Add automatic link alternative description parsing

Dmitry Radchuk · iText-CI · commit 7a9ad3dd4951 · 2024-01-09T09:50:03.000Z
DEVSIX-7996

Autoported commit.
Original commit hash: [7d77290d6]
diff --git a/itext.tests/itext.html2pdf.tests/itext/html2pdf/HtmlConverterPdfUA2Test.cs b/itext.tests/itext.html2pdf.tests/itext/html2pdf/HtmlConverterPdfUA2Test.cs
@@ -58,10 +58,9 @@ public virtual void SimpleLinkTest() {
             converterProperties.SetFontProvider(fontProvider);
             HtmlConverter.ConvertToPdf(new FileStream(sourceHtml, FileMode.Open, FileAccess.Read), pdfDocument, converterProperties
                 );
-            /* TODO: DEVSIX-7996 - Links created from html2pdf are not ua-2 compliant
-            * Two verapdf errors are generated here:
-            * 1. clause="8.9.4.1", Link annotation neither has a Contents entry nor alternate description.
-            * 2. clause="8.5.1", Real content that does not possess the semantics of text objects and does not have
+            /* TODO: DEVSIX-5700 - Links created from html2pdf are not ua-2 compliant
+            * One verapdf error is generated here:
+            * 1. clause="8.5.1", Real content that does not possess the semantics of text objects and does not have
             *    an alternate textual representation is not enclosed within Figure or Formula structure elements.
             */
             CompareAndCheckCompliance(destinationPdf, cmpPdf, false);
@@ -80,10 +79,31 @@ public virtual void BackwardLinkTest() {
             converterProperties.SetFontProvider(fontProvider);
             HtmlConverter.ConvertToPdf(new FileStream(sourceHtml, FileMode.Open, FileAccess.Read), pdfDocument, converterProperties
                 );
-            /* TODO: DEVSIX-7996 - Links created from html2pdf are not ua-2 compliant
-            * Two verapdf errors are generated here:
-            * 1. clause="8.9.4.1", Link annotation neither has a Contents entry nor alternate description.
-            * 2. clause="8.5.1", Real content that does not possess the semantics of text objects and does not have
+            /* TODO: DEVSIX-5700 - Links created from html2pdf are not ua-2 compliant
+            * One verapdf error is generated here:
+            * 1. clause="8.5.1", Real content that does not possess the semantics of text objects and does not have
+            *    an alternate textual representation is not enclosed within Figure or Formula structure elements.
+            */
+            CompareAndCheckCompliance(destinationPdf, cmpPdf, false);
+        }
+
+        [NUnit.Framework.Test]
+        public virtual void ImageLinkTest() {
+            String sourceHtml = SOURCE_FOLDER + "imageLink.html";
+            String cmpPdf = SOURCE_FOLDER + "cmp_imageLink.pdf";
+            String destinationPdf = DESTINATION_FOLDER + "imageLink.pdf";
+            PdfDocument pdfDocument = new PdfDocument(new PdfWriter(destinationPdf, new WriterProperties().SetPdfVersion
+                (PdfVersion.PDF_2_0)));
+            CreateSimplePdfUA2Document(pdfDocument);
+            ConverterProperties converterProperties = new ConverterProperties();
+            FontProvider fontProvider = new DefaultFontProvider(false, true, false);
+            converterProperties.SetFontProvider(fontProvider);
+            converterProperties.SetBaseUri(SOURCE_FOLDER);
+            HtmlConverter.ConvertToPdf(new FileStream(sourceHtml, FileMode.Open, FileAccess.Read), pdfDocument, converterProperties
+                );
+            /* TODO: DEVSIX-5700 - Links created from html2pdf are not ua-2 compliant
+            * One verapdf error is generated here:
+            * 1. clause="8.5.1", Real content that does not possess the semantics of text objects and does not have
             *    an alternate textual representation is not enclosed within Figure or Formula structure elements.
             */
             CompareAndCheckCompliance(destinationPdf, cmpPdf, false);
diff --git a/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/backwardLink.html b/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/backwardLink.html
@@ -7,7 +7,7 @@
 
