Provide API or mechanism for reusing the already embedded fonts instead of creating a new one when editing an existing document. 

This causes an issue in adobe preflight (font name is not unique).  

The issue is observed when performing merge with RadFixedDocument as well.

Cell borders are not set when the border is set on the table:

 <table border="1">

Such fonts are:

• ArialMT - should fallback to Arial
• TimesNewRomanPS-ItalicMT, TimesNewRomanPS-BoldMT, TimesNewRomanPS-BoldItalicMT - should fallback to the Times New Roman variants accordingly. 

When exporting a document that contains a Widget created with FixedContentEditor an exception is thrown: System.ArgumentNullException: 'Value cannot be null. Parameter name: context'

It can be reproduced using the following code snippet:

RadFixedDocument document = new RadFixedDocument();
var page = document.Pages.AddPage();
FixedContentEditor editor = new FixedContentEditor(page);

PushButtonField pushButton = new PushButtonField("button");
editor.Position.Translate(20, 450);
editor.DrawWidget(pushButton, new Size(100, 20));

string path = "Exported.pdf";
File.Delete(path);

using (Stream output = File.OpenWrite(path))
{
	PdfFormatProvider provider = new PdfFormatProvider();
	provider.Export(document, output);
}

document.AcroForm.FormFields.Add(pushButton);

With the current implementation when passing the font file using the FontsRepository the entire font file is embedded.

In order to embed only a subset of characters used in the document, the font should be installed and the font shouldn't be registered in the FontsRepository.

When Separation color space with "name" set as None it is falling back to an "alternateSpace" (color space) instead.

From the Pdf specification:
A Separation color space is defined as follows:
[ /Separation name alternateSpace tintTransform ]

Observed:

The None value is ignored and the color space fallbacks to the Lab color space.

When importing a document containing an EI keyword not followed by space or return an exception is thrown: EndOfStreamException: 'Attempted to read past the end of the stream.'

When importing a PDF document and parsing a name start character followed by a dictionary end one an exception is thrown: System.InvalidOperationException: 'Stack empty.'

With the current implementation, the permissions are ignored on import.

As a side note, there is an option to set permissions on export: UserAccessPermissions.

Importing a document containing an EI keyword within the image stream leads to an endless loop and the document is not imported successfully.