This spec helper was introduced as a temporary solution to the problem
of mismatched types between primary key and foreign key columns. All
plugins have been migrated and the only remaining use of this helper is
in core Discourse.
This patch adds a new step to services named `try`.
It’s useful to rescue exceptions that some steps could raise. That way,
if an exception is caught, the service will stop its execution and can
be inspected like with any other steps.
Just wrap the steps that can raise with a `try` block:
```ruby
try do
step :step_that_can_raise
step :another_step_that_can_raise
end
```
By default, `try` will catch any exception inheriting from
`StandardError`, but we can specify what exceptions to catch:
```ruby
try(ArgumentError, RuntimeError) do
step :will_raise
end
```
An outcome matcher has been added: `on_exceptions`. By default it will
be executed for any exception caught by the `try` step.
Here also, we can specify what exceptions to catch:
```ruby
on_exceptions(ArgumentError, RuntimeError) do |exception|
…
end
```
Finally, an RSpec matcher has been added:
```ruby
it { is_expected.to fail_with_exception }
# or
it { is_expected.to fail_with_exception(ArgumentError) }
```
As part of a previous fix we changed which groups are serialized for a user, in order to fix a bug in the default group selector under user preferences.
However, we should only change this when serializing the current user. This change combines the old code-path and the new based on who is serializing.
* DEV: Gracefully handle remaps which violate DB column constraints
This change implements length constraint enforcement to skip remaps
which exceed column max lengths
* DEV: Only perform skipped column stats lookup when verbose is true
* DEV: Tidy up specs
* DEV: Make skipping violating remap behaviour opt-in
This change introduces a new `skip_max_length_violations` param for
`remap`, set to `false` by default to ensure we still continue to fail
hard when max lenth constraints are violated.
To aid in quick resolution when remaps fail, this change also
adds more context to the exception message to include the offending table
and column information
* Apply suggestions from code review
Co-authored-by: Gerhard Schlager <gerhard.schlager@discourse.org>
* FIX: Various fixes
- Linter errors
- Remap status "logger" early return condition
---------
Co-authored-by: Gerhard Schlager <gerhard.schlager@discourse.org>
Firstly, we need to understand that ActiveRecord can be
connected to a role which prevent writes and this happens in Discourse when a
replica database has been setup for failover purposes. When a role
prevent writes from happening, ActiveRecord will raise the
`ActiveRecord::ReadOnlyError` if a write query is attempted.
Secondly, theme fields are baked at runtime within GET requests. The
baking process involves writing the baked value to the
`ThemeField#baked_value` column in the database.
If we combine the two points above, we can see how the writing of the
baked value to the database will trigger a `ActiveRecord::ReadOnlyError`
in a GET requests when the database is connected to a role preventing
writes. However, failing to bake a theme is not the end of the world and
should not cause GET requests to fail. Therefore, this commit adds a rescue
for `ActiveRecord::ReadOnlyError` in the `ThemeField#ensure_baked!`
method.
Followup 0568d36133
Followup 97cf069a06
Due to the S3 dualstack endpoint change, sites with
S3 backups configured but _not_ S3 uploads were erroring,
with admins unable to access the backups page. This
commit fixes the error by not enabling S3 dualstack
endpoints if S3 uploads have not been enabled, backups
don't need to use them.
c.f. https://meta.discourse.org/t/unable-to-backup-or-navigate-to-backups/335899
When using chat in drawer mode, after you've clicked on a chat bookmark in the user menu, clicking any other chat bookmark would "do nothing".
In 8b18fd1556 we added an optimization to prevent the same route from being reloaded, but it ended up breaking the bookmarks.
This commit reverts the changed made the above commit and adds a system specs that ensure we can click two chat bookmarks in the user menu when using chat in drawer mode.
Internal ref - t/134362
* FEATURE: Add skip notification option to group invite to topic
* DEV: rename `skip_notification` to `should_notify`
* DEV: update `should_notify` param to be default `true` in controllers
* DEV: update spec to use `greater than` instead of `equal to` to prevent flakiness
* Update app/controllers/topics_controller.rb
Co-authored-by: David Taylor <david@taylorhq.com>
* DEV: merged two `#invite_group` specs into one
* DEV: Added test case for `invite-group` in requests spec
---------
Co-authored-by: David Taylor <david@taylorhq.com>
When a parent category shows topics from subcategories, dismissing
should dismiss posts in both parent and subcategories.
