Another attempt at fixing https://meta.discourse.org/t/discourse-with-a-screen-reader/178105/88?u=osama. Previous PR (reverted): #16240.
The problems with the previous PR were:
1. As you scrolled down a topics list, the first topic of every new batch of topics would receive focus and the indicator would show up.
2. Similar to 1, clicking the `See X new or updated topics` notice would also focus a random topic from the new topics that were just loaded.
3. Topics in the suggested topics list received focus too
4. Our custom focus indicator appeared on mobile, but it shouldn't.
This commit should have none of these problems.
This commit is a redo of2f1ddadff7dd47f824070c8a3f633f00a27aacde
which we reverted because it blew up an internal CI check. I looked
into it, and it happened because the old migration to add the bookmark
columns still existed, and those columns were dropped in a post migrate,
so the two migrations to add the columns were conflicting before
the post migrate was run.
------
This commit only includes the creation of the new columns and index,
and does not add any triggers, backfilling, or new data.
A backfill will be done in the final PR when we switch this over.
Intermediate PRs will look something like this:
Add an experimental site setting for using polymorphic bookmarks,
and make sure in the places where bookmarks are created or updated
we fill in the columns. This setting will be used in subsequent
PRs as well.
Listing and searching bookmarks based on polymorphic associations
Creating post and topic bookmarks using polymorphic associations,
and changing special for_topic logic to just rely on the Topic
bookmarkable_type
Querying bookmark reminders based on polymorphic associations
Make sure various other areas like importers, bookmark guardian,
and others all rely on the associations
Prepare plugins that rely on the Bookmark model to use polymorphic
associations
The final core PR will remove all the setting gates and switch over
to using the polymorphic associations, backfill the bookmarks
table columns, and ignore the old post_id and for_topic colummns.
Then it will just be a matter of dropping the old columns down the
line.
This reverts commit 5d77f485cb.
There are some edge cases that we need to handle better. Reverting this
commit because we're going to do a beta release later today.
This is done by defining a `/all` route for use when a category's default filter is 'none'. This was defined for regular category routes in 3e7f7fdd, but not for tag routes.
This commit also corrects the route name TagsShowNoneCategory*Route -> TagsShowCategoryNone*Route, which fixes an error when setting subcategories=none while filtering by tags.
Previously we would issue a 403 for all invalid routes under `/tags/c/...`, which is not semantically correct. In some cases, these 403'd routes would then be handled successfully in the Ember app, leading to some very confusing behavior.
These methods have been natively supported in all our target browsers for many years. We're now feature-detecting `String.prototype.replaceAll`, which is a much more recent addition. If a browser has `replaceAll`, it'll have `padStart` and `padEnd`
This commit is a redo of e21c640a3c
which we reverted to not include half-done work in a release.
This commit is slightly different though, in that it only includes
the creation of the new columns and index, and does not add any
triggers, backfilling, or new data.
A backfill will be done in the final PR when we switch this over.
Intermediate PRs will look something like this:
1. Add an experimental site setting for using polymorphic bookmarks,
and make sure in the places where bookmarks are created or updated
we fill in the columns. This setting will be used in subsequent
PRs as well.
2. Listing and searching bookmarks based on polymorphic associations
3. Creating post and topic bookmarks using polymorphic associations,
and changing special for_topic logic to just rely on the Topic
bookmarkable_type
4. Querying bookmark reminders based on polymorphic associations
5. Make sure various other areas like importers, bookmark guardian,
and others all rely on the associations
6. Prepare plugins that rely on the Bookmark model to use polymorphic
associations
The final core PR will remove all the setting gates and switch over
to using the polymorphic associations, backfill the bookmarks
table columns, and ignore the old post_id and for_topic colummns.
Then it will just be a matter of dropping the old columns down the
line.
Plugins like chat add custom score type to override the title in the UI, but that should be reserved for situations when you need to manage the flag priority separately, which is configurable in the queue settings page.
Currently, if a plugin creates a custom score type, it won't be able to associate a priority, so there's no real gain from doing so. Priorities are tightly related to post-action types, which is something we might want to revise. For now, this change lets plugins move away from custom score types without compromises.
Previously we were loading almost all the data in an afterModel hook, storing it temporarily in route properties, and then passing it to the controller in `setupController`.
This does not follow Ember best-practices, and causes a number of unexpected behaviours. For example, Ember only calls `setupController` **when the model value changes**. Since `model()` was only returning the tag, that meant that category changes and `additionalTag` changes wouldn't always trigger a `setupController` call, and things would get into a very weird state. This is visible when using the 'loading-slider' component because the category navigation dropdown gets 'stuck' when switching categories.
This commit moves all the data-fetching into `model()`. To make things cleaner, it also:
- removes most uses of route-level variables
- introduces async/await in the model() function
- removes some unneeded `get()` usage
- re-uses DiscoverySortableController for queryParam default handling
- Removes override of `renderTemplate()` so that queryParams are correctly passed through to the controller
- Removes some `transitionToRoute` hacks which were working around the queryParams issue
- Switches to `@action`
In certain cases (like chat quoting) we need to be able
to call the API with an async AJAX call before copying
the results to the clipboard. The only way to reliably
do this is by handing off the AJAX promise to a ClipboardItem.
This commit introduces a new clipboardCopyAsync function
to handle this, which will stand alongside the existing
clipboardCopy function which can be used when no AJAX
request is necessary.
The meaning of reminder_at and reminder_last_sent_at changed after
commit 6d422a8033. A bookmark reminder
will fire only if reminder_last_sent_at is null, but before that it
fired everytime reminder_at was set. This is no longer true because
sometimes reminder_at continues to exist even after a reminder fired.
When using the loading-slider, the component instance is re-used across different pages and so the didInsertElement/willDestroyElement hooks are not fired during page transitions. Instead, we can lean on `didReceiveAttrs`.
Similar fix to 87b98e2862
Note that the `scrollTop` feature is still problematic under the loading slider. That will need to be addressed in a future commit.
