User creates XML file with name on native language. During composition XML into WEB format Arbortext produces subfolder with HTML files. If XML file in non-ascii, composed HTML file name is encoded into UTF-8 kind: D0A3D182D0B8D0BBD0B8D0B7D0B0D186D0B8D18F.html.
This happens for such non-ascii letters like Cyrillic. For files with XML names which lenght is about 60 letters resulting files is (as krazy example: D09ED0B1D18AD0B5D0BCD0B8D0BFD0BED181D0BBD0B5D0B4D0BED0B2D0B0D182D0B5D0BBD18CD0BDD0BED181D182D18CD0B2D0BDD0B5D188D0BDD0B5D0B3D0BED0BED181D0BCD0BED182D180D0B0D181D182D0BED0BBD0B0). Arbortext composes xml files with names longer than 65 cyrillic symbols into HTML files with names more than 255 characters in name. Such files corrupt OS: such files exists, but cannot be deleted.
Explanation of encoding is: escape to produce wrong URL reference and corrupt application.
But I don't see a reason to encode russian file names into entities. Cyrillic letters are supported in links, URLs and HTML files. File names can be longer than 60 symbols often to help user to remember content.
For my opinion support of native cyrillic for WEB composition is necessary and easy to do.
