redirect-to-canonical

/** 
 * This service worker was written to address an issue where product pages were accessible from multiple URLs.  
 * While we were able to get appropriate canonicals in place, we were limited in that we could not create redirect rules for these.
 * 
 * So.. we made a service worker that checks the canonical tag against the request, and then redirects if it doesn't match.
 * 
 * You can test this here: https://cloudflareworkers.com/?_ga=2.121767817.342796392.1562183042-1d4c17e1b6d2561f25ac42f49d0553dc#4fbcee4f7582595bd32fa2787a91fcad:https://www.contentkingapp.com/academy/canonical/
 * 
 */

addEventListener('fetch', event => {
  event.passThroughOnException()
  event.respondWith(handleRequest(event.request))
})

async function handleRequest(request) {

// Get the requested URL for comparison to canonical
  let url = new URL(request.url)
  //console.log("\nPresented URL: " + url)
  let requestedurl = url.protocol + "//" + url.host + url.pathname
  //console.log("\nRequested URL: " + requestedurl)

// Read response body
  let response = await fetch(request)

// Make sure this is text and not an image, etc. 
  let type = response.headers.get("Content-Type") || ""
    if (!type.startsWith("text/html")) {
      // Not text. Don't modify.
      return response
    }

// Get text of the page
  let text = await response.text()
  // console.log("\nText: " + text)

// Extract canonical tag from the page
  let regexp = new RegExp('<link rel="canonical" href="(.*)" />', 'gm')

  let canonical
  while (canonical = regexp.exec(text)) {
    //console.log("\nMatch: " + canonical[0])
    let canonicalurl = canonical[1]
    //console.log("\nCanonical URL: " + canonical[1])

    // Check to see if canonical matches request
      if (canonicalurl === requestedurl){
        console.log("\nMatching URLs: " + requestedurl + " equals " + canonicalurl)

        return new Response(text, {
          status: response.status,
          statusText: response.statusText,
          headers: response.headers
        })

      }else{
        console.log("\nUnMatched URLs: " + requestedurl + " does not equal " + canonicalurl)
        // Note, we append query parameters in case there was something important there...
        return Response.redirect(canonicalurl+url.search, 301);
      }      
  }
}