- Hello !
- je suis du texte
-
écrit en oblique.
-
-
public void SetContentType(System.Web.HttpRequest request, System.Web.HttpResponse response, String reportName)
+
+
+
delta parameter (d)
+ Looks how cool is
Open Xml.
+
+ Now with
HtmlToOpenXml, it never been so easy to convert html.
+
+ If you like it, add me a rating on github
+
+
+ simple text
+
+ Hello !
+ je suis du texte
+
écrit en oblique.
+
+
public void SetContentType(System.Web.HttpRequest request, System.Web.HttpResponse response, String reportName)
{
if (request.Browser.Browser.Contains("IE"))
{
- // Replace the %20 to obtain a clean name when saving the file from Word.
+ // Replace the %20 to obtain a clean name when saving the file from Word.
encodedFilename =
Uri.EscapeDataString(Path.GetFileNameWithoutExtension(encodedFilename)).Replace("%20", " ")
+ Path.GetExtension(encodedFilename);
}
}
-
-
+
+
Some <Pre> starting one
line below!
-
-
Header placeholder:
-
- - Item 1
- - Item 2
-
-
Footer Placeholder
-
+
+
Header placeholder:
+
+ - Item 1
+ - Item 2
+
+
Footer Placeholder
+
Lorem Ipsum
-
diff --git a/src/Html2OpenXml/Collections/HtmlAttributeCollection.cs b/src/Html2OpenXml/Collections/HtmlAttributeCollection.cs
index bdccb5f..0c1c686 100755
--- a/src/Html2OpenXml/Collections/HtmlAttributeCollection.cs
+++ b/src/Html2OpenXml/Collections/HtmlAttributeCollection.cs
@@ -20,8 +20,7 @@ namespace HtmlToOpenXml;
///
sealed class HtmlAttributeCollection
{
- private static readonly Regex stripStyleAttributesRegex = new(@"(?
.+?):\s*(?[^;]+);*\s*");
-
+ private static readonly Regex stripStyleAttributesRegex = new(@"(?[^;\s]+)\s?(&\#58;|:)\s?(?[^;&]+)\s?(;|&\#59;)*");
private readonly Dictionary attributes = [];
@@ -37,13 +36,7 @@ public static HtmlAttributeCollection ParseStyle(string? htmlTag)
// Encoded ':' and ';' characters are valid for browser but not handled by the regex (bug #13812 reported by robin391)
// ex=
- MatchCollection matches = stripStyleAttributesRegex.Matches(
-#if NET5_0_OR_GREATER
- System.Web.HttpUtility.HtmlDecode(htmlTag)
-#else
- HttpUtility.HtmlDecode(htmlTag)
-#endif
- );
+ MatchCollection matches = stripStyleAttributesRegex.Matches(htmlTag);
foreach (Match m in matches)
collection.attributes[m.Groups["name"].Value] = m.Groups["val"].Value;
diff --git a/src/Html2OpenXml/Expressions/BlockElementExpression.cs b/src/Html2OpenXml/Expressions/BlockElementExpression.cs
index 7ae7026..62bc8ce 100644
--- a/src/Html2OpenXml/Expressions/BlockElementExpression.cs
+++ b/src/Html2OpenXml/Expressions/BlockElementExpression.cs
@@ -168,8 +168,8 @@ protected override void ComposeStyles (ParsingContext context)
};
}
- var attrValue = styleAttributes!["text-align"];
- JustificationValues? align = Converter.ToParagraphAlign(attrValue);
+ JustificationValues? align = Converter.ToParagraphAlign(styleAttributes!["text-align"]);
+ if (!align.HasValue) align = Converter.ToParagraphAlign(node.GetAttribute("align"));
if (align.HasValue)
{
paraProperties.Justification = new() { Val = align };
diff --git a/src/Html2OpenXml/Expressions/HyperlinkExpression.cs b/src/Html2OpenXml/Expressions/HyperlinkExpression.cs
index 46fdc9a..62221ec 100644
--- a/src/Html2OpenXml/Expressions/HyperlinkExpression.cs
+++ b/src/Html2OpenXml/Expressions/HyperlinkExpression.cs
@@ -113,7 +113,7 @@ public override IEnumerable Interpret (ParsingContext context)
h = new Hyperlink() { History = true, Anchor = "_top" };
}
// is it an anchor?
