bytes(): bool;
-	public function get_bytes(): ?string;
-	public function get_last_error(): ?string;
-}
diff --git a/packages/playground/data-liberation/src/byte-readers/WP_File_Reader.php b/packages/playground/data-liberation/src/byte-readers/WP_File_Reader.php
deleted file mode 100644
index 9100ed7f86..0000000000
--- a/packages/playground/data-liberation/src/byte-readers/WP_File_Reader.php
+++ /dev/null
@@ -1,78 +0,0 @@
-<?php
-
-class WP_File_Reader implements WP_Byte_Reader {
-
-	const STATE_STREAMING = '#streaming';
-	const STATE_FINISHED  = '#finished';
-
-	protected $file_path;
-	protected $chunk_size;
-	protected $file_pointer;
-	protected $offset_in_file;
-	protected $output_bytes    = '';
-	protected $last_chunk_size = 0;
-	protected $last_error;
-	protected $state = self::STATE_STREAMING;
-
-	public function __construct( $file_path, $chunk_size = 8096 ) {
-		$this->file_path  = $file_path;
-		$this->chunk_size = $chunk_size;
-	}
-
-	public function tell(): int {
-		// Save the previous offset, not the current one.
-		// This way, after resuming, the next read will yield the same $output_bytes
-		// as we have now.
-		return $this->offset_in_file - $this->last_chunk_size;
-	}
-
-	public function seek( $offset_in_file ): bool {
-		if ( ! is_int( $offset_in_file ) ) {
-			_doing_it_wrong( __METHOD__, 'Cannot set a file reader cursor to a non-integer offset.', '1.0.0' );
-			return false;
-		}
-		if ( $this->file_pointer ) {
-			_doing_it_wrong( __METHOD__, 'Cannot set a file reader cursor on a file reader that is already initialized.', '1.0.0' );
-			return false;
-		}
-		$this->offset_in_file  = $offset_in_file;
-		$this->last_chunk_size = 0;
-		return true;
-	}
-
-	public function is_finished(): bool {
-		return ! $this->output_bytes && $this->state === static::STATE_FINISHED;
-	}
-
-	public function get_bytes(): string {
-		return $this->output_bytes;
-	}
-
-	public function get_last_error(): ?string {
-		return $this->last_error;
-	}
-
-	public function next_bytes(): bool {
-		$this->output_bytes    = '';
-		$this->last_chunk_size = 0;
-		if ( $this->last_error || $this->is_finished() ) {
-			return false;
-		}
-		if ( ! $this->file_pointer ) {
-			$this->file_pointer = fopen( $this->file_path, 'r' );
-			if ( $this->offset_in_file ) {
-				fseek( $this->file_pointer, $this->offset_in_file );
-			}
-		}
-		$bytes = fread( $this->file_pointer, $this->chunk_size );
-		if ( ! $bytes && feof( $this->file_pointer ) ) {
-			fclose( $this->file_pointer );
-			$this->state = static::STATE_FINISHED;
-			return false;
-		}
-		$this->last_chunk_size = strlen( $bytes );
-		$this->offset_in_file += $this->last_chunk_size;
-		$this->output_bytes   .= $bytes;
-		return true;
-	}
-}
diff --git a/packages/playground/data-liberation/src/byte-readers/WP_GZ_File_Reader.php b/packages/playground/data-liberation/src/byte-readers/WP_GZ_File_Reader.php
deleted file mode 100644
index 1216e7f7cd..0000000000
--- a/packages/playground/data-liberation/src/byte-readers/WP_GZ_File_Reader.php
+++ /dev/null
@@ -1,26 +0,0 @@
-<?php
-
-class WP_GZ_File_Reader extends WP_File_Reader {
-
-	public function next_bytes(): bool {
-		$this->output_bytes = '';
-		if ( $this->last_error || $this->is_finished() ) {
-			return false;
-		}
-		if ( ! $this->file_pointer ) {
-			$this->file_pointer = gzopen( $this->file_path, 'r' );
-			if ( $this->offset_in_file ) {
-				gzseek( $this->file_pointer, $this->offset_in_file );
-			}
-		}
-		$bytes = gzread( $this->file_pointer, $this->chunk_size );
-		if ( ! $bytes && gzeof( $this->file_pointer ) ) {
-			gzclose( $this->file_pointer );
-			$this->state->finish();
-			return false;
-		}
-		$this->offset_in_file += strlen( $bytes );
-		$this->output_bytes   .= $bytes;
-		return true;
-	}
-}
diff --git a/packages/playground/data-liberation/src/byte-readers/WP_Remote_File_Ranged_Reader.php b/packages/playground/data-liberation/src/byte-readers/WP_Remote_File_Ranged_Reader.php
deleted file mode 100644
index 34ac703b18..0000000000
--- a/packages/playground/data-liberation/src/byte-readers/WP_Remote_File_Ranged_Reader.php
+++ /dev/null
@@ -1,187 +0,0 @@
-<?php
-/**
- * Streams bytes from a remote file. Supports seeking to a specific offset and
- * requesting sub-ranges of the file.
- *
- * Usage:
- *
- * $file = new WP_Remote_File_Ranged_Reader('https://example.com/file.txt');
- * $file->seek(0);
- * $file->request_bytes(100);
- * while($file->next_chunk()) {
- *     var_dump($file->get_bytes());
- * }
- * $file->seek(600);
- * $file->request_bytes(40);
- * while($file->next_chunk()) {
- *     var_dump($file->get_bytes());
- * }
- *
- * @TODO: Verify that the remote server supports range requests.
- * @TODO: Support requesting multiple ranges in a single request.
- * @TODO: Abort in-progress requests when seeking to a new offset.
- */
-class WP_Remote_File_Ranged_Reader {
-
-	/**
-	 * @var WordPress\AsyncHttp\Client
-	 */
-	private $client;
-	private $url;
-	private $remote_file_length;
-
-	private $current_request;
-	private $offset_in_remote_file   = 0;
-	private $offset_in_current_chunk = 0;
-	private $current_chunk;
-	private $expected_chunk_size;
-
-	public function __construct( $url, $options = array() ) {
-		$this->client = new WordPress\AsyncHttp\Client();
-		$this->url    = $url;
-	}
-
-	public function request_bytes( $bytes ) {
-		if ( null === $this->remote_file_length ) {
-			$content_length = $this->resolve_content_length();
-			if ( false === $content_length ) {
-				// The remote server won't tell us what the content length is
-				// @TODO: What should we do in this case? Content-length is critical for
-				//        stream-decompressing remote zip files, but we may not need it
-				//        for other use-cases.
