Link Checker - Single Account

Stay organized with collections Save and categorize content based on your preferences.

Alerts icon

As a website evolves, new pages get added and old pages get taken down, leading to possible broken links. Keeping a Google Ads campaign in sync with the website can be challenging for many advertisers. Live advertisements could be pointing to non-existent pages, leading to poor user experience and inaccurate Google Ads stats.

Link Checker addresses this problem by iterating through all of your ads, keywords, and sitelinks, checking that their URLs do not produce "Page not found" or other types of error responses, emailing you when error responses are found, and saving the results of its analysis to a spreadsheet.

Some accounts might have a large number of URLs that cannot all be checked in a single run due to execution time limits or quotas. The script is designed to handle this case by tracking its progress over multiple runs and only checking URLs that it has not checked previously. The script uses the term "analysis" to refer to one full pass through all of your URLs, even if it takes multiple runs to complete.

Configuration

The script's main options can be set in the spreadsheet.

Spreadsheet screenshot

Scope
Determines whether the script checks ads, keywords, and sitelinks; and whether the script checks them even if they are paused. Most users should include all three to ensure all of their URLs are checked, but it's normally not necessary to check paused entities.
Valid Response Codes
The list of HTTP response codes that are considered valid. Most users should consider only 200 a valid response code.
Email after each script execution
If you enable this option, the script emails you with a summary of the URLs it checked after every run. This can provide you with an early alert about URL errors instead of having to wait for all of your URLs to be checked (which might take multiple runs).
Email after finishing entire analysis
If you enable this option, the script emails you a consolidated summary after it finishes checking all URLs.
Email even if no errors are found
If you enable this option, the script emails you (based on the previous two options) even if it did not find any errors. Most users prefer to be emailed only when there are errors, but receiving an email even if there are no errors can be a useful way to ensure the script is running as scheduled.
Save OK URLs to spreadsheet
If you enable this option, the script saves every URL it checks to the spreadsheet, not just the ones with errors. Most users prefer to save only broken URLs, but some might like to see the full population of URLs that the script has checked.
Days between analyses
Use this option to control how often the script starts a fresh analysis of the entire URL list. Note that this option controls the minimum number of days between analyses. The actual number could be higher if an analysis takes more time (for example, if you have a large number of URLs). See Scheduling for more details.

For some use cases, it might not be enough just to know that the URL is returning a valid page. For example, consider the case where a product page contains the text "product discontinued". In this case, you could flag the page as a broken link.

The script offers two validation options:

Failure strings

Set Use simple failure string matching to Yes to search for occurrences of the list of strings defined in failure strings. Any occurrence found in a web page marks the page as broken.

Custom validation

For even more flexibility, a custom JavaScript validation function can be used on each URL and response. For example, you can confirm that the title of a web page contains your brand name as follows:

  1. Set Use custom validation function to Yes.
  2. Insert your custom validation logic into a function near the top of the script:

    function isValidResponse(url, response, options, entityDetails) {
      // Your custom logic here, returning true for valid links, false otherwise.
      // ...
    }
    

A basic implementation is already in place, with examples of how to access the response text.

Scheduling

Each time the script runs, it automatically detects whether it should resume an analysis already in progress or whether the last analysis finished and it is time to start a fresh one (based on the Days between analyses option). As a result, regardless of how often you want to launch a fresh analysis, schedule the script to run Hourly so that each analysis iterates through your URLs as quickly as possible.

For example, set Days between analyses to 1 to cause the script to launch a new analysis no more than once per day. If the script is scheduled to run Hourly and it finishes checking all of your URLs in less than a day, subsequent runs will immediately terminate until the next day when it is time to start a fresh analysis.

How it works

To track its progress, the script creates a label and applies it to your ads, keywords, and sitelinks after it checks them. The script can then identify checked URLs on its next run. Once an analysis is complete (all URLs have been checked), the script clears the label for a new analysis.

The script uses UrlFetchApp to check final URLs and mobile final URLs.

Setup

  • Click the button below to create the spreadsheet-based script in your Google Ads account.

    Install the script template

  • Click the button below to make a copy of the template spreadsheet.

    Copy the template spreadsheet

  • Update spreadsheet_url and recipient_emails in your script.

  • Schedule the script to run Hourly.

Source code

// Copyright 2016, Google Inc. All Rights Reserved.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

/**
 * @name Link Checker
 *
 * @overview The Link Checker script iterates through the ads, keywords, and
 *     sitelinks in your account and makes sure their URLs do not produce "Page
 *     not found" or other types of error responses. See
 *     https://developers.google.com/google-ads/scripts/docs/solutions/link-checker
 *     for more details.
 *
 * @author Google Ads Scripts Team [adwords-scripts@googlegroups.com]
 *
 * @version 3.1
 *
 * @changelog
 * - version 3.1
 *   - Split into info, config, and code.
 * - version 3.0
 *   - Updated to use new Google Ads scripts features.
 * - version 2.2
 *   - Added support for failure strings and custom validation functions.
 * - version 2.1
 *   - Added expansion of conditional ValueTrack parameters (e.g. ifmobile).
 *   - Added expanded text ad and other ad format support.
 * - version 2.0.3
 *   - Added validation for external spreadsheet setup.
 * - version 2.0.2
 *   - Allow the custom tracking label to include spaces.
 * - version 2.0.1
 *   - Catch and output all UrlFetchApp exceptions.
 * - version 2.0
 *   - Completely revised the script to work on larger accounts.
 *   - Check URLs in campaign and ad group sitelinks.
 * - version 1.2
 *   - Released initial version.
 */
/**
 * Configuration to be used for the Link Checker.
 */

