High ASCII characters not being escaped by the XmlFormat function

dev
1

According to ACF documentation, the XmlFormat function escapes high ASCII characters in the range of 159-255 as well as the usual suspects of > < ’ " and &

As an example, the “e” in Zoë (ASCII 235) is not escaped to Zo&#xeb; by Lucee’s XmlFormat function.

Instead of exporting data as a CSV I export in “XML Spreadsheet 2003” format that appears like XLSX with formatting, multiple tabs etc. Unlike XLSX the content is just text. The content of the data cells must be in XML format and is unforgiving of any errors. I understand that “XML Spreadsheet 2003” requires ASCII 128 and above to be escaped and that line feed (ASCII 10) to be escaped to &#10;

While this may not be elegant, Xml2Format below is my workaround alternative to XmlFormat

<cfscript>
function XMLHighSafe(string) {
    var ii = 0 ;
    var tmp = "" ;
    while(ReFind("[^\x00-\x7F]", string, ii, false)) {
        ii = ReFind("[^\x00-\x7F]", string, ii, false) ; 
        tmp = "&##x" & FormatBaseN(Asc(Mid(string, ii, 1)), 16) & ";" ;
        string = Insert(tmp, string, ii) ; 
        string = RemoveChars(string, ii, 1) ; 
        ii = ii + Len(tmp) ; 
     }
     return string;
}
function Xml2Format(string) {
    string = XMLHighSafe(Replace(Replace(Replace(Replace(Replace(Replace(Replace(string, "&", "&amp;", "all"), "'", "&##39;", "all"), Chr(13), "", "all"), chr(34), "&##34;", "all"), Chr(10), "&##10;", "all"), ">", "&gt;", "all"), "<", "&lt;", "all")) ;
    return string;
}
</cfscript>