no issue
- unsued code:
- there are no public assets anymore, might need to use this instead of shared in future, but for now lets remove it to reduce confusion
- the `input password` box was incorrectly registered as an admin helper, thinking that was needed in order to render the default template. This isn't needed.
- apps:
- small structure & comment update to amp app
- moving input_password helper into private blogging app
- refactor helpers in subscribers app
- 🛠 add bunyan and prettyjson, remove morgan
- ✨ add logging module
- GhostLogger class that handles setup of bunyan
- PrettyStream for stdout
- ✨ config for logging
- @TODO: testing level fatal?
- ✨ log each request via GhostLogger (express middleware)
- @TODO: add errors to output
- 🔥 remove errors.updateActiveTheme
- we can read the value from config
- 🔥 remove 15 helper functions in core/server/errors/index.js
- all these functions get replaced by modules:
1. logging
2. error middleware handling for html/json
3. error creation (which will be part of PR #7477)
- ✨ add express error handler for html/json
- one true error handler for express responses
- contains still some TODO's, but they are not high priority for first implementation/integration
- this middleware only takes responsibility of either rendering html responses or return json error responses
- 🎨 use new express error handler in middleware/index
- 404 and 500 handling
- 🎨 return error instead of error message in permissions/index.js
- the rule for error handling should be: if you call a unit, this unit should return a custom Ghost error
- 🎨 wrap serve static module
- rule: if you call a module/unit, you should always wrap this error
- it's always the same rule
- so the caller never has to worry about what comes back
- it's always a clear error instance
- in this case: we return our notfounderror if serve static does not find the resource
- this avoid having checks everywhere
- 🎨 replace usages of errors/index.js functions and adapt tests
- use logging.error, logging.warn
- make tests green
- remove some usages of logging and throwing api errors -> because when a request is involved, logging happens automatically
- 🐛 return errorDetails to Ghost-Admin
- errorDetails is used for Theme error handling
- 🎨 use 500er error for theme is missing error in theme-handler
- 🎨 extend file rotation to 1w
This release includes:
Ghost Editor, this is required to get access to the built in cards
Ghost Editor includes:
Responsive toolbars 🔨🔧
Both Ember and Plain javascript cards 🎴
An embeded HTML card ✍️
This is still an early release, but things are moving in the right direction. :)
We're still defining the spec for the UI, so expect drastic changes over the next couple of weeks.
This is going to be a great writing experience and we can't wait to show what we have planned.
refs #7452
- remove references to 'patronus' in favour of GhostAuth, Note: this will require databases to be deleted ;)
- remove email addresses from test data
refs #2001, #7116
- added debug and wired it up:
- across several key parts of the boot process
- throughout the middleware loading
- for requests
- at render points for key routes
* 🎨 run database population in transaction
refs #6574, refs #7432
- create transaction for creating tables
- if an error occurs or a container get's destroyed before population finishes, transaction is rolled back
* 🎨 simplify transaction creation and test
issue #7452
Remote oauth2 authentication with Ghost.org.
This PR supports:
- oauth2 login or local login
- authentication on blog setup
- authentication on invite
- normal authentication
- does not contain many, many tests, but we'll improve in the next alpha weeks
no issue
- anonymous functions are hard to debug in memory traces etc
- having anonymous middleware functions makes it hard to inspect or debug the middleware stack (something I like to do)
- these 2 are the only ones atm, including all 3rd party middleware
refs #7189
- we had a memory leak after upgrading to knex 0.11.x
- knex has published a new version 0.12.x
- the memory leak does not longer exists
- knex has reverted their pool logic, see https://github.com/tgriesser/knex/pull/1665
closes#7423
- Extend our dirty theme override cache clear hack to also reset the asset hash
_ This brings alpha into line with the LTS branch
- This still needs a rewrite for Ghost 1.0.0 🙄
- Don't let people start Ghost Alpha with non-alpha databases.
