ref https://linear.app/tryghost/issue/ENG-790/remove-use-of-sub-queries-in-email-analytics
- the `delivered_at` column is typically entirely/nearly entirely filled with values meaning the `IS NOT NULL` query matches a huge number of rows that MySQL has to fetch from the index to count
- using `IS NULL` switches that behaviour around as it will now match very few rows which has been shown in testing to be considerably quicker
- after switching to `IS NULL` the query returns an "undelivered" count rather than a "delivered" count, in order to keep the rest of the system behaviour the same we can calculate the delivered count by subtracting the query result from the total number of emails sent which we can fetch using a very fast primary key lookup query on the `emails` table
ref ENG-728
ref https://linear.app/tryghost/issue/ENG-728
This is NOT a functionality change. The Post#permissible method unit
tests have been updated to pass `true` as `hasUserPermission` and we can
see that the permission functionality remains the same.
The permissible method of the post model is responsible for removing
permission based on the data that is being modified, but the permissions
module is setup to allow the permissible method to grant permission -
this means that we call permissible, even if the current actor doesn't
have permission, this results in code that is hard to understand and
manage.
We are going to be instead returning early if an actor does not have
permission, this will allow permissible method signatures to be greatly
simplified (removing the need for hasUserPermission, hasApiKeyPermission
& hasMemberPermission arguments).
closes https://linear.app/tryghost/issue/ENG-721
ref https://linear.app/tryghost/issue/ENG-708
Comments-UI loads `/ghost/admin-frame/` in an iframe to check if a Staff User is authenticated in order to show moderation options. That iframe request loads a HTML page which in turn contains a script that fires off an API request that attempts to fetch the logged-in user details, resulting in a 403 "error" showing up when not authenticated. In the vast majority of cases there will be no staff user authenticated so lots of extra requests and "errors" are seen unnecessarily.
- adjusted the `/ghost/auth-frame/` endpoint to check if the request contains an Admin session cookie
- if it does, continue as before with rendering the HTML page so the script is loaded
- if it doesn't, return an empty 204 response avoiding the script request and subsequent 403-generating API request
- eliminates the 403 error being generated for all typical visitor traffic, the error should only be seen when an Admin was previously logged in but their cookie is no longer valid (either from logging out, or going past the 6month validity period)
closes ENG-627
We were using `cheerio` to parse+modify+serialize our rendered HTML to modify links for member attribution. Cheerio's serializer has a [long-standing issue](https://github.com/cheeriojs/cheerio/issues/720) (that we've [had to deal with before](https://github.com/TryGhost/SDK/issues/124)) where it replaces single-quote attributes with double-quote attributes. That was resulting in broken rendering when content used single-quotes such as in HTML cards that have JSON data inside a `data-` attribute or otherwise used single-quotes to avoid escaping double-quotes in an attribute value.
- swapped the implementation that uses `cheerio` for one that uses `html5parser` to tokenize the html string, from there we can loop over the tokens and replace the href attribute values in the original string without touching any other part of the content. Avoids a full parse+serialize process which is both more costly and can result unexpected content changes due to serializer opinions.
- fixes the quote change bug
- uses tokenization directly to avoid cost of building a full AST
- updated Content API Posts snapshot
- one of our fixtures has a missing closing tag which we're no longer "fixing" with a full parse+serialize step in the link replacer (keeps modified src closer to original and better matches behaviour elsewhere in the app / without member-attribution applied)
- the link replacer no longer converts `attr=""` to `attr` (these are equivalent in the HTML spec so no change in behaviour other than preserving the original source html)
- added a benchmark test file comparing the two implementations because the link replacer runs on render so it's used in a hot path
- new implementation has a 3x performance improvement
- the separate files with the old/new implementations have been cleaned up but I've left the benchmark test file in place for future reference
Benchmark results comparing implementations:
```
❯ node test/benchmark.js
LinkReplacer
├─ cheerio: 5.03K /s ±2.20%
├─ html5parser: 16.5K /s ±0.43%
Completed benchmark in 0.9976526670455933s
┌─────────────┬─────────┬────────────┬─────────┬───────┐
│ (index) │ percent │ iterations │ current │ max │
├─────────────┼─────────┼────────────┼─────────┼───────┤
│ cheerio │ '' │ '5.03K/s' │ 5037 │ 5037 │
│ html5parser │ '' │ '16.5K/s' │ 16534 │ 16534 │
└─────────────┴─────────┴────────────┴─────────┴───────┘
```
refs TryGhost/Product#4125
This PR adds two new integration tests to ensure all our Koenig cards
are rendered properly after going through the EmailRenderer. Although we
have thorough tests for the cards themselves in the Koenig repo, the
EmailRenderer does post-processing on the rendered HTML, such as
inlining CSS, which can adversely impact the rendered output of our
cards in email clients (usually Outlook).
