When exporting a document referencing PdfProcessing`s .NET Standard assemblies with PdfComplianceLevel set an exception is thrown.

The common annotation properties are described in table "Entries common to all annotation dictionaries" of the Pdf specification.

Declined: Handling a partial annotation import-export could lead to exporting invalid PDF documents.

When merging or splitting files that contain the "159 '\u009f'" char, ArgumentException("The encoding is not supported.") is thrown.

This issue can also result in ArgumentException in client applications with the message: currentIndirectObject should be null.

Workaround:

Use PdfFormatProvider or reflection to register the char (check the comments).

A Link annotation must have an associated action, destination, or named destination specifying what should happen when the annotation is activated. When merging documents with invalid Link annotations, an ArgumentException is thrown with the message 'Value cannot be null. Parameter name: namedDestination'

Workaround: Remove the invalid annotations:

foreach (var annotation in document.Annotations.ToList())
{
    var link = annotation as Link;
    if (link != null && link.Action == null && link.Destination == null && link.NamedDestination == null)
    {
        (annotation.Parent as RadFixedPage).Annotations.Remove(annotation);
    }
}

When merging documents containing the same embedded font every time a document is merged the same font is repeatedly embedded in the merged document. This leads to very large files.

A possible workaround before this feature is developed could be to use the PdfProcessing model to import and merge the documents. After the documents are merged to iterate the document`s content in order to cache the fonts using their name and if there are duplicated names to set the already cached one (check the attached project). 

As a side note, the suggested workaround doesn't cover all possible cases (e.g. if there are fonts with a common name but different font sets (modified fonts or font subsets) this may lead to missing/different characters in the produced document). So in order to use this workaround, you will need to ensure the font names are unique for every different font.

When using the Telerik.Documents.ImageUtils.JpegImageConverter default implementation in Blazor WebAssembly application the Magick.NET library doesn't load all of its assemblies.

Workaround:
Create a custom implementation of the JpegImageConverterBase class: Create Custom JpegImageConverter in .Net Standard.