AWS recommends running buckets without ACLs, and to use resource policies to manage access control instead.
This is not a bad idea, because S3 ACLs are whack, and while resource policies are also whack, they're a more constrained form of whack.
Further, some compliance regimes get antsy if you don't go with the vendor's recommended settings, and arguing that you need to enable ACLs on a bucket just to store images in there is more hassle than it's worth.
The new site setting (s3_use_acls) cannot be disabled when secure
uploads is enabled -- the latter relies on private ACLs for security
at this point in time. We may want to reexamine this in future.
Why does this change do?
This commit updates the educate message displayed when there are no new
topics on the `/new` route when the experimental new new view site setting is enabled.
The commit also fixes a couple of bugs:
1. Correct default auto track minutes used in the copy for unread
topics from the 4 minutes to 5 minutes.
2. Correct link to user's preference in copy to go to tracking tab
instead of notifications tab.
Failure message on CI
```
Failures:
1) Network Disconnected Doesn't show the offline indicator when the site setting isn't present
Failure/Error: expect(page).to have_css("html.message-bus-offline")
expected to find css "html.message-bus-offline" but there were no matches
[Screenshot Image]: /__w/discourse/discourse/tmp/capybara/failures_r_spec_example_groups_network_disconnected_doesn_t_show_the_offline_indicator_when_the_site_setting_isn_t_present_764.png
~~~~~~~ JS LOGS ~~~~~~~
http://localhost:31338/uploads/default/test_0/optimized/1X/_129430568242d1b7f853bb13ebea28b3f6af4e7_2_512x512.png - Failed to load resource: net::ERR_INTERNET_DISCONNECTED
http://localhost:31338/categories - Error while trying to use the following icon from the Manifest: http://localhost:31338/uploads/default/test_0/optimized/1X/_129430568242d1b7f853bb13ebea28b3f6af4e7_2_512x512.png (Download error or resource isn't a valid image)
~~~~~ END JS LOGS ~~~~~
# ./spec/system/network_disconnected_spec.rb:35:in `block (3 levels) in <main>'
# ./spec/system/network_disconnected_spec.rb:8:in `with_network_disconnected'
# ./spec/system/network_disconnected_spec.rb:34:in `block (2 levels) in <main>'
# ./spec/rails_helper.rb:380:in `block (3 levels) in <top (required)>'
# ./spec/rails_helper.rb:380:in `block (2 levels) in <top (required)>'
# ./spec/rails_helper.rb:372:in `block (3 levels) in <top (required)>'
# ./vendor/bundle/ruby/3.2.0/gems/timeout-0.3.2/lib/timeout.rb:189:in `block in timeout'
# ./vendor/bundle/ruby/3.2.0/gems/timeout-0.3.2/lib/timeout.rb:196:in `timeout'
# ./spec/rails_helper.rb:367:in `block (2 levels) in <top (required)>'
# ./spec/rails_helper.rb:356:in `block (2 levels) in <top (required)>'
# ./vendor/bundle/ruby/3.2.0/gems/webmock-3.18.1/lib/webmock/rspec.rb:37:in `block (2 levels) in <top (required)>'
```
Not all revisions involve changes to the actual post/topic content. We
may want to know if a revisions includes the topic title or post raw.
Specifically introducing these for use in the Akismet plugin to
conditionally queue checks.
What is the problem?
The user messages routes are currently routed by the server to
`UserActionsController#private_messages`. However, the method is
essentially a no-op and does not do any preloading. As a result, when we
load the user private messages routes, the client ends up having to
issue another request to the server to get more information about the
user profile currently being viewed. This extra request is triggered by
the `user` model's `findDetails` method that is called from the `user`
route in the `afterModel` hook.
What is the solution?
The `user` model's `findDetails` method actually checks the preload
store to see if the `user_${username}` key is present in the store and
if it is, it will use the preloaded data instead of triggering another
request. Since the user private messages routes are nested under the
user route on the client side, we have to rely on the
`UsersController#show` controller action on the server side for the user private
messages route as the `UsersController#show` controller action preloads
the required user information for the client side.
Before, the review button was shown in `primary section` when there were items to review. Otherwise, it was hidden in `more section`.
Because we are allowing admins to customize community section and reorder link, it makes sense to simplify that logic and review link should follow admin's decision.
What is the problem?
When opening the composer, we are seeing multiple requests made to
the `/composer_messages` endpoint. This is due to our use of the
`transitionend` event on the `#reply-control` element. The event is
fired once for each transition event and the `#reply-control` element
has multiple transition events.
What is the solution?
Since are only interested in the `height` transition event, we add a
condition to check that the callback function is only triggered when the
`propertyName` of the `transitionend` event is `height`.
Why is there no tests for this change?
