A Java runner for helmfile.
jhelmfile is a java library which can be used to run the helmfile cli tool.
It offers a convenient API to create and execute helmfile commands like build
, template
, or sync
.
The commands can be executed using a runtime which can be whether a locally installed helmfile binary or a docker image containing helmfile.
The output of helmfile is mapped to java objects which can then be used for further processing.
For example, you could run verifications against them, as you can see in the example project which uses jhelmfile to test a helmfile using junit.
<dependency>
<groupId>io.github.paichinger</groupId>
<artifactId>jhelmfile</artifactId>
<version>1.12</version>
</dependency>
implementation 'io.github.paichinger:jhelmfile:1.12'
For each supported helmfile command (build
, template
and sync
) you can find a builder. See here:
The build
command currently supports the following parameters:
--environment
--state-values-set
--state-values-file
Here an example with the corresponding builder methods:
BuildCommand buildCommand = BuildCommand
.builder()
.helmfileYaml(new File("./my-project/helmfile.yaml")) // mandatory
.environment("acceptance") // optional
.stateValuesFiles(List.of(new File("common-values.yaml"), new File("common-values.yaml"))) // optional
.stateValuesSet(Map.of("myService.version", "v1.2", "myService.logLevel", "DEBUG")) // optional
.build();
The template
command currently supports the following parameters:
--environment
--state-values-set
--state-values-file
--selector
--skip-deps
--skip-needs
--include-transitive-needs
Here an example with the corresponding builder methods:
TemplateCommand templateCommand = TemplateCommand
.builder()
.helmfileYaml(new File("./my-project/helmfile.yaml")) // mandatory
.environment("acceptance") // optional
.stateValuesFiles(List.of(new File("common-values.yaml"), new File("common-values.yaml"))) // optional
.stateValuesSet(Map.of("myService.version", "v1.2", "myService.logLevel", "DEBUG")) // optional
.selectors(List.of(new Selector("group", true, "progressive"))) // optional
.skipDeps(true) // optional - default=false
.skipNeeds(true) // optional - default=false
.includeTransitiveNeeds(true) // optional - default=false
.build();
The sync
command currently supports the following parameters:
--environment
--state-values-set
--state-values-file
--selector
--skip-deps
--skip-needs
--include-transitive-needs
Here an example with the corresponding builder methods:
SyncCommand templateCommand = SyncCommand
.builder()
.helmfileYaml(new File("./my-project/helmfile.yaml")) // mandatory
.environment("acceptance") // optional
.stateValuesFiles(List.of(new File("common-values.yaml"), new File("common-values.yaml"))) // optional
.stateValuesSet(Map.of("myService.version", "v1.2", "myService.logLevel", "DEBUG")) // optional
.selectors(List.of(new Selector("group", true, "progressive"))) // optional
.skipDeps(true) // optional - default=false
.skipNeeds(true) // optional - default=false
.includeTransitiveNeeds(true) // optional - default=false
.build();
The above commands can be executed using one of the available runtimes:
This runtime can be used if you have the helmfile binary available on your machine. Usage:
BinaryRuntime runtime = BinaryRuntime
.builder()
.helmfileBinaryPath("/bin/helmfile")
.build();
You can use the docker runtime if you have docker installed on your machine.
DockerRuntime runtime = DockerRuntime
.builder()
.helmfileBinaryPath("/bin/helmfile") // mandatory - points to to the helmfile binary inside the docker image
.dockerHost("unix:///var/run/docker.sock") // optional - default on linux: "unix:///var/run/docker.sock" on win and mac: "tcp://localhost:2376"
.imageRepository("ghcr.io/helmfile/helmfile") // optional - default "ghcr.io/helmfile/helmfile"
.imageTag("v0.151.0") // optional - default "v0.151.0"
.build();
The runtimes offer methods to execute the supported operations (build
,template
,sync
).
Those methods return POJOs representing the corresponding helmfile output (HelmfileBuild and HelmfileTemplate). Usage:
HelmfileTemplate templateOutput = runtime.template(templateCmmand);
// or
String syncOutput = runtime.sync(syncCommand);
// or
HelmfileBuild buildOutput = runtime.build(buildCommand);
jhelfmile doesn't support all the features the helmifle cli offers yet. However, if you're missing a feature or a command don't hesitate to open an issue or create a PR.
In the examples folder you can find a project which uses jhelmfile to run tests against a helmfile. There, the output of jhelmfile is used to run verifications using junit.