In RichTextBox, only the background (the fill attribute) is supported and the val and color attributes (determining the pattern and its color) are not respected while importing.

The attribute is currently not supported when importing HTML documents. It can be replaced with the corresponding CSS property applied to each cell of the table.

NullReferenceException is thrown when measuring a document with floating image as the first element on a second page. The same problem can be observed during copy/paste of content, but with slightly different call stack: 

System.NullReferenceException: Object reference not set to an instance of an object.
   at Telerik.Windows.Documents.Layout.LayoutBox.get_ValidParent()
   at Telerik.Windows.Documents.DocumentPosition.GetCurrentParagraphBox()
   at Telerik.Windows.Controls.RichTextBoxUI.RadRichTextBoxRulerController.GetCurrentParagraphBox()
   at Telerik.Windows.Controls.RichTextBoxUI.RadRichTextBoxRulerController.AssociatedRichTextBox_LayoutUpdated(Object sender, EventArgs e)

Depending on the document structure, an InvalidOperationException with message "Removed LayoutBox should not be split. Such operation will result in invalid document structure." could be also thrown. A sample call stack is:

Telerik.Windows.Documents.Layout.LayoutBox.CreateNewFragment(Telerik.Windows.Documents.Layout.LayoutBox currentLineStartNode) 
Telerik.Windows.Documents.Layout.TableCellLayoutBox.MeasureOverrideInternal(Telerik.Windows.Documents.Model.SizeF availableSize) 
Telerik.Windows.Documents.Layout.TableCellLayoutBox.Telerik.Windows.Documents.Model.IBlockContainerLayoutBox.CallMeasureOverrideInternal(Telerik.Windows.Documents.Model.SizeF availableSize) 
Telerik.Windows.Documents.Layout.BlockContainerLayoutHelper.CheckAllFloatingBlocksAreIncluded(Telerik.Windows.Documents.Model.SizeF availableSize) 

Fix available in LIB Version 2018.1.312.

The formatting(font size) of a line break is changed to the document default after adding a single paragraph before the line break. However, the current span style will be changed to the default formatting after adding several paragraphs with a text before a line break. This results in adding the next paragraphs with document default formatting instead of the current span style. The problem is reproducable only in flow mode.
Note: the document default style formatting should be different from the formatting of the line breaks.

Consider introducing a property in the XAML that allows the user to set whether the gallery should be loaded asynchronously or not in order to handle scenarios in which the TableStylesGallery is not in a contextual tab.

When the Chinese IME is used with Microsoft Pinyin input, FatalExecutionEngineError is thrown in the FocusCaret() method of DocumentWebLayoutPresenter. The exception is not reproducible in Paged layout mode.

The style of the table fallbacks to the default one and when the content of a cell is deleted, the current font family and font size are changed. The scenario is working as expected in versions before R3 2016.

This field retrieves a character through a code value specified in the field-argument. Such documents are common and are created when symbols (e.g. with font Windings through Insert -> Symbol dialog) are inserted in RTF documents in MS Word. 
This functionality is also used when content containing <sym> element is copied from DOCX and pasted in RadRichTextBox - as the content is converted to RTF containing SYMBOL field during the operation.

Workarounds: 
- Implement such field with the API, see the documentation article (http://docs.telerik.com/devtools/wpf/controls/radrichtextbox/features/fields-and-document-variables/custom-fields ) and the SDK example (https://github.com/telerik/xaml-sdk/tree/master/RichTextBox/CustomField/ ). Note that MS Word doesn't show field menus for these fields in RTF documents.
- If RTF format is used and you have control over document creation, save the documents initially as docx, and then save them to RTF. This way the symbols will be encoded without the SYMBOL field construct, and are successfully imported in RadRichTextBox.

Available in R3 2018 Official Release.

If the spell checking dialog is opened when the caret is at the end of the paragraph, the spell check stops after the fist span in the paragraph after which there is a punctuation (e.g. comma or a dot).

However, if the caret is at the beginning of the paragraph (on the first misspelled word for example), everything works as expected.

The layout of the document goes to infinite cycle and the whole application freezes when specific combination (it's not exactly identified) of sections and section column breaks is present in the document. Sometimes UI may freeze when the document is trying to render for first time.

When a document is imported from HTML, we set locally to the span elements Times New Roman as a font family nevertheless it is not declared anywhere. The same applies for the font size as well - it is set to 12.

Workaround: Clear the FontFamily property after import

foreach (Span span in this.radRichTextBox.Document.EnumerateChildrenOfType<Span>())
{
    span.GetStyleProperty(Span.FontFamilyProperty).ClearValue();
    span.GetStyleProperty(Span.FontSizeProperty).ClearValue();
}

"http://" prefix is automatically added by the Insert Hyperlink Dialog when the provided URI is to local path or mapped drive, e.g. "Z:\temp".

Workaround 1: Set HyperlinkPattern to something that matches file paths, e.g.:
(this.radRichTextBox.InsertHyperlinkDialog as RadInsertHyperlinkDialog).HyperlinkPattern = ".*";

Workaround 2: Insert such paths with the "file://" prefix.

When there is a Section with columns layout and it is separated with continuous section break, inserting column break in some of the columns doesn't get into account when the calculations of the available height are made.

When one column section is selected, "Line between" and "Equal column width" checkboxes should be disabled.

Improve measure and arrange of a document and its child document elements respectively - Span, Paragraph, Table, Section and etc. This will also improve the import time for large documents.

Improve the performance when a document is printed.

When documents containing images (e.g. few JPEGs with size of 5 MB) are loaded, the memory usage of the process goes up pretty quickly and eventually OutOfMemoryException is thrown.

This is not conventional memory leak, as managed types are not leaking. After some investigation, we believe that GC cannot collect some MemoryStream objects (used for the conversion of the image from Base64 to PNG) from the Large Object Heap fast enough, or LOH becomes fragmented.



Available in LIB version: 2017.1.213

Text in a section before continuous section break is not evenly distributed among columns when Left column type is applied.  
As a result the first column's height is bigger than the height of the second column. 

Performing Undo of bolded selection when the selection includes two tables separated by a paragraph is nesting the tables which were siblings.