CONFIG = {
  // URL of the spreadsheet template.
  // This should be a copy of https://docs.google.com/spreadsheets/d/1iO1iEGwlbe510qo3Li-j4KgyCeVSmodxU6J7M756ppk/copy.
  'spreadsheet_url': 'YOUR_SPREADSHEET_URL',

  // Array of addresses to be alerted via email if issues are found.
  'recipient_emails': [
    'YOUR_EMAIL_HERE'
  ],

  // Label to use when a link has been checked. Label will be created if it doesn't exist.
  'label': 'LinkChecker_Done',

  // Number of seconds to sleep after each URL request. If your URLs are
  // all on one or a few domains, use this throttle to reduce the load that the
  // script imposes on your web server(s).
  'throttle': 0,

  // Number of seconds before timeout that the script should stop checking URLs
  // to make sure it has time to output its findings.
  'timeout_buffer': 120,

  'advanced_options': {
    /**
     * Parameters controlling the script's behavior after hitting a UrlFetchApp
     * QPS quota limit.
     */
    'quota_config': {
      'INIT_SLEEP_TIME': 250,
      'BACKOFF_FACTOR': 2,
      'MAX_TRIES': 5
    }
  }
};
/**
 * Performs custom validation on a URL, with access to details such as the URL,
 * the response from the server, configuration options and entity Details.
 *
 * To use, the "Use Custom Validation" option in the configuration spreadsheet
 * must be set to "Yes", and your custom validation code implemented within the
 * below function.
 *
 * See the documentation for this solution for further details.
 *
 * @param {string} url The URL being checked.
 * @param {!HTTPResponse} response The response object for the request.
 * @param {!Object} options Configuration options.
 * @param {!Object} entityDetails Details of the associated Ad / Keywords etc.
 * @return {boolean} Return true if the URL and response are deemed valid.
 */
function isValidResponse(url, response, options, entityDetails) {
  /*
     Some examples of data that can be used in determining the validity of this
     URL. This is not exhaustive and there are further properties available.
  */

  // The HTTP status code, e.g. 200, 404
  // const responseCode = response.getResponseCode();

  // The HTTP response body, e.g. HTML for web pages:
  // const responseText = response.getContentText();

  // The failure strings from the configuration spreadsheet, as an array:
  // const failureStrings = options.failureStrings;

  // The type of the entity associated with the URL, e.g. Ad, Keyword, Sitelink.
  // const entityType = entityDetails.entityType;

  // The campaign name
  // const campaignName = entityDetails.campaign;

  // The ad group name, if applicable
  // const adGroupName = entityDetails.adGroup;

  // The ad text, if applicable
  // const adText = entityDetails.ad;

  // The keyword text, if applicable
  // const keywordText = entityDetails.keyword;

  // The sitelink link text, if applicable
  // const sitelinkText = entityDetails.sitelink;

  /*
     Remove comments and insert custom logic to determine whether this URL and
     response are valid, using the data obtained above.

     If valid, return true. If invalid, return false.
  */

  // Placeholder implementation treats all URLs as valid
  return true;
}

const QUOTA_CONFIG = CONFIG.advanced_options.quota_config;

/**
 * Exceptions that prevent the script from finishing checking all URLs in an
 * account but allow it to resume next time.
 */
const EXCEPTIONS = {
  QPS: 'Reached UrlFetchApp QPS limit',
  LIMIT: 'Reached UrlFetchApp daily quota',
  TIMEOUT: 'Approached script execution time limit'
};

/**
 * Named ranges in the spreadsheet.
 */
const NAMES = {
  CHECK_AD_URLS: 'checkAdUrls',
  CHECK_KEYWORD_URLS: 'checkKeywordUrls',
  CHECK_SITELINK_URLS: 'checkSitelinkUrls',
  CHECK_PAUSED_ADS: 'checkPausedAds',
  CHECK_PAUSED_KEYWORDS: 'checkPausedKeywords',
  CHECK_PAUSED_SITELINKS: 'checkPausedSitelinks',
  VALID_CODES: 'validCodes',
  EMAIL_EACH_RUN: 'emailEachRun',
  EMAIL_NON_ERRORS: 'emailNonErrors',
  EMAIL_ON_COMPLETION: 'emailOnCompletion',
  FAILURE_STRINGS: 'failureStrings',
  SAVE_ALL_URLS: 'saveAllUrls',
  FREQUENCY: 'frequency',
  DATE_STARTED: 'dateStarted',
  DATE_COMPLETED: 'dateCompleted',
  DATE_EMAILED: 'dateEmailed',
  NUM_ERRORS: 'numErrors',
  RESULT_HEADERS: 'resultHeaders',
  ARCHIVE_HEADERS: 'archiveHeaders',
  USE_SIMPLE_FAILURE_STRINGS: 'useSimpleFailureStrings',
  USE_CUSTOM_VALIDATION: 'useCustomValidation'
};

const MILLISECONDS_PER_SECOND = 1000;

const SPREADSHEET_URL = CONFIG.spreadsheet_url;
const RECIPIENT_EMAILS = CONFIG.recipient_emails;
const LABEL = CONFIG.label;
const THROTTLE = CONFIG.throttle;
const TIMEOUT_BUFFER = CONFIG.timeout_buffer;