-				return false;
-			}
-			$this->remote_file_length = $content_length;
-		}
-
-		if ( $this->offset_in_remote_file < 0 || $this->offset_in_remote_file + $bytes > $this->remote_file_length ) {
-			// TODO: Think through error handling
-			return false;
-		}
-
-		$this->seek( $this->offset_in_remote_file );
-
-		$this->current_request         = new WordPress\AsyncHttp\Request(
-			$this->url,
-			array(
-				'headers' => array(
-					'Range' => 'bytes=' . $this->offset_in_remote_file . '-' . ( $this->offset_in_remote_file + $bytes - 1 ),
-				),
-			)
-		);
-		$this->expected_chunk_size     = $bytes;
-		$this->offset_in_current_chunk = 0;
-		if ( false === $this->client->enqueue( $this->current_request ) ) {
-			// TODO: Think through error handling
-			return false;
-		}
-		return true;
-	}
-
-	public function seek( $offset ) {
-		$this->offset_in_remote_file = $offset;
-		// @TODO cancel any pending requests
-		$this->current_request = null;
-	}
-
-	public function tell() {
-		return $this->offset_in_remote_file;
-	}
-
-	public function resolve_content_length() {
-		if ( null !== $this->remote_file_length ) {
-			return $this->remote_file_length;
-		}
-
-		$request = new WordPress\AsyncHttp\Request(
-			$this->url,
-			array( 'method' => 'HEAD' )
-		);
-		if ( false === $this->client->enqueue( $request ) ) {
-			// TODO: Think through error handling
-			return false;
-		}
-		while ( $this->client->await_next_event() ) {
-			switch ( $this->client->get_event() ) {
-				case WordPress\AsyncHttp\Client::EVENT_GOT_HEADERS:
-					$response = $request->response;
-					if ( false === $response ) {
-						return false;
-					}
-					$content_length = $response->get_header( 'Content-Length' );
-					if ( false === $content_length ) {
-						return false;
-					}
-					return (int) $content_length;
-			}
-		}
-		return false;
-	}
-
-	public function next_chunk() {
-		while ( $this->client->await_next_event() ) {
-			/**
-			 * Only process events related to the most recent request.
-			 * @TODO: Support redirects.
-			 * @TODO: Cleanup resources for stale requests.
-			 */
-			if ( $this->current_request->id !== $this->client->get_request()->id ) {
-				continue;
-			}
-
-			if ( $this->offset_in_current_chunk >= $this->expected_chunk_size ) {
-				// The remote server doesn't support range requests and sent us a chunk larger than expected.
-				// @TODO: Handle this case. Should we stream the entire file, or give up?
-				//        Should we cache the download locally, or request the entire file again every
-				//        time we need to seek()?
-				return false;
-			}
-
-			switch ( $this->client->get_event() ) {
-				case WordPress\AsyncHttp\Client::EVENT_GOT_HEADERS:
-					$request = $this->client->get_request();
-					if ( ! $request ) {
-						return false;
-					}
-					$response = $request->response;
-					if ( false === $response ) {
-						return false;
-					}
-					if (
-						$response->status_code !== 206 ||
-						false === $response->get_header( 'Range' )
-					) {
-						// The remote server doesn't support range requests
-						// @TODO: Handle this case. Should we stream the entire file, or give up?
-						//        Should we cache the download locally, or request the entire file again every
-						//        time we need to seek()?
-						return false;
-					}
-					break;
-				case WordPress\AsyncHttp\Client::EVENT_BODY_CHUNK_AVAILABLE:
-					$chunk = $this->client->get_response_body_chunk();
-					if ( ! is_string( $chunk ) ) {
-						// TODO: Think through error handling
-						return false;
-					}
-					$this->current_chunk            = $chunk;
-					$this->offset_in_remote_file   += strlen( $chunk );
-					$this->offset_in_current_chunk += strlen( $chunk );
-
-					return true;
-				case WordPress\AsyncHttp\Client::EVENT_FAILED:
-					// TODO: Think through error handling. Errors are expected when working with
-					//       the network. Should we auto retry? Make it easy for the caller to retry?
-					//       Something else?
-					return false;
-				case WordPress\AsyncHttp\Client::EVENT_FINISHED:
-					// TODO: Think through error handling
-					return false;
-			}
-		}
-	}
-
-	public function get_bytes() {
-		return $this->current_chunk;
-	}
-}
diff --git a/packages/playground/data-liberation/src/byte-readers/WP_Remote_File_Reader.php b/packages/playground/data-liberation/src/byte-readers/WP_Remote_File_Reader.php
deleted file mode 100644
index d55846f7b8..0000000000
--- a/packages/playground/data-liberation/src/byte-readers/WP_Remote_File_Reader.php
+++ /dev/null
@@ -1,109 +0,0 @@
-<?php
-
-/**
- * Streams bytes from a remote file.
- */
-class WP_Remote_File_Reader implements WP_Byte_Reader {
-
-	/**
-	 * @var WordPress\AsyncHttp\Client
-	 */
-	private $client;
-	private $url;
-	private $request;
-	private $current_chunk;
-	private $last_error;
-	private $is_finished = false;
-	private $bytes_already_read;
-	private $skip_bytes = 0;
-
-	public function __construct( $url ) {
-		$this->client = new WordPress\AsyncHttp\Client();
-		$this->url    = $url;
-	}
-
-	public function tell(): int {
-		return $this->bytes_already_read + $this->skip_bytes;
-	}
-
-	public function seek( $offset_in_file ): bool {
-		if ( $this->request ) {
-			_doing_it_wrong( __METHOD__, 'Cannot set a remote file reader cursor on a remote file reader that is already initialized.', '1.0.0' );
-			return false;
-		}
-		$this->skip_bytes = $offset_in_file;
-		return true;
-	}
-
-	public function next_bytes(): bool {
-		if ( null === $this->request ) {
-			$this->request = new WordPress\AsyncHttp\Request(
-				$this->url
-			);
-			if ( false === $this->client->enqueue( $this->request ) ) {
-				// TODO: Think through error handling
-				return false;
-			}
-		}
-
-		$this->after_chunk();
-
-		while ( $this->client->await_next_event() ) {
-			switch ( $this->client->get_event() ) {
-				case WordPress\AsyncHttp\Client::EVENT_BODY_CHUNK_AVAILABLE:
-					$chunk = $this->client->get_response_body_chunk();
-					if ( ! is_string( $chunk ) ) {
-						// TODO: Think through error handling
-						return false;
-					}
-					$this->current_chunk = $chunk;
-
-					/**
-					 * Naive seek() implementation – redownload the file from the start
-					 * and ignore bytes until we reach the desired offset.
-					 *
-					 * @TODO: Use the range requests instead when the server supports them.
-					 */
-					if ( $this->skip_bytes > 0 ) {
-						if ( $this->skip_bytes < strlen( $chunk ) ) {
-							$this->current_chunk       = substr( $chunk, $this->skip_bytes );
-							$this->bytes_already_read += $this->skip_bytes;
-							$this->skip_bytes          = 0;
-						} else {
-							$this->skip_bytes -= strlen( $chunk );
-							continue 2;
-						}
-					}
-					return true;
-				case WordPress\AsyncHttp\Client::EVENT_FAILED:
-					// TODO: Think through error handling. Errors are expected when working with
-					//       the network. Should we auto retry? Make it easy for the caller to retry?
-					//       Something else?