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
- Uses a more appropriate image, with immutable tag (so update prompts work correctly)
- Updates port forwarding
- Improves mount setup (inc. persistant PG/Redis when rebuilding)
- Fixes ember-cli live reload
- Automatically configures VSCode & extensions
Redesign the permalinks page to follow the UX guide. In addition, the ability to edit permalinks was added.
This change includes:
- move to RestModel
- added Validations
- update endpoint and clear old values after the update
- system specs and improvements for unit tests
Follow up to 6f8f6a7726
Prior to the linked commit, the `uploads#create` endpoint had a `upload_type` and `type` param that acted as aliases for each other and raised an error if both of them were missing. In the linked commit, we removed the `type` param and always required the `upload_type` param which break API consumers that only included `type` in their requests.
This commit adds back the `type` param temporarily and introduces a deprecation message for it so that API consumers are made aware of the eventual removal of the `type` param.
These URLs allow the state of a headless browser to be viewed and debugged using any other browser, without needing to restart the test with `SELENIUM_HEADLESS=0`.
It splits the hide_profile_and_presence user option and the default_hide_profile_and_presence site setting for more granular control. It keeps the option to hide the profile under /u/username/preferences/interface and adds the presence toggle in the quick user menu.
Co-authored-by: Régis Hanol <regis@hanol.fr>
This is a very simple change, which creates a permanent table in the DB, rather than generating a temporary table when moving posts. This change is about capturing data and any usage will appear in a follow-up.
I did include a new column created_new_topic in the new table, so that it can be easily audited without having to compare destination topic created_at with moved_post records.
This commit modernizes the post menu by migrating it from the existing widget-based implementation to Glimmer components. This transition aims to improve the maintainability, performance, and overall developer experience.
It also introduces a new DAG-based transformer API for customizations that aims to be more flexible than the widget base one.
---------
Co-authored-by: David Taylor <david@taylorhq.com>
This commit removes the new optimized category style introduced in
previous commits (d37a0d40, 9a80d718 and 430c42ac), in favour of the
existent `categories_only`.
When performing bulk dismissal in Unread and New views, the dismiss button stays at the top of the UI. Because of this we want to provide the dismiss action also in the "sticky" menu that's always in view, even when scrolling a long list of topics.
This reverts commit 5a00a041f1.
Implementation is currently not correct. Multiple uploads can share the
same etag but have different paths in the S3 bucket.
Follow-up to a5497b74be
In the linked commit, as part of simplifying the invite modal, we removed the option to skip sending an email when creating an invite restricted to a specific address. This has caused confusion about whether an email will be sent by Discourse or not, so we're adding back the option to create a restricted invite without emailing.
Internal topic: t/134023/48.
Followup 0568d36133
S3 itself and other S3-compatible providers do not
allow using an S3 custom endpoint and dualstack at
the same time, so this commit fixes that by not using
dualstack when the endpoint is present.
A "bad upload" in this context is a upload with a mismatched URL. This can happen when changing the S3 bucket used for uploads and the upload records in the database have not been remapped correctly.
"Resume editing" would do nothing when going through the `/new-message` flow.
This seems to be broken since [this commit](b0f6d074be). which moved `this._setModel` calls around – the same we're doing now, but to different places: the first one needs to happen after the `draft.data` has been set , while the second needs to happen before the `this.open` call.
When we added direct S3 uploads to Discourse, which use
presigned URLs, we never took into account the dualstack
endpoints for IPv6 on S3.
This commit fixes the issue by using the dualstack endpoints
for presigned URLs and requests, which are used in the
get-presigned-put and batch-presign-urls endpoints used when
directly uploading to S3.
It also makes regular S3 requests for `put` and so on use
dualstack URLs. It doesn't seem like there is a downside to
doing this, but a bunch of specs needed to be updated to reflect this.
This PR adds a small visual change to the new feature item on the `/admin/whats-new` page. When features are marked with an experimental site setting, they should show an indication on the feature item that it is "Experimental"
Currently, if an association is added as a tracked field in
`PostRevisor`, the `PostRevisionSerializer` class will try to serialize
it somehow. This will raise an error as ActiveRecord collection proxies
can't be serialized.
This patch addresses this issue by skipping any association tracked by
the `PostRevisor` class.
