refs PA-37
As we add Posthog to the stack we want to send the existing events we
track to it, as well as opening up the method to teams to use in their
initiatives.
---------
Co-authored-by: Chris Raible <chris@ghost.org>
no issue
- We were checking if Posthog was loaded in a few separate places using
the same logic — this PR consolidates that logic into a simple utility
function so we can ensure consistency and change this logic more easily
if we need to in the future
refs PA-36
- Since Posthog is loaded outside of the main Admin app bundle, we need
to check to make sure it exists before calling it. This way it will only
run on Pro and not locally or on self-hosted instances
- Previously we were checking that `window.posthog` existed, but there
are some cases where `window.posthog` may exist, but the `posthog`
object is not fully loaded yet.
- This change fixes this by checking for `window.posthog.__loaded`
instead, which is set to `true` once the `posthog` object is fully
loaded — at this point, we should be able to call whatever functions we
need to on `window.posthog`
refs PA-24
- Added PostHog identify() calls using the user's hashed email address
when a user is logged into admin
- Added PostHog reset() calls to reset PostHog's distinct_id when a user
logs out of admin
- These events will only be sent in Admin running on Ghost(Pro), and won't impact self-hosted instances.
refs PA-32
- The PostHog toolbar relies on a value that is passed via a hash in the
URL to launch successfully
- Admin overwrites the hash (since it uses hash based routing) before
the toolbar has a chance to read the value
- This change checks for the hash and if it exists, it launches the
toolbar using the hash value
no issue
When we open the editor, we fire 4 requests to fetch member counts. This
commit fixes this by replacing those calls with the members count cache
service.
refs TryGhost/Product#4175
- Added error handling to Sentry's beforeSend function in both Admin and
Core, so if there is any error in beforeSend, we will still send the
unmodified event to Sentry
- This is in response to an incident yesterday wherein the beforeSend
function threw an error due to an unexpected missing value in the
exception. The event sent to Sentry was the error in the beforeSend
function, and the original error never reached Sentry.
- If the original event had reached Sentry, even if unmodified by the
logic in beforeSend, we could have been alerted to the issue sooner and
more easily identified all affected sites.
- Also added defensive logic to protect for certain values in the
exception passed to beforeSend not existing and added unit tests for the
beforeSend function in admin and core
no issue
- removed labs flag
- removed main editor component and all associated components
- switched usage of `<KoenigBasicHtmlInput>` and `<KoenigBasicHtmlTextarea>` over to use plain `textarea`, the only uses were in settings modals that have been replaced in AdminX
- cleaned up unused editor CSS
refs https://github.com/TryGhost/DevOps/issues/83
- this will now continue use the dev server assets if we tell it to,
or copy the dependency package files to the built folder otherwise
- removes `editor` from config API because it's no longer needed
- removes dependency on `editor.url` in tests, as this no longer exists
- edits dev script to pass dev server URL as env var
- adds `@tryghost/koenig-lexical` dependency to Admin
refs https://github.com/TryGhost/DevOps/issues/80
- as part of moving Admin-X-Setting towards GA, we want to change it from
loading the settings externally via a CDN, to bundling it in with
Admin
- the bulk of the changes here are removing the config in Ghost, setting
up the copy to the Admin assets dir, and loading the new path in Admin
- several other changes have come along the way as I've cleaned up
unneeded code
no issue
- Stripe imposes different minimum charges based on the currency used,
see
https://stripe.com/docs/currencies#minimum-and-maximum-charge-amounts
- Our validation uses double that limit, to take into account
conversions to the main currency in use in the Stripe account
closes https://github.com/TryGhost/Product/issues/3695
- Stripe has minimum charge amounts per currency, see
https://stripe.com/docs/currencies#minimum-and-maximum-charge-amounts
- We double these limits in our validation, to take into account
currency conversions to the main currency used on the Stripe account. If
the publishers enters a suggested amount below the limit, they will see
an error in Admin
- 0 is permitted as default value and corresponds to "no suggested
amount"
refs https://github.com/TryGhost/Product/issues/3657
- When reverting a post to a draft, the meta, `email_only` remained
true.
