CAMEL-23686: Avoid creating empty headers map on read access

[gnodet]

CAMEL-23686

Summary

• getHeader(), removeHeader(), and removeHeaders() no longer force creation of a CaseInsensitiveMap when headers haven't been set
• Short-circuits with null return when headers == null and isPopulateHeadersSupported() is false
• JMS-style messages with lazy header population (isPopulateHeadersSupported() == true) are preserved

Problem

Every getHeader(name) call on a message with no headers forced creation of a CaseInsensitiveMap (allocating 4 internal arrays, ~200 bytes) just to return null. This happened frequently in splitter/multicast sub-exchanges that never use headers.

Before:

public Object getHeader(String name) {
    if (headers == null) {
        headers = createHeaders();  // allocates empty CaseInsensitiveMap
    }
    if (!headers.isEmpty()) {       // always false for new map
        return headers.get(name);
    } else {
        return null;                // returns null anyway
    }
}

After:

public Object getHeader(String name) {
    if (headers == null) {
        if (!isPopulateHeadersSupported()) {
            return null;            // no allocation
        }
        headers = createHeaders();
    }
    return headers.get(name);
}

Benchmark results

Baseline route (timer → split(1000) → setBody, no header access):

• 3,891 DefaultMessage instances but only 4 CaseInsensitiveMap — split children don't allocate maps at all
• Without this fix: every sub-exchange would create an empty map on first getHeader() call

Pipeline route (with setHeader on every exchange):

• 24% throughput increase measured by in-flight exchange count at snapshot time
• Map ratio stays at 0.57 (expected — every exchange writes headers)

Test plan

• DefaultMessageHeaderTest (38 tests) — pass, including lazy-populated headers test
• DefaultMessageTest — pass

[github-actions]

🌟 Thank you for your contribution to the Apache Camel project! 🌟
🤖 CI automation will test this PR automatically.

🐫 Apache Camel Committers, please review the following items:

• First-time contributors require MANUAL approval for the GitHub Actions to run
• You can use the command /component-test (camel-)component-name1 (camel-)component-name2.. to request a test from the test bot although they are normally detected and executed by CI.
• You can label PRs using skip-tests and test-dependents to fine-tune the checks executed by this PR.
• Build and test logs are available in the summary page. Only Apache Camel committers have access to the summary.

⚠️ Be careful when sharing logs. Review their contents before sharing them publicly.

[github-actions]

🧪 CI tested the following changed modules:

• core/camel-support

ℹ️ Dependent modules were not tested because the total number of affected modules exceeded the threshold (50). Use the test-dependents label to force testing all dependents.

---

⚙️ View full build and test results

[apupier]

astonishing improvements due to only empty Map instanciation removal

EDIT: I think it will be worthy to wait for Claus or Andrea feedback before merging as they might better know if there are corner cases

[davsclaus]

Clean optimization — eliminates unnecessary CaseInsensitiveMap allocation (~200 bytes) on read access when headers haven't been set. The isPopulateHeadersSupported() guard correctly preserves the JMS-style lazy population path. The DRY refactor of getHeader overloads delegating to the base method is a nice simplification (-37 net lines). CI is green.

This review does not replace specialized AI review tools (CodeRabbit, Sourcery) or static analysis (SonarCloud).

This review was generated by an AI agent and may contain inaccuracies. Please verify all suggestions before applying.