 <div id="15=15">Target div</div>
 <div style="height:1000px">Some content</div>
-<a href="#15=15">Long link text to bottom div. Long link text to bottom div. Long link text to bottom div.
+<a href="#15=15" title="alternative description">Long link text to bottom div. Long link text to bottom div. Long link text to bottom div.
     Long link text to bottom div. Long link text to bottom div. Long link text to bottom div.</a>
 
 </body>
diff --git a/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/blue96x96.png b/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/blue96x96.png
diff --git a/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/cmp_backwardLink.pdf b/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/cmp_backwardLink.pdf
diff --git a/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/cmp_imageLink.pdf b/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/cmp_imageLink.pdf
diff --git a/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/cmp_simpleLink.pdf b/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/cmp_simpleLink.pdf
diff --git a/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/imageLink.html b/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/imageLink.html
@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<html>
+<head>
+
+</head>
+<body>
+
+<a href="#15=15" title="alternative description"><img alt="blue96x96" src="blue96x96.png"></a>
+<div style="height:1000px">Some content</div>
+<div id="15=15">Target div</div>
+
+</body>
+</html>
diff --git a/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/simpleLink.html b/itext.tests/itext.html2pdf.tests/resources/itext/html2pdf/HtmlConverterPdfUA2Test/simpleLink.html
@@ -5,7 +5,7 @@
 </head>
 <body>
 
-<a href="#15=15">Long link text to bottom div. Long link text to bottom div. Long link text to bottom div.
+<a href="#15=15" title="alternative description">Long link text to bottom div. Long link text to bottom div. Long link text to bottom div.
     Long link text to bottom div. Long link text to bottom div. Long link text to bottom div.</a>
 <div style="height:1000px">Some content</div>
 <div id="15=15">Target div</div>
diff --git a/itext/itext.html2pdf/itext/html2pdf/attach/impl/tags/ABlockTagWorker.cs b/itext/itext.html2pdf/itext/html2pdf/attach/impl/tags/ABlockTagWorker.cs
@@ -69,7 +69,7 @@ public override void ProcessEnd(IElementNode element, ProcessorContext context)
                     }
                 }
                 ((Div)GetElementResult()).GetAccessibilityProperties().SetRole(StandardRoles.LINK);
-                LinkHelper.ApplyLinkAnnotation(GetElementResult(), url, context);
+                LinkHelper.ApplyLinkAnnotation(GetElementResult(), url, context, element);
             }
             if (GetElementResult() != null) {
                 String name = element.GetAttribute(AttributeConstants.NAME);
diff --git a/itext/itext.html2pdf/itext/html2pdf/attach/impl/tags/ATagWorker.cs b/itext/itext.html2pdf/itext/html2pdf/attach/impl/tags/ATagWorker.cs
@@ -98,7 +98,7 @@ public override void ProcessEnd(IElementNode element, ProcessorContext context)
                         }
                         GetAllElements()[i] = simulatedDiv;
                     }
-                    LinkHelper.ApplyLinkAnnotation(GetAllElements()[i], url, context);
+                    LinkHelper.ApplyLinkAnnotation(GetAllElements()[i], url, context, element);
                 }
             }
             if (!GetAllElements().IsEmpty()) {
diff --git a/itext/itext.html2pdf/itext/html2pdf/attach/util/LinkHelper.cs b/itext/itext.html2pdf/itext/html2pdf/attach/util/LinkHelper.cs
@@ -41,6 +41,7 @@ You should have received a copy of the GNU Affero General Public License
 using iText.Layout.Properties;
 using iText.Layout.Tagging;
 using iText.StyledXmlParser.Node;
+using iText.StyledXmlParser.Node.Impl.Jsoup.Node;
 
