Release 3.2.0

CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## 3.2.0
+
+- Add new public API to allow parsing into Header and Footer #162. Some API methods as been flagged as obsolete with a clear message of what to use instead.
+  This is not a breaking changes as it keep existing behaviour.
+- Add support for `SVG` format (either from img src or the SVG node tag)
+- Automatically create the `_top` bookmark if needed
+- Fix a crash when a hyperlink contains both `img` and `figcation`
+- Fix a crash when `li` is empty #161
+
 ## 3.1.1
 
 - Fix respecting layout with `div`/`p` ending with line break #158

HtmlToOpenXml.sln

@@ -1,17 +1,17 @@
 
 Microsoft Visual Studio Solution File, Format Version 12.00
-# Visual Studio 15
-VisualStudioVersion = 15.0.26730.16
+# Visual Studio 17
+VisualStudioVersion = 17.8.34511.84
 MinimumVisualStudioVersion = 10.0.40219.1
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "HtmlToOpenXml", "src\Html2OpenXml\HtmlToOpenXml.csproj", "{EF700F30-C9BB-49A6-912C-E3B77857B514}"
 EndProject
 Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "src", "src", "{58520A98-BA53-4BA4-AAE3-786AA21331D6}"
 EndProject
 Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "test", "test", "{84EA02ED-2E97-47D2-992E-32CC104A3A7A}"
 EndProject
-Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "Demo", "examples\Demo\Demo.csproj", "{A1ECC760-B9F7-4A00-AF5F-568B5FD6F09F}"
+Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Demo", "examples\Demo\Demo.csproj", "{A1ECC760-B9F7-4A00-AF5F-568B5FD6F09F}"
 EndProject
-Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "HtmlToOpenXml.Tests", "test\HtmlToOpenXml.Tests\HtmlToOpenXml.Tests.csproj", "{CA0A68E0-45A0-4A01-A061-F951D93D6906}"
+Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "HtmlToOpenXml.Tests", "test\HtmlToOpenXml.Tests\HtmlToOpenXml.Tests.csproj", "{CA0A68E0-45A0-4A01-A061-F951D93D6906}"
 EndProject
 Global
     GlobalSection(SolutionConfigurationPlatforms) = preSolution

README.md

@@ -2,9 +2,9 @@
 ![Download Counts](https://img.shields.io/nuget/dt/HtmlToOpenXml.dll.svg)
 [![MIT License](https://img.shields.io/badge/license-MIT-blue.svg)](https://github.com/onizet/html2openxml/blob/dev/LICENSE)
 
-# What is Html2OpenXml?
+# What is HtmlToOpenXml?
 
-Html2OpenXml is a small .Net library that convert simple or advanced HTML to plain OpenXml components. This program has started in 2009, initially to convert user's comments into Word.
+HtmlToOpenXml is a small .Net library that convert simple or advanced HTML to plain OpenXml components. This program has started in 2009, initially to convert user's comments into Word.
 
 This library supports both **.Net Framework 4.6.2**, **.NET Standard 2.0** and **.NET 8** which are all LTS.
 
@@ -28,7 +28,7 @@ Refer to [w3schools’ tag](http://www.w3schools.com/tags/default.asp) list to s
 * `abbr` and `acronym`
 * `b`, `i`, `u`, `s`, `del`, `ins`, `em`, `strike`, `strong`
 * `br` and `hr`
-* `img`, `figcaption`
+* `img`, `figcaption` and `svg`
 * `table`, `td`, `tr`, `th`, `tbody`, `thead`, `tfoot`, `caption` and `col`
 * `cite`
 * `div`, `span`, `time`, `font` and `p`

examples/Demo/Demo.csproj

@@ -6,14 +6,19 @@
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="DocumentFormat.OpenXml" Version="3.0.2" />
+    <PackageReference Include="DocumentFormat.OpenXml" Version="3.1.0" />
     <PackageReference Include="System.Diagnostics.Process" Version="4.3.0" />
   </ItemGroup>
 
   <ItemGroup>
     <ProjectReference Include="..\..\src\Html2OpenXml\HtmlToOpenXml.csproj" />
   </ItemGroup>
 