- This fix switches `email_only` back to false when the `revertToDraftTask` is executed.
refs https://github.com/TryGhost/Team/issues/2677
- This extends the bulk editing UI to pages.
- New endpoints for editing pages in bulk
- Support for type in bulk edit UI
- Fixed empty messages for lists
- Minor bugfixes (e.g. save button when adding tags became red because
task didn't return true)
---
This pull request adds support for bulk editing and deleting of pages in
the admin UI and the API. It refactors the context menu component and
the list templates to handle different types of content (posts or pages)
dynamically. It also updates the selection list utility and the no posts
box component to work with the new feature. It modifies the `posts.js`
and `pages.js` API files and the corresponding input and output
serializers and routes.
refs https://github.com/TryGhost/Team/issues/2677
- When shift clicking on the first item, it no longer will select from top to that item. It will now just select that item.
- Updates event listeners to use mousedown to prevent text selection glitch
refs https://github.com/TryGhost/Team/issues/2677
Fixed a bug that the current selection is deselected when clicking inside a modal when performing a context menu action.
Refactored the context menu component and its usage in the posts list to
improve the user experience and code quality. Introduced a `state`
property and a `setState` method to the `gh-context-menu` component to
handle different scenarios. Used the `gh-context-menu` component in the
`posts-list/context-menu` component to simplify the modal and loading
logic. Added a `#frozen` property and methods to the `selection-list`
utility to prevent the selection of posts from changing while the
context menu is active.
---------
Co-authored-by: Simon Backx <simon@ghost.org>
refs https://github.com/TryGhost/Team/issues/2677
When deselecting all items, and then using shift, it still tries to shift between the previous selection. This fixes that issue.
fixes https://github.com/TryGhost/Team/issues/2938
- Multi selection is disabled for contributors and authors (no actions available)
- Delete action is only available for admins and owners
fixes https://github.com/TryGhost/Team/issues/2919
This pull request implements a new feature that allows bulk editing of
posts by a filter. It adds a new `bulkEdit` endpoint to the posts API
and new `PostsService` methods to handle the bulk actions.
The posts list component is duplicated, so we can keep working in a
copied version without affecting the old version without a flag. It
temporarily adds a star icon to indicate featured posts in the posts
list.
refs https://github.com/TryGhost/Team/issues/2906
Adds a way to select posts using CMD, shift and CMD+A. And adds a placeholder context menu.
Behind the making it rain feature flag.
- closes https://github.com/TryGhost/Team/issues/2786
- when trying to deselect all recipients, the UI would pass null as the filter to publish-options
- publish-options would then fallback to the default options, which is all recipients, and then the UI wouldn't update which made things even weirder
- we want to fallback to the default recipients when the recipientFilter is undefined (e.g. hasn't been set at all), but not when it is explicitly set to null (e.g. when the user has deselected all recipients)
fixes https://github.com/TryGhost/Team/issues/2560
When an email fails, and you reschedule the post, the error dialog was
shown (from the previous try). The retry button on that page allowed you
to retry sending the email immediately, which could be very confusing.
- The email error dialog is no longer shown for scheduled emails
- The email status is no longer polled for scheduled emails
- Retrying an email is not possible via the API if the post status is
not published or sent
- Added some extra snapshot tests
- When retrying an email, we immediately update the email status to
'pending' to have a better API response (instead of still returning
failed).
- Disabled email sending retrying in development (otherwise very hard to
test failed emails if it takes 10 mins before it gives up automatic
retrying)
refs https://github.com/TryGhost/Ghost/issues/16125
The previous fix for incorrect recipient details being shown when
re-sending a failed email introduced another bug that prevented the
"Match post visibility" default recipients setting from working.