 namespace iText.Html2pdf.Attach.Util {
     /// <summary>Helper class for links.</summary>
@@ -70,7 +71,30 @@ public static void ApplyLinkAnnotation(IPropertyContainer container, String url)
         /// <param name="container">the containing object.</param>
         /// <param name="url">the destination.</param>
         /// <param name="context">the processor context.</param>
+        [Obsolete]
         public static void ApplyLinkAnnotation(IPropertyContainer container, String url, ProcessorContext context) {
+            ApplyLinkAnnotation(container, url, context, "");
+        }
+
+        private static String RetrieveAlternativeDescription(IElementNode element) {
+            IList<INode> children = element.ChildNodes();
+            //if there is an img tag under the link then prefer the alt attribute as a link description
+            if (children.Count == 1 && children[0].ChildNodes().IsEmpty() && children[0] is JsoupElementNode && ((JsoupElementNode
+                )children[0]).GetAttribute(AttributeConstants.ALT) != null) {
+                return ((JsoupElementNode)children[0]).GetAttribute(AttributeConstants.ALT);
+            }
+            //return title attribute value in case of regular link
+            return element.GetAttribute(AttributeConstants.TITLE);
+        }
+
+        /// <summary>Applies a link annotation.</summary>
+        /// <param name="container">the containing object.</param>
+        /// <param name="url">the destination.</param>
+        /// <param name="context">the processor context.</param>
+        /// <param name="alternateDescription">description for a link.</param>
+        [Obsolete]
+        public static void ApplyLinkAnnotation(IPropertyContainer container, String url, ProcessorContext context, 
+            String alternateDescription) {
             if (container != null) {
                 PdfLinkAnnotation linkAnnotation;
                 if (url.StartsWith("#")) {
@@ -86,6 +110,9 @@ public static void ApplyLinkAnnotation(IPropertyContainer container, String url,
                     linkAnnotation = (PdfLinkAnnotation)new PdfLinkAnnotation(new Rectangle(0, 0, 0, 0)).SetAction(PdfAction.CreateURI
                         (url)).SetFlags(PdfAnnotation.PRINT);
                 }
+                if (container is IAccessibleElement && alternateDescription != null) {
+                    ((IAccessibleElement)container).GetAccessibilityProperties().SetAlternateDescription(alternateDescription);
+                }
                 linkAnnotation.SetBorder(new PdfArray(new float[] { 0, 0, 0 }));
                 container.SetProperty(Property.LINK_ANNOTATION, linkAnnotation);
                 if (container is ILeafElement && container is IAccessibleElement) {
@@ -94,6 +121,17 @@ public static void ApplyLinkAnnotation(IPropertyContainer container, String url,
             }
         }
 
+        /// <summary>Applies a link annotation.</summary>
+        /// <param name="container">the containing object.</param>
+        /// <param name="url">the destination.</param>
+        /// <param name="context">the processor context.</param>
+        /// <param name="element">the element node.</param>
+        public static void ApplyLinkAnnotation(IPropertyContainer container, String url, ProcessorContext context, 
+            IElementNode element) {
+            iText.Html2pdf.Attach.Util.LinkHelper.ApplyLinkAnnotation(container, url, context, RetrieveAlternativeDescription
+                (element));
+        }
+
         /// <summary>Creates a destination</summary>
         /// <param name="tagWorker">the tagworker that is building the (iText) element</param>
         /// <param name="element">the (HTML) element being converted</param>
diff --git a/port-hash b/port-hash
@@ -1 +1 @@
-ac7912d76dbb46abc4fdb50394ba3b8765978227
+7d77290d62153623ac14ecae7b820119eb0e9a88

Original file line number	Diff line number	Diff line change
`@@ -69,7 +69,7 @@ public override void ProcessEnd(IElementNode element, ProcessorContext context)`
`69`	`69`	`}`
`70`	`70`	`}`
`71`	`71`	`((Div)GetElementResult()).GetAccessibilityProperties().SetRole(StandardRoles.LINK);`
`72`		`- LinkHelper.ApplyLinkAnnotation(GetElementResult(), url, context);`
	`72`	`+ LinkHelper.ApplyLinkAnnotation(GetElementResult(), url, context, element);`
`73`	`73`	`}`
`74`	`74`	`if (GetElementResult() != null) {`
`75`	`75`	`String name = element.GetAttribute(AttributeConstants.NAME);`
Original file line number	Diff line number	Diff line change
`@@ -98,7 +98,7 @@ public override void ProcessEnd(IElementNode element, ProcessorContext context)`
`98`	`98`	`}`
`99`	`99`	`GetAllElements()[i] = simulatedDiv;`
`100`	`100`	`}`
`101`		`- LinkHelper.ApplyLinkAnnotation(GetAllElements()[i], url, context);`
	`101`	`+ LinkHelper.ApplyLinkAnnotation(GetAllElements()[i], url, context, element);`
`102`	`102`	`}`
`103`	`103`	`}`
`104`	`104`	`if (!GetAllElements().IsEmpty()) {`