Repository navigation
[Sync] develop with upstream (HDR/IBL validation on scene restore, flat Editor exports: #1367, #1369) and the foveation branch's test fix - #78
Merged
Conversation
…Animations (untoldengine#1369) getResourceURL only probed the nested Models/<name>/<name>.<ext> form, but the Editor's asset exporter (RuntimeExportLocation.swift) has always cooked runtime assets flat into the category root (Models/<name>.<ext>), beside the per-model support folder rather than inside it. External projects that load an asset by filename via setEntityMeshAsync never matched that layout and failed with "Filename not found", even though the identical file loaded fine through the Editor. Add the flat form as an additional probe so both cases resolve through the same code path.
…ene restore (untoldengine#1367) * [Bugfix] Fix HDR/IBL validation falling back to stale directory on scene restore serializeScene() only checked assetBasePath/HDR when validating the active HDR, incorrectly treating the engine's built-in default HDR as missing and nulling it out. deserializeScene() then cleared hdrURL but left hdrDirectoryURL stale, so the next IBL bake resolved an empty texture name against a leftover directory and failed to open it. - Validate HDR existence through LoadingSystem's full resource search (project assets, app bundle, engine bundle) instead of only assetBasePath/HDR. - Clear hdrDirectoryURL alongside hdrURL in deserializeScene. - Guard loadImage against an empty texture name. Fixes untoldengine#1364 * [Bugfix] Guard makeHDRTexture against a nil render device The untoldengine#1364 HDR-validation fix made serializeScene/deserializeScene more accurate about whether the active HDR actually resolves, which meant a bare ECS unit test (no real Metal device set up) started exercising generateHDR -> makeHDRTexture for the first time. renderInfo.device is an implicitly-unwrapped optional there, so a nil device crashed the test suite instead of failing gracefully through the existing LoadError.textureCreationFailed path.
…ore, assets the Editor exports flat (untoldengine#1367, untoldengine#1369)
miogds
changed the title
This was referenced Oct 16, 2026
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
2 participants
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.
Brings the fork's
developup to upstreamdevelopat72150232. The merge was clean: the two upstream commits touchSceneSerializer,LoadingSystemandFuncUtils, none of them in the lines the fork changes.What upstream brings (2 commits since the last sync, #76)
What was already on the fork
Everything #76 and #77 merged for testing: the head of untoldengine#1359 at
2832d5fb(foveation, TAA, the forward path, the look in a mixed space), thexr_shader_costchain of #73–#75, and the large-scene stage 2 chain.Also merged: the foveation branch's head,
e7118fa8The one commit on untoldengine#1359's branch that
developstill lacked, merged asd1825d5f. Test files only:RenderGraphBuilderTest.testSupportedRenderExtensionStagesResolveAcrossImmersionModesloops the immersion modes and leftrenderInfo.immersionStyleat.ar, and nothing reset it, so every later render test in the same process drew the AR graph; behind that class the temporal rate-map test failed in serial runs.BaseRenderSetupnow resets the style with the other per-test globals, and the case restores what it found. CI never saw it because the render tests run one process per method there.