Prior to Uppy, the `uploads#create` endpoint used to receive a `type` param that indicated the purpose/target of the upload, such as `avatar`, `site_setting` and so on. With the introduction of Uppy, the `type` param became the MIME type of the file being uploaded, and the purpose/target of the upload became a new param called `upload_type`, however the backend could still use the `type` param (which now contains MIME type) as the purpose/target of the upload if `upload_type` is absent.
We technically don't need to send the MIME type over the network, but it seems like it's done by Uppy and we have no control over the `type` param that Uppy includes:
758de8167b/app/assets/javascripts/discourse/app/lib/uppy/uppy-upload.js (L146-L151)
This commit does a couple of things:
1. It amends the `uploads#create` endpoint so it always requires the `upload_type` param and doesn't fallback to `type` if `upload_type` is absent
2. It forces consumers of the `UppyUpload` class (and by extension `UppyImageUploader`) to specify `type` of the upload
Internal topic: t/140945.
The normalize_emails setting makes it so that only canonical e-mails are considered for validation purposes. This means disallowing "plus addressing". For example, with this enabled, bob@discourse.org and bob+foo@discourse.org are considered the same address, and you can only sign up with one of them.
Currently this is disabled by default, leading to a lot of spam sign-ups. It's healthier to consider this an opt-out setting.
This commit adds an API `upload_image` to `FormKitField` page object for setting an image file on an `Image` field in FormKit. Usage is like this:
```ruby
form.field("image_field").upload_image(image_path)
```
The `value` API also now supports `Image` fields; it returns an `Upload` record if the field has an uploaded image.
Recently we added a new feature for automatically gridding images in the composer (https://github.com/discourse/discourse/pull/29260). After testing this feature under a setting for a short period of time, the feature is no longer experimental anymore.
This PR removes the site setting `experimental_auto_grid_images`.
The primary key is usually a bigint column, but the foreign key columns
are usually of integer type. This can lead to issues when joining these
columns due to mismatched types and different value ranges.
This was using a temporary plugin / test API to make tests pass. After
more careful consideration, we concluded that it is safe to alter the
tables directly.
Docking is a leftover from older header code, it looks like it is no
longer used in the app. This helper was registering a scroll event
listener to check if the header should be docked or not. Initially, a
"docked" class was added to the body element. This class persisted
through the lifecycle of the app and the scroll event was doing no
useful work.
Some older themes may still use it in CSS, that will cause a regression,
from a quick look at existing code, the surface area should be small
(2-3 themes). It's worth removing the event listener for performance
reasons. We could possibly add the class "docked" statically to the body
element, but it's redundant. It's best to clean up the relevant CSS in
themes, where applicable.
This commit adds a new "Community title" field to the about config page. This field controls the `short_site_description` setting, which is shown in the browser tab for key pages such categories pages and topic lists.
Internal topic: t/140812.
This commit changes the uploads:secure_upload_analyse_and_update
and uploads:disable_secure_uploads to no longer rebake affected
posts inline. This just took way too long, and if the task stalled
you couldn't be sure if the rest of it completed.
Instead, we can update the baked_version of affected posts and
utilize our PeriodicalUpdates job to gradually rebake them. I added
warnings about increasing the site setting rebake_old_posts_count and
the global setting max_old_rebakes_per_15_minutes before doing this
as well.
For good measure, the affected post IDs are written to a JSON file too.
This commit contains two changes to how our site setting
keyword system works:
1. Crowdin, our translation provider, does not support YAML lists,
so we are changing site setting keywords in server.en.yml to
be pipe-separated (|)
2. It's unclear to translators what they are supposed to do with
aliases of site settings where the name has changed, e.g.
min_trust_level_for_here_mention. Instead of getting these as
keywords from the yml file, we can discern these from
SiteSettings::DeprecatedSettings automatically, and still use
them for client-side search
These changes should help improve the situation for translators.
Followup bd4e8422fe
In the previous commit, we introduced the `page_view_legacy_total_reqs`
report. However this was not tested properly, and due to a typo
the report returned no data.
This commit fixes the issue and adds a spec to catch this.
This commit adds a new "Invite" link to the sidebar for all users who can invite to the site. Clicking the link opens the invite modal without changing the current route the user is on. Admins can customize the new link or remove it entirely if they wish by editing the sidebar section.
Internal topic: t/129752.
This commit removes the feature flag for the new /about page, enabling it for all sites, and removes the code for old the /about page.
Internal topic: t/140413.
