demonstrates adding an instance-level endpoint for Conan to workhorse. You can also see the Maven project level endpoint
implemented in the same file.
Once the route has been added, you will need to add an additional `/authorize` version of the upload endpoint to your API file.
[Here is an example](https://gitlab.com/gitlab-org/gitlab/blob/398fef1ca26ae2b2c3dc89750f6b20455a1e5507/ee/lib/api/maven_packages.rb#L164)
of the additional endpoint added for Maven. The `/authorize` endpoint verifies and authorizes the request from workhorse,
then the normal upload endpoint is implemented below, consuming the metadata that workhorse provides in order to
create the package record. Workhorse provides a variety of file metadata such as type, size, and different checksum formats.
For testing purposes, you may want to [enable object storage](https://gitlab.com/gitlab-org/gitlab-development-kit/blob/master/doc/howto/object_storage.md)
in your local development environment.
## Services and finders
### Services and finders
Logic for performing tasks such as creating package or package file records or finding packages should not live
within the API file, but should live in services and finders. Existing services and finders should be used or
extended when possible to keep the common package logic grouped as much as possible.
## Configuration
### Configuration
GitLab has a `packages` section in its configuration file (`gitlab.rb`).
It applies to all package systems supported by GitLab. Usually you don't need
...
...
@@ -119,7 +96,96 @@ to add anything there.
Packages can be configured to use object storage, therefore your code must support it.
## Database and handling metadata
## MVC Approach
The way new package systems are integrated in GitLab is using an [MVC](https://about.gitlab.com/handbook/values/#minimum-viable-change-mvc). Therefore, the first iteration should support the bare minimum user actions:
- Authentication
- Uploading a package
- Pulling a package
- Required actions
Required actions are all the additional requests that GitLab will need to handle so the corresponding package manager CLI can work properly. It could be a search feature or an endpoint providing meta information about a package. For example:
- For NuGet, the search request was implemented during the first MVC iteration, to support Visual Studio.
- For NPM, there is a metadata endpoint used by `npm` to get the tarball url.
For the first MVC iteration, it's recommended to stay at the project level of the [remote hierarchy](#remote-hierarchy). Other levels can be tackled with [future Merge Requests](#future-work).
There are usually 2 phases for the MVC:
-[Analysis](#analysis)
-[Implementation](#implementation)
### Keep iterations small
When implementing a new package manager, it is tempting to create one large merge request containing all of the
necessary endpoints and services necessary to support basic usage. Instead, put the
API endpoints behind a [feature flag](feature_flags/development.md) and
submit each endpoint or behavior (download, upload, etc) in a different merge request to shorten the review
process.
### Analysis
During this phase, the idea is to collect as much information as possible about the API used by the package system. Here some aspects that can be useful to include:
-**Authentication**: What authentication mechanisms are available (OAuth, Basic
Authorization, other). Keep in mind that GitLab users will often want to use their
Although not needed for the MVC first iteration, the [CI job tokens](../user/project/new_ci_build_permissions_model.md#job-token)
have to be supported at some point in the future.
-**Requests**: Which requests are needed to have a working MVC. Ideally, produce
a list of all the requests needed for the MVC (including required actions). Further
investigation could provide an example for each request with the request and the response bodies.
-**Upload**: Carefully analyse how the upload process works. This will probably be the most
complex request to implement. A detailed analysis is desired here as uploads can be
encoded in different ways (body or multipart) and can even be in a totally different
format (for example, a JSON structure where the package file is a Base64 value of
a particular field). These different encodings lead to slightly different implementations
on GitLab and GitLab Workhorse. For more detailed information, review [file uploads](#file-uploads).
-**Endpoints**: Suggest a list of endpoint URLs that will be implemented in GitLab.
-**Split work**: Suggest a list of changes to do to incrementally build the MVC.
This will give a good idea of how much work there is to be done. Here is an example
list that would need to be adapted on a case by case basis:
1. Empty file structure (API file, base service for this package)
1. Authentication system for "logging in" to the package manager
1. Identify metadata and create applicable tables
1. Workhorse route for [object storage direct upload](uploads.md#direct-upload)
1. Endpoints required for upload/publish
1. Endpoints required for install/download
1. Endpoints required for required actions
The analysis usually takes a full milestone to complete, though it's not impossible to start the implementation in the same milestone.
In particular, the upload request can have some [requirements in the GitLab Workhorse project](#file-uploads). This project has a different release cycle than the rails backend. It's **strongly** recommended that you open an issue there as soon as the upload request analysis is done. This way GitLab Worhorse is already ready when the upload request is implemented on the rails backend.
### Implementation
The implementation of the different Merge Requests will vary between different package system integrations. Contributors should take into account some important aspects of the implementation phase.
