Parse EPubs as XHTML

Author: adamziel

packages/playground/data-liberation/src/block-markup/WP_HTML_To_Blocks.php

@@ -45,7 +45,6 @@ public function convert() {
 		}
 
 		while ( $this->markup_processor->next_token() ) {
-			var_dump( $this->markup_processor->get_token_type() );
 			switch ( $this->markup_processor->get_token_type() ) {
 				case '#text':
 					if ( $this->ignore_text ) {
@@ -58,7 +57,6 @@ public function convert() {
 					break;
 			}
 		}
-		var_dump( $this->markup_processor->get_last_error() );
 
 		if ( $this->markup_processor->get_last_error() ) {
 			$this->last_error = $this->markup_processor->get_last_error();
@@ -90,8 +88,8 @@ private function handle_tag() {
 		$tag           = strtoupper( $html->get_tag() );
 		$tag_lowercase = strtolower( $tag );
 
-		$is_tag_opener = ! $html->is_tag_closer();
-		if ( ! $html->expects_closer() ) {
+		$is_void_tag = ! $html->expects_closer() && ! $html->is_tag_closer();
+		if ( $is_void_tag ) {
 			switch ( $tag ) {
 				case 'META':
 					$key   = $html->get_attribute( 'name' );
@@ -119,7 +117,7 @@ private function handle_tag() {
 					//        Just insert an HTML block or what?
 					break;
 			}
-		} elseif ( $is_tag_opener ) {
+		} elseif ( ! $html->is_tag_closer() ) {
 			switch ( $tag ) {
 				// Block elements
 				case 'SCRIPT':

packages/playground/data-liberation/src/entity-readers/WP_EPub_Entity_Reader.php

@@ -93,8 +93,8 @@ public function next_entity() {
 				return false;
 			}
 
-			$html_file = array_shift( $this->remaining_html_files );
-			$html      = $this->zip->read_file( $html_file );
+			$html_file                 = array_shift( $this->remaining_html_files );
+			$html                      = $this->zip->read_file( $html_file );
 			$this->current_html_reader = new WP_HTML_Entity_Reader(
 				WP_XML_Processor::create_from_string( $html ),
 				$this->current_post_id

packages/playground/data-liberation/src/xml-api/WP_XML_Processor.php

@@ -1628,13 +1628,13 @@ private function parse_next_tag() {
 						return false;
 					}
 
-					$closer_at = $at;
+					$closer_at                  = $at;
 					$this->parser_state         = self::STATE_DOCTYPE_NODE;
 					$this->token_length         = $closer_at + 1 - $this->token_starts_at;
 					$this->bytes_already_parsed = $closer_at + 1;
 					return true;
 				}
-				
+
 				/*
 				 * Anything else here is either unsupported at this point or invalid
 				 * syntax. See the class-level @TODO annotations for more information.
@@ -1644,7 +1644,6 @@ private function parse_next_tag() {
 				return false;
 			}
 
-
 			/*
 			 * An `<?xml` token at the beginning of the document marks a start of an
 			 * xml declaration.
@@ -2537,7 +2536,7 @@ public function expects_closer() {
 			return false;
 		}
 
-		return ! $this->is_empty_element() && ! $this->is_closing_tag;
+		return $this->is_tag_opener() && ! $this->is_empty_element();
 	}
 
 	/**

packages/playground/data-liberation/tests/WPEPubEntityReaderTests.php

@@ -13,19 +13,17 @@ public function test_entity_reader( $reader ) {
         $entities = [];
         while ( $reader->next_entity() ) {
             $data = $reader->get_entity()->get_data();
-            if(isset($data['content'])) {
-                $data['content'] = $this->normalize_markup( $data['content'] );
-            }
             $entities[] = [
                 'type' => $reader->get_entity()->get_type(),
                 'data' => $data,
             ];
         }
         $this->assertNull( $reader->get_last_error() );
         $this->assertEquals( 3, count($entities) );
-        $this->assertEquals( 117, strlen($entities[0]['data']['content']) );
+        $this->assertGreaterThan( 100, strlen($entities[0]['data']['content']) );
         $this->assertGreaterThan( 1000, strlen($entities[1]['data']['content']) );
         $this->assertGreaterThan( 1000, strlen($entities[2]['data']['content']) );
+        echo $entities[2]['data']['content'];
     }
 
     public function epub_byte_reader_data_provider() {
@@ -39,20 +37,4 @@ public function epub_byte_reader_data_provider() {
         ];
     }
 
-    private function normalize_markup( $markup ) {
-        $processor = new WP_HTML_Processor( $markup );
-        $serialized = $processor->serialize();
-        // Naively remove parts of the HTML that serialize()
-        // adds that we don't want.
-        $serialized = str_replace(
-            [
-                '<html><head></head><body>',
-                '</body></html>',
-            ],
-            '',
-            $serialized
-        );
-        return $serialized;
-    }
-
 }

packages/playground/data-liberation/tests/WPHTMLToBlocksTests.php

@@ -146,7 +146,29 @@ public function test_html_to_blocks_excerpt() {
         }
 
         $this->assertEquals( file_get_contents( $output_file ), $blocks );
-        
+    }
+
+    public function test_xhtml_to_blocks_conversion() {
+        $input = <<<XML
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE html>
+<html>
+    <body>
+        <h1>Hello, world!</h1>
+        <p>And some content</p>
+    </body>
+</html>
+XML;
+        $converter = new WP_HTML_To_Blocks( WP_XML_Processor::create_from_string( $input ) );
+        $converter->convert( $input );
+        $blocks = $converter->get_block_markup();
+        $expected = <<<HTML
+<!-- wp:heading {"level":1} --><h1>Hello, world! </h1><!-- /wp:heading --><!-- wp:paragraph --><p>And some content </p><!-- /wp:paragraph -->
+HTML;
+        $this->assertEquals(
+            $this->normalize_markup( $expected ),
+            $this->normalize_markup( $blocks )
+        );
     }
 
 }