Revert "Revert "Katie/practice cookies"" #45652
Conversation
KatieShipley
force-pushed
the
revert-45604-revert-44979-katie/practice-cookies
branch
from
0ad6140
to
2141f28
Compare
|
|
||
| %link{ :href => "/css/cookies.css", :rel => "stylesheet" } | ||
|
|
||
| = render inline: File.read(deploy_dir('shared/partials/cookie_notice.partial')), type: :partial |
There was a problem hiding this comment.
I'm curious why you use deploy_dir here but Rails.root.join in other places?
There was a problem hiding this comment.
deploy_dir finds the root path in pegasus, Rails.root finds the root path in studio. This was a weird gotcha I ran into, Rails.root works fine locally for both pegasus and dashboard because of how we run the dashboard-server locally, but in all other environments pegasus does not have access to the Rails.root.
There was a problem hiding this comment.
Does deploy_dir work for both Pegasus and Dashboard? IMO, deploy_dir('shared/haml/onetrust_cookie_scripts.haml') is easier to read than
Rails.root.join('..', 'shared', 'haml', 'onetrust_cookie_scripts.haml').
This is just an observation, not a suggestion to change.
| @@ -0,0 +1,10 @@ | |||
| -# OneTrust Cookies Consent Notice scripts for code.org | |||
| - if rack_env?(:production) | |||
| %script{src: 'https://cdn.cookielaw.org/consent/27cca70a-7db3-4852-9ef0-a6660fd0977d/OtAutoBlock.js', type: 'text/javascript'} | |||
There was a problem hiding this comment.
This link seems to point to a very specific version of the js file. How do we know when to use a newer version?
There was a problem hiding this comment.
And the js version in shared/haml/hoc_onetrust_cookie_scripts.haml is different from the one used here.
There was a problem hiding this comment.
The hash there isn't locked to a version, it just identifies our script for the *.code.org domains. When we publish a new version of the script via OneTrust, that same .js url will be updated to pull the newest version without a code change. If for some reason we do make a change on the OneTrust side that changes the link (like what happened in this commit), OneTrust notifies us that the script link has changed so we know to make the code change.
It is different from hoc_onetrust_cookie_scripts because *.code.org and hourofcode.com are separate domains, so they unfortunately need separate cookie management. The behavior configuration for both on the OneTrust side is identical though.
There was a problem hiding this comment.
Got it, thanks for the explanation.
KatieShipley
deleted the
revert-45604-revert-44979-katie/practice-cookies
branch
Reverts #45604
Let's try #44979 again...