Final URL Health Checker: catch the broken landing pages before your spend does

/**
 * Final URL Health Checker
 * The SEM Dispatch Vault, semdispatch.com/vault
 *
 * Collects final URLs from enabled keywords and ads, de-duplicates them,
 * requests each one, and flags any 4xx/5xx into a Google Sheet and an
 * optional email. Read-only: this script never changes the account.
 */

var CONFIG = {
  // URL of a Google Sheet to write to. Leave empty to create a new one
  // on the first run; its URL is logged so you can paste it back here.
  SHEET_URL: "",

  // Maximum URLs to fetch per run. Keeps a single run inside the daily
  // URL-fetch quota; the rest are checked on the next run.
  MAX_URLS: 400,

  // Optional. An email address for the broken-URL digest.
  // Leave empty to log only.
  EMAIL: ""
};

function main() {
  var urls = collectUrls();
  Logger.log("Collected " + urls.length + " distinct final URL(s).");

  var toCheck = urls.slice(0, CONFIG.MAX_URLS);
  if (urls.length > CONFIG.MAX_URLS) {
    Logger.log(
      "Capped at " + CONFIG.MAX_URLS + " this run; " +
        (urls.length - CONFIG.MAX_URLS) + " left for next time."
    );
  }

  var broken = [];
  for (var i = 0; i < toCheck.length; i++) {
    var result = checkUrl(toCheck[i]);
    if (result.broken) broken.push(result);
  }

  Logger.log("Checked " + toCheck.length + ", " + broken.length + " broken.");
  for (var b = 0; b < broken.length; b++) {
    Logger.log("  " + broken[b].status + "  " + broken[b].url);
  }

  writeSheet(broken, toCheck.length);
  maybeEmail(broken, toCheck.length);
}

function collectUrls() {
  var seen = {};
  var urls = [];

  function add(url) {
    if (!url) return;
    if (seen[url]) return;
    seen[url] = true;
    urls.push(url);
  }

  // Enabled keywords in enabled ad groups and campaigns.
  var keywords = AdsApp.keywords()
    .withCondition("Status = ENABLED")
    .withCondition("CampaignStatus = ENABLED")
    .withCondition("AdGroupStatus = ENABLED")
    .get();
  while (keywords.hasNext()) {
    add(keywords.next().urls().getFinalUrl());
  }

  // Enabled ads in enabled ad groups and campaigns.
  var ads = AdsApp.ads()
    .withCondition("Status = ENABLED")
    .withCondition("CampaignStatus = ENABLED")
    .withCondition("AdGroupStatus = ENABLED")
    .get();
  while (ads.hasNext()) {
    add(ads.next().urls().getFinalUrl());
  }

  return urls;
}

function checkUrl(url) {
  try {
    var response = UrlFetchApp.fetch(url, {
      muteHttpExceptions: true,
      followRedirects: true,
      validateHttpsCertificates: true
    });
    var status = response.getResponseCode();
    return {
      url: url,
      status: status,
      broken: status >= 400
    };
  } catch (e) {
    // A fetch that throws (bad host, timeout, refused connection) is as
    // broken to a visitor as a 5xx, so record it rather than skip it.
    return {
      url: url,
      status: "ERROR: " + e.message,
      broken: true
    };
  }
}

function writeSheet(broken, checkedCount) {
  var ss;
  if (CONFIG.SHEET_URL) {
    ss = SpreadsheetApp.openByUrl(CONFIG.SHEET_URL);
  } else {
    ss = SpreadsheetApp.create("SEM Dispatch: Final URL Health");
    Logger.log("Created a new sheet. Paste this URL into CONFIG.SHEET_URL:");
    Logger.log(ss.getUrl());
  }

  var stamp = Utilities.formatDate(
    new Date(),
    AdsApp.currentAccount().getTimeZone(),
    "yyyy-MM-dd HHmm"
  );
  var sheet = ss.insertSheet("broken " + stamp);

  var header = ["Status", "Final URL"];
  var data = [header];
  for (var i = 0; i < broken.length; i++) {
    data.push([broken[i].status, broken[i].url]);
  }
  if (broken.length === 0) {
    data.push(["OK", "No broken URLs found in " + checkedCount + " checked."]);
  }

  sheet.getRange(1, 1, data.length, header.length).setValues(data);
  sheet.getRange(1, 1, 1, header.length).setFontWeight("bold");
  sheet.setFrozenRows(1);

  Logger.log("Wrote " + broken.length + " broken URL(s) to the Sheet.");
}

function maybeEmail(broken, checkedCount) {
  if (CONFIG.EMAIL === "" || broken.length === 0) return;

  var account = AdsApp.currentAccount();
  var lines = [
    "Final URL Health Checker, " + account.getName(),
    checkedCount + " URLs checked, " + broken.length + " broken:",
    ""
  ];
  for (var i = 0; i < broken.length; i++) {
    lines.push(broken[i].status + "  " + broken[i].url);
  }

  MailApp.sendEmail(
    CONFIG.EMAIL,
    "[URLs] " + account.getName() + ": " + broken.length + " broken landing page(s)",
    lines.join("\n")
  );
}