/**
 * The Link Checker script iterates through the ads, keywords, and sitelinks
 * in your account and makes sure their URLs do not produce "Page not found"
 * or other types of error responses.
 */
function main() {
  const spreadsheet = validateAndGetSpreadsheet(SPREADSHEET_URL);
  validateEmailAddresses(RECIPIENT_EMAILS);
  spreadsheet.setSpreadsheetTimeZone(AdsApp.currentAccount().getTimeZone());

  const options = loadOptions(spreadsheet);
  const status = loadStatus(spreadsheet);

  if (!status.dateStarted) {
    // This is the very first execution of the script.
    startNewAnalysis(spreadsheet);
  } else if (status.dateStarted > status.dateCompleted) {
    console.log('Resuming work from a previous execution.');
  } else if (dayDifference(status.dateStarted, new Date()) <
             options.frequency) {
    console.log(`Waiting until ${options.frequency} ` +
            `days have elapsed since the start of the last analysis.`);
    return;
  } else {
    // Enough time has passed since the last analysis to start a new one.
    removeLabels([LABEL]);
    startNewAnalysis(spreadsheet);
  }

  const results = analyzeAccount(options);
  outputResults(results, options);
}

/**
 * Checks as many new URLs as possible that have not previously been checked,
 * subject to quota and time limits.
 *
 * @param {Object} options Dictionary of options.
 * @return {!Object} An object with fields for the URLs checked and an
 *     indication if the analysis was completed (no remaining URLs to check).
 */
function analyzeAccount(options) {
  // Ensure the label exists before attempting to retrieve already checked URLs.
  ensureLabels([LABEL]);

  const checkedUrls = getAlreadyCheckedUrls(options);
  const urlChecks = [];
  let didComplete = false;

  try {
    // If the script throws an exception, didComplete will remain false.
    didComplete = checkUrls(checkedUrls, urlChecks, options);
  } catch(e) {
    if (e == EXCEPTIONS.QPS ||
        e == EXCEPTIONS.LIMIT ||
        e == EXCEPTIONS.TIMEOUT) {
      console.warn(`Stopped checking URLs early because: ${e}`);
      console.log('Checked URLs will still be output.');
    } else {
      throw e;
    }
  }

  return {
    urlChecks, didComplete
  };
}

/**
 * Outputs the results to a spreadsheet and sends emails if appropriate.
 *
 * @param {Object} results An object with fields for the URLs checked and an
 *     indication if the analysis was completed (no remaining URLs to check).
 * @param {Object} options Dictionary of options.
 */
function outputResults(results, options) {
  const spreadsheet = SpreadsheetApp.openByUrl(SPREADSHEET_URL);

  const numErrors = countErrors(results.urlChecks, options);
  console.log(`Found ${numErrors} this execution.`);

  saveUrlsToSpreadsheet(spreadsheet, results.urlChecks, options);

  // Reload the status to get the total number of errors for the entire
  // analysis, which is calculated by the spreadsheet.
  status = loadStatus(spreadsheet);

  if (results.didComplete) {
    spreadsheet.getRangeByName(NAMES.DATE_COMPLETED).setValue(new Date());
    console.log(`Found ${status.numErrors} across the entire analysis.`);
  }

  if (RECIPIENT_EMAILS) {
    if (!results.didComplete && options.emailEachRun &&
        (options.emailNonErrors || numErrors > 0)) {
      sendIntermediateEmail(spreadsheet, numErrors);
    }

    if (results.didComplete &&
        (options.emailEachRun || options.emailOnCompletion) &&
        (options.emailNonErrors || status.numErrors > 0)) {
      sendFinalEmail(spreadsheet, status.numErrors);
    }
  }
}

/**
 * Loads data from a spreadsheet based on named ranges. Strings 'Yes' and 'No'
 * are converted to booleans. One-dimensional ranges are converted to arrays
 * with blank cells omitted. Assumes each named range exists.
 *
 * @param {Object} spreadsheet The spreadsheet object.
 * @param {Array.<string>} names A list of named ranges that should be loaded.
 * @return {!Object} A dictionary with the names as keys and the values
 *     as the cell values from the spreadsheet.
 */
function loadDatabyName(spreadsheet, names) {
  const data = {};

  for (const name of names) {
    const range = spreadsheet.getRangeByName(name);

    if (range.getNumRows() > 1 && range.getNumColumns() > 1) {
      // Name refers to a 2d range, so load it as a 2d array.
      data[name] = range.getValues();
    } else if (range.getNumRows() == 1 && range.getNumColumns() == 1) {
      // Name refers to a single cell, so load it as a value and replace
      // Yes/No with boolean true/false.
      data[name] = range.getValue();
      data[name] = data[name] === 'Yes' ? true : data[name];
      data[name] = data[name] === 'No' ? false : data[name];
    } else {
      // Name refers to a 1d range, so load it as an array (regardless of
      // whether the 1d range is oriented horizontally or vertically).
      const isByRow = range.getNumRows() > 1;
      const limit = isByRow ? range.getNumRows() : range.getNumColumns();
      const cellValues = range.getValues();

      data[name] = [];
      for (let j = 0; j < limit; j++) {
        const cellValue = isByRow ? cellValues[j][0] : cellValues[0][j];
        if (cellValue) {
          data[name].push(cellValue);
        }
      }
    }
  }
  return data;
}

