dependabot[bot] opened a new pull request, #3617: URL: https://github.com/apache/cxf/pull/3617
Bumps [org.atmosphere:atmosphere-runtime](https://github.com/Atmosphere/atmosphere) from 3.1.0 to 4.0.74. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/Atmosphere/atmosphere/releases";>org.atmosphere:atmosphere-runtime's releases</a>.</em></p> <blockquote> <h2>Atmosphere 4.0.74</h2> <h3>Added</h3> <ul> <li>deprecate the static AI readers and binders; samples read their own AiEnvironment</li> <li>give each application its own native clients in every runtime adapter</li> <li>give each application its own AI knobs, settings and runtime instances</li> <li>enforce both an explicit skill scopeTier and a restrictive <a href="https://github.com/AgentScope";><code>@AgentScope</code></a>, stricter wins</li> <li>let one ScopePolicy enforce several scope legs that a request must all pass</li> <li>mark the pagehide close beacon going-away, keep close() a leave</li> <li>give a page reload over SSE, long-polling or streaming the reattach grace</li> <li>mark the disconnect of a close message whose page is going away</li> <li>let an annotation processor fail the framework init instead of being skipped</li> <li>keep a run going for a reattach grace after its client loses the connection</li> <li>end a run cut short with one cancelled frame naming why, on every path and bridge</li> <li>honour atmosphere.ai.* knobs from Spring and Quarkus config and select the runtime by name</li> </ul> <h3>Fixed</h3> <ul> <li>shut down the clients routing rules build for their own endpoint</li> <li>count each AI-state read and flag Kotlin imports, not constant imports</li> <li>build a routing rule's own client with its context's cache-key and logprobs modes</li> <li>shut down an environment's built client after a client was installed over it</li> <li>flag the process default, split-line reads and resolver static imports in the AI-state gate</li> <li>summarize, rerank and spawn subagents on the runtime of their own application</li> <li>offer the auto-configured executor and model so an executor the application sets wins</li> <li>keep a runtime instance with its application after that application resets its resolution</li> <li>have AgentRuntimeResolver.hasExplicitClientBinding read the calling application</li> <li>report each application's own runtime in console, admin and interactions</li> <li>resolve decision models, memory, caches and evaluators per application</li> <li>shut down the HTTP clients of the settings an AI environment built when it closes</li> <li>bind a runtime's application to the calling thread in execute as executeWithHandle does</li> <li>resolve the running application's embedding runtimes through EmbeddingRuntimeResolver</li> <li>stop the LLM_* environment fallback from replacing settings or publishing them per request</li> <li>keep a workspace RUNTIME.md's generation values on its own settings</li> <li>keep the prompt-cache-key and logprobs modes an OpenAI-compatible client was built with</li> <li>stop counting the room frames a poll is writing against their reply's bounds</li> <li>keep a room frame off a member's waiting poll under out-of-order broadcast</li> <li>a long-polling member's blocked poll write no longer stalls the room</li> <li>resume a room member's poll outside its mailbox lock</li> <li>keep room frames in per-member mailboxes, counted once and bounded per reply</li> <li>deliver a room reply whole to every long-polling member of the room</li> <li>deliver a stream-end frame a second filter defers again, instead of dropping it</li> <li>run <a href="https://github.com/Ready";><code>@Ready</code></a> once when a long-polling client re-polls over a suspended poll</li> <li>keep long-polling polls waiting on a run out of the streaming reattach slots</li> <li>read each frame of a multi-frame message when capturing the run id</li> <li>forget long-polling clients that stopped polling; run <a href="https://github.com/Ready";><code>@Ready</code></a> once for racing polls</li> <li>answer a long-polling run-id poll only with what its tracking id was not sent</li> <li>run <a href="https://github.com/Ready";><code>@Ready</code></a> once per long-polling client, not on every poll</li> <li>answer a long-polling reattach poll with its run's replay at once</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/Atmosphere/atmosphere/blob/main/CHANGELOG.md";>org.atmosphere:atmosphere-runtime's changelog</a>.</em></p> <blockquote> <h2>[4.0.74] - 2026-10-16</h2> <h3>Added</h3> <ul> <li>deprecate the static AI readers and binders; samples read their own AiEnvironment</li> <li>give each application its own native clients in every runtime adapter</li> <li>give each application its own AI knobs, settings and runtime instances</li> <li>enforce both an explicit skill scopeTier and a restrictive <a href="https://github.com/AgentScope";><code>@AgentScope</code></a>, stricter wins</li> <li>let one ScopePolicy enforce several scope legs that a request must all pass</li> <li>mark the pagehide close beacon going-away, keep close() a leave</li> <li>give a page reload over SSE, long-polling or streaming the reattach grace</li> <li>mark the disconnect of a close message whose page is going away</li> <li>let an annotation processor fail the framework init instead of being skipped</li> <li>keep a run going for a reattach grace after its client loses the connection</li> <li>end a run cut short with one cancelled frame naming why, on every path and bridge</li> <li>honour atmosphere.ai.