-					$this->last_error = $this->client->get_request()->error;
-					return false;
-				case WordPress\AsyncHttp\Client::EVENT_FINISHED:
-					$this->is_finished = true;
-					return false;
-			}
-		}
-	}
-
-	private function after_chunk() {
-		if ( $this->current_chunk ) {
-			$this->bytes_already_read += strlen( $this->current_chunk );
-		}
-		$this->current_chunk = null;
-	}
-
-	public function get_last_error(): ?string {
-		return $this->last_error;
-	}
-
-	public function get_bytes(): ?string {
-		return $this->current_chunk;
-	}
-
-	public function is_finished(): bool {
-		return $this->is_finished;
-	}
-}
diff --git a/packages/playground/data-liberation/src/entity-readers/WP_Directory_Tree_Entity_Reader.php b/packages/playground/data-liberation/src/entity-readers/WP_Directory_Tree_Entity_Reader.php
new file mode 100644
index 0000000000..fcbcd70133
--- /dev/null
+++ b/packages/playground/data-liberation/src/entity-readers/WP_Directory_Tree_Entity_Reader.php
@@ -0,0 +1,342 @@
+<?php
+
+/**
+ * Data Liberation: Directory tree entity reader.
+ *
+ * This exploration accompanies the WXR reader to inform a generic
+ * data importing pipeline that's not specific to a single input format.
+ *
+ * @TODO: Explore supporting a cursor to allow resuming from where we left off.
+ */
+class WP_Directory_Tree_Entity_Reader implements \Iterator {
+	private $file_visitor;
+	private $filesystem;
+	private $entity;
+
+	private $pending_directory_index;
+	private $pending_files = array();
+	private $parent_ids    = array();
+	private $next_post_id;
+	private $is_finished          = false;
+	private $entities_read_so_far = 0;
+	private $allowed_extensions   = array();
+	private $index_file_patterns  = array();
+	private $markup_converter_factory;
+
+	public static function create(
+		\WordPress\Filesystem\WP_Abstract_Filesystem $filesystem,
+		$options
+	) {
+		if ( ! isset( $options['root_dir'] ) ) {
+			throw new \Exception( 'Missing required options: root_dir' );
+		}
+		if ( ! isset( $options['first_post_id'] ) ) {
+			throw new \Exception( 'Missing required options: first_post_id' );
+		}
+		if ( ! isset( $options['allowed_extensions'] ) ) {
+			throw new \Exception( 'Missing required options: allowed_extensions' );
+		}
+		if ( ! isset( $options['index_file_patterns'] ) ) {
+			throw new \Exception( 'Missing required options: index_file_patterns' );
+		}
+		/**
+		 * @TODO: Use `sub_entity_reader_factory` instead of `markup_converter_factory`
+		 *        and expect a WP_Entity_Reader factory, not a WP_Markup_Converter factory.
+		 *        This way we'll source all the relevant entity data such as post_meta
+		 *        from the files, not just the post_content.
+		 */
+		if ( ! isset( $options['markup_converter_factory'] ) ) {
+			throw new \Exception( 'Missing required options: markup_converter_factory' );
+		}
+		return new self( $filesystem, $options );
+	}
+
+	private function __construct(
+		\WordPress\Filesystem\WP_Abstract_Filesystem $filesystem,
+		$options
+	) {
+		$this->file_visitor             = new \WordPress\Filesystem\WP_Filesystem_Visitor( $filesystem, $options['root_dir'] );
+		$this->filesystem               = $filesystem;
+		$this->next_post_id             = $options['first_post_id'];
+		$this->allowed_extensions       = $options['allowed_extensions'];
+		$this->index_file_patterns      = $options['index_file_patterns'];
+		$this->markup_converter_factory = $options['markup_converter_factory'];
+	}
+
+	public function next_entity() {
+		while ( true ) {
+			if ( null !== $this->pending_directory_index ) {
+				$dir       = $this->file_visitor->get_event()->dir;
+				$depth     = $this->file_visitor->get_current_depth();
+				$parent_id = $this->parent_ids[ $depth - 1 ] ?? null;
+
+				if ( null === $parent_id && $depth > 1 ) {
+					// There's no parent ID even though we're a few levels deep.
+					// This is a scenario where `next_file()` skipped a few levels
+					// of directories with no relevant content in them:
+					//
+					// - /docs/
+					//   - /foo/
+					//     - /bar/
+					//       - /baz.md
+					//
+					// In this case, we need to backtrack and create the missing
+					// parent pages for /bar/ and /foo/.
+
+					// Find the topmost missing parent ID
+					$missing_parent_id_depth = 1;
+					while ( isset( $this->parent_ids[ $missing_parent_id_depth ] ) ) {
+						++$missing_parent_id_depth;
+					}
+
+					// Move up to the corresponding directory
+					$missing_parent_path = $dir;
+					for ( $i = $missing_parent_id_depth; $i < $depth; $i++ ) {
+						$missing_parent_path = dirname( $missing_parent_path );
+					}
+
+					$this->parent_ids[ $missing_parent_id_depth ] = $this->emit_post_entity(
+						array(
+							'content' => '',
+							'source_path' => $missing_parent_path,
+							'parent_id' => $this->parent_ids[ $missing_parent_id_depth - 1 ],
+							'title_fallback' => WP_Import_Utils::slug_to_title( basename( $missing_parent_path ) ),
+						)
+					);
+				} elseif ( false === $this->pending_directory_index ) {
+					// No directory index candidate – let's create a fake page
+					// just to have something in the page tree.
+					$this->parent_ids[ $depth ] = $this->emit_post_entity(
+						array(
+							'content' => '',
+							'source_path' => $dir,
+							'parent_id' => $parent_id,
+							'title_fallback' => WP_Import_Utils::slug_to_title( basename( $dir ) ),
+						)
+					);
+					// We're no longer looking for a directory index.
+					$this->pending_directory_index = null;
+				} else {
+					$file_path                  = $this->pending_directory_index;
+					$this->parent_ids[ $depth ] = $this->emit_post_entity(
+						array(
+							'content' => $this->filesystem->read_file( $file_path ),
+							'source_path' => $file_path,
+							'parent_id' => $parent_id,
+							'title_fallback' => WP_Import_Utils::slug_to_title( basename( $file_path ) ),
+						)
+					);
+					// We're no longer looking for a directory index.
+					$this->pending_directory_index = null;
+				}
+				return true;
+			}
+
+			while ( count( $this->pending_files ) ) {
+				$parent_id = $this->parent_ids[ $this->file_visitor->get_current_depth() ] ?? null;
+				$file_path = array_shift( $this->pending_files );
+				$this->emit_post_entity(
+					array(
+						'content' => $this->filesystem->read_file( $file_path ),
+						'source_path' => $file_path,
+						'parent_id' => $parent_id,
+						'title_fallback' => WP_Import_Utils::slug_to_title( basename( $file_path ) ),
+					)
+				);
+				return true;
+			}
+
+			if ( false === $this->next_file() ) {
+				break;
+			}
+		}
+		$this->is_finished = true;
+		return false;
+	}
+
+	public function get_entity(): ?\WP_Imported_Entity {
+		return $this->entity;
+	}
+
+	protected function emit_post_entity( $options ) {
+		$factory   = $this->markup_converter_factory;
+		$converter = $factory( $options['content'] );
+		$converter->convert();
+		$block_markup = $converter->get_block_markup();
+
+		$post_title = null;
+		if ( ! $post_title ) {
+			$removed_title = WP_Import_Utils::remove_first_h1_block_from_block_markup( $block_markup );
+			if ( false !== $removed_title ) {
+				$post_title   = $removed_title['title'];
+				$block_markup = $removed_title['remaining_html'];
+			}
+		}
+		if ( ! $post_title ) {
+			// In Markdown, the frontmatter title can be a worse title candidate than
+			// the first H1 block. In block markup exports, it will be the opposite.