- Provide a new welcome message for development mode (a little bit of positive reinforcment)
- Provide a RED WARNING when in production mode (will still be used for developing, but we can ignore)
- Change package.json to 1.0.0-alpha.0, we won't relelase this, will bump to .1 for release
refs #6982
- create config util fn: getContentPath
- we can later let the user change the folder names in contentPath
- get rid of custom/default storage paths
[ci skip]
refs #6982
- this was handled before by the old ConfigManager
- this logic belongs into the storage logic itself
- for now we only allow uploading themes via the local-file-store
[ci skip]
closes#7313
- Adds `getSanitizedFileName` function to storage/base.js which replaces non A-Z0-9@. chacracters with -
- modifies /api/theme.js so that zip.shortName is consistent throughout.
closes#7192
- add 008 migration
- added script to re-run 006/01
- re-run 006/01 migration for postgres in any timezone (transform formats only)
- re-run 006/01 migration for sqlite (transform formats only)
- rely on format checks for sqlite, do not check server TZ
closes#7350
- When the active theme is overridden, ensure that the activateTheme middleware gets called by removing the `req.app.activeTheme` value.
- Additionally, ensure that the full cache is invalidated
no issue
`{{amp_content}}` helper can handle error now, if returned from `Amperize` module. In case of on error, we return the unprocessed HTML, which will then get validated by the `Sanitize` functionality.
The unprocessed HTML will be stored in the cache, until the post is updated.
Points to Amperize fork of AileenCGN as dependency to have include the error handling changes incl. timeouts.
* 🐛 fix direct update
closes#7297
- move sitemap initialisation into sitemap handler
- initialise sitemap on first request to sitemap
* 🐛 fix how we pass options to migration files
refs #7317
- clone options when passing them into the migration/fixture files
- do not use default sequence, because it does not clone the arguments
refs #7305
* 🎨 display upgrade alerts with the correct "info" style
* 💄 update use of notifications status/type/location attrs to reflect current usage
no issue
Uses `allowedAttributes` functionality of `Sanitize` HTML and whitelists attributes for certain tags, regarding
AMP validation rules.
This PR fixes issues with inline style like `border`, `bgcolor`, `align` and so on.
closes#7277
Adds `.zip` to `extensions` and `application/zip` to `contentTypes` in config, specificly for uploads to `db`, to allow .zip-file file uploads from labs.
no issue
Video tags aren't supported in Amperize yet, therefore, we strip them out. If a `<video>` tag has nested `<source>` elements, they would stay because they are whitelisted regarding `Sanitize`, as we use them for `<audio>` tags as well.
This PR uses `cheerio` to strip out in `<video>` nested `<source>` tags, without removing the fallback text.
It also removes prohibites attributes for `<amp-iframe>` which are e. g. used by Vimeo embeds.
Removes every kind of inline `style` attributes, as they will cause validation errors as well.
no issue
Fixes a bug with displaying small media types like images or gif. Two reasons for that:
1. In many cases, we only have the relative URL instead of the absolute URL for the media source and therefore, `Amperize` module wasn't able to detect the image size and set the default image size of `width="600"` and `height="400"`.
2. Even if we have detected the correct image size, the attribute `layout="default"` would still make it strech. This issue is fixed in `Amperize`, but it wasn't merged at this time, so I set the dependency on my fork.
Adds `amp-anim` to the `.post-content` class, to have same CSS style as an image.
closes#7266, closes#7267
- Adds node-archiver as a dependency
- Adds new zip-folder utility
- Switch out exec 'zip' for zip folder utility
- Store generated zips in os.tmpdir
- Don't delete zips from content/themes when uploading or deleting
- Fixes path resolution for delete
no issue
This PR takes the existing function `processUrls` in `data/xml/rss` and refactors it to be a stand-alone util.
The change is needed, as this functionality will be accessed from `apps/amp` to convert relative URLs.
closes#7186
- Add a concept of validity to each generator
- Refactor base generator to handle invalid (empty) nodes for both events & the initial generation
- Update the tests a bit, to fix some bugs in the tests
- Ensure the homepage is always present
refs #7204
- Adds a new ThemeValidationError class
- This error has a top level message, but will also contain all the individual errors within the `errorDetails` property
- Updated the API error handling to return `errorDetails` if it is present
no issue
- while testing different situations for custom storage adapters after switching from 0.9 to 0.10, it turned out the error output was not optimal
- this PR improves that
no issue
Double slashes are treated as a HTTP calls as specified in [RFC1801](http://www.ietf.org/rfc/rfc1808.txt). Because of this behaviour the uncapitalise created an open redirect. By removing double slashes in the path we ensure open redirects cannot be created.