+  <PropertyGroup Condition="'$(Configuration)' == 'Debug'">
+    <DefineConstants>$(DefineConstants);DEBUG</DefineConstants>
+    <Optimize>false</Optimize>
+  </PropertyGroup>
+
   <ItemGroup>
     <EmbeddedResource Include="Resources\*" />
   </ItemGroup>

examples/Demo/Program.cs

@@ -15,7 +15,7 @@ static class Program
         static async Task Main(string[] args)
         {
             const string filename = "test.docx";
-            string html = ResourceHelper.GetString("Resources.CompleteRunTest.html");
+            string html = ResourceHelper.GetString("Resources.AdvancedTable.html");
             if (File.Exists(filename)) File.Delete(filename);
 
             using (MemoryStream generatedDocument = new MemoryStream())
@@ -42,7 +42,7 @@ static async Task Main(string[] args)
                     converter.RenderPreAsTable = true;
                     Body body = mainPart.Document.Body;
 
-                    await converter.ParseHtml(html);
+                    await converter.ParseBody(html);
                     mainPart.Document.Save();
 
                     AssertThatOpenXmlDocumentIsValid(package);

src/Html2OpenXml/Expressions/AbbreviationExpression.cs

@@ -32,8 +32,9 @@ public override IEnumerable<OpenXmlElement> Interpret(ParsingContext context)
         var childElements = base.Interpret(context);
 
         // Transform the inline acronym/abbreviation to a reference to a foot note.
+        // Footnote or endnote are invalid inside header and footer
         string? description = node.Title;
-        if (string.IsNullOrEmpty(description))
+        if (string.IsNullOrEmpty(description) || context.HostingPart is not MainDocumentPart)
             return childElements;
 
         string runStyle;

src/Html2OpenXml/Expressions/BlockElementExpression.cs

@@ -325,7 +325,7 @@ private static Paragraph CreateParagraph(ParsingContext context, IList<OpenXmlEl
     /// <summary>
     /// Resolve the next available <see cref="BookmarkStart.Id"/> (they must be unique).
     /// </summary>
-    private static int IncrementBookmarkId(ParsingContext context)
+    protected static int IncrementBookmarkId(ParsingContext context)
     {
         var bookmarkRef = context.Properties<int?>("bookmarkRef");
 

src/Html2OpenXml/Expressions/BlockQuoteExpression.cs

@@ -13,6 +13,7 @@
 using System.Linq;
 using AngleSharp.Html.Dom;
 using DocumentFormat.OpenXml;
+using DocumentFormat.OpenXml.Packaging;
 using DocumentFormat.OpenXml.Wordprocessing;
 
 namespace HtmlToOpenXml.Expressions;
@@ -26,15 +27,19 @@ sealed class BlockQuoteExpression(IHtmlElement node) : BlockElementExpression(no
     /// <inheritdoc/>
     public override IEnumerable<OpenXmlElement> Interpret(ParsingContext context)
     {
-        string? description = node.GetAttribute("cite");
-
         var childElements = base.Interpret(context);
         if (!childElements.Any())
             return [];
+
+        // Footnote or endnote are invalid inside header and footer
+        if (context.HostingPart is not MainDocumentPart)
+            return childElements;
 
         // Transform the inline acronym/abbreviation to a reference to a foot note.
         if (childElements.First() is Paragraph paragraph)
         {
+            string? description = node.GetAttribute("cite");
+
             paragraph.ParagraphProperties ??= new();
             if (paragraph.ParagraphProperties.ParagraphStyleId is null)
                 paragraph.ParagraphProperties.ParagraphStyleId = 

src/Html2OpenXml/Expressions/BodyExpression.cs

@@ -10,6 +10,7 @@
  * PARTICULAR PURPOSE.
  */
 using System.Collections.Generic;
+using System.Globalization;
 using System.Linq;
 using AngleSharp.Dom;
 using AngleSharp.Html.Dom;
@@ -24,28 +25,51 @@ namespace HtmlToOpenXml.Expressions;
 /// </summary>
 sealed class BodyExpression(IHtmlElement node) : BlockElementExpression(node)
 {
+    private bool shouldRegisterTopBookmark;
+
     public override IEnumerable<OpenXmlElement> Interpret(ParsingContext context)
     {
         MarkAllBookmarks();
 