+			//
+			// @TODO: Enable the API consumer to customize the title resolution.
+			$post_title = $converter->get_meta_value( 'post_title' );
+		}
+		if ( ! $post_title ) {
+			$post_title = $options['title_fallback'];
+		}
+
+		$entity_data = array(
+			'post_id' => $this->next_post_id,
+			'post_type' => 'page',
+			'guid' => $options['source_path'],
+			'post_title' => $post_title,
+			'post_content' => $block_markup,
+			'post_excerpt' => $converter->get_meta_value( 'post_excerpt' ) ?? '',
+			'post_status' => 'publish',
+		);
+
+		/**
+		 * Technically `source_path` isn't a part of the WordPress post object,
+		 * but we need it to resolve relative URLs in the imported content.
+		 *
+		 * This path is relative to the root directory traversed by this class.
+		 */
+		if ( ! empty( $options['source_path'] ) ) {
+			$source_path = $options['source_path'];
+			$root_dir    = $this->file_visitor->get_root_dir();
+			if ( str_starts_with( $source_path, $root_dir ) ) {
+				$source_path = substr( $source_path, strlen( $root_dir ) );
+			}
+			$source_path                = ltrim( $source_path, '/' );
+			$entity_data['source_path'] = $source_path;
+		}
+
+		if ( $converter->get_meta_value( 'slug' ) ) {
+			$slug                     = $converter->get_meta_value( 'slug' );
+			$last_segment             = substr( $slug, strrpos( $slug, '/' ) + 1 );
+			$entity_data['post_name'] = $last_segment;
+		}
+
+		if ( $converter->get_meta_value( 'post_order' ) ) {
+			$entity_data['post_order'] = $converter->get_meta_value( 'post_order' );
+		}
+
+		if ( $options['parent_id'] ) {
+			$entity_data['post_parent'] = $options['parent_id'];
+		}
+
+		$this->entity = new \WP_Imported_Entity( 'post', $entity_data );
+		++$this->next_post_id;
+		++$this->entities_read_so_far;
+		return $entity_data['post_id'];
+	}
+
+	private function next_file() {
+		$this->pending_files = array();
+		$this->entity        = null;
+		while ( $this->file_visitor->next() ) {
+			$event = $this->file_visitor->get_event();
+
+			if ( $event->is_exiting() ) {
+				// Clean up stale IDs to save some memory when processing
+				// large directory trees.
+				unset( $this->parent_ids[ $event->dir ] );
+				continue;
+			}
+
+			if ( $event->is_entering() ) {
+				$abs_paths = array();
+				foreach ( $event->files as $filename ) {
+					$abs_paths[] = $event->dir . '/' . $filename;
+				}
+				$this->pending_files = $this->choose_relevant_files( $abs_paths );
+				if ( ! count( $this->pending_files ) ) {
+					// Only consider directories with relevant files in them.
+					// Otherwise we'll create fake pages for media directories
+					// and other directories that don't contain any content.
+					//
+					// One corner case is when there's a few levels of directories
+					// with a single relevant file at the bottom:
+					//
+					// - /docs/
+					//   - /foo/
+					//     - /bar/
+					//       - /baz.md
+					//
+					// In this case, `next_entity()` will backtrack at baz.md and
+					// create the missing parent pages.
+					continue;
+				}
+				$directory_index_idx = $this->choose_directory_index( $this->pending_files );
+				if ( -1 === $directory_index_idx ) {
+					$this->pending_directory_index = false;
+				} else {
+					$this->pending_directory_index = $this->pending_files[ $directory_index_idx ];
+					unset( $this->pending_files[ $directory_index_idx ] );
+				}
+				return true;
+			}
+
+			return false;
+		}
+		return false;
+	}
+
+	protected function choose_directory_index( $files ) {
+		foreach ( $files as $idx => $file ) {
+			if ( $this->looks_like_directory_index( $file ) ) {
+				return $idx;
+			}
+		}
+		return -1;
+	}
+
+	protected function looks_like_directory_index( $path ) {
+		$filename = basename( $path );
+		foreach ( $this->index_file_patterns as $pattern ) {
+			if ( preg_match( $pattern, $filename ) ) {
+				return true;
+			}
+		}
+		return false;
+	}
+
+	protected function choose_relevant_files( $paths ) {
+		return array_filter( $paths, array( $this, 'is_valid_file' ) );
+	}
+
+	protected function is_valid_file( $path ) {
+		$extension = pathinfo( $path, PATHINFO_EXTENSION );
+		return in_array( $extension, $this->allowed_extensions, true );
+	}
+
+	/**
+	 * @TODO: Either implement this method, or introduce a concept of
+	 *        reentrant and non-reentrant entity readers.
+	 */
+	public function get_reentrancy_cursor() {
+		return '';
+	}
+
+	public function current(): mixed {
+		if ( null === $this->entity && ! $this->is_finished ) {
+			$this->next();
+		}
+		return $this->get_entity();
+	}
+
+	public function next(): void {
+		$this->next_entity();
+	}
+
+	public function key(): int {
+		return $this->entities_read_so_far - 1;
+	}
+
+	public function valid(): bool {
+		return ! $this->is_finished;
+	}
+
+	public function rewind(): void {
+		// @TODO: Either implement this method, or formalize the fact that
+		//        entity readers are not rewindable.
+	}
+}
diff --git a/packages/playground/data-liberation/src/entity-readers/WP_HTML_Entity_Reader.php b/packages/playground/data-liberation/src/entity-readers/WP_HTML_Entity_Reader.php
index 531534a73f..b01bd0c875 100644
--- a/packages/playground/data-liberation/src/entity-readers/WP_HTML_Entity_Reader.php
+++ b/packages/playground/data-liberation/src/entity-readers/WP_HTML_Entity_Reader.php
@@ -1,7 +1,11 @@
 <?php
 
+use WordPress\Data_Liberation\Block_Markup\WP_HTML_To_Blocks;
+
 /**
- * Converts a single HTML file into a stream of WordPress posts and post meta.
+ * Converts a single HTML file into a stream of WordPress entities.
+ *
+ * @TODO: Support post meta.
  */
 class WP_HTML_Entity_Reader extends WP_Entity_Reader {
 
@@ -15,11 +19,6 @@ public function __construct( $html, $post_id ) {
 		$this->post_id = $post_id;
 	}
 
-	/**
-	 * Advances to the next entity.
-	 *
-	 * @return bool Whether the next entity was found.
-	 */
 	public function next_entity() {
 		// If we're finished, we're finished.
 		if ( $this->finished ) {
@@ -79,11 +78,6 @@ public function next_entity() {
 		return true;
 	}
 
-	/**
-	 * Returns the current entity.