* DEV: Allow params to be passed on topic redirects
There are several places where we redirect a url to a standard topic url
like `/t/:slug/:topic_id` but we weren't always passing query parameters
to the new url.
This change allows a few more query params to be included on the
redirect. The new params that are permitted are page, print, and
filter_top_level_replies. Any new params will need to be specified.
This also prevents the odd trailing empty page param that would
sometimes appear on a redirect. `/t/:slug/:id.json?page=`
* rubocop: fix missing space after comma
* fix another page= reference
If a user copies a gif from a website into their clipboard and then
tries to paste it into the Discourse composer, we would only paste a
static single frame of the original gif. This happens because the
browser doesn't store the original image in the clipboard, but two
entries:
1. image/png with the frame of the copy moment
2. text/html with the markup of the gif img element
This commit adds an heuristic that detects this and makes us pick the
clipboard content of text/html instead of the image/png when this
happens.
From there our existing HTML paste logic handles and converts the HTML
img tag into markdown, preserving even the alt text.
See https://meta.discourse.org/t/-/218720 for context.
share-topic modal is used everywhere expect when clicking on the top
right corner of the post. This changes standardize on share-topic modal
and add the missing features from share-popup.
The `testem.scss` include triggers a live reload locally. We need these
styles when running `ember test --server`, so this loads that stylesheet
only in that scenario.
* FEATURE: use canonical links in posts.rss feed
Previously we used non canonical links in posts.rss
These links get crawled frequently by crawlers when discovering new
content forcing crawlers to hop to non canonical pages just to end up
visiting canonical pages
This uses up expensive crawl time and adds load on Discourse sites
Old links were of the form:
`https://DOMAIN/t/SLUG/43/21`
New links are of the form
`https://DOMAIN/t/SLUG/43?page=2#post_21`
This also adds a post_id identified element to crawler view that was
missing.
Note, to avoid very expensive N+1 queries required to figure out the
page a post is on during rss generation, we cache that information.
There is a smart "cache breaker" which ensures worst case scenario is
a "page drift" - meaning we would publicize a post is on page 11 when
it is actually on page 10 due to post deletions. Cache holds for up to
12 hours.
Change only impacts public post RSS feeds (`/posts.rss`)
* DEV: Deprecate /posts/:id/reply-ids/all
It was added in ed4c0c4a63 and its only use was removed in b58867b6e9
Nothing in all-the* seems to be using this endpoint.
* Update app/controllers/posts_controller.rb
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
Eventually we want to remove JQuery, but that's a long way off. Installing this package will stop ember-cli printing the deprecation warning on every boot
```
DEPRECATION: The integration of jQuery into Ember has been deprecated and will be removed with Ember 4.0. You can either opt-out of using jQuery, or install the `@ember/jquery` addon to provide the jQuery integration. Please consult the deprecation guide for further details: https://emberjs.com/deprecations/v3.x#toc_jquery-apis
```
The `blocked onebox domains` setting lets site owners change what sites
are allowed to be oneboxed. When a link is entered into a post,
Discourse checks the domain of the link against that setting and blocks
the onebox if the domain is blocked. But if there's a chain of
redirects, then only the final destination website is checked against
the site setting.
This commit amends that behavior so that every website in the redirect
chain is checked against the site setting, and if anything is blocked
the original link doesn't onebox at all in the post. The
`Discourse-No-Onebox` header is also checked in every response and the
onebox is blocked if the header is set to "1".
Additionally, Discourse will now include the `Discourse-No-Onebox`
header with every response if the site requires login to access content.
This is done to signal to a Discourse instance that it shouldn't attempt
to onebox other Discourse instances if they're login-only. Non-Discourse
websites can also use include that header if they don't wish to have
Discourse onebox their content.
Internal ticket: t59305.
In the API keys page where admins can create API keys with restricted scopes, each scope shows a list of URLs that it allows. But currently, this list of allowed URLs shows incomplete URLs for scopes that are added by plugins. For example, the allowed URL for the "run queries" scope of the data-explorer plugin is shown as `/queries/:id/run` when the correct URL for this scope is `/admin/plugins/explorer/queries/:id/run`. The first 3 segments of the path are the mount path of the plugin's engine and it's missing because the routes set of the engine doesn't include the mount path. To fix this, this commit gets the mount path and prepends it to the URL so the complete URL is shown to the user.
It's not possible to write tests for this change because plugins are not loaded in the test environment by default when core's tests suite is running.
Previously, if an admin user tried to add/remove
users to another user's ignored list, it would
be added to their own ignore list because the
controller used current_user. Now for admins only
a source_user_id parameter can be passed through,
which will be used to ignore the target user for
that source user.
The user can select what happens with a bookamrk after it expires. New
option allow bookmark's reminder to be kept even after it has expired.
After a bookmark's reminder notification is created, the reminder date
will be highlighted in red until the user resets the reminder date.
User can do that using the new Clear Reminder button from the dropdown.
The search_ignore_accents site setting can be used to make the search
indexer remove the accents before indexing the content. The unaccent
function from PostgreSQL is better than Ruby's unicode_normalize(:nfkd).
When creating files with create-multipart, if the file
size was somehow zero we were showing a very unhelpful
error message to the user. Now we show a nicer message,
and proactively don't call the API if we know the file
size is 0 bytes in JS, along with extra console logging
to help with debugging.
This change adds support for the categories endpoint to have an api
scope. Only adds GET scope for listing categories and for fetching a
single category.
See: https://meta.discourse.org/t/218080/4
This PR adds an extra description to the 2FA page when granting a user admin access. It also introduces a general system for adding customized descriptions that can be used by future actions.
(Follow-up to dd6ec65061)
Stop Discourse from prompting for push notification on latest iOS beta
where the navigators exposes a broken Push object.
We had some feature detection functions that where outside our
pre-initializer that it dedicated for this stuff. All feature detection
now lives on sniff-capabilities file.
Also removed some old browser detection from the push notifications
code, and simplified the function signature because of it.
