URL parameter blocklist/allowlist in GTM.md

- README - URL parameter blocklist OR allowlist in GTM.md

Useful for preventing PII from being tracked.

Usage:

• In GA4 config tag set page_location={{Page URL (PII Stripped)}}.
• In UA's GA Settings, set page={{UA Pageview URL (PII Stripped)}} and location={{Page URL (PII Stripped)}}.

GTM Var - fn.sanitizeUrl.js

/**
 * @function sanitizeUrl
 * @param {string} url - Input URL to be sanitized.
 * @return {string} URL with parameters modified according to predefined allow/block -list.
 */
function(){
  var CONFIG = {
    
      // Uses allowlist if true, otherwise uses blocklist.
      useAllowlist: false,
    
      // List of parameter names to be blocked (used if useAllowlist=false).
      blocklist: 'name,email,phone,address'.split(','),
    
      // List of parameter names to be allowed (used if useAllowlist=true).
      allowlist: 'utm_campaign,utm_content,utm_medium,utm_source,utm_term,utm_creative_format,utm_marketing_tactic,gbraid,wbraid,gclid,dclid'.split(','),
      
      // Forces blocked parameters to this value.
      // If not provided, blocked parameters will be dropped entirely.
      //replaceString:'[REDACTED]'
  };

  /**
   * @function screenQueryParams
   * @author <https://gist.github.com/smhmic/fca71a9914a755cde241498c8a11d254>
   * @param {string} url - Input string containing a query string to be screened (i.e. URL).
   * @param {string[]} paramList - List of parameter names to be blocked or allowed (depending on isAllowList).
   * @param {boolean} isAllowList - If true, parameters will blocked unless specified in paramList.
   * @param {undefined|string} [replaceString] - If falsy, blocked parameters will be dropped entirely, otherwise this will be the value of blocked params.
   * @return {string} Modified url with parameters screened as indicated.
   * 
   * @example
   *     screenQueryParams( '?a=1&b=2&c=3',  ['a','c']       ); //= ?b=2
   *     screenQueryParams( '?a=1&b=2&c=3',  ['a','c'], true ); //= ?a=1&c=3
   *     screenQueryParams( '/example#?a=1', ['a']           ); //= /example#
   *     screenQueryParams( '/example#a=1',  ['a']           ); //= /example#a=1
   *     screenQueryParams( location.href, ['email'],      false, '[PII]' ); // Example blocklist
   *     screenQueryParams( location.href, ['type','tag'], true,  '*'     ); // Example allowlist
   */
  var screenQueryParams = function( url, paramList, isAllowList, replaceString ){
    if( !url || !url.replace ) return url;
    paramList = paramList || []; 
    if( paramList.split ) paramList = paramList.split(/\s*(,|\n)\s*/g);
    return url.replace( /((\?)|&)([^#&=]+)(?:=([^#&]*))?/g, function(fullMatch,delim,qmark,key,m3){
      if( isAllowList ? paramList.indexOf(key)+1 : !(paramList.indexOf(key)+1) )
        return fullMatch;
      else return replaceString ? delim+key+'='+replaceString : qmark||'';
    // Trim trailing "&"s, and remove "?" if all params removed.
    }).replace(/\?&*$|(\?)&+/,'$1'); // .replace(/&+$/,'').replace(/^\?$/,''); 
  }
  
  return function sanitizeUrl( url ){
    return screenQueryParams( url, CONFIG.useAllowlist ? CONFIG.allowlist : CONFIG.blocklist, CONFIG.useAllowlist, CONFIG.replaceString );
  }
}

GTM Var - Page URL (PII Stripped).js

function(){
  return {{fn.sanitizeUrl}}( location.href );
}

GTM Var - UA Pageview URL (PII Stripped).js

function(){
  return {{fn.sanitizeUrl}}( 
     
    location.pathname + location.search //+ location.hash //<-- Uncomment to include anchor in pageview URL
    
  );
}

Could improve perf by moving regex out of loop using method shown in this solution - https://stackoverflow.com/a/37949642/445295

Could improve perf by moving regex out of loop using method shown in this solution - https://stackoverflow.com/a/37949642/445295

Done!