- else if (!context.Converter.ExcludeLinkAnchor && linkNode.Hash.Length > 1 && linkNode.Hash[0] == '#')
+ else if (context.Converter.SupportsAnchorLinks && linkNode.Hash.Length > 1 && linkNode.Hash[0] == '#')
{
h = new Hyperlink(
) { History = true, Anchor = linkNode.Hash.Substring(1) };
diff --git a/src/Html2OpenXml/Expressions/Image/ImageExpressionBase.cs b/src/Html2OpenXml/Expressions/Image/ImageExpressionBase.cs
index 3cf0422..06fb089 100644
--- a/src/Html2OpenXml/Expressions/Image/ImageExpressionBase.cs
+++ b/src/Html2OpenXml/Expressions/Image/ImageExpressionBase.cs
@@ -12,7 +12,6 @@
using System.Collections.Generic;
using System.Linq;
using DocumentFormat.OpenXml;
-using DocumentFormat.OpenXml.Packaging;
using DocumentFormat.OpenXml.Wordprocessing;
using a = DocumentFormat.OpenXml.Drawing;
@@ -26,6 +25,12 @@ namespace HtmlToOpenXml.Expressions;
///
abstract class ImageExpressionBase(AngleSharp.Dom.IElement node) : HtmlDomExpression
{
+ private readonly RunProperties runProperties = new();
+ private readonly ParagraphProperties paraProperties = new();
+ // some style attributes, such as borders, will convert this node to a framed container
+ private bool renderAsFramed;
+
+
///
public override IEnumerable Interpret (ParsingContext context)
{
@@ -35,16 +40,17 @@ public override IEnumerable Interpret (ParsingContext context)
return [];
Run run = new(drawing);
- Border border = ComposeStyles();
- if (border.Val?.Equals(BorderValues.None) == false)
- {
- run.RunProperties ??= new();
- run.RunProperties.Border = border;
- }
+ ComposeStyles();
+
+ if (runProperties.HasChildren)
+ run.RunProperties = runProperties;
+
+ if (renderAsFramed)
+ return [new Paragraph(paraProperties, run)];
return [run];
}
- private Border ComposeStyles ()
+ private void ComposeStyles ()
{
var styleAttributes = node.GetStyles();
var border = new Border() { Val = BorderValues.None };
@@ -66,7 +72,22 @@ private Border ComposeStyles ()
border.Size = (uint) borderWidth.ValueInPx * 4;
}
}
- return border;
+
+ if (border.Val?.Equals(BorderValues.None) == false)
+ {
+ runProperties.Border = border;
+ }
+
+ // if the layout is not inline and both left and right are auto, image appears centered
+ // https://developer.mozilla.org/en-US/docs/Web/CSS/margin-left
+ var margin = styleAttributes.GetMargin("margin");
+ if (margin.Left.Type == UnitMetric.Auto
+ && margin.Right.Type == UnitMetric.Auto
+ && !AngleSharpExtensions.IsInlineLayout(styleAttributes["display"], "inline-block"))
+ {
+ paraProperties.Justification = new() { Val = JustificationValues.Center };
+ renderAsFramed = true;
+ }
}
///
diff --git a/src/Html2OpenXml/Expressions/Numbering/HeadingElementExpression.cs b/src/Html2OpenXml/Expressions/Numbering/HeadingElementExpression.cs
index 2c70032..9a9f6ff 100644
--- a/src/Html2OpenXml/Expressions/Numbering/HeadingElementExpression.cs
+++ b/src/Html2OpenXml/Expressions/Numbering/HeadingElementExpression.cs
@@ -40,9 +40,9 @@ public override IEnumerable Interpret (ParsingContext context)
paragraph.ParagraphProperties ??= new();
paragraph.ParagraphProperties.ParagraphStyleId =
context.DocumentStyle.GetParagraphStyle(context.DocumentStyle.DefaultStyles.HeadingStyle + level);
-
+
var runElement = childElements.FirstOrDefault();
- if (runElement != null && IsNumbering(runElement))
+ if (runElement != null && context.Converter.SupportsHeadingNumbering && IsNumbering(runElement))
{
var abstractNumId = GetOrCreateListTemplate(context, HeadingNumberingName);
var instanceId = GetListInstance(abstractNumId);
diff --git a/src/Html2OpenXml/Expressions/Numbering/ListExpression.cs b/src/Html2OpenXml/Expressions/Numbering/ListExpression.cs