Co-authored-by: Jarek Radosz <jradosz@gmail.com>
Discourse users and associated accounts are created or updated when a
user logins or connects the account using their account preferences.
This new API can be used to create associated accounts and users too,
if necessary.
Our @mention user search prioritized users based on prefix matches.
So if searching for `sa` we will display `sam`, `asam` in that order
Previously, we did not prioritize group matches based on prefix. This change ensures better parity.
Implementation notes:
1. User search only prioritizes based on username prefix, not name prefix. TBD if we want to change that.
2. @mention on client side will show 0 group matches if we fill up all the spots with user matches. TBD if we want to unconditionally show the first / second group match.
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
Include a calculation of offset in main - this allows docking
to be calculated (again) using global offsets, in case there is a
custom header outside of Discourse that pushes forum content down.
The current implementation ties the filter query params tightly to the
`summary` attribute on the post stream model making it hard to support
other filters.
This feature was rarely used, could be used for spamming users and was
impossible to add a context to why the user was notified of a topic. A
simple private messages that includes the link and personalized message
can be used instead.
* DEV: Show only top level replies
Adds a new query param to the topic view so that we can filter out posts
that aren't top level replies. If a post is a reply to another post
instead of the original topic post we should not include it in the
response if the `filter_top_level_replies` query param is present.
* add rspec test
This commit extends the original copy-codeblocks initializer,
renaming it to codeblock-buttons, and adding another button
to make the code block fullscreen in a modal window. The fullscreen
code is then run through highlight.js.
This commit also moves much of the code out of the initializer
and into a reusable CodeblockButtons class, so it can also be used
in the fullscreen code modal for the copy + paste button.
The fullscreen button will not be shown if there is no scroll overflow
in the code block, nor will it be shown on mobile. This commit also
changes the fullscreen table button to not show on mobile.
This will make long lines of code much easier to read and interact
with. This is gated behind the same `show_copy_button_on_codeblocks`
site setting.
This allows text editors to use correct syntax coloring for the heredoc sections.
Heredoc tag names we use:
languages: SQL, JS, RUBY, LUA, HTML, CSS, SCSS, SH, HBS, XML, YAML/YML, MF, ICS
other: MD, TEXT/TXT, RAW, EMAIL
When returning the customRenderFn from within buildCustomMarkdownCookFunction
for custom markdown engines (such as the one used by the [chat] transcripts)
we were not hoisting/unhoisting the `html_raw` tokens created by the
transcript, which meant that opts.discourse.hoisted could end up in
a state where it was null, and which caused errors and general unpleasantness.
Instead, we can just call the `cook` function that is already exported
from discourse-markdown-it, that takes care of what we did previously
plus the hoisting.
There is a companion chat commit that adds tests for this, there are
no custom markdown engine usages in core to test with.
* FEATURE: upload an avatar option for uploading avatars with selectable avatars
Allow staff or users at or above a trust level to upload avatars even when the site
has selectable avatars enabled.
Everyone can still pick from the list of avatars. The option to upload is shown
below the selectable avatar list.
refactored boolean site setting into an enum with the following values:
disabled: No selectable avatars enabled (default)
everyone: Show selectable avatars, and allow everyone to upload custom avatars
tl1: Show selectable avatars, but require tl1+ and staff to upload custom avatars
tl2: Show selectable avatars, but require tl2+ and staff to upload custom avatars
tl3: Show selectable avatars, but require tl3+ and staff to upload custom avatars
tl4: Show selectable avatars, but require tl4 and staff to upload custom avatars
staff: Show selectable avatars, but only allow staff to upload custom avatars
no_one: Show selectable avatars. No users can upload custom avatars
Co-authored-by: Régis Hanol <regis@hanol.fr>
Currently, providing things like `filter[%24acunetix]=1` to
`UserActionsController#index` will throw an exception because instead of
getting a string as expected, we get a hash instead.
This patch simply uses `#permit` from strong parameters properly: first
we apply it on the whole parameters, this way it filters the keys we’re
interested in. By doing this, if the value is a hash for example, the
whole key/value pair will be ignored completely.
This commit handles the edge case where a draft is lost with no warnings if the user edits the title (or category/tags) of a topic while they're replying.to the same topic. Repro steps are as follows:
1. Start replying to a topic and type enough to get a draft saved.
2. Scroll up to the topic title and click the pencil icon next to the topic title, change the title, category and/or tags, and then save the changes.
3. Reload the page and you'll see that the draft is gone.
This happens because we only allow 1 draft per topic per user and when you edit the title of a topic that you're replying to, from the server perspective it'll look like as if you've submitted your reply so it will advance the draft sequence for the topic and delete the draft.
The fix in this commit makes `PostRevisor` skip advancing the draft sequence when a topic's title is edited using the pencil button next to the title.
Internal ticket: t60854.
Co-authored-by: Robin Ward <robin.ward@gmail.com>
Meta topic: https://meta.discourse.org/t/rtl-direction-is-broken-in-quotes/217639?u=osama.
Posts in Discourse are by default always rendered in the same direction as the rest of site, for example if the site is RTL, a post in that site is always rendered RTL even if it's made of an LTR language entirely. However, this behavior can be changed by enabling the `support mixed text direction` site setting which makes our posts rendering engine consider each "paragraph" in the post and apply an appropriate direction (using the `dir` attribute) on it based on its content/language.
I put paragraph in quotes because technically we only loop through the immediate children of the HTML element that contains the post cooked HTML and do this direction check on them. Most of the time the immediate children are actually paragraphs, but not always. The direction of an element is determined by checking its `textContent` property against a regular expression that checks all characters are RTL characters and based on the regular expression result the `dir` attribute is set on the element.
This technique doesn't work so well on quotes because they may contain multiple paragraphs which may be in different languages/directions. For example: if a site's language is Arabic (RTL language) and the `support mixed text direction` setting is enabled, regular paragraphs outside quotes are rendered as expected with the right direction depending on the paragraph's language. However, paragraphs within a quote are all (incorrectly) rendered in a single direction, LTR or RTL, regardless of whether they're of different languages/directions or not.
