CFLib.org – Common Function Library Project

Description:
This UDF will return the first N characters of a text block (string). If the break lands in the middle of a word or HTML tag, the function will extend the selection until the end for the word/tag (so it will not break in the middle of a word/tag). It will also attempt to close any un-closed HTML tags (either due to bad HTML or due to the closing tags being cut off). This is useful for automatically generating "teaser" text for news items, event listings, etc. Based on JavaScript function of the same name by Steven Levithan ( http://blog.stevenlevithan.com/archives/get-html-summary ).

Return Values:
Returns a string.

Example:

<cfsavecontent variable="text">
This is my input for the UDF. It will test the UDF and show me how well it crops and <b>ignores</b> HTML.
</cfsavecontent>
<cfoutput>#getLeadingHTMl(text, 50)#</cfoutput>

Parameters:

Full UDF Source:

/**
 * Display N leading characters of a text block which may include HTML.
 * 
 * @param input      String to parse. (Required)
 * @param maxChars      Maximum number of characters. (Required)
 * @return Returns a string. 
 * @author Max Paperno (max@wdg.us) 
 * @version 1, June 18, 2010 
 */
function getLeadingHtml(input, maxChars) {
    // token matches a word, tag, or special character
    var    token = "[[:word:]]+|[^[:word:]<]|(?:<(\/)?([[:word:]]+)[^>]*(\/)?>)|<";
    var    selfClosingTag = "^(?:[hb]r|img)$";
    var    output = "";
    var    charCount = 0;
    var    openTags = ""; var strPos = 0; var tag = "";
    var i = 1;

    var    match = REFind(token, input, i, "true");

    while ( (charCount LT maxChars) AND match.pos[1] ) {
        // If this is an HTML tag
        if (match.pos[3]) {
            output = output & Mid(input, match.pos[1], match.len[1]);
            tag = Mid(input, match.pos[3], match.len[3]);
            // If this is not a self-closing tag
            if ( NOT ( match.pos[4] OR REFindNoCase(selfClosingTag, tag) ) ) {
                // If this is a closing tag
                if ( match.pos[2] AND ListFindNoCase(openTags, tag) ) {
                    openTags = ListDeleteAt(openTags, ListFindNoCase(openTags, tag)); 
                } else {
                    openTags = ListAppend(openTags, tag);
                }
            }
        } else {
            charCount = charCount +  match.len[1];
            if (charCount LTE maxChars) output = output & Mid(input, match.pos[1], match.len[1]);
        }
        
        i = i + match.len[1];
        match = REFind(token, input, i, "true");
    }

    // Close any tags which were left open
    while ( ListLen(openTags) ) {
        output = output & "</" & ListLast(openTags) & ">";
        openTags = ListDeleteAt(openTags, ListLen(openTags));
    }

    if ( Len(input) GT Len(output) )
        output = output & "...";
    
    return output;
}