Followup 30fdd7738e
Adds a new site setting and corresponding user preference
to disable smart lists. By default they are enabled, because
this is a better experience for most users. A small number of
users would prefer to not have this enabled.
Smart lists automatically append new items to each
list started in the composer when enter is pressed. If
enter is pressed on an empty list item, it is cleared.
This setting will be removed when the new composer is complete.
This commit allows themes to define up to 2 screenshots
in about.json. These should be paths within the theme's
git repository, images with a 1MB max file size and max width 3840x2160.
These screenshots will be downloaded and stored against a theme
field, and we will use these in the redesigned theme grid UI.
These screenshots will be updated when the theme is updated
in the same way the additional theme files are.
For now this is gated behind a hidden `theme_download_screenshots`
site setting, to allow us to test this on a small number of sites without
making other sites make unnecessary uploads.
**Future considerations:**
* We may want to have a specialized naming system for screenshots. E.g. having light.png/dark.png/some_palette.png
* We may want to show more than one screenshot for the theme, maybe in a carousel or reacting to dark mode or color palette changes
* We may want to allow clicking on the theme screenshot to show a lightbox
* We may want to make an optimized thumbnail image for the theme grid
---------
Co-authored-by: Ted Johansson <ted@discourse.org>
Followup 9762e65758
When we added the Revise... option for posts/new topics
in the review queue, which sends a PM to the user, we used
`SystemMessage.create_from_system_user`, which always sends
the PM from the system user. However, this makes it so if the
user replies to the PM, which they are encouraged to do,
no one will see it unless they actively monitor the system inbox.
This commit changes it so `SystemMessage.create` is used,
which uses the `site_contact_username` and `site_contact_group`
site settings as participants in the sent PM. Then, when the
user replies, it will send to that inbox instead.
If `site_contact_username` is blank, the system user is used.
# Context
Add `disableDefaultKeyboardShortcuts` function to the plugin API to allow for disabling [default bindings](e4941278b2/app/assets/javascripts/discourse/app/lib/keyboard-shortcuts.js (L49)).
# Details
This function is used to disable a "default" keyboard shortcut. You can pass an array of shortcut bindings as strings to disable them.
**Please note that this function must be called from a pre-initializer.**
Example:
```js
api.disableDefaultKeyboardShortcuts(['command+f', 'shift+c']);
```
- Added system spec, displaying intended behavior
This patch replaces the parameters provided to a service through
`params` by the contract object.
That way, it allows better consistency when accessing input params. For
example, if you have a service without a contract, to access a
parameter, you need to use `params[:my_parameter]`. But with a contract,
you do this through `contract.my_parameter`. Now, with this patch,
you’ll be able to access it through `params.my_parameter` or
`params[:my_parameter]`.
Some methods have been added to the contract object to better mimic a
Hash. That way, when accessing/using `params`, you don’t have to think
too much about it:
- `params.my_key` is also accessible through `params[:my_key]`.
- `params.my_key = value` can also be done through `params[:my_key] =
value`.
- `#slice` and `#merge` are available.
- `#to_hash` has been implemented, so the contract object will be
automatically cast as a hash by Ruby depending on the context. For
example, with an AR model, you can do this: `user.update(**params)`.
Currently in services, we don’t make a distinction between input
parameters, options and dependencies.
This can lead to user input modifying the service behavior, whereas it
was not the developer intention.
This patch addresses the issue by changing how data is provided to
services:
- `params` is now used to hold all data coming from outside (typically
user input from a controller) and a contract will take its values from
`params`.
- `options` is a new key to provide options to a service. This typically
allows changing a service behavior at runtime. It is, of course,
totally optional.
- `dependencies` is actually anything else provided to the service (like
`guardian`) and available directly from the context object.
The `service_params` helper in controllers has been updated to reflect
those changes, so most of the existing services didn’t need specific
changes.
The options block has the same DSL as contracts, as it’s also based on
`ActiveModel`. There aren’t any validations, though. Here’s an example:
```ruby
options do
attribute :allow_changing_hidden, :boolean, default: false
end
```
And here’s an example of how to call a service with the new keys:
```ruby
MyService.call(params: { key1: value1, … }, options: { my_option: true }, guardian:, …)
```
Background
When creating webhooks on a site without the Discourse Category Experts plugin installed, the category_experts_unapproved_event and category_experts_approved_event webhook events are getting automatically added to webhooks without a way to disable them.
