refs #6534
- don't output publisher for the 'People' type on the author page
- change publisher to a full 'Organisation' for the 'Article' type on posts
Note:
Google's structured data validator also wants image & publisher.logo inside of Article to be full 'ImageObject's.
Currently, it output's an error for them: 'The attribute itemtype has an invalid value.'
However, the spec on Schema.org says a url is valid: https://schema.org/Article,
which is slightly different to Google's spec here: https://developers.google.com/structured-data/rich-snippets/articles#article_markup_properties
Ideally, we would output a full 'ImageObject', however we don't currently have the width & height info required.
Therefore, I think what we have is valid strictly speaking, but we should aim to fix this when we have better image tools.
no issue
- minor optical fix for schema.org metadata
- sameAs property was showing `null` value in array, if no data was provided
- instead of showing `null`, it will be empty, if no data (author website, facebook or twitter) it will be an empty array
no issue
- if the session is "authenticated" but is invalid, the initial load of the app would fail. It could be replicated by:
1. Authenticating in a running ghost instance
2. Stopping ghost
3. Deleting the database
4. Re-starting ghost
5. Closing the browser
6. Re-opening the browser and visiting http://localhost:2368/ghost
- this fix stores the `transition` object for the duration of the initial load so that `sessionInvalidated` method can trigger actions before the transition has finished
no issue
- always ensure we load a full settings object so that we don't risk saving a partial settings object back to the server
- should fix the issues reported in Slack of disappearing private settings over time
no issue
- Check for title, user_name and user_email in the top level of config.
- If they exist, return them as part of the setup check, so that the setup screen can be prepopulated
closes#4556
- when <kbd>CMD-S</kbd> is used, if the focused element is an input, trigger it's `focusout` handler then schedule the save action to happen after any actions resulting from the trigger
- split out read CSV function into utility and add tests
- update API response to follow JSONAPI more closely
- update the UI to match the new API response
Update for synchronous feature service
Add client-side handling of server-side errors when adding subscribers
- display server-provided error message when we get a server error
- fix the ajax util's `getRequestErrorMessage` method so that it works correctly with Ember's `InvalidError` object instead of the previous request object that it was receiving (*TODO:* this really needs looking at properly so we aren't losing details and Ember Data can do it's stuff)
Styling updates
- proper icon for ascending/descending
- change hover colour to green for "Import CSV" button
Delete subscriber button with confirm modal
- display delete button when hovering over a subscriber row (WARN: really ugly button, styles definitely want looking at)
- show confirm modal when clicking the delete button
- delete subscriber, remove from table, and update total on confirm
no issue
- add some more tests, optimise tests and finish tests
- subscriber model checks external context permissions in permissible fn
- add missing permissions for subscriber csv
Form:
- add confirm, location & referrer hidden fields
- add script to populate location & referrer
- add helper for creating the email field
- pass through input class and placeholder for email from top level form helper
- rename subscribe_form template & helper as it sounds more natural
- handle success and error cases differently
- improve error message display
- ensure useful data is passed back so that we can show nice messages
- check for honeypot value being filled out
- refactor error handler to set an error and always still render
Initial Subscribers screen
- set up mocked api endpoints
- basic subscribers screen with data loading, infinite scroll
"Add Subscriber" screen
- uses modal to display a new subscriber form
- validates subscriber e-mail address
- moves pagination from route into controller to use filtered/sorted CPs on top of a live-query so that new subscribers are added to the list and the total can be properly managed
TODO:
- there is currently a pretty serious performance issue where the whole table is re-rendered when the live-query is updated. `ember-light-table` doesn't allow for live-binding and has no options to easily manipulate it's rows using an external interface - it's possible to move the page loading into the component so we only render new rows but that leaves it difficult to react to new subscribers being added through the UI. I believe the number of components used within the table is also adding to the performance problems.
- most likely solution is to drop `ember-light-table` in favour of rendering the table directly - glimmer should do a good job of fast updates even though the underlying array will be completely swapped out
"Import subscribers" screen
- uses modal to display an import subscribers CSV file upload form
- displays upload progress
- displays import stats and reloads subscribers table once import has completed
- adds `gh-file-uploader` component (NB. pared down copy of `gh-image-uploader`, ripe for some refactoring)
- fixes subscribers acceptance test failing because fixtures did not have the labs flag enabled
Unfortunately this doesn't have 100% test coverage as we're limited in how we can simulate file uploads 😞
Fix performance issues with subscribers table
- moves the table definition from the component up to the controller
- switches back to manually manipulating table rows instead of using a live-query
This is a quick-fix in that it allows us to continue using the `ember-light-table` component but it does mean that we lose some flexibility that the live-query gave us. For now it's not much of an issue and it allows us to defer deeper performance/flexibility work until we have a concrete need and requirements.
Hook up Export CSV button
- use a hidden iFrame to trigger the browser to hit the CSV export endpoint and download the file
Re-order subscribers table by clicking column headers
- displays currently sorted column and sort direction
- clicking a column header re-fetches the data from the server with the appropriate query params
Fix scroll triggers for infinite pagination + icon change
- adds a debounce as well as the throttle so that we always get a final scroll trigger once scrolling has stopped
- changes the subscribers icon from the temporary team icon to the mail icon
refs #6640
- add `NotFoundError` to ajax service and test against that when catching errors in `gh-profile-image`
- don't use `passthrough` in Mirage for gravatar requests to avoid network calls during testing
- add additional tests for `gh-profile-image` and put the debounced gravatar test back in place
Closes#5882
* If a gravatar image is available, remove the default image behind it
* If gravatar image is not available, keep or replace the default image
closes#6534
- new input fields in general settings incl. validation
- facebook and twitter as new models in settings.js
- adds values for facebook and twitter to default-settings.js
- adds blog helpers for facebook and twittter
- rather than saving the whole URL, the Twitter username incl. '@' will be extracted from URL and saved in the settings. The User will still input the full URL. After saving the blog setting, the stored Twitter username will be parsed again as the full URL and available in the input field. A custom transform is used for this.
- adding meta fields to be rendered in {{ghost_head}}:
- '<meta property="article:publisher" content="https://www.facebook.com/page" />' and
- '<meta name="twitter:site" content="@user"/>'
- adds facebook and twitter to unit test for structured data
- adds unit test for general settings
- adds acceptance test for new input fields in general settings
- adds a custom transform for twitter model to save only the username to the server
- adds unit test for transform