/**
 * Loads options from the spreadsheet.
 *
 * @param {Object} spreadsheet The spreadsheet object.
 * @return {!Object} A dictionary of options.
 */
function loadOptions(spreadsheet) {
  return loadDatabyName(spreadsheet,
      [NAMES.CHECK_AD_URLS, NAMES.CHECK_KEYWORD_URLS,
       NAMES.CHECK_SITELINK_URLS, NAMES.CHECK_PAUSED_ADS,
       NAMES.CHECK_PAUSED_KEYWORDS, NAMES.CHECK_PAUSED_SITELINKS,
       NAMES.VALID_CODES, NAMES.EMAIL_EACH_RUN,
       NAMES.EMAIL_NON_ERRORS, NAMES.EMAIL_ON_COMPLETION,
       NAMES.SAVE_ALL_URLS, NAMES.FREQUENCY,
       NAMES.FAILURE_STRINGS, NAMES.USE_SIMPLE_FAILURE_STRINGS,
       NAMES.USE_CUSTOM_VALIDATION]);
}

/**
 * Loads state information from the spreadsheet.
 *
 * @param {Object} spreadsheet The spreadsheet object.
 * @return {!Object} A dictionary of status information.
 */
function loadStatus(spreadsheet) {
  return loadDatabyName(spreadsheet,
      [NAMES.DATE_STARTED, NAMES.DATE_COMPLETED,
       NAMES.DATE_EMAILED, NAMES.NUM_ERRORS]);
}

/**
 * Saves the start date to the spreadsheet and archives results of the last
 * analysis to a separate sheet.
 *
 * @param {Object} spreadsheet The spreadsheet object.
 */
function startNewAnalysis(spreadsheet) {
  console.log('Starting a new analysis.');

  spreadsheet.getRangeByName(NAMES.DATE_STARTED).setValue(new Date());

  // Helper method to get the output area on the results or archive sheets.
  function getOutputRange(rangeName) {
    const headers = spreadsheet.getRangeByName(rangeName);
    return headers.offset(1, 0, headers.getSheet().getDataRange().getLastRow());
  }

  getOutputRange(NAMES.ARCHIVE_HEADERS).clearContent();

  const results = getOutputRange(NAMES.RESULT_HEADERS);
  results.copyTo(getOutputRange(NAMES.ARCHIVE_HEADERS));

  getOutputRange(NAMES.RESULT_HEADERS).clearContent();
}

/**
 * Counts the number of errors in the results.
 *
 * @param {Array.<Object>} urlChecks A list of URL check results.
 * @param {Object} options Dictionary of options.
 * @return {number} The number of errors in the results.
 */
function countErrors(urlChecks, options) {
  let numErrors = 0;

  for (const urlCheck of urlChecks) {
    if (options.validCodes.indexOf(urlCheck.responseCode) == -1) {
      numErrors++;
    }
  }

  return numErrors;
}

/**
 * Saves URLs for a particular account to the spreadsheet starting at the first
 * unused row.
 *
 * @param {Object} spreadsheet The spreadsheet object.
 * @param {Array.<Object>} urlChecks A list of URL check results.
 * @param {Object} options Dictionary of options.
 */
function saveUrlsToSpreadsheet(spreadsheet, urlChecks, options) {
  // Build each row of output values in the order of the columns.
  const outputValues = [];
  for (const urlCheck of urlChecks) {

    if (options.saveAllUrls ||
        options.validCodes.indexOf(urlCheck.responseCode) == -1) {
      outputValues.push([
        urlCheck.customerId,
        new Date(urlCheck.timestamp),
        urlCheck.url,
        urlCheck.responseCode,
        urlCheck.entityType,
        urlCheck.campaign,
        urlCheck.adGroup,
        urlCheck.ad,
        urlCheck.keyword,
        urlCheck.sitelink
      ]);
    }
  }

  if (outputValues.length > 0) {
    // Find the first open row on the Results tab below the headers and create a
    // range large enough to hold all of the output, one per row.
    const headers = spreadsheet.getRangeByName(NAMES.RESULT_HEADERS);
    const lastRow = headers.getSheet().getDataRange().getLastRow();
    const outputRange = headers.offset(lastRow - headers.getRow() + 1,
                                     0, outputValues.length);
    outputRange.setValues(outputValues);
  }

  for (const email of RECIPIENT_EMAILS) {
    spreadsheet.addEditor(email);
  }
}

/**
 * Sends an email to a list of email addresses with a link to the spreadsheet
 * and the results of this execution of the script.
 *
 * @param {Object} spreadsheet The spreadsheet object.
 * @param {boolean} numErrors The number of errors found in this execution.
 */
function sendIntermediateEmail(spreadsheet, numErrors) {
  spreadsheet.getRangeByName(NAMES.DATE_EMAILED).setValue(new Date());

  MailApp.sendEmail(RECIPIENT_EMAILS.join(','),
    'Link Checker Results',
    `The Link Checker script found ${numErrors} URLs with errors in ` +
    `an execution that just finished. See ${spreadsheet.getUrl()} for details.`
    );
}

/**
 * Sends an email to a list of email addresses with a link to the spreadsheet
 * and the results across the entire account.
 *
 * @param {Object} spreadsheet The spreadsheet object.
 * @param {boolean} numErrors The number of errors found in the entire account.
 */
function sendFinalEmail(spreadsheet, numErrors) {
  spreadsheet.getRangeByName(NAMES.DATE_EMAILED).setValue(new Date());