-	 *
-	 * @return WP_Imported_Entity|false The current entity, or false if there are no entities left.
-	 */
 	public function get_entity() {
 		if ( $this->is_finished() ) {
 			return false;
@@ -91,20 +85,10 @@ public function get_entity() {
 		return $this->entities[0];
 	}
 
-	/**
-	 * Checks if this reader has finished yet.
-	 *
-	 * @return bool Whether the reader has finished.
-	 */
 	public function is_finished(): bool {
 		return $this->finished;
 	}
 
-	/**
-	 * Returns the last error that occurred when processing the HTML.
-	 *
-	 * @return string|null The last error, or null if there was no error.
-	 */
 	public function get_last_error(): ?string {
 		return null;
 	}
diff --git a/packages/playground/data-liberation/src/wxr/WP_WXR_Reader.php b/packages/playground/data-liberation/src/entity-readers/WP_WXR_Entity_Reader.php
similarity index 93%
rename from packages/playground/data-liberation/src/wxr/WP_WXR_Reader.php
rename to packages/playground/data-liberation/src/entity-readers/WP_WXR_Entity_Reader.php
index 25c21ff608..398983c370 100644
--- a/packages/playground/data-liberation/src/wxr/WP_WXR_Reader.php
+++ b/packages/playground/data-liberation/src/entity-readers/WP_WXR_Entity_Reader.php
@@ -1,6 +1,9 @@
 <?php
+
+use WordPress\ByteReader\WP_Byte_Reader;
+
 /**
- * Data Liberation API: WP_WXR_Reader class
+ * Data Liberation API: WP_WXR_Entity_Reader class
  *
  * Reads WordPress eXtended RSS (WXR) files and emits entities like posts,
  * comments, users, and terms. Enables efficient processing of large WXR
@@ -11,7 +14,7 @@
  *
  * ## Design goals
  *
- * WP_WXR_Reader is built with the following characteristics in mind:
+ * WP_WXR_Entity_Reader is built with the following characteristics in mind:
  *
  * * Speed – it should be as fast as possible
  * * No PHP extensions required – it can run on any PHP installation
@@ -21,7 +24,7 @@
  *
  * ## Implementation
  *
- * `WP_WXR_Reader` uses the `WP_XML_Processor` to find XML tags representing meaningful
+ * `WP_WXR_Entity_Reader` uses the `WP_XML_Processor` to find XML tags representing meaningful
  * WordPress entities. The reader knows the WXR schema and only looks for relevant elements.
  * For example, it knows that posts are stored in `rss > channel > item` and comments are
  * stored in `rss > channel > item > `wp:comment`.
@@ -33,7 +36,7 @@
  *
  * Example:
  *
- *     $reader = WP_WXR_Reader::create_for_streaming();
+ *     $reader = WP_WXR_Entity_Reader::create_for_streaming();
  *
  *     // Add data as it becomes available
  *     $reader->append_bytes( fread( $file_handle, 8192 ) );
@@ -64,24 +67,24 @@
  *     }
  *
  * The next_entity() -> fread -> break usage pattern may seem a bit tedious. This is expected. Even
- * if the WXR parsing part of the WP_WXR_Reader offers a high-level API, working with byte streams
+ * if the WXR parsing part of the WP_WXR_Entity_Reader offers a high-level API, working with byte streams
  * requires reasoning on a much lower level. The StreamChain class shipped in this repository will
  * make the API consumption easier with its transformation–oriented API for chaining data processors.
  *
- * Similarly to `WP_XML_Processor`, the `WP_WXR_Reader` enters a paused state when it doesn't
+ * Similarly to `WP_XML_Processor`, the `WP_WXR_Entity_Reader` enters a paused state when it doesn't
  * have enough XML bytes to parse the entire entity.
  *
  * ## Caveats
  *
  * ### Extensibility
  *
- * `WP_WXR_Reader` ignores any XML elements it doesn't recognize. The WXR format is extensible
+ * `WP_WXR_Entity_Reader` ignores any XML elements it doesn't recognize. The WXR format is extensible
  * so in the future the  reader may start supporting registration of custom handlers for unknown
  * tags in the future.
  *
  * ### Nested entities intertwined with data
  *
- * `WP_WXR_Reader` flushes the current entity whenever another entity starts. The upside is
+ * `WP_WXR_Entity_Reader` flushes the current entity whenever another entity starts. The upside is
  * simplicity and a tiny memory footprint. The downside is that it's possible to craft a WXR
  * document where some information would be lost. For example:
  *
@@ -101,7 +104,7 @@
  * </rss>
  * ```
  *
- * `WP_WXR_Reader` would accumulate post data until the `wp:post_meta` tag. Then it would emit a
+ * `WP_WXR_Entity_Reader` would accumulate post data until the `wp:post_meta` tag. Then it would emit a
  * `post` entity and accumulate the meta information until the `</wp:postmeta>` closer. Then it
  * would advance to `<wp:post_id>` and **ignore it**.
  *
@@ -122,7 +125,7 @@
  *
  * @since WP_VERSION
  */
-class WP_WXR_Reader implements Iterator {
+class WP_WXR_Entity_Reader extends WP_Entity_Reader {
 
 	/**
 	 * The XML processor used to parse the WXR file.
@@ -355,7 +358,7 @@ public static function create( WP_Byte_Reader $upstream = null, $cursor = null )
 			if ( false === $cursor ) {
 				_doing_it_wrong(
 					__METHOD__,
-					'Invalid cursor provided for WP_WXR_Reader::create().',
+					'Invalid cursor provided for WP_WXR_Entity_Reader::create().',
 					null
 				);
 				return false;
@@ -364,7 +367,7 @@ public static function create( WP_Byte_Reader $upstream = null, $cursor = null )
 		}
 
 		$xml    = WP_XML_Processor::create_for_streaming( '', $xml_cursor );
-		$reader = new WP_WXR_Reader( $xml );
+		$reader = new WP_WXR_Entity_Reader( $xml );
 		if ( null !== $cursor ) {
 			$reader->last_post_id    = $cursor['last_post_id'];
 			$reader->last_comment_id = $cursor['last_comment_id'];
@@ -627,7 +630,7 @@ private function read_next_entity() {
 			/**
 			 * Custom adjustment: the Accessibility WXR file uses a non-standard
 			 * wp:wp_author tag.
-			 * @TODO: Should WP_WXR_Reader care about such non-standard tags when
+			 * @TODO: Should WP_WXR_Entity_Reader care about such non-standard tags when
 			 *        the regular WXR importer would ignore them? Perhaps a warning
 			 *        and an upstream PR would be a better solution.
 			 */
@@ -672,7 +675,7 @@ private function read_next_entity() {
 			 *         Long time ago...
 			 *     </wp:comment_content>
 			 *
-			 * The semantics of such a structure is not clear. The WP_WXR_Reader will
+			 * The semantics of such a structure is not clear. The WP_WXR_Entity_Reader will
 			 * enter an error state when it encounters such a structure.
