A family of Microsoft word processing software products for creating web, email, and print documents.
There is a related problem. Some keyboard layouts allow two methods for producing a character with a diacritical mark : 1) first type the dead key combination assigned to the mark, then type the base character; or 2) first type the base character, then type the key combination assigned to the mark. The two key combinations are not identical. The second method seems to be more general, allowing any mark or sequence of marks to be applied to any base character, but not all fonts properly render the compound shape.
Example: Using the "US-Extended" keyboard, you can get “ḍ” by 1) typing RightALT-x followed by “d”; or by 2) typing “d” followed by SHIFT-RightALT-x. You can similarly type Ḍ, ṭ, ṃ, ḥ in both ways. This keyboard includes diacritical marks for quite a few languages written or transliterated using the Latin script.
The problem I've noticed is this: MS Word accepts without complaint the words containing the accents produced by the first method, but flags with spelling error any words containing the accents produced by the second method. Since some other text editors, including Libre Office and Google mail, accept without complaint the accents typed by either method, could the Word behavior be a bug?