PVCs and CloudNativePG Clusters need S3 buckets and backup schedules
provisioned consistently. This operator watches the
backups.unkin.net/{schedule,destination} annotations on those objects and
provisions everything needed to back them up, with no new CRDs.
- Add a PVC controller that provisions cephrgw ObjectStoreUser/Bucket/BucketAccess,
auto-generates a restic repo-password Secret and creates a k8up Schedule scoped
to the PVC via spec.backup.volumes[].persistentVolumeClaim.claimName.
- Add a CNPG Cluster controller that provisions the same bucket stack, idempotently
patches spec.backup.barmanObjectStore (leaving a user-set destinationPath alone
with a Warning event) and creates a ScheduledBackup.
- Resolve destinations through a ConfigMap lookup table; requeue until the
BucketAccess is Ready before creating schedule resources; own-reference created
resources and retain bucket data by default.
- Add schedule-mapping helpers (k8up 5-field/shortcut pass-through, CNPG 6-field
seconds-first) and deterministic, length-bounded name derivation.
- Add unit tests (schedule mapping, name derivation, destination resolution) and
envtest controller tests for both paths, wiring the external CRDs into envtest.
- Add kubebuilder-generated RBAC, a Dockerfile (distroless/nonroot), Woodpecker
lint/test/build pipelines and a tag-triggered image push to the artifactapi
docker-internal registry, plus a version-bump Makefile and deploy manifests.
jsonutils
jsonutils exposes a few tools to work with JSON:
- a fast, simple
Concatto concatenate (not merge) JSON objects and arrays FromDynamicJSONto convert a data structure into a "dynamic JSON" data structureReadJSONandWriteJSONbehave likejson.Unmarshalandjson.Marshal, with the ability to use another underlying serialization library through anAdapterconfigured at runtime- a
JSONMapSlicestructure that may be used to store JSON objects with the order of keys maintained
Dynamic JSON
We call "dynamic JSON" the go data structure that results from unmarshaling JSON like this:
var value any
jsonBytes := `{"a": 1, ... }`
_ = json.Unmarshal(jsonBytes, &value)
In this configuration, the standard library mappings are as follows:
| JSON | go |
|---|---|
number |
float64 |
string |
string |
boolean |
bool |
null |
nil |
object |
map[string]any |
array |
[]any |
Map slices
When using JSONMapSlice, the ordering of keys is ensured by replacing
mappings to map[string]any by a JSONMapSlice which is an (ordered)
slice of JSONMapItems.
Notice that a similar feature is available for YAML (see yamlutils),
with a YAMLMapSlice type based on the JSONMapSlice.
JSONMapSlice is similar to an ordered map, but the keys are not retrieved
in constant time.
Another difference with the the above standard mappings is that numbers don't always map
to a float64: if the value is a JSON integer, it unmarshals to int64.
See also some examples
Adapters
ReadJSON, WriteJSON and FromDynamicJSON (which is a combination of the latter two)
are wrappers on top of json.Unmarshal and json.Marshal.
By default, the adapter merely wraps the standard library.
The adapter may be used to register other JSON serialization libraries, possibly several ones at the same time.
If the value passed is identified as an "ordered map" (i.e. implements ifaces.Ordered
or ifaces.SetOrdered, the adapter favors the "ordered" JSON behavior and tries to
find a registered implementation that support ordered keys in objects.
Our standard library implementation supports this.
As of v0.25.0, we support through such an adapter the popular mailru/easyjson
library, which kicks in when the passed values support the easyjson.Unmarshaler
or easyjson.Marshaler interfaces.
In the future, we plan to add more similar libraries that compete on the go JSON serializers scene.
Registering an adapter
In package github.com/go-openapi/swag/easyjson/adapters, several adapters are available.
Each adapter is an independent go module. Hence you'll pick its dependencies only if you import it.
At this moment we provide:
stdlib: JSON adapter based on the standard libraryeasyjson: JSON adapter based on thegithub.com/mailru/easyjson
The adapters provide the basic Marshal and Unmarshal capabilities, plus an implementation
of the MapSlice pattern.
You may also build your own adapter based on your specific use-case. An adapter is not required to implement all capabilities.
Every adapter comes with a Register function, possibly with some options, to register the adapter
to a global registry.
For example, to enable easyjson to be used in ReadJSON and WriteJSON, you would write something like:
import (
"github.com/go-openapi/swag/jsonutils/adapters"
easyjson "github.com/go-openapi/swag/jsonutils/adapters/easyjson/json"
)
func init() {
easyjson.Register(adapters.Registry)
}
You may register several adapters. In this case, capability matching is evaluated from the last registered adapters (LIFO).