As an example, please click the following URL: https://dev.ghost.org///Google.com/.
This issue has been reported by pentesters of our product [LearningSpaces.io](http://learningspaces.io).
refs #2852
- improvement: ensure custom storage adapter has required functions
- serve, save and exists are from now on required functions for a custom storage adapter
- add delete as required storage function
closes#6588, #7095
* `ImageObject` with image dimensions (#7152, #7151, #7153)
- Returns meta data as promise
- returns a new Promise from meta data
- uses `Promise.props()` to resolve `getClient()` and `getMetaData()`
- Adds 'image-size' util
The util returns an object like this
```
{
height: 50,
url: 'http://myblog.com/images/cat.jpg',
width: 50
};
```
if the dimensions can be fetched and rejects with error, if not.
In case we get a locally stored image or a not complete url (like `//www.gravatar.com/andsoon`), we add the protocol to the incomplete one and use `urlFor()` to get the absolute URL. If the request fails or `image-size` is not able to read the file, we reject with error.
- adds 'image-size' module to dependencies
- adds `getImageSizeFromUrl` function that returns image dimensions
- In preparation of AMP support and to improve our schema.org JSON-LD and structured data, I made the following changes:
- Changes the following properties to be `Objects`, which have a `url` property by default and a `dimensions` property, if `width` and `height` are available:
- `metaData.coverImage`
- `metaData.authorImage`
- `metaData.blog.logo`
- Checks cache by calling `getCachedImageSizeFromUrl`. If image dimensions were fetched already, returns them from cache instead of fetching them again.
- If we have image dimensions on hand, the output in our JSON-LD changes from normal urls to be full `ImageObjects`. Applies to all images and logos.
- Special case for `publisher.logo` as it has size restrictions: if the image doesn't fulfil the restrictions (<=600 width and <=60 height), we simply output the url instead, so like before.
- Adds new property for schema.org JSON-LD: `mainEntityOfPage` as an Object.
- Adds additional Open Graph data (if we have the image size): `og:image:width` and `og:image:height`
- Adds/updates tests
* AMP router and controller (#7171, #7157)
Implements AMP in `/apps/`:
- renders `amp.hbs` if route is `/:slug/amp/`
- updates `setResponseContext` to set context to `['amp', 'post']` for a amp post and `['amp', 'page']` for a page, but will not render amp template for a page
- updates `context_spec`
- registers 'amp' as new internal app
- adds the `amp.hbs` template to `core/server/apps/amp` which will be the default template for AMP posts.
- adds `isAmpURL` to `post-lookup`
* 🎨 Use `context` in meta as array (#7205)
Instead of reading the first value of the context array, we're checking if it includes certain context values.
This is a preparation change for AMP, where the context will be delivered as `['amp', 'post']`.
* ✨ AMP helpers (#7174, #7216, #7215, #7223)
- Adds AMP helpers `{{amp_content}}`, `{{amp_component}}` and `{{amp_ghost_head}}` to support AMP:
- `{{amp_content}}`:
- Adds `Amperize` as dependency
- AMP app uses new helper `{{amp_content}}` to render AMP HTML
- `Amperize` transforms regular HTML into AMP HTML
- Adds test for `{{amp_content}}` helper
- Adds 'Sanitize-HTML` as dependendy
- After the HTML get 'amperized' we still might have some HTML tags, which are prohibited in AMP HTML, so we use `sanitize-html` to remove those. With every update, `Amperize` gets and it is able to transform more HTML tags, they valid AMP HTML tags (e. g. `video` and `amp-video`) and will therefore not be removed.