-        return base.Interpret(context);
+        var elements = base.Interpret(context);
+
+        if (shouldRegisterTopBookmark && elements.Any())
+        {
+            // Check whether it already exists
+            var body = context.MainPart.Document.Body!;
+            if (body.Descendants<BookmarkStart>().Where(b => b.Name?.Value == "_top").Any())
+            {
+                return elements;
+            }
+
+            var bookmarkId = IncrementBookmarkId(context).ToString(CultureInfo.InvariantCulture);
+            // this is expected to stand in the 1st paragraph
+            Paragraph? p = body.FirstChild as Paragraph;
+            p ??= body.PrependChild(new Paragraph());
+            p.InsertAfter(new BookmarkEnd() { Id = bookmarkId }, p.ParagraphProperties);
+            p.InsertAfter(new BookmarkStart() { Id = bookmarkId, Name = "_top" }, p.ParagraphProperties);
+        }
+
+        return elements;
     }
 
     protected override void ComposeStyles(ParsingContext context)
     {
         base.ComposeStyles(context);
 
+        var mainPart = context.MainPart;
+
         // Unsupported W3C attribute but claimed by users. Specified at <body> level, the page
         // orientation is applied on the whole document
         string? attr = styleAttributes!["page-orientation"];
         if (attr != null)
         {
             PageOrientationValues orientation = Converter.ToPageOrientation(attr);
 
-            var sectionProperties = context.MainPart.Document.Body!.GetFirstChild<SectionProperties>();
+            var sectionProperties = mainPart.Document.Body!.GetFirstChild<SectionProperties>();
             if (sectionProperties == null || sectionProperties.GetFirstChild<PageSize>() == null)
             {
-                context.MainPart.Document.Body.Append(ChangePageOrientation(orientation));
+                mainPart.Document.Body.Append(ChangePageOrientation(orientation));
             }
             else
             {
@@ -61,10 +85,10 @@ protected override void ComposeStyles(ParsingContext context)
 
         if (paraProperties.BiDi is not null)
         {
-            var sectionProperties = context.MainPart.Document.Body!.GetFirstChild<SectionProperties>();
+            var sectionProperties = mainPart.Document.Body!.GetFirstChild<SectionProperties>();
             if (sectionProperties == null || sectionProperties.GetFirstChild<PageSize>() == null)
             {
-                context.MainPart.Document.Body.Append(sectionProperties = new());
+               mainPart.Document.Body.Append(sectionProperties = new());
             }
 
             sectionProperties.AddChild(paraProperties.BiDi.CloneNode(true));
@@ -105,10 +129,17 @@ private void MarkAllBookmarks()
         var links = node.QuerySelectorAll("a[href^='#']");
         if (links.Length == 0) return;
 
-        foreach (var link in links.Cast<IHtmlAnchorElement>())
+        foreach (var link in links.Cast<IHtmlAnchorElement>().Where(l => l.Hash.Length > 0))
         {
+            if (link.IsTopAnchor())
+            {
+                shouldRegisterTopBookmark = true;
+                return;
+            }
+
             var id = link.Hash.Substring(1);
             var target = node.Owner!.GetElementById(id);
+
             // `id` attribute is preferred but `name` is also valid
             target ??= node.Owner!.GetElementsByName(id).FirstOrDefault();
 

src/Html2OpenXml/Expressions/HtmlDomExpression.cs

@@ -67,6 +67,7 @@ private static Dictionary<string, Func<IElement, HtmlDomExpression>> InitKnownTa
             { TagNames.Strong, el => new PhrasingElementExpression((IHtmlElement) el, new Bold()) },
             { TagNames.Sub, el => new PhrasingElementExpression((IHtmlElement) el, new VerticalTextAlignment() { Val = VerticalPositionValues.Subscript }) },
             { TagNames.Sup, el => new PhrasingElementExpression((IHtmlElement) el, new VerticalTextAlignment() { Val = VerticalPositionValues.Superscript }) },
+            { TagNames.Svg, el => new SvgExpression((AngleSharp.Svg.Dom.ISvgSvgElement) el) },
             { TagNames.Table, el => new TableExpression((IHtmlTableElement) el) },
             { TagNames.Time, el => new PhrasingElementExpression((IHtmlElement) el) },
             { TagNames.U, el => new PhrasingElementExpression((IHtmlElement) el, new Underline() { Val = UnderlineValues.Single }) },

src/Html2OpenXml/Expressions/HyperlinkExpression.cs

@@ -14,6 +14,7 @@
 using System.Linq;
 using AngleSharp.Html.Dom;
 using DocumentFormat.OpenXml;
+using DocumentFormat.OpenXml.Packaging;
 using DocumentFormat.OpenXml.Wordprocessing;
 
