This might seem a bit stupid, but I got bitten by this today - I was under the
impression the command is required to make the changes visible in the running
app (despite the section above saying it isn't). Hopefully this helps any
potential contributor to get their mind sorted better than mine.
Can't cache the event as the data is changed the scheduling js.
We also can't verify the token in the EventSchedules controller
as we are making susequent ajax calls and have no means to refresh
the token for the second request.
Make draggables scrollable.
There also is no need to spam the console with debug code...
Partly fixes#2353
The postgres image does not come with pre-configured osem
databases per environment. Calling db:create doesn't hurt if the
databases already exist, so let's do it every time we bootstrap.
From version 3.0.0 it is needed to specfied which test framework you're
using and which portion of the matchers you want to use in
`rails_helper.rb` (or the old `spec_helper.rb` in our case).
Also do not set memore_store in the production environment if
there is no memcached. File based cache is default for a reason,
it even works for most normal traffic pages
Requiring a ticket after registration is convoluted; while it may be neccesary
for some workflows, it definitely isn't for others, and the core of osem
doesn't need it... so let's make it optional.
We are using apache as reverse proxy and two systemd units
(puma & delayed job) for events.opensuse.org. Track them
also in this project, might be useful for others.
The ChangeVisitIdTypeOfAhoyEventsToInteger from 364be5541 is not very well done.
You can not change the schema without changing possibly existing data.
Ahoy events used some kind of string as `visit_id`. Changing this column
to integrer will just barf up `Mysql2::Error: Truncated incorrect INTEGER value`.
So we need a way for people who are in this situation now to get out of it.
Luckily we completely dropped ahoy events a while ago so we can just delete them.
* Clean up legacy charting imports
* Update existing donut and line charts
* Create helpers for parsing out existing chart data
* Move chart partials to a more common path