  MailApp.sendEmail(RECIPIENT_EMAILS.join(','),
      'Link Checker Results',
      `The Link Checker script found ${numErrors} URLs with errors ` +
      `across its entire analysis. See ${spreadsheet.getUrl()} for details.`);
}

/**
 * Retrieves all final URLs and mobile final URLs in the account across ads,
 * keywords, and sitelinks that were checked in a previous run, as indicated by
 * them having been labeled.
 *
 * @param {Object} options Dictionary of options.
 * @return {!Object} A map of previously checked URLs with the URL as the key.
 */
function getAlreadyCheckedUrls(options) {
  const urlMap = {};

  function addToMap(items) {
    for (const item of items) {
      for(const url of expandUrlModifiers(item)){
        urlMap[url] = true;
      }
    }
  }

  if (options.checkAdUrls) {
    addToMap(getUrlsBySelector(AdsApp.ads().
                               withCondition(labelCondition(true))));
  }

  if (options.checkKeywordUrls) {
    addToMap(getUrlsBySelector(AdsApp.keywords().
                               withCondition(labelCondition(true))));
  }

  if (options.checkSitelinkUrls) {
    addToMap(getAlreadyCheckedSitelinkUrls());
  }

  return urlMap;
}

/**
 * Retrieves all final URLs and mobile final URLs for campaign and ad group
 * sitelinks.
 *
 * @return {Array.<string>} An array of URLs.
 */
function getAlreadyCheckedSitelinkUrls() {
  const urls = [];

  // Helper method to get campaign or ad group sitelink URLs.
  function addSitelinkUrls(selector) {
    const iterator = selector.withCondition(labelCondition(true)).get();

    for (const entity of iterator) {
      const sitelinks = entity.extensions().sitelinks();
      urls.push(...getUrlsBySelector(sitelinks));
    }
  }

  addSitelinkUrls(AdsApp.campaigns());
  addSitelinkUrls(AdsApp.adGroups());

  return urls;
}

/**
 * Retrieves all URLs in the entities specified by a selector.
 *
 * @param {Object} selector The selector specifying the entities to use.
 *     The entities should be of a type that has a urls() method.
 * @return {!Array.<string>} An array of URLs.
 */
function getUrlsBySelector(selector) {
  const urls = [];
  const entities = selector.get();

  // Helper method to add the url to the list if it exists.
  function addToList(url) {
    if (url) {
      urls.push(url);
    }
  }

  for (const entity of entities){
    addToList(entity.urls().getFinalUrl());
    addToList(entity.urls().getMobileFinalUrl());
  }

  return urls;
}

/**
 * Retrieves all final URLs and mobile final URLs in the account across ads,
 * keywords, and sitelinks, and checks their response code. Does not check
 * previously checked URLs.
 *
 * @param {Object} checkedUrls A map of previously checked URLs with the URL as
 *     the key.
 * @param {Array.<Object>} urlChecks An array into which the results of each URL
 *     check will be inserted.
 * @param {Object} options Dictionary of options.
 * @return {boolean} True if all URLs were checked.
 */
function checkUrls(checkedUrls, urlChecks, options) {
  let didComplete = true;

  // Helper method to add common conditions to ad group and keyword selectors.
  function addConditionsForAds(selector, includePaused) {
    const statuses = ['ENABLED'];
    if (includePaused) {
      statuses.push('PAUSED');
    }

    const predicate = ` IN (${statuses.join(',')})`;
    return selector.withCondition(labelCondition(false)).
        withCondition(`ad_group_ad.status${predicate}`).
        withCondition(`campaign.status${predicate}`).
        withCondition(`ad_group.status${predicate}`);
  }

  function addConditionsForKeywords(selector, includePaused) {
    const statuses = ['ENABLED'];
    if (includePaused) {
      statuses.push('PAUSED');
    }

    const predicate = ` IN (${statuses.join(',')})`;
    return selector.withCondition(labelCondition(false)).
        withCondition(`ad_group_criterion.status${predicate}`).
        withCondition(`campaign.status${predicate}`).
        withCondition(`ad_group.status${predicate}`);
  }

  if (options.checkAdUrls) {
    didComplete = didComplete && checkUrlsBySelector(checkedUrls, urlChecks,
        addConditionsForAds(
                      AdsApp.ads().withCondition('CreativeFinalUrls != ""'),
                      options.checkPausedAds), options);
  }

  if (options.checkKeywordUrls) {
    didComplete = didComplete && checkUrlsBySelector(checkedUrls, urlChecks,
        addConditionsForKeywords(
                      AdsApp.keywords().withCondition('FinalUrls != ""'),
                      options.checkPausedKeywords), options);
  }

  if (options.checkSitelinkUrls) {
    didComplete = didComplete &&
        checkSitelinkUrls(checkedUrls, urlChecks, options);
  }