The reason for this is that when we're determining the direction for the quote, it's considered as one element and the direction is set on the whole quote. But for complex quotes that contain mixed paragraphs, we need to be more surgical and apply direction on individual paragraphs/elements within the quote.
This commit adds special handling for quotes to ensure that:
* the quote top bar (the avatar plus the chevron and arrow) always match the site direction
* each immediate paragraph (`<p>` elements) under `<blockquote>` in the quote gets a direction based on its content.
For before/after screenshots, see PR #16004.
This option will make it so the [quote] bbcode will always
include the HTML link to the quoted post, even if a topic_id
is not provided in the PrettyText#cook options. This is so
[quote] bbcode can be used in other places, like chat messages,
that always need the link and do not have an "off-topic" ID
to use.
The 'new' tab doesn't exist for anonymous users. Every 'new' topic also publishes a message on the `/latest` channel, so the blue banner at the top of the topic-list will still be functional
* DEV: remove duplicate code in button component template
* DEV: refactor is-loading state of d-button component
Before this change on initialisation `forceDisabled` is set `false` and then might change to `undefined` - depending on the use of the button component. This change ensures a boolean value for `forceDisabled`.
The added test works with and without the new change. The test is added as it represents the default use case for most buttons.
Previously cached counting made redis calls in main thread and performed
the flush in main thread.
This could lead to pathological states in extreme heavy load.
This refactor reduces load and cleans up the interface
Previously we were publishing one messagebus message per user which was 'tracking' a topic. On large sites, this can easily be 1000+ messages. The important information in the message is common between all users, so we can manage with a single message on a shared channel, which will be much more efficient.
For user-specific values (notification_level and last_read_post_number), the JS app can infer values which are 'good enough'. Correct values will be loaded as soon as a topic-list containing the topic is visited.
aa1442fdc3 split theme stylesheets so that every component gets its own stylesheet. Therefore, there is now no need for parent themes to collate the settings/variables of its children during scss compilation.
Technically this is a breaking change for any themes which depend on the settings/variables of their child components. That was never a supported/recommended arrangement, so we don't expect this to cause issues.
If a theme is updated to introduce a new setting AND immediately make use of it in a stylesheet, then an error was being shown. This is because the stylesheet compilation was using the theme's cached settings, and the cache is only cleared **after** the theme has finished compiling.
This commit updates the SCSS compilation to use uncached values for settings. A similar fix was applied to other parts of theme compilation back in 2020: (a51b8d9c66)
The previous Discourse.cache usage was different to how other theme-related caching is handled, and also requires reaching out to redis every time. The common theme cache is held in memory (as a DistributedCache)
This will only work under Ember CLI, and a small hack is required to make the Resolver work in development mode. In future, when we move to a more recent version of the Ember Resolver, this hack will not be required.
The use of a `/g` regex was causing some surprising, seemingly random, behavior. (https://stackoverflow.com/a/1520853/5913559)
There was also a known issue which would cause inconsistent class behavior when running the 'loading slider' theme component.
This commit takes the opportunity to refactor the component to remove the use of observers and remove the regex-based class parsing.
This commit makes sure that the email log's bounce_error_code
conforms to the SMTP error code RFC on save, so that
it is always in the format X.X.X or XXX without any
additional string details. Also included is a migration
to fix this issue for past records.
Similar site settings exist for likes and edits and the new ones work
in a similar way.
By default, users below TL2 have a limit of 20, the limit is increased
by 1.5 for TL2 users up to 30, by 2 for TL3 users up to 40 and by 3 for
TL4 users up to 60.
We validate the *format* of email addresses in many places with a match against
a regex, often with very slightly different syntax.
Adding a separate EmailAddressValidator simplifies the code in a few spots and
feels cleaner.
Deprecated the old location in case someone is using it in a plugin.
No functionality change is in this commit.
Note: the regex used at the moment does not support using address literals, e.g.:
* localpart@[192.168.0.1]
* localpart@[2001:db8::1]
This commit moves _getMultilineContents and _applySurround into
TextareaTextManipulation, so other text area components using
that mixin can benefit from them (such as the chat composer).
It also creates a public function wrapper for many TextareaTextManipulation
functions that should not have underscore prefixes because they are
used outside the file. Will make follow-up PRs for each plugin/theme using
those functions then a final follow-up core PR to fix these up.
After leaving a group, it is trying to reload its member list. Previously, when the members_visibility_level attribute has a value of 2 or higher, it displayed an error popup since the can_see_members attribute was not updated.
2FA support in Discourse was added and grown gradually over the years: we first
added support for TOTP for logins, then we implemented backup codes, and last
but not least, security keys. 2FA usage was initially limited to logging in,
but it has been expanded and we now require 2FA for risky actions such as
adding a new admin to the site.
As a result of this gradual growth of the 2FA system, technical debt has
accumulated to the point where it has become difficult to require 2FA for more
actions. We now have 5 different 2FA UI implementations and each one has to
support all 3 2FA methods (TOTP, backup codes, and security keys) which makes
it difficult to maintain a consistent UX for these different implementations.
Moreover, there is a lot of repeated logic in the server-side code behind these
5 UI implementations which hinders maintainability even more.
This commit is the first step towards repaying the technical debt: it builds a
system that centralizes as much as possible of the 2FA server-side logic and
UI. The 2 main components of this system are:
1. A dedicated page for 2FA with support for all 3 methods.
2. A reusable server-side class that centralizes the 2FA logic (the
`SecondFactor::AuthManager` class).
From a top-level view, the 2FA flow in this new system looks like this:
1. User initiates an action that requires 2FA;
2. Server is aware that 2FA is required for this action, so it redirects the
user to the 2FA page if the user has a 2FA method, otherwise the action is
performed.
3. User submits the 2FA form on the page;
4. Server validates the 2FA and if it's successful, the action is performed and
the user is redirected to the previous page.
A more technically-detailed explanation/documentation of the new system is
available as a comment at the top of the `lib/second_factor/auth_manager.rb`
file. Please note that the details are not set in stone and will likely change
in the future, so please don't use the system in your plugins yet.
