The simplest way to do this would be to set up a different cache invalidation policy for the F/E cache on Helioviewer.org.
Currently we pre-cache all feature/event data to prevent the several second load time that would otherwise result when attempting to display features and events from the HEK. The script that builds the cache is run every 10 minutes to pick up new entries. Because some data may not appear in the HEK until several hours, days, or weeks after the fact, the caching script refreshes recent cache entries with decreasing frequency according to the age of the cache file. (Once cache file per day of data, plus locally generated auxiliary data for region polygon PNGs).
This mechanism could be used to retrieve new revisions of data periodically without any extra logic.
Note: rebuilding the entire F/E cache takes of order a day.
The simplest way to do this would be to set up a different cache invalidation policy for the F/E cache on Helioviewer.org.
Currently we pre-cache all feature/event data to prevent the several second load time that would otherwise result when attempting to display features and events from the HEK. The script that builds the cache is run every 10 minutes to pick up new entries. Because some data may not appear in the HEK until several hours, days, or weeks after the fact, the caching script refreshes recent cache entries with decreasing frequency according to the age of the cache file. (Once cache file per day of data, plus locally generated auxiliary data for region polygon PNGs).
This mechanism could be used to retrieve new revisions of data periodically without any extra logic.
Note: rebuilding the entire F/E cache takes of order a day.