index 811976c..3fa20af 100644
--- a/src/Html2OpenXml/Expressions/Numbering/ListExpression.cs
+++ b/src/Html2OpenXml/Expressions/Numbering/ListExpression.cs
@@ -50,9 +50,20 @@ readonly struct ListContext(string listName, int absNumId, int instanceId, int l
public override IEnumerable Interpret(ParsingContext context)
{
- var liNodes = node.Children.Where(n => n.LocalName == "li");
+ var liNodes = node.Children.Where(n => n.LocalName.Equals("li", StringComparison.OrdinalIgnoreCase));
if (!liNodes.Any()) yield break;
+ // W3C requires that nested list stands below a `li` element but some editors
+ // don't care to respect the standard. Let's reparent those lists
+ var nestedList = node.Children.Where(n =>
+ n.LocalName.Equals("ol", StringComparison.OrdinalIgnoreCase) ||
+ n.LocalName.Equals("ul", StringComparison.OrdinalIgnoreCase));
+ if (nestedList.Any())
+ {
+ foreach (var list in nestedList)
+ list.PreviousElementSibling?.AppendChild(list);
+ }
+
var listContext = context.Properties("listContext");
var parentContext = listContext;
var listStyle = GetListName(node, listContext.Name);
@@ -79,17 +90,38 @@ public override IEnumerable Interpret(ParsingContext context)
var expression = new BlockElementExpression(liNode);
var childElements = expression.Interpret(context);
if (!childElements.Any()) continue;
- Paragraph p = (Paragraph) childElements.First();
-
- p.ParagraphProperties ??= new();
- p.ParagraphProperties.ParagraphStyleId = GetStyleIdForListItem(context.DocumentStyle, liNode);
- p.ParagraphProperties.NumberingProperties = new NumberingProperties {
- NumberingLevelReference = new() { Val = level - 1 },
- NumberingId = new() { Val = listContext.InstanceId }
- };
- if (listContext.Dir.HasValue) {
- p.ParagraphProperties.BiDi = new() {
- Val = OnOffValue.FromBoolean(listContext.Dir == DirectionMode.Rtl)
+
+ // ensure to filter out any non-paragraph like any nested table
+ var paragraphs = childElements.OfType();
+ var listItemStyleId = GetStyleIdForListItem(context.DocumentStyle, liNode);
+
+ if (paragraphs.Any())
+ {
+ var p = paragraphs.First();
+ p.ParagraphProperties ??= new();
+ p.ParagraphProperties.ParagraphStyleId = listItemStyleId;
+ p.ParagraphProperties!.NumberingProperties ??= new NumberingProperties {
+ NumberingLevelReference = new() { Val = level - 1 },
+ NumberingId = new() { Val = listContext.InstanceId }
+ };
+ if (listContext.Dir.HasValue) {
+ p.ParagraphProperties.BiDi = new() {
+ Val = OnOffValue.FromBoolean(listContext.Dir == DirectionMode.Rtl)
+ };
+ }
+ }
+
+ // any standalone paragraphs must be aligned (indented) along its current level
+ foreach (var p in paragraphs.Skip(1))
+ {
+ // if this is a list item paragraph, skip it
+ if (p.ParagraphProperties?.NumberingProperties is not null)
+ continue;
+
+ p.ParagraphProperties ??= new();
+ p.ParagraphProperties.ParagraphStyleId ??= (ParagraphStyleId?) listItemStyleId!.CloneNode(true);
+ p.ParagraphProperties.Indentation = new() {
+ Left = (level * Indentation * 2).ToString()
};
}
diff --git a/src/Html2OpenXml/Expressions/Table/TableCellExpression.cs b/src/Html2OpenXml/Expressions/Table/TableCellExpression.cs
index 23bd2e2..4e816f8 100644
--- a/src/Html2OpenXml/Expressions/Table/TableCellExpression.cs
+++ b/src/Html2OpenXml/Expressions/Table/TableCellExpression.cs
@@ -1,4 +1,4 @@
-/* Copyright (C) Olivier Nizet https://github.com/onizet/html2openxml - All Rights Reserved
+/* Copyright (C) Olivier Nizet https://github.com/onizet/html2openxml - All Rights Reserved
*
* This source is subject to the Microsoft Permissive License.
