README

This module extends wilr/silverstripe-googlesitemaps and symbiote/silverstripe-queuedjobs to generate static sitemap XML files (chunks and index) in a queued and scalable way.

Features

• Generates static sitemap chunks and index files using wilr/silverstripe-googlesitemaps
• Runs as a background job using symbiote/silverstripe-queuedjobs (supports long-running LARGE jobs)
• Supports multilingual sites using tractorcow/silverstripe-fluent, with per-locale domain context
• Saves output to public/sitemap/{locale}
• Optionally uploads all sitemap files (XML + XSL) to an S3-compatible bucket such as DigitalOcean Spaces
• Includes a proxy controller so sitemap files can be served under your app domain for Google Search Console compliance

Installation

Install via Composer:

composer require hudhaifas/silverstripe-googlesitemaps-queued

Then run dev/build:

vendor/bin/sake dev/build flush=all

Configuration

You can schedule the job using the CMS UI or declare it in YAML as a default recurring job:

SilverStripe\Core\Injector\Injector:
  Symbiote\QueuedJobs\Services\QueuedJobService:
    properties:
      defaultJobs:
        DailyEnglishSitemapJob:
          type: Hudhaifas\GoogleSitemapsQueued\Job\GenerateSitemapJob
          filter:
            JobTitle: 'Generate en Sitemap XML (queued)'
          construct:
            0: 'en'
            1: 'https://en.example.com'
          startDateFormat: 'Y-m-d H:i:s'
          startTimeString: 'tomorrow 01:00'
          recreate: true

        DailyArabicSitemapJob:
          type: Hudhaifas\GoogleSitemapsQueued\Job\GenerateSitemapJob
          filter:
            JobTitle: 'Generate ar Sitemap XML (queued)'
          construct:
            0: 'ar'
            1: 'https://ar.example.com'
          startDateFormat: 'Y-m-d H:i:s'
          startTimeString: 'tomorrow 02:00'
          recreate: true

Environment Variables

If uploading to S3-compatible storage (e.g. DigitalOcean Spaces):

AWS_ACCESS_KEY_ID=your-key
AWS_SECRET_ACCESS_KEY=your-secret
AWS_REGION=fra1
AWS_BUCKET_NAME=my-bucket-name
AWS_PUBLIC_BUCKET_PREFIX=public/assets
AWS_ENDPOINT=https://fra1.digitaloceanspaces.com
AWS_PUBLIC_CDN_PREFIX=https://cdn.example.com/public/

If AWS_BUCKET_NAME is not set, the sitemap files will only be saved locally to public/sitemap/{locale}.

How Domain and URL Generation Works

• All links and asset references in the generated XML (including <loc> and <?xml-stylesheet?>) are created using SitemapBase::AbsoluteLink().
• If AWS_PUBLIC_CDN_PREFIX is set, it is used as the base for all URLs. Otherwise, the app’s Director::absoluteBaseURL() is used.
• For correct domain generation in worker contexts (e.g., localhost in cloud CI runners or App Platform), you must explicitly pass the correct domain to the job constructor (e.g., https://en.example.com).
• This domain is injected during generation using SitemapHelper::withAlternateBaseURL().

XSL Stylesheet Support

Generated sitemap XML files include a reference to an XSL file so they are human-readable in the browser. These files:

• styleSheet.xsl – used by individual sitemap chunks
• styleSheetIndex.xsl – used by the main sitemap index

Both files are rendered dynamically during the job’s first step and uploaded alongside the XML files. URLs are made relative and CDN-compatible to avoid cross-origin issues.

Proxying Sitemap URLs for Google Search Console

Google Search Console (GSC) only accepts sitemap URLs that are served from the same domain as the site being verified.

If your sitemap files are hosted on a CDN like DigitalOcean Spaces, this module includes a SitemapProxyController that allows you to serve those files via your app domain.

Example

If a file is uploaded to:

https://cdn.example.com/public/sitemap/en/sitemap.xml

And your site domain is:

https://en.example.com

Then you should submit this URL to Google Search Console:

https://en.example.com/sitemap/en/sitemap.xml

This request is handled by SitemapProxyController and will issue a 301 redirect to the CDN-hosted file:

https://cdn.example.com/public/sitemap/en/sitemap.xml

Running Large Jobs

This job is registered as a LARGE job type and must be processed using the large queue. To run it manually or via cron:

vendor/bin/sake dev/tasks/ProcessJobQueueTask queue=large

Example cron job (every 15 minutes):

*/15 * * * * /path/to/vendor/bin/sake dev/tasks/ProcessJobQueueTask queue=large

Access Control

By default, the GenerateSitemapJob runs as an anonymous (non-authenticated) user:

public function getRunAsMemberID()
{
    return 0;
}

This ensures that the sitemap only includes publicly accessible pages, avoiding pages that require login or specific member roles. If your application requires indexing private pages for a secured crawler or internal search engine, you may override this behavior by extending the job and modifying getRunAsMemberID() to return a valid member ID.

Testing Locally

To queue and run a job manually in code:

use Symbiote\QueuedJobs\Services\QueuedJobService;
use Hudhaifas\GoogleSitemapsQueued\Job\GenerateSitemapJob;

singleton(QueuedJobService::class)->queueJob(
    new GenerateSitemapJob('en', 'https://en.example.com')
);

To process it:

vendor/bin/sake dev/tasks/ProcessJobQueueTask queue=large

License

MIT