 			 *
 			 * Such nesting wasn't found in any WXR files analyzed when building
@@ -901,38 +904,4 @@ private function after_entity() {
 		$this->text_buffer            = '';
 		$this->last_opener_attributes = array();
 	}
-
-	public function current(): object {
-		if ( null === $this->entity_data && ! $this->is_finished() && ! $this->get_last_error() ) {
-			$this->next();
-		}
-		return $this->get_entity();
-	}
-
-	private $last_next_result = null;
-	public function next(): void {
-		// @TODO: Don't keep track of this. Just make sure the next_entity()
-		//        call will make the is_finished() true.
-		$this->last_next_result = $this->next_entity();
-	}
-
-	public function key(): string {
-		return $this->get_reentrancy_cursor();
-	}
-
-	public function valid(): bool {
-		return false !== $this->last_next_result && ! $this->is_finished() && ! $this->get_last_error();
-	}
-
-	public function rewind(): void {
-		// Haven't started yet.
-		if ( null === $this->last_next_result ) {
-			return;
-		}
-		_doing_it_wrong(
-			__METHOD__,
-			'WP_WXR_Reader does not support rewinding.',
-			null
-		);
-	}
 }
diff --git a/packages/playground/data-liberation/src/import/WP_Import_HTML_Processor.php b/packages/playground/data-liberation/src/import/WP_Import_HTML_Processor.php
new file mode 100644
index 0000000000..eced41d5f9
--- /dev/null
+++ b/packages/playground/data-liberation/src/import/WP_Import_HTML_Processor.php
@@ -0,0 +1,20 @@
+<?php
+
+/**
+ * A copy of the WP_Interactivity_API_Directives_Processor class
+ * from the Gutenberg plugin.
+ *
+ * @package WordPress
+ * @subpackage Markdown API
+ */
+
+final class WP_Import_HTML_Processor extends \WP_HTML_Processor {
+
+	public function get_string_index_after_current_token() {
+		$name = 'current_token';
+		$this->set_bookmark( $name );
+		$bookmark = $this->bookmarks[ '_' . $name ];
+		$this->release_bookmark( $name );
+		return $bookmark->start + $bookmark->length;
+	}
+}
diff --git a/packages/playground/data-liberation/src/import/WP_Import_Session.php b/packages/playground/data-liberation/src/import/WP_Import_Session.php
index a731f3f9fc..931dbd1b70 100644
--- a/packages/playground/data-liberation/src/import/WP_Import_Session.php
+++ b/packages/playground/data-liberation/src/import/WP_Import_Session.php
@@ -510,6 +510,7 @@ public function bump_frontloading_progress( $frontloading_progress, $events = ar
 
 			$attempts     = get_post_meta( $placeholder->ID, 'attempts', true );
 			$new_attempts = $attempts;
+			$new_status   = $placeholder->post_status;
 			switch ( $event->type ) {
 				case WP_Attachment_Downloader_Event::SUCCESS:
 					$new_status   = self::FRONTLOAD_STATUS_SUCCEEDED;
diff --git a/packages/playground/data-liberation/src/import/WP_Stream_Importer.php b/packages/playground/data-liberation/src/import/WP_Stream_Importer.php
index 41f08a4e7c..77ebd09c00 100644
--- a/packages/playground/data-liberation/src/import/WP_Stream_Importer.php
+++ b/packages/playground/data-liberation/src/import/WP_Stream_Importer.php
@@ -2,6 +2,9 @@
 
 use WordPress\AsyncHTTP\Client;
 use WordPress\AsyncHTTP\Request;
+use WordPress\ByteReader\WP_File_Reader;
+use WordPress\ByteReader\WP_Remote_File_Reader;
+use WordPress\Filesystem\WP_Byte_Reader;
 
 /**
  * Idea:
@@ -35,12 +38,12 @@ class WP_Stream_Importer {
 	/**
 	 * Populated from the WXR file's <wp:base_blog_url> tag.
 	 */
-	private $source_site_url;
+	protected $source_site_url;
 	/**
 	 * A list of [original_url, migrated_url] pairs for rewriting the URLs
 	 * in the imported content.
 	 */
-	private $site_url_mapping = array();
+	protected $site_url_mapping = array();
 	/**
 	 * A list of candidate base URLs that have been spotted in the WXR file.
 	 *
@@ -66,8 +69,8 @@ class WP_Stream_Importer {
 	 * Once the API consumer decides on the mapping, it can call
 	 * add_site_url_mapping() to tell the importer what to map that domain to.
 	 */
-	private $site_url_mapping_candidates = array();
-	private $entity_iterator_factory;
+	protected $site_url_mapping_candidates = array();
+	protected $entity_iterator_factory;
 	/**
 	 * @param array|string|null $query {
 	 *     @type string      $uploads_path  The directory to download the media attachments to.
@@ -76,7 +79,7 @@ class WP_Stream_Importer {
 	 *                                      after the import. E.g. http://127.0.0.1:9400/wp-content/uploads/
 	 * }
 	 */
-	private $options;
+	protected $options;
 
 	const STAGE_INITIAL          = '#initial';
 	const STAGE_INDEX_ENTITIES   = '#index_entities';
@@ -98,7 +101,7 @@ class WP_Stream_Importer {
 	 * The current state of the import process.
 	 * @var string
 	 */
-	private $stage = self::STAGE_INITIAL;
+	protected $stage = self::STAGE_INITIAL;
 	/**
 	 * The next stage of the import process. An explicit call to
 	 * next_stage() is required to advance the importer.
@@ -108,13 +111,13 @@ class WP_Stream_Importer {
 	 * failed to download.
 	 * @var string
 	 */
-	private $next_stage;
+	protected $next_stage;
 
 	/**
 	 * Iterator that streams entities to import.
 	 */
-	private $entity_iterator;
-	private $resume_at_entity;
+	protected $entity_iterator;
+	protected $resume_at_entity;
 	/**
 	 * A map of currently downloaded resources for each entity in
 	 * the following format:
@@ -123,13 +126,13 @@ class WP_Stream_Importer {
 	 *
 	 * @var array<string,array<string,bool>>
 	 */
-	private $active_downloads = array();
-	private $downloader;
+	protected $active_downloads = array();
+	protected $downloader;
 
 	public static function create_for_wxr_file( $wxr_path, $options = array(), $cursor = null ) {
 		return static::create(
 			function ( $cursor = null ) use ( $wxr_path ) {
-				return WP_WXR_Reader::create( new WP_File_Reader( $wxr_path ), $cursor );
+				return WP_WXR_Entity_Reader::create( WP_File_Reader::create( $wxr_path ), $cursor );
 			},
 			$options,
 			$cursor
@@ -139,7 +142,7 @@ function ( $cursor = null ) use ( $wxr_path ) {
 	public static function create_for_wxr_url( $wxr_url, $options = array(), $cursor = null ) {
 		return static::create(
 			function ( $cursor = null ) use ( $wxr_url ) {
-				return WP_WXR_Reader::create( new WP_Remote_File_Reader( $wxr_url ), $cursor );
+				return WP_WXR_Entity_Reader::create( new WP_Remote_File_Reader( $wxr_url ), $cursor );
 			},
 			$options,
 			$cursor
@@ -152,14 +155,14 @@ public static function create(
 		$cursor = null
 	) {
 		$options  = static::parse_options( $options );
-		$importer = new WP_Stream_Importer( $entity_iterator_factory, $options );
+		$importer = new static( $entity_iterator_factory, $options );
 		if ( null !== $cursor && true !== $importer->initialize_from_cursor( $cursor ) ) {
 			return false;
 		}
 		return $importer;
 	}
 
-	private function initialize_from_cursor( $cursor ) {
+	protected function initialize_from_cursor( $cursor ) {
 		$cursor = json_decode( $cursor, true );
 		if ( ! is_array( $cursor ) ) {
 			_doing_it_wrong( __METHOD__, 'Cannot resume an importer with a non-array cursor.', '1.0.0' );
@@ -182,7 +185,7 @@ private function initialize_from_cursor( $cursor ) {
 		return true;
 	}
 
-	private function set_source_site_url( $source_site_url ) {
+	protected function set_source_site_url( $source_site_url ) {
 		$this->source_site_url = $source_site_url;
 		// -1 is a well-known index for the source site URL.