* Please see the License.txt file for more information.
@@ -10,6 +10,7 @@
* PARTICULAR PURPOSE.
*/
using System.Collections.Generic;
+using System.Globalization;
using System.Linq;
using AngleSharp.Html.Dom;
using DocumentFormat.OpenXml;
@@ -59,6 +60,27 @@ protected override void ComposeStyles(ParsingContext context)
{
base.ComposeStyles(context);
+ Unit width = styleAttributes!.GetUnit("width");
+ if (!width.IsValid)
+ {
+ var widthValue = cellNode.GetAttribute("width");
+ if (!string.IsNullOrEmpty(widthValue))
+ {
+ width = Unit.Parse(widthValue);
+ }
+ }
+
+ if (width.IsValid)
+ {
+ cellProperties.TableCellWidth = new TableCellWidth
+ {
+ Type = width.Type == UnitMetric.Percent ? TableWidthUnitValues.Pct : TableWidthUnitValues.Dxa,
+ Width = width.Type == UnitMetric.Percent
+ ? ((int) (width.Value * 50)).ToString(CultureInfo.InvariantCulture)
+ : width.ValueInDxa.ToString(CultureInfo.InvariantCulture)
+ };
+ }
+
// Manage vertical text (only for table cell)
string? direction = styleAttributes!["writing-mode"];
if (direction != null)
@@ -98,4 +120,4 @@ internal static bool IsValidRowSpan(int rowSpan)
// 0 means it extends until the end of the table grouping section
return rowSpan == 0 || rowSpan > 1;
}
-}
\ No newline at end of file
+}
diff --git a/src/Html2OpenXml/Expressions/Table/TableExpression.cs b/src/Html2OpenXml/Expressions/Table/TableExpression.cs
index ecb0b9c..84c1d02 100644
--- a/src/Html2OpenXml/Expressions/Table/TableExpression.cs
+++ b/src/Html2OpenXml/Expressions/Table/TableExpression.cs
@@ -1,4 +1,4 @@
-/* Copyright (C) Olivier Nizet https://github.com/onizet/html2openxml - All Rights Reserved
+/* Copyright (C) Olivier Nizet https://github.com/onizet/html2openxml - All Rights Reserved
*
* This source is subject to the Microsoft Permissive License.
* Please see the License.txt file for more information.
@@ -175,17 +175,11 @@ protected override void ComposeStyles (ParsingContext context)
switch (width.Type)
{
case UnitMetric.Percent:
- if (width.Value == 100)
+ tableProperties.TableWidth = new TableWidth
{
- // Use Auto=0 instead of Pct=auto
- // bug reported by scarhand (https://html2openxml.codeplex.com/workitem/12494)
- tableProperties.TableWidth = new() { Type = TableWidthUnitValues.Auto, Width = "0" };
- }
- else
- {
- tableProperties.TableWidth = new() { Type = TableWidthUnitValues.Pct,
- Width = (width.Value * 50).ToString(CultureInfo.InvariantCulture) };
- }
+ Type = TableWidthUnitValues.Pct,
+ Width = (width.Value * 50).ToString(CultureInfo.InvariantCulture)
+ };
break;
case UnitMetric.Point:
case UnitMetric.Pixel:
@@ -287,4 +281,4 @@ protected override void ComposeStyles (ParsingContext context)
}
}
}
-}
\ No newline at end of file
+}
diff --git a/src/Html2OpenXml/HtmlConverter.cs b/src/Html2OpenXml/HtmlConverter.cs
index 7cb7a38..f834ecf 100755
--- a/src/Html2OpenXml/HtmlConverter.cs
+++ b/src/Html2OpenXml/HtmlConverter.cs
@@ -339,7 +339,8 @@ private TPart ResolveHeaderFooterPart(HeaderFooterValues? type)
public AcronymPosition AcronymPosition { get; set; }
///
- /// Gets or sets whether anchor links are included or not in the convertion.