Since this is a new system that needs to be tested, we've decided to migrate
only the 2FA for adding a new admin to the new system at this time (in this
commit). Our plan is to gradually migrate the remaining 2FA implementations to
the new system.
For screenshots of the 2FA page, see PR #15377 on GitHub.
* FIX: Don't accept accents in slug if generation_method == 'ascii'
Fixes bug reported in:
- https://meta.discourse.org/t/404-when-trying-to-edit-category-with-accent-in-slug/214762
- https://meta.discourse.org/t/formatting-and-accents-in-urls/215734/5
Assuming `SiteSetting.slug_generation_method == 'ascii'.
If the user provides a slug containing non-ascii characters while
creating the category, the user will receive a 404 error just
after saving the category since the slug will be escaped anyway but
Category.find_by_slug_path won't escape the category slug
causing the Edit Page of the category to be inaccessible.
This commit checks the provided slug and raises an error if the
provided slugcontains non-ascii characters ensuring that the
provided value is consistent with the site settings.
It also changes Category.find_by_slug_path to always escape the slug,
since if present, it is escaped anyway in Category.ensure_slug to
prevent the 404 in the Edit Category Page in case the user already
have some category with a non-ascii slug.
* Removed trailing whitespace
When parent category or grandparent category is muted, then category should be muted as well.
Still, it can be overridden by setting individual subcategory notification level.
CategoryUser record is not created, mute for subcategories is purely virtual.
This can happen if the topic to which a user is invited is in a private
category and the user was not invited to one of the groups that can see
that specific category.
This used to be a warning and this commit makes it an error.
This was deprecated in Discourse 2.4, but no end version was put on the deprecation. Many plugins/themes are still using it. This commit restores it under ember-cli so that it does not block the Ember CLI rollout, and can be removed in a future commit.
This commit introduces two new APIs for handling unused uploads, one
can be used to exclude uploads in bulk when the data model allow and
the other one excludes uploads one by one.
We added this constraint in 5bd55acf83
but it is causing problems in hosted sites and is catching the
issue too far down the line. This commit removes the constraint
for now, and also fixes an issue found with PostDestroyer
which wasn't using the UserStatCountUpdater when updating post_count
and thus was causing negative numbers to occur.
Previously calls such as `Emoji["smile"]` would force a full dehydration of
objects from Redis.
This introduces a version safe site and global emoji cache so lookups are
cheap. It eliminates iterating through the list of emojis and pulling from
redis.
Distributed cache uses a normalized name as the key and stores an Array tuple
with version and Emoji. Successful hits always confirm version matches.
Interface to Emoji object remains unchanged.
We opted for 2 caches to improve reuse on multisites. misses though will be
stored in both caches. If there is a hit on the global cache we can avoid
looking up in site local cache and storing a miss there.
Previously we were calling `EXPIRE` every time we incremented a given key. Instead, we can call EXPIRE once when the key is first populated. A LUA script is used to make this as efficient as possible.
Consumers of this Concern use daily keys. Since we're now calling EXPIRE only at the beginning of the day, rather than throughout the day, the expire time has been increased from 3 to 4 days.
Whenever we got a bounced email in the Email::Receiver we
previously would just set bounced: true on the EmailLog and
discard the status/diagnostic code. This commit changes this
flow to store the bounce error code (defined in the RFC at
https://www.iana.org/assignments/smtp-enhanced-status-codes/smtp-enhanced-status-codes.xhtml)
not just in the Email::Receiver, but also via webhook events
from other mail services and from SNS.
This commit does not surface the bounce error in the UI,
we can do that later if necessary.
To make this possible in development mode, the `sourceURL=` implementation needs to include something plugin-specific. This has no effect on production.
The asset version is bumped in order to trigger a re-compilation of plugin JS assets.
There is a couple of layers of caching for theme JavaScript in Discourse:
The first layer is the `javascript_caches` table in the database. When a theme
with JavaScript files is installed, Discourse stores each one of the JavaScript
files in the `theme_fields` table, and then concatenates the files, compiles
them, computes a SHA1 digest of the compiled JavaScript and store the results
along with the SHA1 digest in the `javascript_caches` table.
Now when a request comes in, we need to render `<script>` tags for the
activated theme(s) of the site. To do this, we retrieve the `javascript_caches`
records of the activated themes and generate a `<script>` tag for each record.
The `src` attribute of these tags is a path to the `/theme-javascripts/:digest`
route which simply responds with the compiled JavaScript that has the requested
digest.
The second layer is a distributed cache whose purpose is to make rendering
`<script>` a lot more efficient. Without this cache, we'd have to query the
`javascript_caches` table to retrieve the SHA1 digests for every single
request. So we use this cache to store the `<script>` tags themselves so that
we only have to retrieve the `javascript_caches` records of the activated
themes for the first request and future requests simply get the cached
`<script>` tags.
What this commit does it ensures that the SHA1 digest in the
`javascript_caches` table stay the same across compilations by adding an order
by id clause to the query that loads the `theme_fields` records. Currently, we
specify no order when retrieving the `theme_fields` records so the order in
which they're retrieved can change across compilations and therefore cause the
SHA1 to change even though the individual records have not changed at all.
An inconsistent SHA1 digest across compilations can cause the database cache
and the distributed cache to have different digests and that causes the
JavaScript to fail to load (and if the theme heavily customizes the site, it
gives the impression that the site is broken) until the cache is cleared.
This can happen in busy sites when 2 concurrent requests recompile the
JavaScript files of a theme at the same time (this can happen when deploying a
new Discourse version) and request A updates the database cache after request B
did, and request B updates the distributed cache after request A did.
Internal ticket: t60783.
Co-authored-by: David Taylor <david@taylorhq.com>
We serve `service-worker.js` in an unusual way, which means that the sourcemap is not available on an adjacent path. This means that the browser fails to fetch the map, and shows an error in the console.
