Repository navigation
Conversation
…is and series metadata - getLabels() reads attributes.ome.labels for zarr v3 labels groups (spec 0.5/0.6), falling back to the legacy attributes.labels; fix the v0.5 labels fixture to the spec layout. - Axis serializes longName (spec) and still accepts long_name on read. - displacements/coordinates transforms keep their interpolation field. - Model labels, image-label and series in the v0.5 and v0.6 OmeMetadata (new shared ImageLabel type; unknown keys in colors/properties are kept). createScaleLevel no longer drops ome fields in v0.5. - Add the projectAxis transform (droppedInputs/createdOutputs). - byDimension items use inputAxes/outputAxes, accepting the snake_case aliases on read. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
OME-NGFF 0.6 requires the input/output of coordinate transformations to be
JSON objects {"name": ..., "path": ...}. The v0.6 model stored them as
strings, squashing objects into "path#name" on read and writing strings
back, so reading a conformant file and saving it corrupted it. Scale levels
created via createScaleLevel also lacked the mandatory dataset input/output.
- Add CoordinateSystemRef (name/path, NON_NULL) and use it for input/output
on all v0.6 transformations; factory methods take CoordinateSystemRef and
gain ref-less scale/translation and a sequence factory.
- Legacy pre-release string refs are still read: a bare string is a name,
except for the input of a dataset transformation, which is the dataset
path; "<path>#<name>" strings from earlier zarr-java are split. Writing
always emits the object form.
- createScaleLevel writes input {"path": <level>} and output
{"name": <intrinsic>} (from existing datasets, else first coordinate
system; ZarrException if neither). Multiple transformations are wrapped in
a sequence whose nested transformations carry no refs; getMultiscaleNode
flattens such sequences back to [scale, translation].
- Scene graph resolution works on CoordinateSystemRef.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…s as {name, path} objects)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Resolves the conflicts between the two PRs: the projectAxis transform and the convenience constructors of displacements/coordinates from zarr-developers#109 now take CoordinateSystemRef like every other v0.6 transform after zarr-developers#110, and the zarr-developers#109 test for createScaleLevel expects the single sequence zarr-developers#110 writes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every OME-Zarr 0.6 metadata document we have test data for is now read and written back, and the result must equal the original: the RFC-5 examples (bumped to their final 0.6 state), the valid cases of the official ngff-spec conformance suite (new submodule at tag 0.6) and our scene fixtures, which are moved from 0.6.dev3 to final 0.6. The round trip found what was still lost on write: - members zarr-java does not model on ome, multiscales entries and omero channels (@id, @type, wavelength_id) are kept as extra fields - unknown transformation types were written as a nested "raw" object - the axes of an image came from the first coordinate system instead of the intrinsic one the datasets map into Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Scene.openScene listed the children of the scene group to find its images, so opening a scene over HTTP failed with "The underlying store does not support listing". Like ome-zarr-py, the images are now found through the paths the scene transformations reference; listing still adds images no transformation references, where the store supports it. The new OmeZarrV06RealDataTest reads real 0.6 datasets with chunk data over HTTP from the RFC-5 test data bucket of the University of Münster: an image, a plate and four scenes. The expected pixel sums were cross-checked with zarr-python. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Resolves the conflicts with the move of OME-Zarr to dev.zarr.conventions.ome (zarr-developers#78). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Picks up main incl. the move of OME-Zarr to dev.zarr.conventions.ome (zarr-developers#78). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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.
Builds on #116 (and with it #109 and #110); only the last commit is new.
Scene.openScenelisted the children of the scene group to find its images, so opening a scene over HTTP failed with "The underlying store does not support listing". Like ome-zarr-py, the images are now found through thepaths the scene transformations reference. Where the store can list, listing still adds images that no transformation references.The new
OmeZarrV06RealDataTestreads real 0.6 data with chunks over HTTP from the RFC-5 test data bucket of the University of Münster: cells3d, an HCS plate and four scenes (stitched tiles, a CBCT/dose scene, one with displacement fields). It also round-trips the metadata of 26 of its groups. The expected pixel sums were cross-checked with zarr-python; all 62 regions I compared were bit-identical.🤖 Generated with Claude Code