- the server always sets `post.emailSegment` to `'all'` for new posts so
the publish flow recipient filter logic that checked for
`post.emailSegment` being present always defaulted to `'all'` rather
than falling back to the selected default recipients setting
- when a post has been published but the email failed it will have its
`newsletter` value set so we can use that as a check for using the
`post.emailSegment` value in place of the default recipients setting
closes https://github.com/TryGhost/Ghost/issues/16125
We weren't taking into account any existing email segment set on the
post. This is usually not an issue because during the publishing flow
the post.emailSegment and the selectedRecipientFilter are kept in sync,
but it becomes and issue when the email fails to send and is later
retried - we now have an inconsistency between the two values.
refs https://github.com/TryGhost/Team/issues/2371
- Adds a test that schedules a post 5 seconds in the future and waits
for it to be published
- Reduced the time restrictions for scheduling:
- The minimum time in the frontend is now 5 seconds in the future (came
from 5 minutes in the future)
- The time picker now suggests 10 minutes in the future instead of the
minimum scheduling time (came from 5 minutes)
- In the backend, a post will be allowed to be scheduled if it is at
least 2 minutes in the past (came from 2 minutes in the future)
- The scheduler will publish a post if it is at least 5 minutes in the
past, and maximum 5 minutes in the future (came from 2 minutes)
closes https://github.com/TryGhost/Team/issues/2242
Contributors don't have permission to fetch `/newsletters/` but the publish flow was sending a request every time a contributor opened a post in the editor creating noise in event logs and in the developer console.
- disabled the newsletters fetch when the logged in user is a contributor
- contributors can't publish so the "missing" data has no effect on the publish flow as it's not used
refs https://github.com/TryGhost/Team/issues/2110
- dynamically defined properties on the config service did not have
autotracking set up properly if they were accessed in any way before the
property was defined, this caused problems in a number of areas because
we have both "unauthed" and "authed" sets of config and when not logged
in we had parts of the app checking for authed config properties that
don't exist until after sign-in and subsequent config re-fetch
- renamed `config` service to `configManager` and updated to only
contain methods for fetching config data
- added a `config` instance initializer that sets up a `TrackedObject`
instance with some custom properties/methods and registers it on
`config:main`
- uses application instance initializer rather than a standard
initializer because standard initializers are only called once when
setting up the test suite so we'd end up with config leaking across
tests
- added an `@inject` decorator that when used takes the property name
and injects whatever is registered at `${propertyName}:main`, this
allows us to use dependency injection for any object rather than just
services or controllers
- using `application.inject()` in the initializer was initially used but
that only works for objects that extend from `EmberObject`, the
injections weren't available in native-class glimmer components so this
decorator keeps the injection syntax consistent
- swapped all `@service config` uses to `@inject config`
no issue
The `config` service has been a source of confusion when writing with modern Ember patterns because it's use of the deprecated `ProxyMixin` forced all property access/setting to go via `.get()` and `.set()` whereas the rest of the system has mostly (there are a few other uses of ProxyObjects remaining) eliminated the use of the non-native get/set methods.
- removed use of `ProxyMixin` in the `config` service by grabbing the API response after fetching and using `Object.defineProperty()` to add native getters/setters that pass through to a tracked object holding the API response data. Ember's autotracking automatically works across the native getters/setters so we can then use the service as if it was any other native object
- updated all code to use `config.{attrName}` directly for getting/setting instead of `.get()` and `.set()`
- removed unnecessary async around `config.availableTimezones` which wasn't making any async calls
no issue
The `settings` service has been a source of confusion when writing with modern Ember patterns because it's use of the deprecated `ProxyMixin` forced all property access/setting to go via `.get()` and `.set()` whereas the rest of the system has mostly (there are a few other uses of ProxyObjects remaining) eliminated the use of the non-native get/set methods.
- removed use of `ProxyMixin` in the `settings` service by grabbing the attributes off the setting model after fetching and using `Object.defineProperty()` to add native getters/setters that pass through to the model's getters/setters. Ember's autotracking automatically works across the native getters/setters so we can then use the service as if it was any other native object
- updated all code to use `settings.{attrName}` directly for getting/setting instead of `.get()` and `.set()`
- removed use of observer in the `customViews` service because it was being set up before the native properties had been added on the settings service meaning autotracking wasn't able to set up properly
fixes https://github.com/TryGhost/Team/issues/2009
- When an email is sent, but it failed there was no way to retry once you left the retry screen
- There was no indication that the email failed to send in the post list and editor
refs e3db911108
Contributors/Authors/Editors do not have permissions to fetch members so the request to fetch a member count when checking member host limits fails and blocked publishing because it was treated like a failed limit check.