 namespace HtmlToOpenXml.Expressions;
@@ -42,8 +43,10 @@ public override IEnumerable<OpenXmlElement> Interpret (ParsingContext context)
         // Let's see whether the link tag include an image inside its body.
         // If so, the Hyperlink OpenXmlElement is lost and we'll keep only the images
         // and applied a HyperlinkOnClick attribute.
-        var imagesInLink = childElements.Where(e => e.HasChild<Drawing>());
-        if (imagesInLink.Any())
+        IEnumerable<OpenXmlElement> imagesInLink;
+        // Clickable image is only supported in body but not in header/footer
+        if (context.HostingPart is MainDocumentPart &&
+            (imagesInLink = childElements.Where(e => e.HasChild<Drawing>())).Any())
         {
             foreach (var img in imagesInLink)
             {
@@ -56,25 +59,42 @@ public override IEnumerable<OpenXmlElement> Interpret (ParsingContext context)
                 if (enDp.MoveNext()) alt = enDp.Current.Description;
                 else alt = null;
 
-                d.InsertInDocProperties(
-                    new a.HyperlinkOnClick() { Id = h.Id ?? h.Anchor, Tooltip = alt });
+                d.Inline ??= new a.Wordprocessing.Inline();
+                d.Inline.DocProperties ??= new a.Wordprocessing.DocProperties();
+
+                if (h.Anchor == "_top")
+                {
+                    // exception case: clickable image requires the _top bookmark get registred with a relationship
+                    var extLink = context.HostingPart.AddHyperlinkRelationship(new Uri("#_top", UriKind.Relative), false);
+                    d.Inline.DocProperties.Append(
+                        new a.HyperlinkOnClick() { Id = extLink.Id, Tooltip = alt });
+                }
+                else
+                {
+                    d.Inline.DocProperties.Append(
+                        new a.HyperlinkOnClick() { Id = h.Id ?? h.Anchor, Tooltip = alt });
+                }
             }
         }
 
         // can't use GetFirstChild<Run> or we may find the one containing the image
+        List<Run> runs = [];
         foreach (var el in childElements)
         {
-            if (el is Run run && !run.HasChild<Drawing>())
-            {
-                run.RunProperties ??= new();
-                run.RunProperties.RunStyle = context.DocumentStyle.GetRunStyle(
-                        context.DocumentStyle.DefaultStyles.HyperlinkStyle);
-                break;
-            }
+            if (el is Run r) runs.Add(r);
+            // unroll paragraphs. CloneNode is need to unparent the run
+            else runs.AddRange(el.Elements<Run>().Select(r => (Run) r.CloneNode(true)));
+        }
+
+        foreach (var run in runs.Where(run => !run.HasChild<Drawing>()))
+        {
+            run.RunProperties ??= new();
+            run.RunProperties.RunStyle = context.DocumentStyle.GetRunStyle(
+                    context.DocumentStyle.DefaultStyles.HyperlinkStyle);
         }
 
         // Append the processed elements and put them to the Run of the Hyperlink
-        h.Append(childElements);
+        h.Append(runs);
 
         return [h];
     }
@@ -87,20 +107,21 @@ public override IEnumerable<OpenXmlElement> Interpret (ParsingContext context)
         if (string.IsNullOrEmpty(att))
             return null;
 
+        // Always accept _top anchor
+        if (linkNode.IsTopAnchor())
+        {
+            h = new Hyperlink() { History = true, Anchor = "_top" };
+        }
         // is it an anchor?
-        if (att![0] == '#' && att.Length > 1)
+        else if (!context.Converter.ExcludeLinkAnchor && linkNode.Hash.Length > 1 && linkNode.Hash[0] == '#')
         {
-            // Always accept _top anchor
-            if (!context.Converter.ExcludeLinkAnchor || att == "#_top")
-            {
-                h = new Hyperlink(
-                    ) { History = true, Anchor = att.Substring(1) };
-            }
+            h = new Hyperlink(
+                ) { History = true, Anchor = linkNode.Hash.Substring(1) };
         }
         // ensure the links does not start with javascript:
         else if (AngleSharpExtensions.TryParseUrl(att, UriKind.Absolute, out var uri))
         {
-            var extLink = context.MainPart.AddHyperlinkRelationship(uri!, true);
+            var extLink = context.HostingPart.AddHyperlinkRelationship(uri!, true);
 
             h = new Hyperlink(
                 ) { History = true, Id = extLink.Id };