This commit re-writes the source map reference in the static_controller to be an absolute link to the asset (including the appropriate CDN, if enabled), and adds a spec for the behavior.
It's important to do this at runtime, rather than JS precompile time, so that changes to CDN configuration do not require re-compilation to take effect.
The old choose-topic component did not have the same style as the rest
of the create invite modal and was not very suitable to use in the modal
because it introduced the search results in modal's body.
The new topic-chooser is built using select-kit and provides a more
polished user experience.
This makes a small improvement to 'cold cache' ember-cli build times, and a large improvement to 'warm cache' build times
The ember-auto-import update means that vendor is now split into multiple files for efficiency. These are named `chunk.*`, and should be included immediately after the `vendor.js` file. This commit also updates the rails app to render script tags for these chunks.
This change was previously merged, and caused memory-related errors on RAM-constrained machines. This was because Webpack 5 switches from multiple worker processes to a single multi-threaded process. This meant that it was hitting node's default heap size limit (~500mb on a 1GB RAM server). Discourse's standard install procedure recommends adding 2GB swap to 1GB-RAM machines, so we can afford to override's Node's default via the `--max-old-space-size` flag.
- Update UI to improve contrast
- Make it clear that the message is only shown to administrators
- Add theme name and id to the console output
- Parse the error backtrace to identify the theme-id for post-decoration errors
- Improve console output to include the theme name / URL
- Add `?safe_mode=no_custom` to the admin panel link, so that it will work even if the theme is causing the site to break
The chat quoting mechanism will need to be able to generate
markdown for all kinds of uploads. The UploadMarkdown class
was missing generation for video and audio uploads. This
commit adds that in, and also expands the server-side regex
recognition of FileHelper types to match those in uploads.js,
and adds a spec for UploadMarkdown
In 8e5b945b0f, we reverted the commit but
at the same time resulted in Theme::BASE_COMPILER_VERSION going
backwards which caused problems with themes caching.
This commit bumps the version to clear all the caches.
Follow-up to 8e5b945b0f
Breakdown of fixes in this commit:
* `UserStat#topic_count` was not updated when visibility of
the topic changed.
* `UserStat#post_count` was not updated when post was hidden or
unhidden.
* `TopicConverter` was only incrementing or decrementing the counts by 1
even if a user has multiple posts in the topic.
* The commit turns off the verbose logging by default as it is just
noise to normal users who are not debugging this problem.
Another use case for focusComposer() is if the user is
already inside a topic but another component (such as the
floating chat window) needs to open the composer. This
commit also fixes the appendText option to only prepend
2 new lines if there is text before the text to be appended.
Follow up 7850ee318f
- Update UI to improve contrast
- Make it clear that the message is only shown to administrators
- Add theme name and id to the console output
- Parse the error backtrace to identify the theme-id for post-decoration errors
- Improve console output to include the theme name / URL
- Add `?safe_mode=no_custom` to the admin panel link, so that it will work even if the theme is causing the site to break
accept HTML attribute is not fully supported on iOS yet and can contain
only MIME types. This changes the input to allow all files and the
extension check is performed later in JavaScript.
Instead of relaying on /timings request, we should cache last read post number. That should protect from having incorrect unread counter when going back to topic list.
This additional cache is very temporary as once /timings request is finished, serializer will have a correct result.
Simplified flow is:
1. Store in cache information about last seen post number before /timings request is sent
2. When getting back to topic list compare value of last seen post number returned by /latest request and information in cache. If cache number is higher, than use it instead of information returned by /latest. In addition delete cache item as there is high chance that `/timings` request already finished.
3. Optionally, delete cache when timings request is done and topic list was not yet visited.
Keeping cache reasonably small should not affect performance.
TopicTrackingState should correctly set filterCategory and filterTag for all different configurations.
When filterTag exists and new_topic message arrives, it ensures that filterTag is included in payload tags
If filterTag is part of payload tags, message that new topics are available is displayed and after click, new topics are included in the list.
This adds logic to increase an `InvitedUser` record, increase
`redemption_count` and create a `:invitee_accepted` to let the inviter
know that the invitee used the invite.
Initial support for this was implemented in commit 9969631.
In ab5361d69a, we rescue from the PG error
but the transaction is already aborted causing any DB query after to
fail. As such, we avoid triggering the error in the first place by
checking that we would not be insertin a negative number into the
counter cache.
Follow-up to ab5361d69a
Allows to write custom code blocks:
```
```mermaid height=200,foo=bar
test
```
```
Which will then get converted to:
```
<pre data-code-wrap="mermaid" data-code-height="200" data-code-foo="bar">
<code class="lang-nohighlight">
test
</code>
</pre>
```
Sorting group members worked always kept the group owners at the top of
the list. This commit keeps the group owners at the top of the list only
when no order exists.
This commits adds a new advance_draft to PostCreator that controls if
the draft sequence will be advanced or not. If the draft sequence is
advanced then the old drafts will be cleared. This used to happen for
posts created by plugins or through the API and cleared user drafts
by mistake.
We need this in other places, this commit moves clipboardCopy
to the utilities.js lib. Had to remove use of Promise as well because
lib/utilities cannot import it, otherwise it will cause a mini racer error.
This commit adds a new helpful function to the composer controller
which can be used to focus the composer and insert text, regardless
of whether the consumer knows whether the composer is open or has
a draft. This is good for cases where an action needs to copy text
to the composer or open it with text after navigating to a URL.
The inspiration for this addition is the discourse-chat plugin,
which needs to be able to copy quoted markdown from the chat
and insert it into the composer, and unlike in the topic controller
we have no idea of the state of the composer from chat.
There are still spots in the code base which results in us trying to turn the post and topic count negative. However,
we have a job that runs on a daily basis which will correct the count. Therefore, avoid raising an error for now
and log the exception instead.
* FEATURE: Add external_id to topics
This commit allows for topics to be created and fetched by an
external_id. These changes are API only for now as there aren't any
front changes.