Since email newsletters are a core feature of Ghost, these bugs are
typically fairly urgent, and since it is email, they are also quite
difficult to troubleshoot and fix. These two tests are intended to
prevent bugs of this sort, which in the past have been created by
seemingly harmless changes like bumping dependencies that are used in
the EmailRenderer.
The idea is to create a 'Golden Post' which has at least 1 of every card
from Koenig, run that post through the EmailRenderer, and take a
snapshot of the rendered HTML. In the future, if we make any changes to
the EmailRenderer or the Koenig cards themselves, this will trigger us
to carefully consider the changes, and it provides an 'expected' output
to compare our changes against.
Additionally, the second test simply checks that all cards from
`kg-default-nodes` are included in the 'Golden Post'. This protects
against any new cards that we will add in the future — as soon as we add
them to Koenig and bump `kg-default-nodes` in Ghost, this test will
fail, prompting us to add the new card to the Golden Post and update the
snapshots.
We should also run the 'Golden Post' through a test in Litmus, which
allows us to visually inspect the rendered email across many different
email clients. Ideally we would create a process to review the output of
the 'Golden Post' in Litmus whenever we update the snapshot as well.
fixes PROD-61
This adds a new default plan setting. It defaults to yearly, which is
the current default selected interval in Portal.
Behind the new portal improvements feature flag, the default plan can be
changed. It will also change automatically if the available intervals
are changed.
This PR also wires up passing the new setting to the Portal preview.
ref GRO-54
fixes GRO-63
fixes GRO-62
fixes GRO-69
When the config `hostSettings:managedEmail:enabled` is enabled, or the
new flag (`newEmailAddresses`) is enabled for self-hosters, we'll start
to check the from addresses of all outgoing emails more strictly.
- Current flow: nothing changes if the managedEmail config is not set or
the `newEmailAddresses` feature flag is not set
- When managedEmail is enabled: never allow to send an email from any
chosen email. We always use `mail.from` for all outgoing emails. Custom
addresses should be set as replyTo instead. Changing the newsletter
sender_email is not allowed anymore (and ignored if it is set).
- When managedEmail is enabled with a custom sending domain: if a from
address doesn't match the sending domain, we'll default to mail.from and
use the original as a replyTo if appropriate and only when no other
replyTo was set. A newsletter sender email addresss can only be set to
an email address on this domain.
- When `newEmailAddresses` is enabled: self hosters are free to set all
email addresses to whatever they want, without verification. In addition
to that, we stop making up our own email addresses and send from
`mail.from` by default instead of generating a `noreply`+ `@` +
`sitedomain.com` address
A more in depth example of all cases can be seen in
`ghost/core/test/integration/services/email-addresses.test.js`
Includes lots of new E2E tests for most new situations. Apart from that,
all email snapshots are changed because the from and replyTo addresses
are now included in snapshots (so we can see unexpected changes in the
future).
Dropped test coverage requirement, because tests were failing coverage
locally, but not in CI
Fixed settings test that set the site title to an array - bug tracked in
GRO-68
fixes GRO-34
fixes GRO-33
This also adds a new way to run all tests with enforced numeric ObjectIDs.
These numeric ids cause issues if they are used withing NQL filters. So they
surface tiny bugs in our codebase.
You can run tests using this option via:
NUMERIC_IDS=1 yarn test:e2e
Also removed some defensive logic that could be explained by unquoted ids.
closes https://github.com/TryGhost/Product/issues/4075
- when a member clicks on "Unsubscribe from that list" from Apple Mail,
the member's email is put into Mailgun's Unsubscribe suppression list.
Ghost listens for "Unsubscribe" events from Mailgun, and unsubscribes
the member from all the newsletters
- now, the member is only unsubscribed from the newsletter they
unsubscribe to (not all of them)
- now, the email is also deleted from Mailgun's suppression list, so
that it doesn't affect any other membership
no issue
- flag is no longer used in Admin so we can clean it up in Core too
- updated Post model to set blank document to `lexical` field rather than `mobiledoc` as a default value
- switched over to returning `mobiledoc,lexical` as default formats in Admin API
no issue
- Casper and Source theme files were out of date — this commit updates the theme fixtures, and fixes up a few tests to pass with the updated themes
refs TryGhost/Product#3510
- Added `TryGhost/Source` as a submodule in `ghost/core/content/themes` so `Source` will ship with Ghost (along with Casper)
- With this change, new installs will use `Source` as the default theme. Existing sites will have `Source` installed, but not activated, as this is a large change and we don't want to drastically change existing sites without warning. Users can upgrade to use `Source` simply by clicking 'Activate' in design settings.