+ /// Gets or sets whether anchor links are included or not in the conversion
+ /// (defaults ).
///
/// An anchor is a term used to define a hyperlink destination inside a document.
/// .
@@ -351,7 +352,23 @@ private TPart ResolveHeaderFooterPart(HeaderFooterValues? type)
/// elements
/// and set the value of href to #name of your bookmark.
///
- public bool ExcludeLinkAnchor { get; set; }
+ public bool SupportsAnchorLinks { get; set; } = true;
+
+ ///
+ /// Gets or sets whether anchor links are included or not in the conversion.
+ ///
+ /// An anchor is a term used to define a hyperlink destination inside a document.
+ /// .
+ ///
+ /// It exists some predefined anchors used by Word such as _top to refer to the top of the document.
+ /// The anchor #_top is always accepted regardless this property value.
+ /// For others anchors like refering to your own bookmark or a title, add a
+ /// and
+ /// elements
+ /// and set the value of href to #name of your bookmark.
+ ///
+ [Obsolete("Use SupportsAnchorLink instead, if ExcludeLinkAnchor = true -> SupportsAnchorLink = false")]
+ public bool ExcludeLinkAnchor { get => !SupportsAnchorLinks; set => SupportsAnchorLinks = !value; }
///
/// Gets the Html styles manager mapping to OpenXml style properties.
@@ -367,7 +384,7 @@ public WordDocumentStyle HtmlStyles
public CaptionPositionValues TableCaptionPosition { get; set; }
///
- /// Gets or sets whether the pre tag should be rendered as a table (default ).
+ /// Gets or sets whether the pre tag should be rendered as a table (defaults ).
///
/// The table will contains only one cell.
public bool RenderPreAsTable { get; set; }
@@ -378,6 +395,16 @@ public WordDocumentStyle HtmlStyles
///
public bool ContinueNumbering { get; set; } = true;
+ ///
+ /// Defines whether any headings (h1-h6) could be considered as multi-level numbering, such as
+ /// top-level headings (Heading 1) are numbered 1, 2, 3, for example, and second-level headings (Heading 2) are numbered 1.1, 1.2, 1.3.
+ /// This feature is enabled by default.
+ ///
+ /// The converter is detecting headings starting with a number (ie: 1. or 1 )
+ /// are considered as numbering.
+ ///
+ public bool SupportsHeadingNumbering { get; set; } = true;
+
///
/// Gets the mainDocumentPart of the destination OpenXml document.
///
diff --git a/src/Html2OpenXml/HtmlToOpenXml.csproj b/src/Html2OpenXml/HtmlToOpenXml.csproj
index 03daeff..bfffc34 100644
--- a/src/Html2OpenXml/HtmlToOpenXml.csproj
+++ b/src/Html2OpenXml/HtmlToOpenXml.csproj
@@ -9,13 +9,13 @@
HtmlToOpenXml
HtmlToOpenXml
HtmlToOpenXml.dll
- 3.2.1
+ 3.2.2
icon.png
Copyright 2009-$([System.DateTime]::Now.Year) Olivier Nizet
- See changelog https://github.com/onizet/html2openxml/blob/master/CHANGELOG.md
+ (Please write the package release notes in CHANGELOG.md)
README.md
office openxml netcore html
- 3.2.1
+ 3.2.2
MIT
https://github.com/onizet/html2openxml
https://github.com/onizet/html2openxml
@@ -64,5 +64,15 @@
true
+
+
+
+
+
+
+
+ @(ReleaseNoteLines, '%0a')
+
+
\ No newline at end of file
diff --git a/src/Html2OpenXml/Utilities/AngleSharpExtensions.cs b/src/Html2OpenXml/Utilities/AngleSharpExtensions.cs
index 3d89624..085f7d4 100644
--- a/src/Html2OpenXml/Utilities/AngleSharpExtensions.cs
+++ b/src/Html2OpenXml/Utilities/AngleSharpExtensions.cs
@@ -153,4 +153,14 @@ public static string CollapseLineBreaks(this string str)
return new string(chars, 0, length);
}
+
+ ///
+ /// Determines whether the layout mode is inline vs block or flex.