 		// Every subsequent call to set_source_site_url() will
@@ -235,7 +238,7 @@ public function get_reentrancy_cursor() {
 		);
 	}
 
-	private static function parse_options( $options ) {
+	protected static function parse_options( $options ) {
 		if ( ! isset( $options['new_site_url'] ) ) {
 			$options['new_site_url'] = get_site_url();
 		}
@@ -255,7 +258,7 @@ private static function parse_options( $options ) {
 		return $options;
 	}
 
-	private function __construct(
+	protected function __construct(
 		$entity_iterator_factory,
 		$options = array()
 	) {
@@ -266,7 +269,7 @@ private function __construct(
 		}
 	}
 
-	private $frontloading_retries_iterator;
+	protected $frontloading_retries_iterator;
 	public function set_frontloading_retries_iterator( $frontloading_retries_iterator ) {
 		$this->frontloading_retries_iterator = $frontloading_retries_iterator;
 	}
@@ -277,7 +280,7 @@ public function set_frontloading_retries_iterator( $frontloading_retries_iterato
 	 *
 	 * @var WP_Entity_Importer
 	 */
-	private $importer;
+	protected $importer;
 
 	public function next_step() {
 		switch ( $this->stage ) {
@@ -328,10 +331,10 @@ public function advance_to_next_stage() {
 		return true;
 	}
 
-	private $indexed_entities_counts = array();
-	private $indexed_assets_urls     = array();
+	protected $indexed_entities_counts = array();
+	protected $indexed_assets_urls     = array();
 
-	private function index_next_entities( $count = 10000 ) {
+	protected function index_next_entities( $count = 10000 ) {
 		if ( null !== $this->next_stage ) {
 			return false;
 		}
@@ -454,7 +457,7 @@ public function get_indexed_assets_urls() {
 		return $this->indexed_assets_urls;
 	}
 
-	private $frontloading_events = array();
+	protected $frontloading_events = array();
 	public function get_frontloading_events() {
 		return $this->frontloading_events;
 	}
@@ -475,7 +478,7 @@ public function get_frontloading_progress() {
 	 * downloader will enqueue B for download and will skip C and D since
 	 * the relevant files already exist in the filesystem.
 	 */
-	private function frontloading_advance_reentrancy_cursor() {
+	protected function frontloading_advance_reentrancy_cursor() {
 		while ( $this->downloader->next_event() ) {
 			$event = $this->downloader->get_event();
 			switch ( $event->type ) {
@@ -510,7 +513,7 @@ private function frontloading_advance_reentrancy_cursor() {
 	 * before import_entities() so that every inserted post already has
 	 * all its attachments downloaded.
 	 */
-	private function frontload_next_entity() {
+	protected function frontload_next_entity() {
 		if ( null === $this->entity_iterator ) {
 			$this->entity_iterator = new WP_Entity_Iterator_Chain();
 			if ( null !== $this->frontloading_retries_iterator ) {
@@ -615,7 +618,7 @@ private function frontload_next_entity() {
 	 *        large datasets, but maybe it could be a choice for
 	 *        the API consumer?
 	 */
-	private function import_next_entity() {
+	protected function import_next_entity() {
 		if ( null !== $this->next_stage ) {
 			return false;
 		}
@@ -717,8 +720,8 @@ private function import_next_entity() {
 		return true;
 	}
 
-	private $imported_entities_counts = array();
-	private function count_imported_entity( $type ) {
+	protected $imported_entities_counts = array();
+	protected function count_imported_entity( $type ) {
 		if ( ! array_key_exists( $type, $this->imported_entities_counts ) ) {
 			$this->imported_entities_counts[ $type ] = 0;
 		}
@@ -728,7 +731,7 @@ public function get_imported_entities_counts() {
 		return $this->imported_entities_counts;
 	}
 
-	private function enqueue_attachment_download( string $raw_url, $options = array() ) {
+	protected function enqueue_attachment_download( string $raw_url, $options = array() ) {
 		$output_filename = $this->new_asset_filename(
 			$options['original_url'] ?? $raw_url,
 			$options['context_path'] ?? null
@@ -776,7 +779,7 @@ private function enqueue_attachment_download( string $raw_url, $options = array(
 	 *   different permissions. Just because Bob deletes his copy, doesn't
 	 *   mean we should delete Alice's copy.
 	 */
-	private function new_asset_filename( string $raw_asset_url, $context_path = null ) {
+	protected function new_asset_filename( string $raw_asset_url, $context_path = null ) {
 		$raw_asset_url = $this->rewrite_attachment_url(
 			$raw_asset_url,
 			$context_path
@@ -798,7 +801,7 @@ private function new_asset_filename( string $raw_asset_url, $context_path = null
 		return $filename;
 	}
 
-	private function rewrite_attachment_url( string $raw_url, $context_path = null ) {
+	protected function rewrite_attachment_url( string $raw_url, $context_path = null ) {
 		if ( WP_URL::can_parse( $raw_url ) ) {
 			// Absolute URL, nothing to do.
 			return $raw_url;
@@ -821,7 +824,7 @@ private function rewrite_attachment_url( string $raw_url, $context_path = null )
 	 * @TODO: How can we process the videos?
 	 * @TODO: What other asset types are there?
 	 */
-	private function url_processor_matched_asset_url( WP_Block_Markup_Url_Processor $p ) {
+	protected function url_processor_matched_asset_url( WP_Block_Markup_Url_Processor $p ) {
 		return (
 			$p->get_tag() === 'IMG' &&
 			$p->get_inspected_attribute_name() === 'src' &&
@@ -829,11 +832,11 @@ private function url_processor_matched_asset_url( WP_Block_Markup_Url_Processor
 		);
 	}
 
-	private function is_child_of_a_mapped_url( $url ) {
+	protected function is_child_of_a_mapped_url( $url ) {
 		return $this->get_url_mapping_target( $url ) !== false;
 	}
 
-	private function get_url_mapping_target( $source_url ) {
+	protected function get_url_mapping_target( $source_url ) {
 		$url = WP_URL::parse( $source_url );
 		foreach ( $this->site_url_mapping as $pair ) {
 			$parsed_base_url = $pair[0];
@@ -844,8 +847,8 @@ private function get_url_mapping_target( $source_url ) {
 		return false;
 	}
 
-	private $first_iterator = true;
-	private function create_entity_iterator() {
+	protected $first_iterator = true;
+	protected function create_entity_iterator() {
 		$factory = $this->entity_iterator_factory;
 		if ( $this->first_iterator ) {
 			$this->first_iterator = false;
diff --git a/packages/playground/data-liberation/src/wxr/WXR_Import_Info.php b/packages/playground/data-liberation/src/wxr/WXR_Import_Info.php
deleted file mode 100644
index 544305ba68..0000000000
--- a/packages/playground/data-liberation/src/wxr/WXR_Import_Info.php
+++ /dev/null
@@ -1,29 +0,0 @@
-<?php
-/**
- * A fork of the WordPress Importer plugin (humanmade/WordPress-Importer).