#### Authentication
The MVC must support [Personal Access Tokens](../user/profile/personal_access_tokens.md) right from the start. We currently support two options for these tokens: OAuth and Basic Access.
OAuth authentication is already supported. You can see an example in the [npm API](https://gitlab.com/gitlab-org/gitlab/-/blob/master/ee/lib/api/npm_packages.rb).
support is done by overriding a specific function in the API helpers, like
[this example in the Conan API](https://gitlab.com/gitlab-org/gitlab/-/blob/master/ee/lib/api/conan_packages.rb).
For this authentication mechanism, keep in mind that some clients can send an unauthenticated
request first, wait for the 401 Unauthorized response with the [`WWW-Authenticate`](https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/WWW-Authenticate)
field, then send an updated (authenticated) request. This case is more involved as
GitLab needs to handle the 401 Unauthorized response. The [Nuget API](https://gitlab.com/gitlab-org/gitlab/-/blob/master/ee/lib/api/nuget_packages.rb)
supports this case.
#### Authorization
There are project and group level permissions for `read_package`, `create_package`, and `destroy_package`. Each
endpoint should
[authorize the requesting user](https://gitlab.com/gitlab-org/gitlab/blob/398fef1ca26ae2b2c3dc89750f6b20455a1e5507/ee/lib/api/conan_packages.rb)
against the project or group before continuing.
#### Database and handling metadata
The current database model allows you to store a name and a version for each package.
Every time you upload a new package, you can either create a new record of `Package`
...
...
@@ -137,44 +203,41 @@ delegate from the package model.
Note that the existing package UI only displays information within the `packages_packages` and `packages_package_files`
tables. If the data stored in the metadata tables need to be displayed, a ~frontend change will be required.
## Authorization
#### File uploads
There are project and group level permissions for `read_package`, `create_package`, and `destroy_package`. Each
endpoint should
[authorize the requesting user](https://gitlab.com/gitlab-org/gitlab/blob/398fef1ca26ae2b2c3dc89750f6b20455a1e5507/ee/lib/api/conan_packages.rb#L84)
against the project or group before continuing.
## Keep iterations small
When implementing a new package manager, it is easy to end up creating one large merge request containing all of the
necessary endpoints and services necessary to support basic usage. If this is the case, consider putting the
API endpoints behind a [feature flag](feature_flags/development.md) and
submitting each endpoint or behavior (download, upload, etc) in different merge requests to shorten the review
process.
File uploads should be handled by GitLab Workhorse using object accelerated uploads. What this means is that
the workhorse proxy that checks all incoming requests to GitLab will intercept the upload request,
upload the file, and forward a request to the main GitLab codebase only containing the metadata
and file location rather than the file itself. An overview of this process can be found in the
demonstrates adding an instance-level endpoint for Conan to workhorse. You can also see the Maven project level endpoint
implemented in the same file.
#### MVC MRs
Once the route has been added, you will need to add an additional `/authorize` version of the upload endpoint to your API file.
[Here is an example](https://gitlab.com/gitlab-org/gitlab/blob/398fef1ca26ae2b2c3dc89750f6b20455a1e5507/ee/lib/api/maven_packages.rb#L164)
of the additional endpoint added for Maven. The `/authorize` endpoint verifies and authorizes the request from workhorse,
then the normal upload endpoint is implemented below, consuming the metadata that workhorse provides in order to
create the package record. Workhorse provides a variety of file metadata such as type, size, and different checksum formats.
These changes represent all that is needed to deliver a minimally usable package management system.
For testing purposes, you may want to [enable object storage](https://gitlab.com/gitlab-org/gitlab-development-kit/blob/master/doc/howto/object_storage.md)
in your local development environment.
1. Empty file structure (API file, base service for this package)
1. Authentication system for 'logging in' to the package manager
1. Identify metadata and create applicable tables
1. Workhorse route for [object storage direct upload](uploads.md#direct-upload)
1. Endpoints required for upload/publish
1. Endpoints required for install/download
1. Endpoints required for remove/delete
### Future Work
#### Possible post-MVC MRs
While working on the MVC, contributors will probably find features that are not mandatory for the MVC but can provide a better user experience. It's generally a good idea to keep an eye on those and open issues.
These updates are not essential to be able to publish and consume packages, but may be desired as the system is
released for general use.
Here are some examples
1. Endpoints required for search
1. Front end updates to display additional package information and metadata
1. Limits on file sizes
1. Tracking for metrics
1. Read more metadata fields from the package to make it available to the front end. For example, it's usual to be able to tag a package. Those tags can be read and saved by backend and then displayed on the packages UI.
1. Endpoints for the upper levels of the [remote hierarchy](#remote-hierarchy). This step might need to create a [naming convention](#naming-conventions)
