Add Paragraph property controlling  whether SpacingBefore / SpacingAfter is applied between paragraphs with the same style.

In MS Word, the corresponding option is controlled with "Don't add space between paragraphs of the same style" checkbox in the font properties dialog - and the paragraph spacing after will not be applied. 

In OOXML, it is described with the w:contextualSpacing element. We don't respect this property on import of such files, and the paragraph spacing is bigger.

Common use case for this feature is that when list is applied to multiple paragraphs, Paragraph List style is applied to the items, and this style has "Don't add space between paragraphs of the same style" applied, so the vertical spacing between the items is smaller.

Currently Find dialog searches only in the editor from which it's started, e.g. if the user presses Ctrl+F in a comment, only the text of this comment is searched.

A table in the body is wider than a table in the header when the document is printed. 

Steps to reproduce:

- Insert a tables in the header and the body.
- Print the document

Expected: The table in the body has the same width.
Actual: The table in the body is wider that the table in the header

Add full support for spell checking of comments. Currently the words are not underlined, and the user cannot show the suggestions in the context menu; but it can trigger the spell checking dialog from the UI command.

Check if this is possible for other child editors like footnotes/endnotes.

In MS Word the users have the ability to embed different objects and files in the document, using the Insert tab -> Text section -> Object. Implement similar behavior in RadRichTextBox, allowing to embed objects as inlines, to edit them, and to re-embed them after editing.

When a symbol is inserted, the font family of RichTextBox is set to the symbol's font family.


Workaround: Create a custom InsertSymbolDialog. The font family could be persisted in the Show() method of the new dialog and reset when the dialog is closing. Attached is a sample project demonstrating the workaround.

When a paragraph between two tables with different columns count is deleted and then the two tables are copy-pasted, the layout of the second table gets broken.

If a table style is applied to a table, and then document is exported/imported to RTF, properties of the style are applied as local properties. Subsequent edits of the style (or changing the style of the table) seems to not work, as the local properties are with higher priority.

When using the "Replace" function or the API, the formatting of the replaced word is lost.

Note: Using "Replace All" does not reset the formatting.

Steps to reproduce:
1. Enter the text "The quick brown fox" in a document
2. Apply formatting to the word "brown" (eg, make it bold)
3. Perform a find/replace on the word brown (eg, replace with blue) using "replace", not "replace all"
4. The formatting of the word is reset (it's no longer bold) 

Horizontal table borders are drawn outside of the table when vertical table borders are not present. The issue is very visible when background is applied to the table cells - the ends of the horizontal borders finish just outside the table cell background.

When the table's background is set, a few lines next to borders appearing in white. 

When the HightlightColorPicker is in DataTemplate its image cannot be set to a different than the default one. When the control is not in template it is working properly.

When a custom list is created and applied over different parts in the text (which are separated from text without list), the export to DOCX and RTF format is wrong as the list numbering is not correct. Instead of having several lists starting from index 1, there is only one list with continuing numbering in the exported document.

Multi-byte Unicode characters can be represented as multiple ANSI characters in the form \'xx, for example:
\'89\'ce can represent the character 火 if multi-byte encoding should be used. Using multi-byte encoding could be set explicitly using the \ucN tag, or implicitly using the \fcharset for the current font (e.g. \fcharset134 is double-byte).

When such text is imported (commonly CJK text [Chinese, Korean, Japanese] the characters are imported incorrectly (as two '?' or as two different characters for each character)

Such RTF documents are produced from Office 2010 (Word, Outlook), WordPad - on saving or on copying content from the applications.

We could export the image absolute position as CSS style, for example: ".img{position: absolute;left: 100px;top: 150px;}"

Additionally we could export the left, top, right and bottom margins of the image, as we already have this information in the model.

We need to support uneditable areas in the document flow. The closes analog in HTML is <div editableContent=false>...</div>.

When floating image is anchored to the first position of the first paragraph on a page, and there are some empty paragraph on the previous page, sometimes the image is positioned on the previous page instead of along the paragraph.