- * It's meant to be rewritten to use the new libraries shipped with Data Liberation:
- *
- * * HTML parser
- * * Streaming XML parser
- * * Streaming multi-request HTTP client
- *
- * Original plugin created by Ryan Boren, Jon Cave (@joncave), Andrew Nacin (@nacin),
- * and Peter Westwood (@westi). Redux project by Ryan McCue and contributors.
- *
- * See https://github.com/humanmade/WordPress-Importer/blob/master/class-wxr-importer.php
- */
-class WXR_Import_Info {
-	public $home;
-	public $siteurl;
-
-	public $title;
-
-	public $users         = array();
-	public $post_count    = 0;
-	public $media_count   = 0;
-	public $comment_count = 0;
-	public $term_count    = 0;
-
-	public $generator = '';
-	public $version;
-}
diff --git a/packages/playground/data-liberation/tests/WPWXRReaderTests.php b/packages/playground/data-liberation/tests/WPWXRReaderTests.php
index d9b131ce3f..9b73c2aad9 100644
--- a/packages/playground/data-liberation/tests/WPWXRReaderTests.php
+++ b/packages/playground/data-liberation/tests/WPWXRReaderTests.php
@@ -2,13 +2,15 @@
 
 use PHPUnit\Framework\TestCase;
 
+use WordPress\ByteReader\WP_File_Reader;
+
 class WPWXRReaderTests extends TestCase {
     
     /**
      * @dataProvider preexisting_wxr_files_provider
      */
     public function test_does_not_crash_when_parsing_preexisting_wxr_files_as_string($path, $expected_entitys) {
-        $wxr = WP_WXR_Reader::create();
+        $wxr = WP_WXR_Entity_Reader::create();
         $wxr->append_bytes(file_get_contents($path));
         $wxr->input_finished();
 
@@ -25,7 +27,7 @@ public function test_does_not_crash_when_parsing_preexisting_wxr_files_as_string
      */
     public function test_does_not_crash_when_parsing_preexisting_wxr_files_as_stream($path, $expected_entitys) {
         $stream = fopen($path, 'r');
-        $wxr = WP_WXR_Reader::create();
+        $wxr = WP_WXR_Entity_Reader::create();
         $found_entities = 0;
         while(true) {
             $chunk = fread($stream, 100);
@@ -64,7 +66,7 @@ public function preexisting_wxr_files_provider() {
 
 
     public function test_simple_wxr() {
-        $importer = WP_WXR_Reader::create();
+        $importer = WP_WXR_Entity_Reader::create();
         $importer->append_bytes(file_get_contents(__DIR__ . '/fixtures/wxr-simple.xml'));
         $importer->input_finished();
         $this->assertTrue( $importer->next_entity() );
@@ -182,7 +184,7 @@ public function test_simple_wxr() {
     }
 
     public function test_attachments() {
-        $importer = WP_WXR_Reader::create();
+        $importer = WP_WXR_Entity_Reader::create();
         $importer->append_bytes(<<<XML
             <?xml version="1.0" encoding="UTF-8"?>
             <rss>
@@ -265,7 +267,7 @@ public function test_attachments() {
     }
 
     public function test_terms() {
-        $importer = WP_WXR_Reader::create();
+        $importer = WP_WXR_Entity_Reader::create();
         $importer->append_bytes(<<<XML
             <?xml version="1.0" encoding="UTF-8"?>
             <rss>
@@ -300,7 +302,7 @@ public function test_terms() {
     }
 
     public function test_category() {
-        $importer = WP_WXR_Reader::create();
+        $importer = WP_WXR_Entity_Reader::create();
         $importer->append_bytes(<<<XML
             <?xml version="1.0" encoding="UTF-8"?>
             <rss>
@@ -331,7 +333,7 @@ public function test_category() {
     }
 
     public function test_tag_string() {
-        $importer = WP_WXR_Reader::create();
+        $importer = WP_WXR_Entity_Reader::create();
         $importer->append_bytes(<<<XML
             <?xml version="1.0" encoding="UTF-8"?>
             <rss>
@@ -379,7 +381,7 @@ public function test_tag_streaming() {
         XML;
         $chunks = str_split($wxr, 10);
 
-        $wxr = WP_WXR_Reader::create();
+        $wxr = WP_WXR_Entity_Reader::create();
         while(true) {
             if(true === $wxr->next_entity()) {
                 break;
@@ -411,7 +413,7 @@ public function test_tag_streaming() {
     }
 
     public function test_parse_comment() {
-        $wxr = WP_WXR_Reader::create();
+        $wxr = WP_WXR_Entity_Reader::create();
         $wxr->append_bytes(<<<XML
             <?xml version="1.0" encoding="UTF-8"?>
             <rss>
@@ -494,7 +496,7 @@ public function test_parse_comment() {
     }
 
     public function test_retains_last_ids() {
-        $wxr = WP_WXR_Reader::create();
+        $wxr = WP_WXR_Entity_Reader::create();
         $wxr->append_bytes(<<<XML
             <?xml version="1.0" encoding="UTF-8"?>
             <rss>
@@ -565,8 +567,8 @@ public function test_scan_entities_without_reentrancy() {
             "post_meta"
         ];
 
-		$wxr = WP_WXR_Reader::create(
-            new WP_File_Reader( $xml_path )
+		$wxr = WP_WXR_Entity_Reader::create(
+            WP_File_Reader::create( $xml_path )
         );
 
         for($i = 0; $i < 11; $i++) {
@@ -595,8 +597,8 @@ public function test_scan_entities_with_reentrancy() {
             "post_meta"
         ];
 
-		$wxr = WP_WXR_Reader::create(
-            new WP_File_Reader( $xml_path )
+		$wxr = WP_WXR_Entity_Reader::create(
+            WP_File_Reader::create( $xml_path )
         );
 
         for($i = 0; $i < 11; $i++) {
@@ -606,8 +608,8 @@ public function test_scan_entities_with_reentrancy() {
                 $wxr->get_entity()->get_type()
             );
             $cursor = $wxr->get_reentrancy_cursor();
-            $wxr = WP_WXR_Reader::create(
-                new WP_File_Reader( $xml_path ),
+            $wxr = WP_WXR_Entity_Reader::create(
+                WP_File_Reader::create( $xml_path ),
                 $cursor
             );
             $this->assertTrue( $wxr->next_entity() );