The category_experts_unapproved_event and category_experts_approved_event webhook events are associated with the Discourse Category Experts plugin so I am moving these webhook events into the Category Experts plugin.
Changes
This PR deletes Category Experts plugin specific webhook event types added into core.
The new style is called `categories_only_optimized` and it is designed
to show only the parent categories, without any subcategories. This
works best for communities with many categories (over a thousand).
Bug introduced in this PR https://github.com/discourse/discourse/pull/29244
When the experiment toggle button was introduced, new features did not look right when the toggle button was not available.
In addition, the plugin name can be an empty string. In that case, information about new features should be displayed.
…or a tip with the highest priority.
This regressed in 597ef11195 where we got rid of `next()` calls, so we'd render the first tip we encounter.
The commit also adds a test and updates existing ones.
Currently, when calling a service with its block form, a `#result`
method is automatically created on the caller object. Even if it never
clashed so far, this could happen.
This patch removes that method, and instead use a more classical way of
doing things: the result object is now provided as an argument to the
main block. This means if we need to access the result object in an
outcome block, it will be done like this from now on:
```ruby
MyService.call(params) do |result|
on_success do
# do something with the result object
do_something(result)
end
end
```
In the same vein, this patch introduces the ability to match keys from
the result object in the outcome blocks, like we already do with step
definitions in a service. For example:
```ruby
on_success do |model:, contract:|
do_something(model, contract)
end
```
Instead of
```ruby
on_success do
do_something(result.model, result.contract)
end
```
Database dumps sometimes reference functions in the `discourse_functions` schema. It's possible that some of these functions have been dropped in a newer version of Discourse. In that case, restoring an older backup will fail with a `ERROR: function discourse_functions.something_something() does not exist` error. The restore functionality contains a workaround for that problem, but it didn't work with functions created in plugin migrations.
This commit adds support for temporarily creating missing `discourse_functions` from plugins. And it adds a simple check if the DB migration file even contains the required `DROPPED_TABLES` or `DROPPED_COLUMNS` constant. We don't need to create an instance of the DB migration class unless one of those constants is used. This makes the restore slightly faster and works around a problem with migrations that execute without `up` or `down` methods (e.g. `BackfillChatChannelAndThreadLastMessageIdsPostMigrate`).
This PR is a follow-up to ea1473e532. When we initially added the experimental feature for automatically adding `[grid]` to images, we add the [grid] surrounding images after all the uploads have been completed.
This can lead to confusion when `[grid]` is delayed to be added in the composer, as users may try to add grid manually leading to breakage. This also leads to issues with Discourse AI's automatic image caption feature.
**In this PR**: we simply move the logic to be added when the images are uploaded and processing. This way, `[grid]` surrounding images is added immediately. We also apply a fix for an edge-case to prevent images from being wrapped in `[grid]` when they are already inside `[grid]` tags.
This commit brings back some reports hidden or changed
by the commit in 14b436923c if
the site setting `use_legacy_pageviews` is false.
* Unhide the old “Consolidated Pageviews” report and rename it
to “Legacy Consolidated Pageviews”
* Add a legacy_page_view_total_reqs report called “Legacy Pageviews”,
which calculates pageviews in the same way the old page_view_total_reqs
report did.
This will allow admins to better compare old and new pageview
stats which are based on browser detection if they have switched
over to _not_ use legacy pageviews.
Toggle the button to enable the experimental site setting from "What's new" announcement.
The toggle button is displayed when:
- site setting exists and is boolean;
- potentially required plugin is enabled.
This PR adds the feature where three or more image uploads in the composer will result in the images being surrounded by `[grid]` tags. This helps take advantage of the grid feature (https://github.com/discourse/discourse/pull/21513) and display images in a more appealing way immediately after upload.
* FIX: participating users statistics...
... was (mis-)counting
- bots
- anonymous users
- suspended users
There's now a "valid_users" function that holds the AR query for valid users and which is used in all "users", "active_users", and "participating_users" queries.
Internal ref - t/138435
This commit simplifies the initial state of the invite modal when it's opened to make it one click away from creating an invite link. The existing options/fields within the invite modal are still available, but are now hidden behind an advanced mode which can be enabled.
On the technical front, this PR also switches the invite modal to use our FormKit library.
Internal topic: t/134023.
When a user is missing required fields, they are required to fill those up before continuing to interact with the forum. This applies to admins as well.
