We are using the Telerik Xamarin UI components, which were released on 18th March 2020, for one of our micro-services hosted in Azure cloud, to convert rtf text to raw text. Currently we have Linux containers in Azure. The code which we are using is as follows -

var rtfFormatProvider = new RtfFormatProvider();

var txtFormatProvider = new TxtFormatProvider();

RadFlowDocument doc = null;

This results in a missing paragraph. For example, having a table with three cells and Page field in the footer of a document and exporting it to PDF, will not export the last paragraph. 

As a workaround add new run after the last field character in the cell's table before exporting the document to PDF.

                BlockCollection footerContent = this.document.Sections.First().Footers.Default.Blocks;
                Table footerTable = footerContent.First() as Table;
                var cells = footerTable.Rows.Last().Cells.Where(x => x.EnumerateChildrenOfType<FieldCharacter>().Any());
                foreach (var cell in cells)
                {
                    cell.Blocks.AddParagraph().Inlines.AddRun();
                }

The current implementation uses System.Windows.Media.Imaging.BitmapImage class in order to take the image pixels for the PDF export. However, BitmapImage class throws NotSupportedException when being initialized with a WMF or EMF image.

WORKAROUND: WMF and EMF images may be converted to PNG images before the PDF export. The following code snippet shows how to convert all inline WMF image to PNG images by using System.Drawing.Image class:

private static void ConvertInlineWmfImagesToPng(RadFlowDocument document)
{
    foreach (ImageInline image in document.EnumerateChildrenOfType<ImageInline>())
    {
        if (image.Image.ImageSource.Extension.Equals("wmf", StringComparison.InvariantCultureIgnoreCase))
        {
            using (MemoryStream wmfImageStream = new MemoryStream(image.Image.ImageSource.Data))
            {
                using (MemoryStream pngImageStream = new MemoryStream())
                {
                    var imageDrawing = System.Drawing.Image.FromStream(wmfImageStream);
                    imageDrawing.Save(pngImageStream, ImageFormat.Png);
                    byte[] pngBytes = pngImageStream.ToArray();
 
                    image.Image.ImageSource = new ImageSource(pngBytes, "png");
                }
            }
        }
    }
}

I have created a very simple template (see attached file), import it, add my contents and later try to do a MailMerge.
However, the call to this function fails with the following exception:
System.ArgumentException
  HResult=0x80070057
  Message=The document element is already associated with a parent.
Parametername: item
  Source=Telerik.Windows.Documents.Flow
  StackTrace:
   at Telerik.Windows.Documents.Flow.Model.Collections.DocumentElementCollection`2.VerifyDocumentElementOnInsert(T item) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Flow\Flow\Model\Collections\DocumentElementCollection.cs:line 69
   at Telerik.Windows.Documents.Core.Data.DocumentElementCollectionBase`2.InsertRange(Int32 index, IEnumerable`1 items) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Core\Core\Core\Data\DocumentElementCollectionBase.cs:line 129
   at Telerik.Windows.Documents.Flow.Model.InlineRangeEditor.InsertInlinesInRange(InlineBase start, IEnumerable`1 inlines) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Flow\Flow\Model\InlineRangeEditor.cs:line 132
   at Telerik.Windows.Documents.Flow.Model.Fields.FieldInfo.UpdateFieldCore(FieldUpdateContext context) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Flow\Flow\Model\Fields\FieldInfo.cs:line 236
   at Telerik.Windows.Documents.Flow.Model.Fields.FieldInfo.UpdateFieldInternal(FieldUpdateContext context) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Flow\Flow\Model\Fields\FieldInfo.cs:line 186
   at Telerik.Windows.Documents.Flow.Model.MailMergeProcessor.ExecuteMailMerge(RadFlowDocument document, Object record) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Flow\Flow\Model\MailMergeProcessor.cs:line 57
   at Telerik.Windows.Documents.Flow.Model.MailMergeProcessor.Execute(RadFlowDocument document, IEnumerable collection) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Flow\Flow\Model\MailMergeProcessor.cs:line 25
   at Telerik.Windows.Documents.Flow.Model.RadFlowDocument.MailMerge(IEnumerable collection) in c:\DeveloperTooling_Agent13\_work\91\s\Documents\Flow\Flow\Model\RadFlowDocument.cs:line 337
   at JOIM.TextExport.DocumentGenerator.Export(String outputPath) in P:\Tolaris\JOIM.Common\JOIM.TextExport\DocumentGenerator.cs:line 581

