refs: https://github.com/TryGhost/Toolbox/issues/146
Switched to @tryghost/logging instead of passing around the library. The main sticking points of this change are jobs. When jobs are launched we don't want them to use a separate @tryghost/logging instance because they would start parallel rotation jobs. @tryghost/logging v2.x passes all logs to the parent process if run in a child process, so that we can use the same patterns in jobs and the rest of the codebase.
refs https://github.com/TryGhost/Team/issues/1001
We fall back to existing behaviour if no API key is present, or if there
is an error communicating with the Twitter API. We're also currently
requesting all the data, which will be thinned down once we understand
what we need.
This also includes a custom renderer for embeds of type "twitter" which
will be used to output the custom HTML for emails
refs https://github.com/TryGhost/Team/issues/1211
This registers the NFT custom OEmbed provider to the OEmbed service for
the canary API. This should probably be done in a centralised place -
but we do not have a single instance of the OEmbed service.
When we have more information about why the OEmbed service is
instantiated like this, we can think about moving it into a singleton
service with an `init` method - which is where we can register custom
providers.
- Ghost has a set of core packages that it is safe to require directly in any file - tpl is one of them!
- This keeps the DI signature smaller and easier to reason about
refs: #13380
- The i18n package is deprecated. It is being replaced with the tpl package.
Co-authored-by: Aleksander Chromik <aleksander.chromik@footballco.com>
no issue
- Logic with slightly more complex structure belongs to the service. Extracting it there also show's how little of an API the oembed service should actually expose
refs 829e8ed010
- i18n is used everywhere but only requires shared or external packages, therefore it's a good candidate for living in shared
- this reduces invalid requires across frontend and server, and lets us use it everywhere until we come up with a better option
- Having these as destructured from the same package is hindering refactoring now
- Events should really only ever be used server-side
- i18n should be a shared module for now so it can be used everywhere until we figure out something better
- Having them seperate also allows us to lint them properly
closes https://github.com/TryGhost/Ghost/issues/12260
- if a card type was not explicitly chosen (i.e. a url was pasted into the editor) then abort fetching the oembed endpoint if we detect it's a `wp-json` oembed and return a bookmark card payload instead
- cleaned up an unused argument in the internal `fetchBookmarkData()` method
no issue
- known failing case was Medium. Some *.medium.com articles triggered redirects but without cookies being stored/sent across the redirect requests it would cause an infinite redirect loop and we'd abort after 10 redirects
- use `got`'s [cookie support](https://github.com/sindresorhus/got/tree/v9.6.0#cookies) via `tough-cookie` so that we can create bookmark cards for medium.com and other sites with similar problems
no issue
- localhost check was causing an immediate "no provider" error when fetching oembed for a localhost url and bypassing the bookmark fallback even when the configured site lives on localhost
- allow `localhost:port` through in oembed endpoints when it matches the configured url
refs https://github.com/TryGhost/Ghost/issues/11212
- if a bookmark card fetch is performed (either directly or from fallback) and the page does not have an extractable title, return a more specific error message than "No provider found for supplied URL."
no issue
- added an `externalRequest` lib
- uses same underlying `got` module as our `request` lib
- uses `got`'s `beforeRequest` and `beforeRedirect` hooks to perform it's own dns resolution for each url that's encountered and aborts with an error if it resolves to a private IP address block
- includes a bypass for Ghost's configured url so that requests to it's own hostname+port are not blocked
- updated v2 and canary oembed controllers to use the `externalRequest` lib
* refactored `core/frontend/apps` to destructure common imports
* refactored `core/frontend/services/{apps, redirects, routing}` to destructure common imports
* refactored `core/frontend/services/settings` to destructure common imports
* refactored remaining `core/frontend/services` to destructure common imports
* refactored `core/server/adapters` to destructure common imports
* refactored `core/server/data/{db, exporter, schema, validation}` to destructure common imports
* refactored `core/server/data/importer` to destructure common imports
* refactored `core/server/models/{base, plugins, relations}` to destructure common imports
* refactored remaining `core/server/models` to destructure common imports
* refactored `core/server/api/canary/utils/serializers/output` to destructure common imports
* refactored remaining `core/server/api/canary/utils` to destructure common imports
* refactored remaining `core/server/api/canary` to destructure common imports
* refactored `core/server/api/shared` to destructure common imports
* refactored `core/server/api/v2/utils` to destructure common imports
* refactored remaining `core/server/api/v2` to destructure common imports
* refactored `core/frontend/meta` to destructure common imports
* fixed some tests referencing `common.errors` instead of `@tryghost/errors`
- Not all of them need to be updated; only updating the ones that are
causing failures
* fixed errors import being shadowed by local scope
no issue
- embed cards now store metadata including thumbnail urls in their payload
- we want to use this metadata to render video cards in emails
- by default oembed endpoints return fairly small thumbnail images that don't look great when blown up
- oembed supports a `maxwidth` query param that will instruct oembed providers to return larger sizes of the content if available
no issue
- added the same 2sec timeout and `Ghost` user-agent header to the `rel="alternate"` oembed request that we use for the initial html page request
no issue
- prevent oembed fetching from accessing IP addresses or localhost domains
- prevent oembed endpoint from passing through fetched responses as-is
- reject any fetched data that does not validate against the oembed spec
- strip any unknown properties from the oembed response before returning
Credits: Nick Mykhailyshyn
fixes#11636
- malformed URLs passed to oembed API would cause `got` or `metascraper`
to throw an error and this would result in a 500 error from Ghost
- this commit catches the errors and returns a reasonable response
no issue
Current metascraper rule for fetching page metadata in case of bookmark card gives preference to publisher logo over icon tags. This PR updates giving first preference to icon link tags followed by logo.
requires https://github.com/TryGhost/Ghost-Admin/pull/1293
- updates `oembed` endpoint behaviour
- if an oembed provider is not found then we use `metascraper` to populate a metadata object
- when metadata is returned rather than an oembed response the payload will look like this:
```json
{
"url": "...",
"type": "bookmark",
"metadata": {
"url": "...",
"title": "...",
"description": "...",
"author": "...",
"publisher": "...",
"thumbnail": "...",
"icon": "..."
}
}
```
- adds a `bookmark` card which generates output for the bookmark card:
```html
<figure class="kg-card kg-bookmark-card">
<a href="[URL]" class="kg-bookmark-container">
<div class="kg-bookmark-content">
<div class="kg-bookmark-title">[TITLE]</div>
<div class="kg-bookmark-description">[DESCRIPTION]</div>
<div class="kg-bookmark-metadata">
<img src="[ICON]" class="kg-bookmark-icon">
<span class="kg-bookmark-author">[AUTHOR]</span>
<span class="kg-bookmark-publisher">[PUBLISHER]</span>
</div>
</div>
<div class="kg-bookmark-thumbnail">
<img src="[THUMBNAIL]">
</div>
</a>
</figure>
```
- if a particular bit of data does not exist then the associated html element will not be present
