I think, currently this app has a function refresh_data - https://github.com/OpenDataServices/org-ids/blob/live/prefix_finder/frontend/views.py#L178
This gets the latest data from the source.
Therefore, it should be called regularly to make sure we get the latest data, and in a sensible way.
Currently it is called from a line in views.py - https://github.com/OpenDataServices/org-ids/blob/live/prefix_finder/frontend/views.py#L178
This means when a new server process for this app is created, it's called then. Actually no - when a new server process is created the first time it gets a request it loads the views file THEN gets the latest data (tested this in UWSGI logs).
Shouldn't this be run from a command, with a cron job updating regularly?
Note this function is also called by calls to URLS ^_preview_branch/([A-Za-z0-9-]+ - some way of allowing people to preview different branches should be provided - maybe remove the URL and add a parameter to the command?
I think, currently this app has a function refresh_data - https://github.com/OpenDataServices/org-ids/blob/live/prefix_finder/frontend/views.py#L178
This gets the latest data from the source.
Therefore, it should be called regularly to make sure we get the latest data, and in a sensible way.
Currently it is called from a line in views.py - https://github.com/OpenDataServices/org-ids/blob/live/prefix_finder/frontend/views.py#L178
This means when a new server process for this app is created, it's called then. Actually no - when a new server process is created the first time it gets a request it loads the views file THEN gets the latest data (tested this in UWSGI logs).
Shouldn't this be run from a command, with a cron job updating regularly?
Note this function is also called by calls to URLS
^_preview_branch/([A-Za-z0-9-]+- some way of allowing people to preview different branches should be provided - maybe remove the URL and add a parameter to the command?