SVMPForge Product Feed for WooCommerce

FAQ

Does this plugin require WooCommerce?

Yes. WooCommerce must be installed and active. The plugin will show a notice and stay inactive until WooCommerce is available.

Which output formats are supported?

XML (RSS 2.0) and CSV — the two formats Google Merchant Center and Meta most commonly ingest. Both are fully available.

How many feeds can I create?

As many as you like. This plugin applies no feed limit, no trial period and no licence check of any kind.

Can I schedule automatic feed updates?

Yes. Each feed can be set to regenerate hourly, twice daily, daily or weekly at a time you pick. Generation runs in the background using Action Scheduler (bundled with WooCommerce).

Are product variations supported?

Yes. Variable products can be exported as individual variations, each with the correct item group id so Google groups them together.

Where are feed files stored?

Generated feed files are written to a protected directory inside your uploads folder and served through a clean feed URL such as https://your-store.com/apfw-feed/{unique-token}.xml. The token is a random identifier generated per feed, so feed URLs cannot be guessed. On sites using Plain permalinks the feed is served through an equivalent query-string URL instead.

Why do some CSV values start with an apostrophe?

Values beginning with =, +, - or @ are escaped with a leading apostrophe in CSV feeds, the same way WooCommerce’s own CSV exporter does, so they cannot execute as formulas when a feed is opened in a spreadsheet. Plain numeric values are never escaped. Developers can disable this via the apfw_csv_escape_formulas filter.

Which currency and language is the feed labelled with?

Products are exported exactly as WooCommerce stores them, so every feed is labelled with your store currency and your site language. Developers can relabel a feed with the apfw_feed_currency and apfw_feed_language filters — for example from a multi-currency plugin that has actually converted the prices.

Does this plugin phone home?

No. This plugin does not contact any external server. Licensing and automatic updates apply only to a separately sold add-on hosted outside WordPress.org.

Where does the Google product category list come from?

The plugin bundles Google’s official product taxonomy file (version 2021-09-21), published by Google at https://www.google.com/basepages/producttype/taxonomy-with-ids.en-US.txt for merchants and feed tools to use. It is redistributed unmodified as reference data so the category mapping screen can search and validate categories offline, without any external request. See data/README.txt in the plugin folder for full provenance. Developers can supply a different or localized list with the apfw_google_taxonomy filter.

How are very large catalogs handled?

Every generation — scheduled, manual or started from the editor — is one resumable run made of short steps. Each step processes products for a bounded number of seconds (it checks the clock and the memory limit after every product), appends them to a temporary file and saves its position; the next step continues from the last product written. Steps are chained in the background through Action Scheduler (bundled with WooCommerce) or, without it, through WP-Cron, so no single request can hit a PHP or web-server timeout and a host with a hard max_execution_time of 60 seconds still completes a 500 MB feed. Products are fetched with keyset pagination (ID > last id), so the millionth product costs the same query as the first.

If a step is killed by a fatal error, the run is not lost: the next step resumes from the saved position with the temporary file trimmed to its last good byte, so no product is ever duplicated. After three failures at the same position the run stops and the log names the product that caused it. When you start a generation from the editor, the browser drives the steps so you see live progress; closing the tab hands the run to the background and the feeds list keeps showing its progress. The finished file is swapped into place atomically, so the previously published feed is never served half-written, and feeds of 1 MB or more are also stored pre-compressed for clients that accept gzip. Feed URLs support HEAD and byte-range requests, and very large files can be handed to the web server (X-Sendfile / X-Accel-Redirect) from Settings.