+ ///
+ [MethodImpl(MethodImplOptions.AggressiveInlining)]
+ public static bool IsInlineLayout(string? displayMode, string defaultLayout)
+ {
+ return (displayMode ?? defaultLayout)
+ .StartsWith("inline", StringComparison.OrdinalIgnoreCase) == true;
+ }
}
\ No newline at end of file
diff --git a/src/Html2OpenXml/Utilities/OpenXmlExtensions.cs b/src/Html2OpenXml/Utilities/OpenXmlExtensions.cs
index 5242f49..905ba21 100755
--- a/src/Html2OpenXml/Utilities/OpenXmlExtensions.cs
+++ b/src/Html2OpenXml/Utilities/OpenXmlExtensions.cs
@@ -9,11 +9,9 @@
* IMPLIED WARRANTIES OF MERCHANTABILITY AND/OR FITNESS FOR A
* PARTICULAR PURPOSE.
*/
-using System;
using System.Runtime.CompilerServices;
using DocumentFormat.OpenXml;
using DocumentFormat.OpenXml.Wordprocessing;
-using DocumentFormat.OpenXml.Drawing.Wordprocessing;
namespace HtmlToOpenXml;
diff --git a/test/HtmlToOpenXml.Tests/HeadingTests.cs b/test/HtmlToOpenXml.Tests/HeadingTests.cs
index c42ed25..b8feed1 100644
--- a/test/HtmlToOpenXml.Tests/HeadingTests.cs
+++ b/test/HtmlToOpenXml.Tests/HeadingTests.cs
@@ -51,6 +51,32 @@ public void OrderedPattern_ReturnsNumberingHeading(string html)
});
}
+ [TestCase("1. Heading 1
1.1 Heading Normal Case")]
+ [TestCase("1. Heading 1
1.1 Heading Double Space
", Description = "Double space after number")]
+ [TestCase("1. Heading 1
1.2 Heading Tab
", Description = "Tab after number")]
+ [TestCase("1. Heading 1
1.3Heading No Space
", Description = "No space after number")]
+ public void OrderedPattern_DisableNumberingSupports_ReturnsSimpleHeading(string html)
+ {
+ converter.SupportsHeadingNumbering = false;
+ var elements = converter.Parse(html);
+
+ var absNum = mainPart.NumberingDefinitionsPart?.Numbering
+ .Elements()
+ .Where(abs => abs.AbstractNumDefinitionName?.Val == NumberingExpressionBase.HeadingNumberingName)
+ .SingleOrDefault();
+ Assert.That(absNum, Is.Null);
+
+ var paragraphs = elements.Cast();
+ Assert.Multiple(() =>
+ {
+ Assert.That(paragraphs.Count(), Is.EqualTo(2));
+ Assert.That(paragraphs.First().InnerText, Is.EqualTo("1. Heading 1"));
+ Assert.That(paragraphs.First().ParagraphProperties?.NumberingProperties?.NumberingLevelReference?.Val,
+ Is.Null,
+ "First paragraph is not a numbering");
+ });
+ }
+
[Test]
public void MaxLevel_ShouldBeIgnored()
{
diff --git a/test/HtmlToOpenXml.Tests/ImgTests.cs b/test/HtmlToOpenXml.Tests/ImgTests.cs
index 15e5596..45aa66d 100644
--- a/test/HtmlToOpenXml.Tests/ImgTests.cs
+++ b/test/HtmlToOpenXml.Tests/ImgTests.cs
@@ -219,6 +219,22 @@ public async Task ParseIntoDocumentPart_ReturnsImageParentedToPart (Type openXml
AssertThatOpenXmlDocumentIsValid();
}
+ [TestCase("display:block", ExpectedResult = true)]
+ [TestCase("display:flex", ExpectedResult = true)]
+ [TestCase("display:inline", ExpectedResult = false)]
+ [TestCase("", ExpectedResult = false)]
+ public bool CenterImg_ReturnsFramedImg(string displayMode)
+ {
+ var elements = converter.Parse($@"![]()
");
+
+ Assert.That(elements, Has.Count.EqualTo(1));
+ Assert.That(elements[0], Is.TypeOf());
+ AssertIsImg(mainPart, elements[0]);
+ return elements[0].GetFirstChild()?.
