Bump openhtmltopdf 1.1.37, pdfbox 3.0.7, jsoup 1.22.1

Allow self-closing unknown HTML tags for compatibility with real-world HTML

Author: zspitzer

CHANGELOG.md

@@ -14,9 +14,9 @@ Major rewrite switching from Flying Saucer/iText to OpenHTMLToPDF + PDFBox 3.x.
 
 | Component | Version | Purpose |
 |-----------|---------|---------|
-| OpenHTMLToPDF | 1.1.24 | HTML/CSS to PDF rendering |
-| PDFBox | 3.0.5 | PDF manipulation (cfpdf actions) |
-| jsoup | 1.18.3 | HTML parsing/cleanup |
+| OpenHTMLToPDF | 1.1.37 | HTML/CSS to PDF rendering |
+| PDFBox | 3.0.7 | PDF manipulation (cfpdf actions) |
+| jsoup | 1.22.1 | HTML parsing/cleanup |
 
 ### New cfpdf Actions
 
@@ -183,6 +183,8 @@ The `scale` attribute (1-100) now works, rendering content at the specified perc
 - **No License Issues**: All open source libraries (Apache 2.0, LGPL)
 - **Better CSS Support**: CSS 2.1 with some CSS3 support via OpenHTMLToPDF
 - **Modern PDFBox**: PDFBox 3.x with improved performance and security
+- **Self-closing HTML tags**: Unknown/custom tags are allowed to self-close for compatibility with real-world HTML
+- **Removed legacy v2 dist jars**: Cleaned out old iText, Flying Saucer, TagSoup, and pre-release PDFBox jars
 
 ### Removed Features
 

source/java/pom.xml

@@ -32,8 +32,8 @@
 
 	<properties>
 		<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
-		<openhtmltopdf.version>1.1.24</openhtmltopdf.version>
-		<pdfbox.version>3.0.5</pdfbox.version>
+		<openhtmltopdf.version>1.1.37</openhtmltopdf.version>
+		<pdfbox.version>3.0.7</pdfbox.version>
 	</properties>
 
 	<build>
@@ -226,7 +226,7 @@
 		<dependency>
 			<groupId>org.jsoup</groupId>
 			<artifactId>jsoup</artifactId>
-			<version>1.18.3</version>
+			<version>1.22.1</version>
 		</dependency>
 
 		<!-- Lucee (provided - not bundled) -->

source/java/src/org/lucee/extension/pdf/PDFDocument.java

@@ -39,6 +39,8 @@
 
 import org.jsoup.Jsoup;
 import org.jsoup.helper.W3CDom;
+import org.jsoup.parser.Parser;
+import org.jsoup.parser.Tag;
 import org.w3c.dom.Document;
 
 import com.openhtmltopdf.outputdevice.helper.BaseRendererBuilder.PageSizeUnits;
@@ -430,8 +432,12 @@ public byte[] render(Dimension dimension, double unitFactor, PageContext pc, boo
 			String baseUrl = getBaseUrl(pc);
 
 			// Parse HTML with JSoup and convert to W3C DOM
-			org.jsoup.nodes.Document jsoupDoc = Jsoup.parse(html);
-			jsoupDoc.outputSettings().syntax(org.jsoup.nodes.Document.OutputSettings.Syntax.xml);
+			Parser parser = Parser.htmlParser();
+			parser.tagSet().onNewTag( tag -> {
+				if ( !tag.isKnownTag() ) tag.set( Tag.SelfClose );
+			});
+			org.jsoup.nodes.Document jsoupDoc = Jsoup.parse( html, parser );
+			jsoupDoc.outputSettings().syntax( org.jsoup.nodes.Document.OutputSettings.Syntax.xml );
 
 			// Convert local file paths to file:// URIs for OpenHTMLToPDF compatibility
 			convertLocalPathsToURIs(jsoupDoc);

source/java/src/org/lucee/extension/pdf/util/XMLUtil.java

@@ -14,6 +14,8 @@
 
 import org.jsoup.Jsoup;
 import org.jsoup.helper.W3CDom;
+import org.jsoup.parser.Parser;
+import org.jsoup.parser.Tag;
 import org.w3c.dom.Document;
 import org.w3c.dom.Element;
 import org.w3c.dom.Node;
@@ -100,8 +102,12 @@ else if (xml.getSystemId() != null) {
 			}
 
 			// Parse with jsoup and convert to W3C DOM
-			org.jsoup.nodes.Document jsoupDoc = Jsoup.parse(html);
-			jsoupDoc.outputSettings().syntax(org.jsoup.nodes.Document.OutputSettings.Syntax.xml);
+			Parser parser = Parser.htmlParser();
+			parser.tagSet().onNewTag( tag -> {
+				if ( !tag.isKnownTag() ) tag.set( Tag.SelfClose );
+			});
+			org.jsoup.nodes.Document jsoupDoc = Jsoup.parse( html, parser );
+			jsoupDoc.outputSettings().syntax( org.jsoup.nodes.Document.OutputSettings.Syntax.xml );
 			W3CDom w3cDom = new W3CDom();
 			return w3cDom.fromJsoup(jsoupDoc);
 		}