* knobs from Spring and Quarkus config and select the runtime by name</li> </ul> <h3>Fixed</h3> <ul> <li>shut down the clients routing rules build for their own endpoint</li> <li>count each AI-state read and flag Kotlin imports, not constant imports</li> <li>build a routing rule's own client with its context's cache-key and logprobs modes</li> <li>shut down an environment's built client after a client was installed over it</li> <li>flag the process default, split-line reads and resolver static imports in the AI-state gate</li> <li>summarize, rerank and spawn subagents on the runtime of their own application</li> <li>offer the auto-configured executor and model so an executor the application sets wins</li> <li>keep a runtime instance with its application after that application resets its resolution</li> <li>have AgentRuntimeResolver.hasExplicitClientBinding read the calling application</li> <li>report each application's own runtime in console, admin and interactions</li> <li>resolve decision models, memory, caches and evaluators per application</li> <li>shut down the HTTP clients of the settings an AI environment built when it closes</li> <li>bind a runtime's application to the calling thread in execute as executeWithHandle does</li> <li>resolve the running application's embedding runtimes through EmbeddingRuntimeResolver</li> <li>stop the LLM_* environment fallback from replacing settings or publishing them per request</li> <li>keep a workspace RUNTIME.md's generation values on its own settings</li> <li>keep the prompt-cache-key and logprobs modes an OpenAI-compatible client was built with</li> <li>stop counting the room frames a poll is writing against their reply's bounds</li> <li>keep a room frame off a member's waiting poll under out-of-order broadcast</li> <li>a long-polling member's blocked poll write no longer stalls the room</li> <li>resume a room member's poll outside its mailbox lock</li> <li>keep room frames in per-member mailboxes, counted once and bounded per reply</li> <li>deliver a room reply whole to every long-polling member of the room</li> <li>deliver a stream-end frame a second filter defers again, instead of dropping it</li> <li>run <a href="https://github.com/Ready";><code>@Ready</code></a> once when a long-polling client re-polls over a suspended poll</li> <li>keep long-polling polls waiting on a run out of the streaming reattach slots</li> <li>read each frame of a multi-frame message when capturing the run id</li> <li>forget long-polling clients that stopped polling; run <a href="https://github.com/Ready";><code>@Ready</code></a> once for racing polls</li> <li>answer a long-polling run-id poll only with what its tracking id was not sent</li> <li>run <a href="https://github.com/Ready";><code>@Ready</code></a> once per long-polling client, not on every poll</li> <li>answer a long-polling reattach poll with its run's replay at once</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/Atmosphere/atmosphere/commit/e969284eb90d32018ac8a6c2db39e22e7d1d6698";><code>e969284</code></a> release: Atmosphere 4.0.74</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/89263a4d90643c499dd6b752351805645f8ada4c";><code>89263a4</code></a> test(adk): give the ADK tests a Gemini key so the per-request runner test run...</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/25b81afc22db9e0bfe1a1131630ae5746fff5119";><code>25b81af</code></a> fix(spring-boot): shut down the clients routing rules build for their own end...</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/26a9dd0fcb74568011e084857dd9f35b4e4b45a5";><code>26a9dd0</code></a> ci: check the gate self-test paths of push and pull_request apart</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/53f6eb2ba6cfa5a69f134b144bc61d0ca0c3dbbf";><code>53f6eb2</code></a> fix(ci): count each AI-state read and flag Kotlin imports, not constant imports</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/496e4c371354b1e0b2a65aea7b2e1158c7318978";><code>496e4c3</code></a> docs(ai): say which reads stay on the static API, what is deprecated, why Qua...</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/07ddfb0a7e52bb6f1064bb07ddf61f79a42629c6";><code>07ddfb0</code></a> refactor(ai): deprecate AiConfig.installHostConfig in favour of the host's ow...</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/edef9e27d2ffc621acadb9596b2dc7f15aef2498";><code>edef9e2</code></a> fix(spring-boot): build a routing rule's own client with its context's cache-...</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/d8290cc2ae2ab18c0f153bf510d8a527f2c9c30a";><code>d8290cc</code></a> fix(ai): shut down an environment's built client after a client was installed...</li> <li><a href="https://github.com/Atmosphere/atmosphere/commit/129226bf56cbcb0f1ad5b090e2a1a7633b193406";><code>129226b</code></a> ci: run the gate self-test when a file one of its cases injects into changes</li> <li>Additional commits viewable in <a href="https://github.com/Atmosphere/atmosphere/compare/atmosphere-project-3.1.0...atmosphere-4.0.74";>compare view</a></li> </ul> </details> <br /> [](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores) Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. [//]: # (dependabot-automerge-start) [//]: # (dependabot-automerge-end) --- <details> <summary>Dependabot commands and options</summary> <br /> You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) </details> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
