Rewrite excerpt macro to support excerpt extension

src/Converter/ConfluenceConverter.php

@@ -16,6 +16,7 @@
 use HalloWelt\MigrateConfluence\Converter\Postprocessor\FixMultilineTable;
 use HalloWelt\MigrateConfluence\Converter\Postprocessor\FixMultilineTemplate;
 use HalloWelt\MigrateConfluence\Converter\Postprocessor\NestedHeadings;
+use HalloWelt\MigrateConfluence\Converter\Postprocessor\RestoreExcerptBlock;
 use HalloWelt\MigrateConfluence\Converter\Postprocessor\RestorePStyleTag;
 use HalloWelt\MigrateConfluence\Converter\Postprocessor\RestoreTimeTag;
 use HalloWelt\MigrateConfluence\Converter\Postprocessor\TasksReportMacro as RestoreTasksReportMacro;
@@ -398,6 +399,7 @@ private function runProcessors( DOMDocument $dom ): void {
 	private function runPostProcessors(): void {
 		$postProcessors = [
 			new RestorePStyleTag(),
+			new RestoreExcerptBlock(),
 			new RestoreTimeTag(),
 			new FixLineBreakInHeadings(),
 			new FixImagesWithExternalUrl(),

src/Converter/Postprocessor/RestoreExcerptBlock.php

@@ -0,0 +1,24 @@
+<?php
+
+namespace HalloWelt\MigrateConfluence\Converter\Postprocessor;
+
+use HalloWelt\MigrateConfluence\Converter\IPostprocessor;
+
+class RestoreExcerptBlock implements IPostprocessor {
+
+	/**
+	 * @inheritDoc
+	 */
+	public function postprocess( string $wikiText ): string {
+		return preg_replace_callback(
+			'/#####EXCERPTBLOCKOPEN\|(.*?)\|(.*?)#####(.*?)#####EXCERPTBLOCKCLOSE#####/si',
+			static function ( $matches ) {
+				$name = $matches[1];
+				$hidden = $matches[2];
+				$content = $matches[3];
+				return "<excerpt-block name=\"$name\" hidden=\"$hidden\">$content</excerpt-block>";
+			},
+			$wikiText
+		);
+	}
+}

src/Converter/Processor/ExcerptMacro.php

@@ -2,11 +2,15 @@
 
 namespace HalloWelt\MigrateConfluence\Converter\Processor;
 
+use DOMNode;
+
 /**
+ * Converts the Confluence excerpt macro to a BlueSpice <excerpt-block> element.
+ * The broken macro category is added because the BlueSpice Excerpt extension is not yet available.
+ *
  * @see https://confluence.atlassian.com/doc/excerpt-macro-148062.html
- * @see https://docs.atlassian.com/DAC/javadoc/confluence/4.0/reference/com/atlassian/confluence/macro/Macro.OutputType.html
  */
-class ExcerptMacro extends ConvertMacroToTemplateWithBodyBase {
+class ExcerptMacro extends StructuredMacroProcessorBase {
 
 	/**
 	 * @inheritDoc
@@ -17,15 +21,47 @@ protected function getMacroName(): string {
 
 	/**
 	 * @inheritDoc
+	 *
+	 * Pandoc strips unknown HTML elements like <excerpt-block> when converting to MediaWiki
+	 * format. To preserve the tag, we insert text placeholders around the content here and
+	 * restore the actual <excerpt-block> tag in the RestoreExcerptBlock postprocessor.
+	 * Placeholders use pipe-separated values to avoid HTML attribute quote encoding issues.
 	 */
-	protected function getWikiTextTemplateStartName(): string {
-		return 'ExcerptStart';
-	}
+	protected function doProcessMacro( DOMNode $node ): void {
+		$macroId = $node->getAttribute( 'ac:macro-id' );
+		$hidden = 'false';
 
-	/**
-	 * @inheritDoc
-	 */
-	protected function getWikiTextTemplateEndName(): string {
-		return 'ExcerptEnd';
+		foreach ( $node->childNodes as $childNode ) {
+			if ( $childNode->nodeName === 'ac:parameter'
+				&& $childNode->getAttribute( 'ac:name' ) === 'hidden' ) {
+				$hidden = trim( $childNode->nodeValue );
+				break;
+			}
+		}
+
+		$parent = $node->parentNode;
+
+		$openTag = $node->ownerDocument->createTextNode(
+			"#####EXCERPTBLOCKOPEN|$macroId|$hidden#####"
+		);
+		$parent->insertBefore( $openTag, $node );
+
+		foreach ( $node->childNodes as $childNode ) {
+			if ( $childNode->nodeName === 'ac:rich-text-body' ) {
+				foreach ( iterator_to_array( $childNode->childNodes ) as $bodyChild ) {
+					$parent->insertBefore( $bodyChild->cloneNode( true ), $node );
+				}
+			}
+		}
+
+		$closeTag = $node->ownerDocument->createTextNode( '#####EXCERPTBLOCKCLOSE#####' );
+		$parent->insertBefore( $closeTag, $node );
+
+		$brokenCategory = $node->ownerDocument->createTextNode(
+			$this->getBrokenMacroCategory()
+		);
+		$parent->insertBefore( $brokenCategory, $node );
+
+		$parent->removeChild( $node );
 	}
 }

tests/phpunit/data/excerpt-macro-input.xml

@@ -1,6 +1,6 @@
 <?xml version="1.0"?>
 <xml xmlns:ac="sample_namespace" xmlns:ri="sample_second_namespace">
-	<ac:structured-macro ac:name="excerpt">
+	<ac:structured-macro ac:name="excerpt" ac:macro-id="b50b3cdb-e32a-447c-8c42-7be043c457ce">
 		<ac:parameter ac:name="hidden">true</ac:parameter>
 		<ac:parameter ac:name="atlassian-macro-output-type">BLOCK</ac:parameter>
 		<ac:rich-text-body>

tests/phpunit/data/excerpt-macro-output.xml

@@ -1,9 +1,6 @@
 <?xml version="1.0"?>
 <xml xmlns:ac="sample_namespace" xmlns:ri="sample_second_namespace">
-	{{ExcerptStart###BREAK###
-|hidden = true###BREAK###
-|atlassian-macro-output-type = BLOCK###BREAK###
-}}
+	#####EXCERPTBLOCKOPEN|b50b3cdb-e32a-447c-8c42-7be043c457ce|true#####
 			<p>This is the <strong>text</strong> I want to reuse in other pages. This text is inside an Excerpt macro.</p>
-  		{{ExcerptEnd}}
+  		#####EXCERPTBLOCKCLOSE#####[[Category:Broken_macro/excerpt]]
 </xml>