+ Justification?.Val?.Value == JustificationValues.Center;
+ }
+
private static (Drawing, ImagePart) AssertIsImg (OpenXmlPartContainer container, OpenXmlElement paragraph)
{
var run = paragraph.GetFirstChild();
diff --git a/test/HtmlToOpenXml.Tests/LinkTests.cs b/test/HtmlToOpenXml.Tests/LinkTests.cs
index 27443ce..f688bef 100644
--- a/test/HtmlToOpenXml.Tests/LinkTests.cs
+++ b/test/HtmlToOpenXml.Tests/LinkTests.cs
@@ -82,7 +82,7 @@ public void Anchoring_WithUnknownTarget_ReturnsHyperlinkWithBookmark ()
[Test]
public void SetExcludeAnchoring_ReturnsSimpleRun ()
{
- converter.ExcludeLinkAnchor = true;
+ converter.SupportsAnchorLinks = false;
// _top is always present and bypass the previous rule
var elements = converter.Parse(@"Anchor2");
diff --git a/test/HtmlToOpenXml.Tests/NumberingTests.cs b/test/HtmlToOpenXml.Tests/NumberingTests.cs
index a7b914e..c48b699 100644
--- a/test/HtmlToOpenXml.Tests/NumberingTests.cs
+++ b/test/HtmlToOpenXml.Tests/NumberingTests.cs
@@ -514,7 +514,7 @@ public void WithRtl_ReturnsBidi(string dir, bool? expectedValue)
}
[Test]
- public void NestedNumberList_ReturnsIncrementalIdentation()
+ public void NestedNumberList_ReturnsIncrementalIndentation()
{
const int maxLevel = 8;
var sb = new System.Text.StringBuilder();
@@ -541,5 +541,75 @@ public void NestedNumberList_ReturnsIncrementalIdentation()
TestContext.Out.WriteLine($"{i}. {ident?.Left?.Value}");
}
}
+
+ [Test(Description = "Nested list must be a children of a `li` tag but some editor are not respecting the W3C standard (issue #173)")]
+ public async Task NestedNumberList_NonCompliant_ReturnsIncrementalIndentation()
+ {
+ await converter.ParseBody(@"
+ - Item1
+ - Item2
+ - Item 2.1
+
");
+
+ var absNum = mainPart.NumberingDefinitionsPart?.Numbering
+ .Elements()
+ .SingleOrDefault();
+ Assert.That(absNum, Is.Not.Null);
+
+ var inst = mainPart.NumberingDefinitionsPart?.Numbering
+ .Elements().Where(i => i.AbstractNumId?.Val == absNum.AbstractNumberId)
+ .SingleOrDefault();
+ Assert.That(inst, Is.Not.Null);
+ Assert.That(inst.NumberID?.Value, Is.Not.Null);
+
+ var elements = mainPart.Document.Body!.ChildElements;
+ Assert.Multiple(() => {
+ Assert.That(elements, Has.Count.EqualTo(3));
+ Assert.That(elements, Is.All.TypeOf());
+ Assert.That(mainPart.NumberingDefinitionsPart?.Numbering, Is.Not.Null);
+ });
+
+ // assert paragraphs linked to numbering instance
+ Assert.Multiple(() =>
+ {
+ Assert.That(elements.Cast().Select(e =>
+ e.ParagraphProperties?.NumberingProperties?.NumberingId?.Val?.Value),
+ Has.All.EqualTo(inst.NumberID.Value),