- Updated protections to prevent users from uploading their own conflicting version of `Source`
no issue
Collection cards contain dynamic data that can change when there's any change to a published post but in Ghost all post/page content is rendered once on save and stored as a static string meaning we need a new approach for triggering a re-render of pages that plays well with caching.
- fixed typo in the relations/authors code that meant we weren't correctly calling the prototype method on the Post model inside the `onFetchedCollection` event handler
- updated Post model to clear the `html` field of all pages when saving or deleting a published post
- updated Post model to re-render `html` fields when fetching individual posts or a collection of posts
- modified `insertExtraPostsTags` fixture util to wrap it's concurrent post edits in a transaction otherwise MySQL errors because it hits a deadlock
refs https://ghost.slack.com/archives/C02G9E68C/p1695296293667689
- We block all outgoing networking by default in tests. When editing a
post/page, Ghost tries to send a webmention. Because of my earlier
changes
(1e3232cf82)
it tries 3 times - all network requests fail instantly when networking
is disabled - but it adds a delay in between those retries. So my change
disables retries during tests.
- Adds a warning when afterEach hook tooks longer than 2s due to
awaiting jobs/events
- We reduce the amount of webmentions that are send, by only sending
webmentions for added or removed urls, no longer when a post html is
changed (unless post is published/removed).
This reverts commit 3e9da6df0c.
- changes introduced an error fetching `/admin/pages/` when using MySQL
- "The values in where clause must not be object or array"
no issue
Collection cards contain dynamic data that can change when there's any change to a published post but in Ghost all post/page content is rendered once on save and stored as a static string meaning we need a new approach for triggering a re-render of pages that plays well with caching.
- fixed typo in the relations/authors code that meant we weren't correctly calling the prototype method on the Post model inside the `onFetchedCollection` event handler
- updated Post model to clear the `html` field of all pages when saving or deleting a published post
- updated Post model to re-render `html` fields when fetching individual posts or a collection of posts
- modified `insertExtraPostsTags` fixture util to wrap it's concurrent post edits in a transaction otherwise MySQL errors because it hits a deadlock
fixes https://github.com/TryGhost/Product/issues/3752
- Added some extra tests for edge cases
- Updated handling of multiple subscriptions so they are handled better
- Canceling a subscription when the member still has other subscriptions will now get handled correctly where the status and products of the member stay intact
fixes https://github.com/TryGhost/Product/issues/3723
This also fixes usage of localhost instead of 127.0.0.1 as a test URL
for playwright. This caused issues for cookies because the member
impersonation navigated to 127.0.0.1 instead of localhost, meaning that
the next page.goto call would go to localhost and lose the cookies.
fixes https://github.com/TryGhost/Product/issues/3687
After this change, relative URLs in emails will be replaced with
absolute URLs using the post URL. Making relative Portal URLs possible
etc.
Updates the test data generator to fix invalid URL encoding (somehow a
backslash + escaped double quote was added when it wasn't required).
refs https://github.com/TryGhost/Product/issues/3676
- add filter for sidebar display of theme errors (angry red box)
- filter specific to each page feature, will need to add each one by this approach
refs https://github.com/TryGhost/Arch/issues/25
- The instance should have two built-in collections "latest" (prviously known as "index") and "featured". These have been filled through in-memory tricks before, now they should come pre-populated through fixtures mechanism.
refs: https://github.com/TryGhost/Toolbox/issues/595
We're rolling out new rules around the node assert library, the first of which is enforcing the use of assert/strict. This means we don't need to use the strict version of methods, as the standard version will work that way by default.
This caught some gotchas in our existing usage of assert where the lack of strict mode had unexpected results:
- Url matching needs to be done on `url.href` see aa58b354a4
- Null and undefined are not the same thing, there were a few cases of this being confused
- Particularly questionable changes in [PostExporter tests](c1a468744b) tracked [here](https://github.com/TryGhost/Team/issues/3505).
- A typo see eaac9c293a
Moving forward, using assert strict should help us to catch unexpected behaviour, particularly around nulls and undefineds during implementation.
refs https://github.com/TryGhost/Toolbox/issues/592
- it turns out that `TRUNCATE` in CI takes ~300ms for all tables, but
`DELETE FROM` takes ~30ms
- whilst truncating is generally known to be faster, I believe it's only
faster on large tables
- this saves 90% of the time it takes to reset the DB in MySQL
refs https://github.com/TryGhost/Toolbox/issues/592
- async-await makes the code easier to read
- also performs a small optimization to only load the foreign_keys
pragma once for SQLite
refs https://github.com/TryGhost/Toolbox/issues/592
- we should reset the URL service to avoid event listeners piling up and
slowing down CI due to the number of events it has to process