  return didComplete;
}

/**
 * Retrieves all final URLs and mobile final URLs in a selector and checks them
 * for a valid response code. Does not check previously checked URLs. Labels the
 * entity that it was checked, if possible.
 *
 * @param {Object} checkedUrls A map of previously checked URLs with the URL as
 *     the key.
 * @param {Array.<Object>} urlChecks An array into which the results of each URL
 *     check will be inserted.
 * @param {Object} selector The selector specifying the entities to use.
 *     The entities should be of a type that has a urls() method.
 * @param {!Object} options Dictionary of options.
 * @return {boolean} True if all URLs were checked.
 */
function checkUrlsBySelector(checkedUrls, urlChecks, selector, options) {
  const customerId = AdsApp.currentAccount().getCustomerId();
  const iterator = selector.get();
  const entities = [];
  // Helper method to check a URL.
  function checkUrl(entity, url) {
    if (!url) {
      return;
    }

    const urlsToCheck = expandUrlModifiers(url);

    for (const expandedUrl of urlsToCheck) {
      if (checkedUrls[expandedUrl]) {
        continue;
      }

      const entityType = entity.getEntityType();
      const entityDetails = {
        entityType: entityType,
        campaign: entity.getCampaign ? entity.getCampaign().getName() : '',
        adGroup: entity.getAdGroup ? entity.getAdGroup().getName() : '',
        ad: entityType == 'Ad' ? getAdAsText(entity) : '',
        keyword: entityType == 'Keyword' ? entity.getText() : '',
        sitelink: entityType.indexOf('Sitelink') != -1 ?
            entity.getLinkText() : ''
      };

      const responseCode = requestUrl(expandedUrl, options, entityDetails);

      urlChecks.push({
        customerId: customerId,
        timestamp: new Date(),
        url: expandedUrl,
        responseCode: responseCode,
        entityType: entityDetails.entityType,
        campaign: entityDetails.campaign,
        adGroup: entityDetails.adGroup,
        ad: entityDetails.ad,
        keyword: entityDetails.keyword,
        sitelink: entityDetails.sitelink
      });

      checkedUrls[expandedUrl] = true;
    }
  }

  entities.push(...iterator);

  for (const entity of entities) {
    checkUrl(entity, entity.urls().getFinalUrl());
    checkUrl(entity, entity.urls().getMobileFinalUrl());

    // Sitelinks do not have labels.
    if (entity.applyLabel) {
      entity.applyLabel(LABEL);
      checkTimeout();
    }
  }

  // True only if we did not breach an iterator limit.
  return entities.length == iterator.totalNumEntities();
}

/**
 * Retrieves a text representation of an ad, casting the ad to the appropriate
 * type if necessary.
 *
 * @param {Ad} ad The ad object.
 * @return {string} The text representation.
 */
function getAdAsText(ad) {
  // There is no AdTypeSpace method for textAd
  if (ad.getType() === 'TEXT_AD') {
    return ad.getHeadline();
  } else if (ad.isType().expandedTextAd()) {
    const eta = ad.asType().expandedTextAd();
    return eta.getHeadlinePart1() + ' - ' + eta.getHeadlinePart2();
  } else if (ad.isType().gmailImageAd()) {
    return ad.asType().gmailImageAd().getName();
  } else if (ad.isType().gmailMultiProductAd()) {
    return ad.asType().gmailMultiProductAd().getHeadline();
  } else if (ad.isType().gmailSinglePromotionAd()) {
    return ad.asType().gmailSinglePromotionAd().getHeadline();
  } else if (ad.isType().html5Ad()) {
    return ad.asType().html5Ad().getName();
  } else if (ad.isType().imageAd()) {
    return ad.asType().imageAd().getName();
  } else if (ad.isType().responsiveDisplayAd()) {
    return ad.asType().responsiveDisplayAd().getLongHeadline();
  } else if (ad.isType().responsiveSearchAd()) {
    return ad.asType().responsiveSearchAd().getHeadlines().map(h => h.text).join(', ');
  }
  return 'N/A';
}


/**
 * Retrieves all final URLs and mobile final URLs for campaign and ad group
 * sitelinks and checks them for a valid response code. Does not check
 * previously checked URLs. Labels the containing campaign or ad group that it
 * has been checked.
 *
 * @param {Object} checkedUrls A map of previously checked URLs with the URL as
 *     the key.
 * @param {Array.<Object>} urlChecks An array into which the results of each URL
 *     check will be inserted.
 * @param {Object} options Dictionary of options.
 * @return {boolean} True if all URLs were checked.
 */
function checkSitelinkUrls(checkedUrls, urlChecks, options) {
  let didComplete = true;

  // Helper method to check URLs for sitelinks in a campaign or ad group
  // selector.
  function checkSitelinkUrls(selector) {
    const iterator = selector.withCondition(labelCondition(false)).get();
    const entities = [];

    for (const iteratorElement of iterator) {
      entities.push(iteratorElement);
    }

    for (const entity of entities) {
      const sitelinks = entity.extensions().sitelinks();

      if (sitelinks.get().hasNext()) {
        didComplete = didComplete &&
            checkUrlsBySelector(checkedUrls, urlChecks, sitelinks, options);
        entity.applyLabel(LABEL);
        checkTimeout();
      }
    }