- prevented the up-front limit check from running for non-Admin staff users. Publishing will still fail for other users if the site is over the hosting plan members limit but the error will be shown later in the flow when the actual publish request is made.
closes https://github.com/TryGhost/Team/issues/1885
- adds limit check for members to the `PublishOptions` class when it's constructed to set a `publishDisabledError` property if the limit check fails
- if `publishOptions.publishDisabledCheck` is present, all publish options in the publish flow are disabled, the underlying error message is shown, and the continue button is removed to prevent filling in everything only to find at the end of the process that publishing fails
- added handling for a `HostLimitError` error from the API when confirming publishing so the proper underlying message is displayed instead of the confusing "Host limit error, cannot edit post" error
- this is a backup measure for any instances where you're under the max members limit when starting the publish flow but are over the limit when you reach the end of the publish flow
no issue
Since `ember-moment@10.0` it's not been necessary to use the `ember-cli-moment-shim` package, with `moment` instead being usable directly via `ember-auto-import`. Getting rid of the shim package is necessary for compatibility with `embroider`, Ember's new build tooling.
- dropped `ember-cli-moment-shim` dependency
- added `moment-timezone` dependency and updated all imports to reflect the different package
- worked around `ember-power-calendar` having `ember-cli-moment-shim` as a sub-dependency
- added empty in-repo-addon `ember-power-calendar-moment` to avoid `ember-power-calendar` complaining about a missing package
- added `ember-power-calendar-utils` in-repo-addon that is a copy of `ember-power-calendar-moment` but without the build-time renaming of the tree for better compatibility with embroider
fixes: https://github.com/TryGhost/Ghost/issues/12219
- the WYSIWYG editor supports ^2^ for superscript and ~2~ for subscript
- with this change, the same syntax is supported in the markdown card, which was missing
Co-authored-by: Hannah Wolfe <github.erisds@gmail.com>
no issue
The default recipients setting "Usually nobody" was being respected with the email recipient list defaulting to no members selected. However the UI for that state was confusing because the default publish options ended up being "Publish and email" and "Not sent as a newsletter" but it was expected to be "Publish" with the newsletter option being disabled.
- updated the `PublishOptions` setup to reflect the desired outcome for "Usually nobody"
- default publish type is set to "Publish"
- default email recipients are set to match post visibility - this means there are fewer clicks required when switching from "Publish" to "Publish and send"
- updated mirage data setup so any members created are automatically assigned to any newsletter instance with `subscribeOnSignup` set
- ensures we get proper member counts in the publish flow
no issue
- fixed incorrect conversion of time when converting between site tz and utc when setting scheduled hour/minute in publish flow
- we initially created a UTC `newDate` moment from `publishOptions.scheduledAtUTC` but then used `newDate.tz(siteTimezone).format()` to get the time which was the source of the bug, it was assumed that was a non-destructive action returning a new date but it actually changed the underlying date resulting in the later calculations causing timezone adjustments to be applied twice
- simplified by not converting to UTC inside the publish-at component as that's already handled inside PublishOptions
- fixed time not resetting to minimum schedule time when set to earlier date than allowed
no issue
Problem:
- the server stores time with second-level precision but we send milliseconds
- when scheduling a post we were storing the selected publish-at time in the PublishOptions instance with non-zero milliseconds as the initial date was based on now+5mins
- when we were saving after the initial schedule that millisecond-containing time was still being used resulting in a perceived time difference in our client-side and server-side validations indicating that the published_at value should be updated, but when that time was <2 mins in the future the published_at change validation was triggered resulting in errors
Solution:
- ensure we only set times on PublishOptions with 0 milliseconds
- update client-side validations so we only trigger published_at validation when it's dirty
Also fixed:
- client-side validation errors were not shown on the confirm step due to a missing `.args`
