feat(dataconverter): add custom DataConverter samples

[Bueller87]

[Which sample(s) or area?]

dataconverter, root README.md, and dataconverter tests.

[What changed?]

• Added a compression sample that wraps JSON payloads with gzip to reduce large repetitive workflow/activity payloads before they are written to Cadence history.
• Added an encryption sample that wraps JSON payloads with AES-256-GCM, including demo key loading, nonce/tag handling, and guidance about logs and other disclosure surfaces.
• Added a BlobStore / S3 claim-check offload sample that stores large payloads outside Cadence history, keeps only a small reference in history, and uses a zero-config local filesystem store by default.

[Why?]

• Compression gives users a copyable pattern for reducing history size and storage overhead when workflows pass large, repetitive JSON payloads.
• Encryption gives users a concrete starting point for keeping sensitive workflow and activity payloads unreadable in Cadence history while making the key-management and logging caveats explicit.
• BlobStore / S3 offload gives users a claim-check pattern for payloads approaching Cadence size limits without forcing workflow or activity code to manage object-store references directly.

[How did you test it?]

Run the targeted tests:

./gradlew test --tests 'com.uber.cadence.samples.dataconverter.*'

Run the samples manually:

1. Start Cadence locally.

2. Register the sample domain if needed:

./gradlew -q execute -PmainClass=com.uber.cadence.samples.common.RegisterDomain

3. Start the DataConverter worker in terminal 1:

./gradlew -q execute -PmainClass=com.uber.cadence.samples.dataconverter.DataConverterWorker

4. Start each sample workflow from terminal 2:

./gradlew -q execute -PmainClass=com.uber.cadence.samples.dataconverter.CompressionStarter

./gradlew -q execute -PmainClass=com.uber.cadence.samples.dataconverter.EncryptionStarter

./gradlew -q execute -PmainClass=com.uber.cadence.samples.dataconverter.S3OffloadStarter

Also verified the final diff with:

git diff --check

[Potential risks]

Low. The changes are additive and isolated to the new dataconverter sample package plus README/test coverage. Existing samples should be unaffected; the main caveat is that users must adapt the demo encryption key and local BlobStore before using these patterns in production.

[Release notes]

• Added a gzip compression DataConverter sample.
• Added an AES-256-GCM encryption DataConverter sample.
• Added a BlobStore / S3 claim-check offload DataConverter sample.

[Documentation Changes]

• Updated the root README.md with the new DataConverter sample entry.
• Added a package README with setup, run commands, and sample selection guidance.
• Documented encryption key configuration, local BlobStore behavior, and S3 swap instructions.

[shijiesheng]

Suggested change

	try {
	ByteArrayOutputStream out = new ByteArrayOutputStream();
	try (GZIPOutputStream gzip = new GZIPOutputStream(out)) {
	gzip.write(jsonBytes);
	}
	return out.toByteArray();
	} catch (IOException e) {
	throw new DataConverterException("Failed to gzip-compress JSON payload", e);
	}
	ByteArrayOutputStream out = new ByteArrayOutputStream();
	try (GZIPOutputStream gzip = new GZIPOutputStream(out)) {
	gzip.write(jsonBytes);
	} catch (IOException e) {
	throw new DataConverterException("Failed to gzip-compress JSON payload", e);
	}
	return out.toByteArray();

[Bueller87]

Great suggestion and taken.

[shijiesheng]

You can use MinIO to integrate with real S3API. Otherwise, rename the example as largeblob to avoid confusion. There are users who are on GCP using GCS.

[Bueller87]

Really good point. The implementation is already storage-agnostic (everything goes through a BlobStore interface), so the "S3" branding ends up being a bit of an apology.

I renamed the sample to something provider-neutral in this PR claimcheck since that's the formal pattern name, but happy to go with largeblob if you prefer.

[gitar-bot]

Code Review ✅ Approved 1 resolved / 1 findings

Introduces Gzip compression, AES-256-GCM encryption, and S3-based claim-check DataConverter samples with updated documentation. The missing GZIPOutputStream import was resolved, and no other issues were found.

✅ 1 resolved

✅ Bug: GZIPOutputStream removed — gzip variable is undefined

📄 src/main/java/com/uber/cadence/samples/compression/CompressedJsonDataConverter.java:68-76
The refactoring removed the GZIPOutputStream creation (try (GZIPOutputStream gzip = new GZIPOutputStream(out))) but left a reference to gzip.write(jsonBytes) on line 71. This code will not compile because the gzip variable is never declared. Additionally, return out.toByteArray() on line 75 is outside the try-catch block, so even if the variable existed, the data written to out would never be gzip-compressed (the stream is never flushed/closed). The original try-with-resources pattern was correct and should be restored.

Rules ✅ All requirements met

Repository Rules

✅ PR Description Quality Standards: The PR description contains all required sections with substantive content, including clear explanations of what changed, why, and copyable test commands.

Options

Auto-apply is off → Gitar will not commit updates to this branch.
Display: compact → Showing less information.

Comment with these commands to change:

Auto-apply	Compact
gitar auto-apply:on	gitar display:verbose

_{Was this helpful? React with 👍 / 👎 | Gitar}

[shijiesheng]

make sense