ci(windows): unblock hidapi source build under CMake 4.x runner image#3509
Merged
Conversation
Every check-windows run since 2026-06-10 ~14:36 UTC fails at the "Setup
hidapi" step — before any AetherSDR code compiles. GitHub's
windows-latest runner image updated to CMake 4.x, which removed
compatibility with cmake_minimum_required(<3.5); hidapi 0.14.0's
CMakeLists declares an older minimum, so its configure now hard-fails:
CMake Error at CMakeLists.txt:1 (cmake_minimum_required):
Compatibility with CMake < 3.5 has been removed from CMake.
Last green main run was 04:10 UTC; every PR branch since (including the
AetherClaude bot's) fails identically, so this blocks the whole queue.
Fix: pass -DCMAKE_POLICY_VERSION_MINIMUM=3.5 in setup-hidapi.ps1 — the
exact override CMake's error message recommends. Configure proceeds with
3.5 policy semantics; the build itself is unaffected. Drop the flag when
hidapi is bumped to a release with a modern declared minimum.
Audited the sibling setup scripts for the same latent failure: opus
1.5.2 declares cmake_minimum_required(3.16) (fine under 4.x — builds
locally with CMake 4.3.3 daily), and setup-onnxruntime.ps1's cmake
mention is an informational echo, not an invocation. hidapi is the only
affected script.
Principle XI.
…=3.5 at the dot The first attempt passed -DCMAKE_POLICY_VERSION_MINIMUM=3.5 unquoted. PowerShell's native-argument tokenizer splits that token at the dot, so CMake received "3" and rejected it: CMake Error: Invalid CMAKE_POLICY_VERSION_MINIMUM value "3". Quoting the whole argument delivers "3.5" intact. The script's other -D flags only survive unquoted because their values contain no dot; comment updated so the next person adding a dotted -D value doesn't trip the same wire. Principle XI.
This was referenced Jun 12, 2026
ea5wa
added a commit
to ea5wa/AetherSDR
that referenced
this pull request
Jun 12, 2026
…CI fix) Conflict resolution in src/gui/MainWindow.cpp followed the aethersdr#3351 extraction: took upstream's refactored constructor and ported the two WFM-branch additions to their new homes — the VfoWidget::wfmActivated connect into wireVfoWidget() (MainWindow_Wiring.cpp) and the HAVE_HIDAPI guard around the FlexControl isTMate2() call (MainWindow_Controllers.cpp). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
G6PWY-Chris
pushed a commit
to G6PWY-Chris/AetherSDR
that referenced
this pull request
Jun 22, 2026
…aethersdr#3509) ## Summary **Repo-wide CI blocker.** Every `check-windows` run since 2026-06-10 ~14:36 UTC fails at the **Setup hidapi** step, before any AetherSDR code compiles. Last green run on main: 04:10 UTC. Every PR branch since fails identically (aethersdr#3508, the AetherClaude bot's `aetherclaude/issue-3506` branch). ## Root cause GitHub's `windows-latest` runner image updated to **CMake 4.x**, which removed compatibility with `cmake_minimum_required(<3.5)`. hidapi 0.14.0's CMakeLists declares an older minimum, so `scripts/setup/setup-hidapi.ps1`'s source build now hard-fails at configure: ``` CMake Error at CMakeLists.txt:1 (cmake_minimum_required): Compatibility with CMake < 3.5 has been removed from CMake. ... Or, add -DCMAKE_POLICY_VERSION_MINIMUM=3.5 to try configuring anyway. ``` ## Fix One flag on the configure line — the exact override CMake's own error message recommends: ```powershell cmake -B $buildDir -S $srcDir.FullName -G "Ninja" ` -DCMAKE_BUILD_TYPE=Release ` -DBUILD_SHARED_LIBS=ON ` -DCMAKE_POLICY_VERSION_MINIMUM=3.5 ``` Configure proceeds with 3.5 policy semantics; the compiled artifact is unaffected. Inline comment documents when the flag can be dropped (hidapi version bump with a modern minimum). ## Sibling-script audit Checked the other Windows setup scripts for the same latent failure: | Script | cmake source build? | Declared minimum | At risk? | |---|---|---|---| | `setup-hidapi.ps1` | yes — hidapi 0.14.0 | < 3.5 | **this PR** | | `setup-opus.ps1` | yes — opus 1.5.2 | 3.16 | no (verified: builds locally under CMake 4.3.3) | | `setup-onnxruntime.ps1` | no (informational echo only) | n/a | no | ## Validation This PR's own `check-windows` run exercises the fixed script — green CI on this PR **is** the validation. Once merged, aethersdr#3508 (and the bot's PR) just need a CI re-run.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Repo-wide CI blocker. Every
check-windowsrun since 2026-06-10 ~14:36 UTC fails at the Setup hidapi step, before any AetherSDR code compiles. Last green run on main: 04:10 UTC. Every PR branch since fails identically (#3508, the AetherClaude bot'saetherclaude/issue-3506branch).Root cause
GitHub's
windows-latestrunner image updated to CMake 4.x, which removed compatibility withcmake_minimum_required(<3.5). hidapi 0.14.0's CMakeLists declares an older minimum, soscripts/setup/setup-hidapi.ps1's source build now hard-fails at configure:Fix
One flag on the configure line — the exact override CMake's own error message recommends:
Configure proceeds with 3.5 policy semantics; the compiled artifact is unaffected. Inline comment documents when the flag can be dropped (hidapi version bump with a modern minimum).
Sibling-script audit
Checked the other Windows setup scripts for the same latent failure:
setup-hidapi.ps1setup-opus.ps1setup-onnxruntime.ps1Validation
This PR's own
check-windowsrun exercises the fixed script — green CI on this PR is the validation.Once merged, #3508 (and the bot's PR) just need a CI re-run.