YAML.dump ignores Boolean values

[saskiawagenaar]

Problem:
When performing a YAML.dump on a model object the value of Boolean fields are missing in the output.
For example, when running a YAML.dump on a V1Deployment which has a V1Container where stdin equals true, the value of stdin will not be present in the YAML.dump of the V1Deployment.

Reproduction path:
Load the following yaml into a V1Deployment via Yaml.loadAs:

---
apiVersion: apps/v1
kind: Deployment
spec:
  template:
    spec:
      containers:
      - image: example
        name: example
        stdin: true

Then run a Yaml.dump on the V1Deployment, the result equals:

---
apiVersion: apps/v1
kind: Deployment
spec:
  template:
    spec:
      containers:
      - image: example
        name: example

The value of stdin is missing in the dumped yaml, although it is read in correctly in the V1Deployment object.

Probable cause:
In version 2.3.0 of the Swagger codegeneration a bug was introduced in generating non-primitive Boolean getters. They now follow the naming conventions for primitive booleans, which equal is<booleanName> instead of get<booleanName>. However, this naming convention does not hold for Boolean primitives, their getters should be named get<BooleanName>.
(See also http://www.oracle.com/technetwork/articles/javaee/spec-136004.html, paragraph 8.3)

Snakeyaml uses getter methods to determine which fields are readable. It cannot find a correct getter for all Boolean fields, and therefore it will not serialize this field.

This issue is introduced in the Swagger codegeneration via swagger-api/swagger-codegen#4882 and fixed via OpenAPITools/openapi-generator#432.

Note 1:
Yaml.loadAs has no issues, because the setters of Boolean fields are named correctly and therefore it can deserialze Yaml to Java beans.

Note 2:
Luckily the Boolean fields seem to be passed correctly to kubernetes, because here Gson is used for serialization, which does not use getters and setters to serialize Java beans.

[brendandburns]

@mbohlool fyi, looks like we should update the version of the generator that we use...

[davidxia]

I just ran into this. I created a test that fails here. I tried to fix but don't know how.

Does this commit hash need to be updated to a commit in OpenAPITools/openapi-generator now?

[mihirnimbalkar]

I think I'm hitting the same problem at V1SecurityContext#privileged field in #531

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[davidxia]

/remove-lifecycle stale

[ghost]

A temporary workaround as following (may not work for some scenario)

//Since Gson serialization is not affected, so we can convert to JSON
Gson gson = new Gson();
String json = gson.toJson(crd);

//you can uses the JSON output in kubectl, so optionally, use the snakeYAML's function to convert JSON to YAML
org.yaml.snakeyaml.Yaml yaml = new org.yaml.snakeyaml.Yaml();
Object result = yaml.load(json);
String output = yaml.dumpAsMap(result);

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[davidxia]

/remove-lifecycle stale

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[brendandburns]

/lifecycle frozen

[brendandburns]

I'm going to close this as I believe that we have advanced the code generator past the commit with the fix for this issue (and this issue is pretty old). We can reopen if it reoccurs.