* add annotations
* add external_id to this spec
* Several PR feedback changes
- Add guardian to find topic
- 403 is returned for not found as well now
- add `include_external_id?`
- external_id is now case insensitive
- added test for posts_controller
- added test for topic creator
- created constant for max length
- check that it redirects to the correct path
- restrain external id in routes file
* remove puts
* fix tests
* only check for external_id in webhook if exists
* Update index to exclude external_id if null
* annotate
* Update app/controllers/topics_controller.rb
We need to check whether the topic is present first before passing it to the guardian.
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
* Apply suggestions from code review
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
Under ember-cli, we rely on the `ember-export-application-global` addon to make `window.Discourse` available. This happens in an initializer. Previously this inititalizer would run after `auto-load-modules`, and so any widget/helper modules would not be able to access it. This commit sets some `after` parameters on the `auto-load-modules` and `inject-objects` initializers to ensure that `export-application-global` is run first.
This reverts commit f4c6a61855 and a8325c9016
This update of ember-auto-import and webpack causes significantly higher memory use during rebuilds. This made ember-cli totally unusable on 1GB RAM / 2GB swap environments. We don't have a specific need for this upgrade right now, so reverting for now.
Random strings can result into much longer tsvectors. For example
parsing a Base64 string of ~600kb can result in a tsvector of over 1MB,
which is the maximum size of a tsvector.
Follow-up-to: 823c3f09d4
This commit introduces our own handling and warning for Sidekiq's new 'non-json-serializable' warning. This decouples us from Sidekiq's own deprecation cycle, and allows us to use our own deprecation system. It also means that the dump/parse happens in test mode, which will help us to catch occurrences before they reach production.
This can be disabled by setting `EMBER_CLI_PROD_ASSETS=0`, but this option will not be available for long. If your theme/plugin/site has issues under Ember CLI, please open a topic on https://meta.discourse.org
In the commit d8bf2810ff we hoisted
the userOptionFields array to a module-level variable, but kept
the code inside save() the same. This causes an issue where if
save() is called twice on the same user with some array of user
option fields, the userOptionFields array is mutated, which means
the second save is likely not saving the fields intended.
This commit fixes the issue by not mutating the array. We cannot
change them into consts though, because we have an API to add more
items to the array.
This commit allows group SMTP emails to be sent with a
different from email address that has been set up as an
alias in the email provider. Emails from the alias will
be grouped correctly using Message-IDs in the mail client,
and replies to the alias go into the correct group inbox.
Ensures that `UserStat#post_count` and `UserStat#topic_count` does not
go below 0. When it does like it did now, we tend to have bugs in our
code since we're usually coding with the assumption that the count isn't
negative.
In order to support the constraints, our post and topic fabricators in
tests will now automatically increment the count for the respective
user's `UserStat` as well. We have to do this because our fabricators
bypasss `PostCreator` which holds the responsibility of updating `UserStat#post_count` and
`UserStat#topic_count`.
Ember tests follows a convention where test files have a postfix of
`-test.js`. This ensures that any files in the tests folder which
follows this pattern is included.
Job arguments go via JSON, and so DateTime objects will appear as strings in the Job's `#execute` method. The latest version of Sidekiq has started warning about this to reduce developer confusion.
Job arguments go via JSON, and so symbols will appear as strings in the Job's `#execute` method. The latest version of Sidekiq has started warning about this to reduce developer confusion.
The latest version of Sidekiq introduced a warning when jobs are queued with arguments which 'do not stringify to JSON safely'. In the vast majority of cases, this is because a hash is passed with symbols as keys. When those args are passed to the job, the keys will be stringified.
Our job wrapper already takes care of this issue by calling '.with_indifferent_access' on the args before passing them to `#execute`, so we don't need to change anything about our use. All we need to do is satisfy Sidekiq's warning system by 'stringifying' all the keys before enqueuing the job.
When uploading an image, we change the uploading placeholder several times. Every time, we correct the position of the cursor after replacing. But we schedule repositioning of cursor to the afterRender queue in Ember Run Loop. As a result, sometimes we replace the placeholder several times but correct the cursor position only once at the end.
It just cannot work correctly with scheduling, we'll always be dealing with cumulative error. Removing scheduling fixes the problem.
Sadly, I cannot make the test work, I skipped it for now, going to give it another try later.
This makes a small improvement to 'cold cache' ember-cli build times, and a large improvement to 'warm cache' build times
The ember-auto-import update means that vendor is now split into multiple files for efficiency. These are named `chunk.*`, and should be included immediately after the `vendor.js` file. This commit also updates the rails app to render script tags for these chunks
This commit adds a requestCustomMarkdownCookFunction function
to the `helper` that is provided to custom markdown rules
via their `setup` function.
The way this works is that once the default markdown engine that
we use for cooking posts has been set up, we loop through all
of the callbacks registered by `requestCustomMarkdownCookFunction`
and call `_buildCustomMarkdownCookFunction`. This creates
a new markdown engine using many of the same settings as the
default one, but will allow for the following options to be
changed by the markdown rule requesting the custom function:
* featuresOverride - The markdown-it features to allow for the engine
* markdownItRules - The markdown-it rules to allow for the engine
After this engine is set up a render function which renders + sanitizes
the output is returned for use by the markdown rule.
The use case for this API is mainly for block BBCode markdown rules
which want to render their content with a limited subset of the
markdown features/rules. Our initial use case for this is chat message
quoting.
This commit also does some minor refactoring of discourse-markdown-it
to accommodate this new engine building.
When changing to uppy for file uploads we forgot to add
these conditions to the paste event from 9c96511ec4
Basically, if you are pasting more than just a file (e.g. text,
html, rtf), then we should not handle the file and upload it, and
instead just paste in the text. This causes issues with spreadsheet
tools, that will copy the text representation and also an image
representation of cells to the user's clipboard.
This also moves the paste event for composer-upload-uppy to the
element found by the `editorClass` property, so it shares the paste
event with d-editor (via TextareaTextManipulation), which makes testing
this possible as the ember paste bindings are not picked up unless both
paste events are on the same element.