    // True only if we did not breach an iterator limit.
    didComplete = didComplete &&
        entities.length  == iterator.totalNumEntities();
  }

  const statuses = ['ENABLED'];
  if (options.checkPausedSitelinks) {
    statuses.push('PAUSED');
  }

  const predicate = ` IN (${statuses.join(',')})`;
  checkSitelinkUrls(AdsApp.campaigns().
                    withCondition(`campaign.status${predicate}`));
  checkSitelinkUrls(AdsApp.adGroups().
                    withCondition(`ad_group.status${predicate}`).
                    withCondition(`campaign.status${predicate}`));

  return didComplete;
}

/**
 * Expands a URL that contains ValueTrack parameters such as {ifmobile:mobile}
 * to all the combinations, and returns as an array. The following pairs of
 * ValueTrack parameters are currently expanded:
 *     1. {ifmobile:<...>} and {ifnotmobile:<...>} to produce URLs simulating
 *        clicks from either mobile or non-mobile devices.
 *     2. {ifsearch:<...>} and {ifcontent:<...>} to produce URLs simulating
 *        clicks on either the search or display networks.
 * Any other ValueTrack parameters or customer parameters are stripped out from
 * the URL entirely.
 *
 * @param {string} url The URL which may contain ValueTrack parameters.
 * @return {!Array.<string>} An array of one or more expanded URLs.
 */
function expandUrlModifiers(url) {
  const ifRegex = /({(if\w+):([^}]+)})/gi;
  const modifiers = {};
  let matches;
  let modifiedUrls;
  while (matches = ifRegex.exec(url)) {
    // Tags are case-insensitive, e.g. IfMobile is valid.
    modifiers[matches[2].toLowerCase()] = {
      substitute: matches[0],
      replacement: matches[3]
    };
  }
  if (Object.keys(modifiers).length) {
    let mobileCombinations;
    if (modifiers.ifmobile || modifiers.ifnotmobile) {
      mobileCombinations =
          pairedUrlModifierReplace(modifiers, 'ifmobile', 'ifnotmobile', url);
    } else {
      mobileCombinations = [url];
    }

    // Store in a map on the offchance that there are duplicates.
    const combinations = {};
    for (const url of mobileCombinations) {
      if (modifiers.ifsearch || modifiers.ifcontent) {
        for (const modifiedUrl of
            pairedUrlModifierReplace(modifiers, 'ifsearch', 'ifcontent', url)) {
          combinations[modifiedUrl] = true;
        }
      } else {
        combinations[url] = true;
      }
    }
    modifiedUrls = Object.keys(combinations);
  } else {
    modifiedUrls = [url];
  }
  // Remove any custom parameters
  return modifiedUrls.map(url => url.replace(/{[\w_+:]+}/g, ''));
}

/**
 * Return a pair of URLs, where each of the two modifiers is mutually exclusive,
 * one for each combination. e.g. Evaluating ifmobile and ifnotmobile for a
 * mobile and a non-mobile scenario.
 *
 * @param {Object} modifiers A map of ValueTrack modifiers.
 * @param {string} modifier1 The modifier to honour in the URL.
 * @param {string} modifier2 The modifier to remove from the URL.
 * @param {string} url The URL potentially containing ValueTrack parameters.
 * @return {!Array.<string>} A pair of URLs, as a list.
 */
function pairedUrlModifierReplace(modifiers, modifier1, modifier2, url) {
  return [
    urlModifierReplace(modifiers, modifier1, modifier2, url),
    urlModifierReplace(modifiers, modifier2, modifier1, url)
  ];
}

/**
 * Produces a URL where the first {if...} modifier is set, and the second is
 * deleted.
 *
 * @param {Object} mods A map of ValueTrack modifiers.
 * @param {string} mod1 The modifier to honour in the URL.
 * @param {string} mod2 The modifier to remove from the URL.
 * @param {string} url The URL potentially containing ValueTrack parameters.
 * @return {string} The resulting URL with substitutions.
 */
function urlModifierReplace(mods, mod1, mod2, url) {
  const modUrl = mods[mod1] ?
      url.replace(mods[mod1].substitute, mods[mod1].replacement) :
      url;
  return mods[mod2] ? modUrl.replace(mods[mod2].substitute, '') : modUrl;
}

/**
 * Requests a given URL. Retries if the UrlFetchApp QPS limit was reached,
 * exponentially backing off on each retry. Throws an exception if it reaches
 * the maximum number of retries. Throws an exception if the UrlFetchApp daily
 * quota limit was reached.
 *
 * @param {string} url The URL to test.
 * @param {!Object} options The options loaded from the configuration sheet.
 * @param {!Object} entityDetails Details of the entity, e.g. type, name etc.
 * @return {number|string} The response code received when requesting the URL,
 *     or an error message.
 */
function requestUrl(url, options, entityDetails) {
  let responseCode;
  let sleepTime = QUOTA_CONFIG.INIT_SLEEP_TIME;
  let numTries = 0;

  while (numTries < QUOTA_CONFIG.MAX_TRIES && !responseCode) {
    try {
      // If UrlFetchApp.fetch() throws an exception, responseCode will remain
      // undefined.
      const response = UrlFetchApp.fetch(url, {muteHttpExceptions: true});
      responseCode = response.getResponseCode();

      if (options.validCodes.indexOf(responseCode) !== -1) {
        if (options.useSimpleFailureStrings &&
            bodyContainsFailureStrings(response, options.failureStrings)) {
          responseCode = 'Failure string detected';
        } else if (options.useCustomValidation && !isValidResponse(url,
            response, options, entityDetails)) {
          responseCode = "Custom validation failed";
        }
      }

      if (THROTTLE > 0) {
        Utilities.sleep(THROTTLE * MILLISECONDS_PER_SECOND);
      }
    } catch(e) {
      if (e.message.indexOf('Service invoked too many times in a short time:')
          != -1) {
        Utilities.sleep(sleepTime);
        sleepTime *= QUOTA_CONFIG.BACKOFF_FACTOR;
      } else if (e.message.indexOf('Service invoked too many times:') != -1) {
        throw EXCEPTIONS.LIMIT;
      } else {
        return e.message;
      }
    }

    numTries++;
  }

  if (!responseCode) {
    throw EXCEPTIONS.QPS;
  } else {
    return responseCode;
  }
}