- `{{amp_ghost_head}}`:
- registers `{{amp_ghost_head}}` helper, but uses `{{ghost_head}}` code
- uses `{{amp_ghost_head}}` in `amp.hbs` instead of `{{ghost_head}}`
- `{{ghost_head}}`:
- Render `amphtml` link in metadata for post, which links to the amp post (`getAmpUrl`)
- Updates all test in metadata to support `amp` context
- Changes context conditionals to work with full array instead of first array value
- Adds conditionals, so no additional javascript gets rendered in `{{ghost_head}}`
- Removes trailing `/amp/` in URLs, so only `amphtml` link on regular post renders it
- Adds a conditional, so no code injection will be included, for an `amp` context.
- `{{amp_components}}`:
- AMP app uses new helper `{{amp_components}}` to render necessary script tags for AMP extended components as `amp-iframe`, `amp-anime` and `amp-form`
- Adds test for `{{amp_components}}`
closes#7182
When calling `config.set()` in the settings api, we want to set the active timezone of the blog to make it available in our `settingsCache`. But because the `theme` object in the `set` prototype was already set to `Etc/UTC` as default, the `_.merge` function would always overwrite our `activeTimezone` with the default value.
This PR changes the code in the way, that we always set 'Etc/UTC' for the timezone as default, _until_ we fetched our settings and therefore the `activeTimezone` setting, so we can overwrite it.
This issue had not only influence on the date helper, but everywhere in our codebase, where we rely on reading the `timezone` from our config, instead of our settings. The `{{@blog.timezone}}` helper reflected that quiet well, as it would always show `Etc/UTC`
closes#7212
When no Blog title is set and there is a `from` mail adress without custom name set up in `config.js`, it will overwrite `config.theme.title` in `GhostMailer.from` and forces to have a (wrong) Blog title after sending a test mail or an invitation.
- Uses a variable `defaultBlogTitle` instead of overwriting `config.theme.title`, when no blog title is set and therefore `config.theme.title` has no value.
no issue
- Source out validation logic into a upload validation middleware for all upload types (csv, image, subscribers). This unit can be later used for Ghost 1.0 as a pre validation core unit.
- More usage of route tests than controller tests. These are use case tests, a use case only changes if the product changes
no issue
This small change swaps over the order in which the app routes, and catch-all `*` routes are processed.
This will have an impact on the behaviour of all of our internal apps. Both the private blogging and subscribers apps have a route where they render their own template. Private blogging has `/private/` which renders `private.hbs` and subscribers has a similar `/subscribe/` route (and in future `/unsubscribe/`).
Because these routes weren't listed in our reserved words list, it is possible for a blog to already have a post or a page that lives at `/private/` or, perhaps more likely `/subscribe/`. Prior to this change, their already setup page would be rendered instead of the app's page. After this change, the app's own route will be correctly rendered.
This is effectively a bug fix, because if you enable these features then you would expect them to work.
Moving forward, this change is absolutely required for the AMP app, because the route for that app is `*/amp/`. If the app routes aren't processed first, then this will not work.
* Updating Ghost-Admin: Gruntfile work
* move remainder of ember tasks to client repo
refs #6977
- move ember build tasks to client repo
- remove unneeded dependencies
no issue
- sets `settings.activeTimezone` to best-guess based on current server time when performing the timezones migration in order to prevent unexpected changes in timezone when upgrading
refs #6976, #7019, #7125
- Ensure maintenance mode flag is set back to what is in config.js rather than defaulted to false on boot
- Remove stack trace from 503 errors
- Add error message to 503 error
- Ensure error page is rendered for Ghost-Admin on reload with 503
closes#5071
- Send application/json requests to UpdateCheck service. New UpdateCheck service accepts JSON request
- If UpdateCheck service respponse has messages[] array, iterate over the array and create custom notifications intended for current version
- Save custom notification if its not already in the store AND its uuid is not in seenNotifications array
- When a custom notification is dismissed, store its uuid in seenNotifications array
- setup test fixtures to trigger tests properly
- api_notification_spec test to ensure custom notification can be added to store and added to seenNotifications when dismissed
- update_check_spec test to ensure custom notification can be displayed for a specific Ghost version
- added test to ensure messages meant for other versions don't create notifications
closes#6948
- the hbs engine was never initialised when server starts
- when you request a page which does not exist, express jumps directly into the error handlers
- delete some dynamic hbs engine setters in theme handler
closes#6625
- "url" and "author" fields depend on {id, published_at, slug, author_id} to construct post url.