We keep a whitelist of paths that can still be visited in this mode: FAQ, About, 2FA setup, and any admin route for admins.
We concluded that admins should still be able to enable safe mode even with missing required fields. Since plugins etc. can potentially mess with the ability to fill those up.
When staff only mode is enabled - Discourse.enable_readonly_mode(Discourse::STAFF_WRITES_ONLY_MODE_KEY)
Staff members couldn't reset their password via the "forgot password" link.
This fixes it.
Internal ref. t/133990
We're seeing errors in logs due to some sites setting the reserved_usernames setting to nil. This is causing multiple use cases upstream of User#reserved_username? to error out.
This commit changes from using the raw #reserved_usernames to using the #reserved_usernames_map helper which exists on list-type site settings. It returns an empty array if the raw value is nil or empty string.
When adding or updating a custom user field to apply to all users (retroactively) we want to alert the admin that this will force all existing users to fill up the field before they are able to access the forum again.
However, we currently show this prompt when making changes only to other attributes on the custom field, i.e. the requirement hasn't changed.
This commit fixes that.
In #29169 we added a NULLS NOT DISTINCT option to the unique index on problem_check_trackers. This is to enforce uniqueness even when the target is NULL. (Postgres considers all NULLs to be distinct by default.)
However, this only works in PG15. In PG13 it does nothing.
This commit adds a default dummy string value __NULL__ to target. Since it's a string, PG13 will be able to correctly identify duplicate records.
We're expecting the period param to be something that neatly coerces into a symbol. If we receive something like a nested parameter, this will blow up.
This commit raises an InvalidParameters exception in the case of a non-stringy period parameter.
Creating or updating flags generates global side effects. Sometimes it
seems the state can leak from the flag specs.
This is probably related to the use of `fab!`. This patch replaces those
calls with standard `let`s. While the overall performances of these
tests will be a little less good, their state should not leak anymore.
Since we recently blocked accidental serialization of AR models, we are getting a 500 error in some cases with thumbnails. We can fix this by serializing the thumbnail, previously we just returned a raw OptimizedImage object.
Thumbnails are now attached to the serializer in core, therefore we no longer need to use add_to_serializer within the chat plugin to use thumbnails within chat message uploads.
We're expecting the ID param to be something that neatly coerces into an ID. If we receive something like a nested parameter, this will blow up. (We already handle the case of arrays.)
This commit raises an InvalidParameters exception in the case of a nested ID.
We're expecting the page param to be something that neatly coerces into an integer. If we receive something like a nested parameter, this will blow up. (I'm sure there are other examples as well.)
This commit falls back to a page value of 1 if the coercion fails.
We want to allow lightboxing of smaller images, even if they are below the minimum size for image thumbnail generation.
This change sets a minimum threshold of 100 x 100 pixels for triggering the lightbox.
---------
Co-authored-by: Régis Hanol <regis@hanol.fr>
This commit is fixing the path which sets a default value to trigger. We were doing `if (!this.model.trigger)` but `this.model.trigger` can have `0` as value, which would trigger this codepath and this codepath was setting the first value of `badgeTriggers` as a default value for trigger.
This patch improves the custom `array` type available in contracts.
It’s now able to split strings on `|` on top of `,`, and to be more
consistent, it also tries to cast the resulting items to integers.
Theme modifiers can now be defined as theme settings, this allows for
site operators to override behavior of theme modifiers.
New syntax is:
```
{
...
"modifiers": {
"modifier_name": {
"type": "setting",
"value": "setting_name"
}
}
}
```
This also introduces a new theme modifier for serialize_post_user_badges. Name of badge must match the name of the badge in the badges table. The client-side is updated to load this new data from the post-stream serializer.
Co-authored-by: David Taylor <david@taylorhq.com>
Technically we don't show the edit custom section button on mobile, but the button is present so I just fixed it so the finder works on mobile. We should probably remove this test or find a way to make the button visible on mobile.
Also used `mobile: true` instead of manual url.
Adding the directory item test causes the default test to fail randomly due to directory items not getting removed properly.
Removing this for now, and also moving this test to the common system folder instead of system/user_page
#29209 introduced a bug where columns to the directory added via add_directory_column are not being translated properly.
This fixes the issue and adds an integration test.
