kubevela.github.io/docs/platform-engineers/cue/patch-strategy.md

281 lines
6.5 KiB
Markdown

---
title: Patch strategy
---
By default, KubeVela will merge patched values with CUE's merge. However, CUE cannot handle conflicting fields currently.
KubeVela provides a series of patching strategies to help resolve conflicting issues. When writing patch traits and workflow steps, you can use these patch strategies to solve conflicting values. Note that the patch strategy is not an official capability provided by CUE, but an extension developed by KubeVela.
> For more information about how to patch definitions, please refer to [Patch in the Definitions](../traits/patch-trait).
Let's write an env-patch trait to show how to use these patch strategies.
## patchKey
If you want to add multiple environment variables for a specific container, you can use the `+patchKey=name` annotation to find the container. In this case, KubeVela will merge these environment variables by default. This means that `patchKey` cannot handle duplicate fields.
> After KubeVela version 1.4, you can use `,` to split multiple patchKeys, such as `patchKey=name,image`.
Apply the following definition to your cluster:
```yaml
myenv: {
type: "trait"
annotations: {}
labels: {
"ui-hidden": "true"
}
description: "Add env on K8s pod for your workload which follows the pod spec in path 'spec.template'"
attributes: appliesToWorkloads: ["*"]
}
template: {
patch: {
spec: template: spec: {
// +patchKey=name
containers: [{
name: context.name
env: [
for k, v in parameter.env {
name: k
value: v
},
]
}]
}
}
parameter: {
env: [string]: string
}
}
```
Use the above `myenv` trait in your application:
```yaml
apiVersion: core.oam.dev/v1beta1
kind: Application
metadata:
name: webservice-app
spec:
components:
- name: express-server
type: webservice
properties:
image: crccheck/hello-world
env:
- name: OLD
value: old
traits:
- type: myenv
properties:
env:
NEW: new
```
Before using the `myenv` patch trait, the `env` in the application is like:
```
spec:
containers:
- env:
- name: OLD
value: old
```
After using the `myenv` patch trait, the `env` in the application is like:
```
spec:
containers:
- env:
- name: OLD
value: old
- name: NEW
value: new
```
Finally, we can see that the application's `env` contains two environment variables: `OLD=old` and `NEW=new`.
## retainKeys
You can use the `+patchStrategy=retainKeys` annotation if you want to be able to override duplicate values while merging variables.
The strategy of this annotation is similar to the Kubernetes official [retainKeys](https://kubernetes.io/docs/tasks/manage-kubernetes-objects/update-api-object-kubectl-patch/#use-strategic-merge-patch- to-update-a-deployment-using-the-retainkeys-strategy) strategy.
> In the following example, `+patchKey=name` specifies which container the patch should be applied to, while `+patchStrategy=retainKeys` specifies that when merge environment variables, if a duplicate environment variable name is specified, the environment variable value will be overwritten.
```yaml
myenv: {
type: "trait"
annotations: {}
labels: {
"ui-hidden": "true"
}
description: "Add env on K8s pod for your workload which follows the pod spec in path 'spec.template'"
attributes: appliesToWorkloads: ["*"]
}
template: {
patch: {
spec: template: spec: {
// +patchKey=name
containers: [{
name: context.name
// +patchStrategy=retainKeys
env: [
for k, v in parameter.env {
name: k
value: v
},
]
}]
}
}
parameter: {
env: [string]: string
}
}
```
Use the above `myenv` trait in your application:
```yaml
apiVersion: core.oam.dev/v1beta1
kind: Application
metadata:
name: webservice-app
spec:
components:
- name: express-server
type: webservice
properties:
image: crccheck/hello-world
env:
- name: OLD
value: old
- name: OLD2
value: old2
traits:
- type: myenv
properties:
env:
NEW: new
OLD2: override
```
Before using the `myenv` patch trait, the `env` in the application is like:
```
spec:
containers:
- env:
- name: OLD
value: old
- name: OLD2
value: old2
```
After using the `myenv` patch trait, the `env` in the application is like:
```
spec:
containers:
- env:
- name: OLD
value: old
- name: OLD2
value: override
- name: NEW
value: new
```
Finally, we can see that the application's `env` contains three environment variables: `OLD=old`, `OLD2=override` and `NEW=new`.
## replace
If you wish to replace the entire env array directly, you can use the `+patchStrategy=replace` annotation.
> In the following example, `+patchKey=name` specifies which container the patch should be applied to, while `+patchStrategy=replace` specifies that when merge the arrays, the entire array of environment variables will be replaced directly.
```yaml
myenv: {
type: "trait"
annotations: {}
labels: {
"ui-hidden": "true"
}
description: "Add env on K8s pod for your workload which follows the pod spec in path 'spec.template'"
attributes: appliesToWorkloads: ["*"]
}
template: {
patch: {
spec: template: spec: {
// +patchKey=name
containers: [{
name: context.name
// +patchStrategy=replace
env: [
for k, v in parameter.env {
name: k
value: v
},
]
}]
}
}
parameter: {
env: [string]: string
}
}
```
Use the above `myenv` trait in your application:
```yaml
apiVersion: core.oam.dev/v1beta1
kind: Application
metadata:
name: webservice-app
spec:
components:
- name: express-server
type: webservice
properties:
image: crccheck/hello-world
env:
- name: OLD
value: old
- name: OLD2
value: old2
traits:
- type: myenv
properties:
env:
NEW: replace
```
Before using the `myenv` patch trait, the `env` in the application is like:
```
spec:
containers:
- env:
- name: OLD
value: old
- name: OLD2
value: old2
```
After using the `myenv` patch trait, the `env` in the application is like:
```
spec:
containers:
- env:
- name: NEW
value: replace
```
Finally, we can see that the application's `env` contains one environment variable: `NEW=replace`.