[BUG][JAVA] Fix generation of complex default values (fixes #24993)
Closes: #24993
Type of change
This is primarily a bug fix: complex default values in OpenAPI specs (nested arrays/objects, oneOf schemas with defaults, nullable fields, containers) previously generated non-compilable Java code (see #24993).
As part of making the generation correct, default rendering was also extended where it silently produced invalid or incomplete output (e.g. date library specific defaults), so minor parts of this PR could be considered new behavior. I'm happy to split any feature-like part into a separate PR if the maintainers prefer smaller, atomic changes.
Problem
Given an OpenAPI spec with complex default values, e.g.:
components:
schemas:
NestedArrayWithDefaultValues:
type: object
properties:
matrix:
type: array
default:
- [1, 2]
- [3, 4]
pet:
oneOf:
- $ref: '#/components/schemas/Pet'
- $ref: '#/components/schemas/Tag'
default:
name: Rex
the previous generator output for Java clients was syntactically invalid Java, such as:
private List<List<Integer>> matrix = new ArrayList<>(Arrays.asList(Arrays.asList(1, 2), [3, 4]));
or a default that referenced a schema class without constructing it, resulting in compile errors in the generated samples.
Root cause
The default-value rendering path in AbstractJavaCodegen only supported scalar defaults. When the default was a complex structure (arrays, nested arrays, objects, oneOf branches, date/time values), the renderer emitted fragments that could not be compiled.
Fix
- Extracted the complex default construction logic into a focused
JavaDefaultValueRenderercomponent (following the pattern of other util components), soAbstractJavaCodegenkeeps only the generator policy/configuration and delegates recursive default rendering through an explicit context (schema resolution, Java naming/type conversion, collection construction, date-library configuration, oneOf handling). - Correct default rendering for:
- arrays and nested arrays (
List/Set,Arrays.asList(...),LinkedHashSetforuniqueItems) - object defaults and oneOf branch selection
- date/time libraries (
java8,threetenbp,joda, legacyjava.util.Date) - nullable / optional fields with and without defaults
- container initialization (previously broken for
Map-like and nested containers)
- arrays and nested arrays (
- Updated affected generator mustache templates and regenerated all impacted Java samples.
Tests
- New directed unit tests in
JavaClientCodegenTest,SpringCodegenTestandJavaPlayFrameworkCodegenTestcovering arrays/objects defaults, oneOf branch selection, date libraries, nullable defaults and container initialization. - Regenerated samples are committed so CI can verify the generator outputs of the HEAD commit.
Affected generators
Java-based generators that share AbstractJavaCodegen / the updated pojo.mustache templates: java client libraries (okhttp-gson, jersey2/3, resttemplate/webclient, native, rest-assured, feign, microprofile, helidon, retrofit2), spring (boot/declarative), JavaJaxRS, java-msf4j-server, and JavaPlayFramework.
PR checklist
- Read the contribution guidelines.
- Run the following to build the project and update samples:
(For Windows users, please run the script in WSL)./mvnw clean package || exit ./bin/generate-samples.sh ./bin/configs/*.yaml || exit ./bin/utils/export_docs_generators.sh || exit
Commit all changed files.
This is important, as CI jobs will verify all generator outputs of your HEAD commit as it would merge with master.
These must match the expectations made by your contribution.
You may regenerate an individual generator by passing the relevant config(s) as an argument to the script, for example./bin/generate-samples.sh bin/configs/java*.
IMPORTANT: Do NOT purge/delete any folders/files (e.g. tests) when regenerating the samples as manually written tests may be removed. - If your PR is targeting a particular programming language, @mention the technical committee members, so they are more likely to review the pull request.
@bbdouglas (2017/07) @sreeshas (2017/08) @jfiala (2017/08) @lukoyanov (2017/09) @cbornet (2017/09) @jeff9finger (2018/01) @karismann (2019/03) @Zomzog (2019/04) @lwlee2608 (2019/10) @martin-mfg (2023/08) @wing328