using Telerik.Windows.Documents.Flow.FormatProviders.Docx;
using Telerik.Windows.Documents.Flow.Model;
using Telerik.Windows.Documents.Spreadsheet.Model;

...

using (StreamfileStream = File.Open(templatePath, FileMode.Open, FileAccess.Read, FileShare.ReadWrite))
{
    var provider = newDocxFormatProvider(); 
    _document = provider.Import(fileStream);
}

...

_document = Document.MailMerge(new [] { MergeFieldData });

When nested div elements are imported the WordsProcessing library creates separate paragraph elements for each div. However, if the outer div doesn't contain any inline children preceding the inner div element, the first created paragraph will be removed and only the second paragraph will be left. This leads to losing any style properties of the outer div.

Example:

<div style="margin-top: 50px;">
    <div>Text</div>
</div>

When the last run from a paragraph is underlined, the associated bullet has underline applied as well. The same applies for the background color.

When importing from HTML, all successive spaces in a spans are trimmed. Instead, in some cases one space should be left, e.g. between words. For example, the importing the following HTML should leave one space after the hyperlink:

<p><a href="www.telerik.com" target="_blank"><span>test</span></a>      and more.</p>

Workaround: After importing, check if the runs after the hyperlinks start with space:
foreach (var hyperlinkEnd in this.document.EnumerateChildrenOfType<FieldCharacter>().Where(f => f.FieldCharacterType == FieldCharacterType.End))
{
    Paragraph currentParagraph = hyperlinkEnd.Paragraph;
    int indexOfNextRun = currentParagraph.Inlines.IndexOf(hyperlinkEnd) + 1;
    if (currentParagraph.Inlines.Count > indexOfNextRun)
    {
        Run run = currentParagraph.Inlines[indexOfNextRun] as Run;
        if (run != null && run.Text[0] != ' ')
        {
            run.Text = " " + run.Text;
        }
    }
}

Currently Run.Shading is not exported to PDF.

Note: This scenario is common when converting HTML to PDF, as Run.Shading is set when construct like <span style="background-color:#ffcc00;"> is used.

Workaround: Iterate all the Runs in the already imported HTML document and set their HighlightColor to Shading.BackgroundColor.LocalValue. Check this code snippet:

foreach (Run run in document.EnumerateChildrenOfType<Run>())
{
	if (!run.Properties.HighlightColor.HasLocalValue)
	{
		run.HighlightColor = run.Shading.BackgroundColor.LocalValue;
	}
}

If HTML document is imported, and it contains image with invalid URL, then the image is imported with this URL in the document model. On subsequent export to Docx, the library tries to download the image data, which throws WebException. Instead, the image should be replaced with generic 'error' image.

Workaround: Manually test the image URL for correctness on HTML import, and replace the data:

        static void Main(string[] args)
        {
            HtmlFormatProvider htmlFormatProvider = new HtmlFormatProvider();
            htmlFormatProvider.ImportSettings.LoadFromUri += (sender, e) =>
            {
                if (!IsValid(e.Uri))
                {
                    e.SetData(File.ReadAllBytes("no-image.png"));
                }
            };
        }

        private static bool IsValid(string uri)
        {
            try
            {
                using (WebClient client = new WebClient())
                {
                    client.DownloadData(uri);
                }
            }
            catch (WebException)
            {
                return false;
            }

            return true;
        }