- implemented a generic solution by defining defaultColumnsToFetch() in
base class for models.
- findPage() calls defaultColumnsToFetch() before loading models
- results are transformed by filtering out additional properties to return just the requested fields
- Added a test case to check for url and author fields
- Renamed allColumns as requestedColumns and used _.map instead of Promise.map
closes#6691
- removes dependency on semver & package.json in preinstall script
- has a simplified proxy of semver to look for the right version numbers
refs #6958 (first task of it)
-includes additional tests
-Instead of removing all slashes "/g" we now specifically remove leading and trailing slashes.
closes#6932
- new default order of posts: scheduled, draft, published
- invent orderDefaultRaw fn for each model
- each model is able to create a default raw order query
- separate count and fetch query for fetchPage, because the count query where group/order statements attached
closes#7060
- changed meta referrer from origin to origin-when-cross-origi
- made referrer policy configurable via referrerPolicy option in config js
- added example to config.example.js
-modified test to reflect new defaul origin-when-cross-origin
-added a test for configuration changed referrerPolicy
closes#6984
-the backlink had a static href to {{@blog.url}} which is now changed to {{#if subscribed_url}}{{subscribed_url}}{{else}}{{@blog.url}}{{/if}} to reflect the referring url.
-This PR makes the back link on the subscribe page to link back to the referring page instead of linking back to /
refs #7023
The template can be used for all cases (different newsletter interval, different amount of blog posts, with or without blog picture).
The template can be filled with the following data structure:
- `blog.logo` is the blog logo from settings
- `blog.title` is the title of the current blog
- `blog.url` is the URL of the blog
- `blog.twitterURL` is the twitter profile URL of the blog from settings
- `blog.facebookURL` is the facebook page URL of the blog from settings
- `blog.unsubscribe` is the link for the reader to unsubscribe from the blog
- `newsletter.interval` is the interval in words like 'weekly', 'daily', 'monthly'
- `newsletter.date` is the date of the newsletter issue in a format like 'June 9th, 2016'
- `blog.post` is expected to be an array:
- `blog.post[i].picture` is the picture of the blog post. There are also conditionals, which change the inline CSS to either show a border-top if there's no picture (in that case the HTML code, that shows the picture is not active)
- `blog.post[i].title` is the title of the current blog post
- `blog.post[i].text` is the text of the current blog post, which needs to be cut down to 278 letters, plus ending with `…`
- `blog.post[i].url` is the URL of the current blog post
- `blog.post[i].tag` is the tag of the current blog post
- `blog.post[i].author` is the author of the current blog post
Important is, that only HEX HTML entities will work, especially in Outlook. So instead of `—` we need to use `—` and `’` instead of `'` or `'` and so on.
Added unit test for newsletter template in `test/unit/mail/utils_spec.js`.
no issue
This is a fix for the default-scheduler.
When a post never had a published_at value, the oldTime for removing the job would be null. And in this case we would try to delete a job with an invalidate date.
refs #6413
- PUT endpoint to publish a post/page for the scheduler
- fn endpoint to get all scheduled posts (with from/to query params) for the scheduler
- hardcoded permission handling for scheduler client
- fix event bug: unscheduled
- basic structure for scheduling
- post scheduling basics
- offer easy option to change adapter
- integrate the default scheduler adapter
- update scheduled posts when blog TZ changes
- safety check before scheduler can publish a post (not allowed to publish in the future or past)
- add force flag to allow publishing in the past
- invalidate cache header for /schedules/posts/:id
no issue
- config.theme.timezone can be undefined, when settings are not loaded from the database
- this PR will define the default blog TZ in config
- use `Etc/UTC` as default instead of `Europe/Dublin`
closes#6406
- created listeners.js connector
- merged listeners.js with events.js (in models/base)
- set a post to draft when published_at would be in the past
- reschedule a post when published_at would be in the future