Adds up and down buttons next to the inputs of value lists when there is more than 1 item present. This helps to re-order the items in the value lists if necessary.
- Limit bulk re-invite to 1 time per day
- Move bulk invite by csv behind a site setting (hidden by default)
- Bump invite expiry from 30 -> 90 days
## Updates to rate_limiter
When limiting reinvites I found that **staff** are never limited in any way. So I updated the **rate_limiter** model to allow for a few things:
- add an optional param of `staff_limit`, which (when included and passed values, and the user passes `.staff?`) will override the default `max` & `secs` values and apply them to the user.
- in the case you **do** pass values to `staff_limit` but the user **does not** pass `staff?` the standard `max` & `secs` values will be applied to the user.
This should give us enough flexibility to
1. continue to apply a strict rate limit to a standard user
2. but also apply a secondary (less strict) limit to staff
In our legacy environment, Ember RFC176 shims are included in `discourse-loader.js` which is part of the `vendor.js` bundle. This meant that the module shims were available as soon as the vendor.js asset was loaded.
Under Ember CLI, we were defining these shims in `discourse-boot.js`. This is loaded by the browser much later, and meant that the shims were not available to themes/plugins that call `require()` before Discourse has booted. This was causing errors under some circumstances.
This commit refactors the Ember CLI implementation so that the shims are included in the vendor.js bundle. This is done via an addon which leans on the ember-rfc176-data NPM package. This will ensure we have all the definitions, without the need for manual copy/paste.
In our legacy environment, Ember RFC176 shims are included in `discourse-loader.js` which is part of the `vendor.js` bundle. This meant that the module shims were available as soon as the vendor.js asset was loaded.
Under Ember CLI, we were defining these shims in `discourse-boot.js`. This is loaded by the browser much later, and meant that the shims were not available to themes/plugins that call `require()` before Discourse has booted. This was causing errors under some circumstances.
This commit refactors the Ember CLI implementation so that the shims are included in the vendor.js bundle. This is done via an addon which leans on the ember-rfc176-data NPM package. This will ensure we have all the definitions, without the need for manual copy/paste.
Previously, `resetSite()` would immediately generate a new `Site` instance, and run all the initialization logic within the model. This included initializing Category objects.
This was problematic because `resetSite()` is called before any initializers have been run. That means that any modifications to the Site or Category classes would not have any effect on the already-initialized Site/Category instances.
This commit makes two main changes so so that the test environment is more production-like:
1. Update `resetSite` so that it simply stores the new data in the PreloadStore, and destroys the old Site instance. Initialization of a new site instance happens 'just in time' (normally during the `inject-discourse-objects` initializer)
2. Update the `helperContext` in tests to use getters. This avoids the need to look up `Site.current()` before initializers have run
It also makes a minor adjustment to one test which was relying on a side-effect of the previous behavior.
This should resolve the failing tests for discourse-category-expert under Ember-CLI: https://github.com/discourse/discourse-category-experts/pull/69
Non-staff users are not allowed to see whisper so this change prevents
non-staff user from seeing a like count that does not make sense to
them. In the future, we might consider adding another like count column
for staff user.
Follow-up to 4492718864
This also switches to using the NPM package for better build stability. And adds a clearer label in the alert that is displayed to show your current timezone (when changing timezones).
* Some are no longer flaky or easily fixed
* Some are out of date or test things we can't do accurately (scroll
position) and are removed.
* Unwinds some uppy tests and makes sure all promises and runloops are
resolved.
Everything has been run in legacy/ember cli multiple times to ensure no
obvious suite regressions.
When the record is not saved, we should display a proper message.
One potential reason can be plugins for example discourse-calendar is specifying that only first post can contain event
This fixes rare cases of layout shift caused by images appearing slightly smaller after being loaded.
For example, a 371x1031 image is uploaded. It gets lightboxed, with the generated thumbnail of size 179x500. `height: auto` changes that thumbnail's size (only after being loaded) to 179x497, causing a 3px shift.
I did not observe any regressions with this change.
* FIX: Remove svg icons from webmanifest shortcuts
While SVGs are valid in the webmanifest, Chromium has not implemented
support for it in this specific manifest member.
Revert when https://bugs.chromium.org/p/chromium/issues/detail?id=1091612
lands.
* fix test
We don't need raw to decide if we can fast edit or not, we will fetch the raw later when we do the replacement, but this step can be done directly from innerHTML.
Sometimes plugins need to have additional data or options available
when rendering custom markdown features/rules that are not available
on the default opts.discourse object. These additional options should
be namespaced to the plugin adding them.
```
Site.markdown_additional_options["chat"] = { limited_pretty_text_markdown_rules: [] }
```
These are passed down to markdown rules on opts.discourse.additionalOptions.
The main motivation for adding this is the chat plugin, which currently stores
chat_pretty_text_features and chat_pretty_text_markdown_rules on
the Site object via additions to the serializer, and the Site object is
not accessible to import via markdown rules (either through
Site.current() or through container.lookup). So, to have this working
for both front + backend code, we need to attach these additional options
from the Site object onto the markdown options object.
This reverts commit f43bba8d59.
Adding randomness has introduced a lot of flakiness in our ember-cli tests. We should fix those issues at the source. However, given the upcoming stable release, this randomness has been reverted so that the stable release includes a stable test suite. Having a stable test suite on stable will make backporting future commits much easier.
Osama Sayegh
Martin Brennan
Mark VanLandingham
David Taylor
Alan Guo Xiang Tan
Robin Ward
Roman Rizzi
Jarek Radosz
Bianca Nenciu
Blake Erickson
Rafael dos Santos Silva
Penar Musaraj
Sam
Ella E
Ayke Halder
Joffrey JAFFEUX
Kris
Jordan Vidrine
Dan Ungureanu
Jeff Wong
Krzysztof Kotlarek
Vinoth Kannan
Loïc Guitaut
Michael Brown
megothss
Gerhard Schlager
janzenisaac
Martin Brennan
Andrei Prigorshnev
Keegan George
Natalie Tay
tshenry
Kyle E. Mitchell