/**
 * Searches the body of a HTTP response for any occurrence of a "failure string"
 * as defined in the configuration spreadsheet. For example, "Out of stock".
 *
 * @param {!HTTPResponse} response The response from the UrlFetchApp request.
 * @param {!Array.<string>} failureStrings A list of failure strings.
 * @return {boolean} Returns true if at least one failure string found.
 */
function bodyContainsFailureStrings(response, failureStrings) {
  const contentText = response.getContentText() || '';
  // Whilst searching for each separate failure string across the body text
  // separately may not be the most efficient, it is simple, and tests suggest
  // it is not overly poor performance-wise.
  return failureStrings.some(
      failureString => contentText.indexOf(failureString) !== -1);
}

/**
 * Throws an exception if the script is close to timing out.
 */
function checkTimeout() {
  if (AdsApp.getExecutionInfo().getRemainingTime() < TIMEOUT_BUFFER) {
    throw EXCEPTIONS.TIMEOUT;
  }
}

/**
 * Returns the number of days between two dates.
 *
 * @param {Object} from The older Date object.
 * @param {Object} to The newer (more recent) Date object.
 * @return {number} The number of days between the given dates (possibly
 *     fractional).
 */
function dayDifference(from, to) {
  return (to.getTime() - from.getTime()) / (24 * 3600 * 1000);
}

/**
 * Builds a string to be used for withCondition() filtering for whether the
 * label is present or not.
 *
 * @param {boolean} hasLabel True if the label should be present, false if the
 *     label should not be present.
 * @return {string} A condition that can be used in withCondition().
 */
function labelCondition(hasLabel) {
  return `LabelNames ${(hasLabel ? `CONTAINS_ANY` : `CONTAINS_NONE`)}` +
         ` ["${LABEL}"]`;
}

/**
 * Retrieves an entity by name.
 *
 * @param {Object} selector A selector for an entity type with a Name field.
 * @param {string} name The name to retrieve the entity by.
 * @return {Object} The entity, if it exists, or null otherwise.
 */
function getEntityByName(selector, name) {
  const entities = selector.withCondition(`Name = "${name}"`).get();

  if (entities.hasNext()) {
    return entities.next();
  } else {
    return null;
  }
}

/**
 * Retrieves a Label object by name.
 *
 * @param {string} labelName The label name to retrieve.
 * @return {Object} The Label object, if it exists, or null otherwise.
 */
function getLabel(labelName) {
  return getEntityByName(AdsApp.labels(), labelName);
}

/**
 * Checks that the account has all provided labels and creates any that are
 * missing. Since labels cannot be created in preview mode, throws an exception
 * if a label is missing.
 *
 * @param {Array.<string>} labelNames An array of label names.
 */
function ensureLabels(labelNames) {
  for (const labelName of labelNames) {
    const label = getLabel(labelName);

    if (!label) {
      if (!AdsApp.getExecutionInfo().isPreview()) {
        AdsApp.createLabel(labelName);
      } else {
        throw `Label ${labelName} is missing and cannot be created in ` +
            `preview mode. Please run the script or create the label manually.`;
      }
    }
  }
}

/**
 * Removes all provided labels from the account. Since labels cannot be removed
 * in preview mode, throws an exception in preview mode.
 *
 * @param {Array.<string>} labelNames An array of label names.
 */
function removeLabels(labelNames) {
  if (AdsApp.getExecutionInfo().isPreview()) {
    throw 'Cannot remove labels in preview mode. Please run the script or ' +
        'remove the labels manually.';
  }

  for (const labelName of labelNames) {
    const label = getLabel(labelName);

    if (label) {
      label.remove();
    }
  }
}

/**
 * Validates the provided spreadsheet URL to make sure that it's set up
 * properly. Throws a descriptive error message if validation fails.
 *
 * @param {string} spreadsheeturl The URL of the spreadsheet to open.
 * @return {Spreadsheet} The spreadsheet object itself, fetched from the URL.
 * @throws {Error} If the spreadsheet URL hasn't been set
 */
function validateAndGetSpreadsheet(spreadsheeturl) {
  if (spreadsheeturl == 'YOUR_SPREADSHEET_URL') {
    throw new Error('Please specify a valid Spreadsheet URL. You can find' +
        ' a link to a template in the associated guide for this script.');
  }
  return SpreadsheetApp.openByUrl(spreadsheeturl);
}

/**
 * Validates the provided email addresses to make sure it's not the default.
 * Throws a descriptive error message if validation fails.
 *
 * @param {Array.<string>} recipientEmails The list of email adresses.
 * @throws {Error} If the list of email addresses is still the default
 */
function validateEmailAddresses(recipientEmails) {
  if (recipientEmails &&
      recipientEmails[0] == 'YOUR_EMAIL_HERE') {
    throw new Error('Please either specify a valid email address or clear' +
        ' the RECIPIENT_EMAILS field.');
  }
}