In QUnit, we have `transition: none !important` set in the stylesheet
with no easy way to disable. We'll have to accept the risk of not
writing test for this performance fix.
What does this change do?
Suggested topics by default are ordered in the following way:
1. Unread topics in current category of topic that is being viewed
2. Unread topics in other categories
3. New topics in current category of topics that is being viewed
4. New topics in other categories
5. Random topics
With the experimental new new view, we want to remove the concept of
read and new so that new order is as such:
1. Topics created by the current user with posts that the user has not
read ordered by topic's bumped date
2. Topics in current category of topic with posts that the user has not
read ordered by topic's bumped date
3. Topics in other categories with posts that the user has not read
ordered by topic's bumped date
4. Random topics ordered by topic's bumped date
Using the runtime information, we will be able to more efficiently group
the test files across the test processes hence leading to better
utilization of resources.
Meta topic: https://meta.discourse.org/t/mention-suggestion-list-box-in-the-rtl-website-in-wrong-place/266763?u=osama.
Our autocomplete box doesn't currently take into account the user's locale and places itself off-screen when using an RTL locale. This commit changes the placement logic for the autocomplete box when an RTL locale is used to make sure that:
1. the autocomplete box's right side is near and to the left of the caret
2. the autocomplete box doesn't go beyond the composer's left side.
When a topic already has multiple synonym tags of a target tag, if we try to update the "`tag_id`" column to target tag id then it will raise a unique violation error since there are multiple synonyms present in the topic. So before doing that action, we must delete the problematic tags so the topic has only one synonym tag to update.
This is not an issue when the topic has a target tag already along with synonyms.
`DiscourseIpInfo` expects zeitwerk auto-loading to be available, so we need to ensure the rake task loads the full rails environment. Normally we run this task as part of assets:precompile, so the app is already initialized. This commit only affects the case where the maxmind task is run directly.
Few weeks ago we implemented `onPresenceChangeCallback` to re-sync chat channels state when going back to a long time inactive tab. This codepath however contained a bug as we were reseting all subscriptions but only restarting global subscriptions and not per channel subscriptions.
This commit should correctly ensure we correctly do so. It's sadly very hard to test time related changes in system specs.
Co-authored-by: Martin Brennan <mjrbrennan@gmail.com>
What is the problem?
We were calling out to methods that calls `has_css?` or `has_selector?`
which returns a boolean. Since we are not using the return value, it
means the methods can be deemed unnecessary. However, we do want those
checks and this commit adds the necessarily assertions to make use of
the return values.
When a user type a message with mentions, the autocomplete popup
may suggest users or groups. We were adding all these object to
the `currentMessage.mentionedUsers` collection, while we should
have been adding only users. A group added to that collection led to
the error later when trying to update user status on mentions.
Test were sometimes failing with similar error to the following:
```
1) UsernameChanger#override when unicode_usernames is off overrides the username if a new name has different case
Failure/Error:
protect { v8.eval(<<~JS) }
__paths = #{paths_json};
__utils.avatarImg({size: #{size.inspect}, avatarTemplate: #{avatar_template.inspect}}, __getURL);
JS
MiniRacer::RuntimeError:
ReferenceError: __optInput is not defined
# JavaScript at exports.helperContext (<anonymous>:21:17)
# JavaScript at getRawAvatarSize (<anonymous>:108:49)
# JavaScript at avatarUrl (<anonymous>:102:21)
# JavaScript at Object.avatarImg (<anonymous>:129:15)
# JavaScript at <anonymous>:2:9
# ./lib/pretty_text.rb:259:in `block in avatar_img'
# ./lib/pretty_text.rb:661:in `block in protect'
# ./lib/pretty_text.rb:661:in `synchronize'
# ./lib/pretty_text.rb:661:in `protect'
# ./lib/pretty_text.rb:259:in `avatar_img'
# ./app/jobs/regular/update_username.rb:14:in `execute'
```
This should not be needed as it should already have been initialised but that should stop the flakey-ness for now while being a safe change.
This will make it simpler to work with this code. This also can make this code more stable and increase stability of our test suite.
Cooked message now will be available immediately after cooking, it wasn't the case before:
await message.cook();
const cooked = message.cooked;
This also removes a call to `message.cook()` from message fabricator. Alternatively we may leave the call there and make the fabricator function async, but I fill it's better this way. If someone needs to test something related to cooked message, they can either pass cooked text to fabricator:
message = fabricators.message({ cooked: "<p>cooked</p>" });
or call `message.cook()` after fabrication:
message = fabricators.message({ message: "raw message" });
await message.cook()
It seems like the overhead of GPU acceleration is not worth it and is
slowing down our system tests. Locally the following command completes
in `2 minutes 8.6 seconds` with GPU disabled as compared to `2 minutes 45.4 seconds` with GPU enabled.