+ "All paragraphs are linked to the same list instance");
+ Assert.That(elements.Take(2).Select(p => p.GetFirstChild()?.NumberingProperties?.NumberingLevelReference?.Val?.Value), Has.All.EqualTo(0));
+ Assert.That(elements.Last().GetFirstChild()?.NumberingProperties?.NumberingLevelReference?.Val?.Value, Is.EqualTo(1));
+ });
+ AssertThatOpenXmlDocumentIsValid();
+ }
+
+ [Test]
+ public void NestedParagraph_ReturnsIndentedItems()
+ {
+ var elements = converter.Parse(@"
+ -
+
Paragraph text
+ Paragraph text
+
+
");
+
+ Assert.That(elements, Is.Not.Empty);
+
+ var inst = mainPart.NumberingDefinitionsPart?.Numbering
+ .Elements()
+ .SingleOrDefault();
+ Assert.That(inst, Is.Not.Null);
+ Assert.Multiple(() => {
+ Assert.That(elements.Last().GetFirstChild()?.NumberingProperties?.NumberingId,
+ Is.Null,
+ "Last paragraph is standalone and not linked to a list instance");
+ Assert.That(elements.Cast().Select(e =>
+ e.ParagraphProperties?.ParagraphStyleId?.Val?.Value),
+ Has.All.EqualTo("ListParagraph"),
+ "All paragraphs use the same paragraph style");
+ Assert.That(elements.Last().GetFirstChild()?.Indentation?.Left?.Value,
+ Is.EqualTo("720"),
+ "Last standalone paragraph is aligned with the level 1");
+ });
+ }
}
}
\ No newline at end of file
diff --git a/test/HtmlToOpenXml.Tests/StyleTests.cs b/test/HtmlToOpenXml.Tests/StyleTests.cs
index 2fb56f5..24afe76 100644
--- a/test/HtmlToOpenXml.Tests/StyleTests.cs
+++ b/test/HtmlToOpenXml.Tests/StyleTests.cs
@@ -165,5 +165,21 @@ public void DuplicateStyle_ReturnsLatter()
var styleAttributes = HtmlAttributeCollection.ParseStyle("color:red;color:blue");
Assert.That(styleAttributes["color"], Is.EqualTo("blue"));
}
+
+ [Test(Description = "Encoded ':' and ';' characters are valid")]
+ public void EncodedStyle_ShouldSucceed()
+ {
+ var styleAttributes = HtmlAttributeCollection.ParseStyle("text-decoration:underline;color:red");
+ Assert.That(styleAttributes["text-decoration"], Is.EqualTo("underline"));
+ Assert.That(styleAttributes["color"], Is.EqualTo("red"));
+ }
+
+ [Test(Description = "Key style with no value should be ignored")]
+ public void EmptyStyle_ShouldBeIgnoredd()
+ {
+ var styleAttributes = HtmlAttributeCollection.ParseStyle("text-decoration;color:red");
+ Assert.That(styleAttributes["text-decoration"], Is.Null);
+ Assert.That(styleAttributes["color"], Is.EqualTo("red"));
+ }
}
}
Heading 5
OfDEXhz/AIRbxtAkV1pYNrJHNC0qSbPlDKQDg8e3tVD4Q6BKPGmp67Y2lxbaJ5ckVo04IMgZ
1KjnqAByfXFezMiv95Q2PUZoCKOigY9BSkA9QDjmgqGBDAEHsaia0tnQxtbxMjdVKAg0W9pb
Wilba3ihB6iNAufyoltLaZw8tvFIw6FkBNSgADA4FLRRRRRRRX//2Q==
- ">
@@ -102,79 +126,81 @@