- Uses a temporary, clean, per-test-process directory for minio data
- Runs a separate minio instance for each test process
- Unskips minio-based tests in CI
Constants should always be only assigned once. The logical OR assignment
of a constant is a relic of the past before we used zeitwerk for
autoloading and had bugs where a file could be loaded twice resulting in
constant redefinition warnings.
- limits security key deletes to second factor keys
- also deletes backup codes (lingering backup codes break login flow entirely)
* Add spec for rake task to disable 2FA for a user
This adds dedicated routes for /login and /signup, replacing the use of modals. Currently, this is behind the experimental_full_page_login feature flag. It also includes some small consistency fixes related to formatting, spacing, icons, and the loading of certain elements
Currently, when the MessageFormat compiler fails on some translations,
we just have the raw output from the compiler in the logs and that’s not
always very helpful.
Now, when there is an error, we iterate over the translation keys and
try to compile them one by one. When we detect one that is failing, it’s
added to a list that is now outputted in the logs. That way, it’s easier
to know which keys are not properly translated, and the problems can be
addressed quicker.
---
The previous implementation of this patch had a bug: it wasn’t handling
locales with country/region code properly. So instead of iterating over
the problematic keys, it was raising an error.
In ed6c9d1545, we started flushing
Redis's database at the end of each test. However, we had something like
this:
```
config.after(:each, type: :system) { teardown system test stuff }
config.after(:each) { # flush redis }
```
When stuff was defined in this order, flushing redis was called before
the teardown of system test. Instead we have to switch the order around
which is what this commit does.
If a plugin's JS fails to load for some reason, most commonly
ad blockers, the entire admin interface would break. This is because
we are adding links to the admin routes for plugins that define
them in the sidebar.
We have a fix for this already in the plugin list which shows a warning
to the admin. This fix just prevents the broken link from rendering
in the sidebar if the route is not valid.
This helps uncover issues with bigint columns that are joined with int
columns. It also introduces a temporary API for plugins to migrate int
columns to bigint in test environment to make tests pass.
We some times get the following failure on Github CI:
```
expected StandardError with message matching /some.host/, got #<Socket::ResolutionError: getaddrinfo: Temporary failure in name resolution> with backtrace:
```
* Add migrations to ensure password hash is synced across users & user_passwords
* Persist password-related data in user_passwords instead of users
* Merge User#expire_old_email_tokens with User#expire_tokens_if_password_changed
* Add post deploy migration to mark password-related columns from users table as read-only
* Refactored UserPassword#confirm_password? and changes required to accommodate hashing the password after validations
There have been too many flaky tests as a result of leaking state in
Redis so it is easier to resolve them by ensuring we flush Redis'
database.
Locally on my machine, calling `Discourse.redis.flushdb` takes around
0.1ms which means this change will have very little impact on test
runtimes.
While using `OpenStruct` is nice, it’s generally not a very good idea as
it usually leads to performance problems.
The `OpenStruct` source code even says basically to avoid it.
Since the context object is crucial in our services, this patch replaces
`OpenStruct` with a custom implementation instead.
Because of unreliability, the spec was temporarily disabled. However, it is ensuring that the custom flags system is working correctly. Therefore it would be great to enable it again.
I made a few fixes to try to mitigate this situation:
- Reduced amount of Redis calls;
- When deleting, ensure that the modal is closed before checking the result;
- Moved duplicated name tests to a separate block;
- Increased wait time to 3 times the default because I noticed that sometimes it gets stuck for a moment. Most of the time it is fast, but sometimes when I run tests in a loop 50 times I see slowness.
Dismissing admin notices is an admin-only action. This is enforced on the back-end both by a routing constraint and a policy in the relevant service.
However, we still unconditionally display the "Dismiss" button to anyone with access to the admin dashboard. When clicked, it results in a 404 modal (due to the routing constraint.)
With this change we only render the dismiss button for admins.
If you have the admin dashboard open, and one of the admin notices listed has already been dismissed (e.g. in another tab, or by another admin) we would show an ugly "FAILED" modal.
This change makes the admin dismiss endpoint idempotent. If the admin notice is already destroyed, then respond with 200. This will also correctly remove it from the list in the front-end.
When a post has some replies, and the user click on the button to show them, we would load ALL the replies. This could lead to DoS if there were a very large number of replies.
This adds support for pagination to these post replies.
Internal ref t/129773
FIX: Duplicated parent posts
DEV: Query refactor
XHR requests are handled differently by the application and the
responses do not have any preloaded data so the cache key needs to
differntiate between those requests.
