Feature/237 raw image import - #58
Open
ThomasMalletCodra wants to merge 5 commits into
Open
Conversation
Image formats impose different data type and value-range constraints, while the existing export path offered no shared way to normalize data, handle non-finite values, or prevent unsafe conversions. Add an opt-in preparation backend that produces format-compatible image data while preserving the exact legacy behavior when no export parameters are provided. * [NEW] : Add ImageExportParam with normalization, range handling, target data type, and non-finite value policies * [NEW] : Add format-aware preparation with supported data type validation and overflow-safe conversion * [CHG] : Extend write_image with optional export parameters while keeping the original delegation path unchanged for param=None * [NEW] : Export the image preparation API through sigima.io, sigima.io.image, and sigima.params * [NEW] : Add focused tests for legacy behavior, source preservation, normalization, non-finite values, conversions, and format constraints * [NEW] : Add French translations for the image export parameters
Raw camera files require explicit binary layout information that ordinary image formats infer from their headers. * [NEW] : Add streaming RAW decoding with dimensions, framing, byte order and cancellation support * [CHG] : Route optional import parameters through the image registry while preserving legacy handlers * [NEW] : Export RAW parameters and cover registry compatibility, validation and translations
Define per-format capabilities and strict writer validation for all built-in image exports. Propagate Pillow, tifffile, text, and MAT options through a shared preparation path, provide exact in-memory previews, preserve coordinated text metadata, and cover the behavior with focused tests and French translations.
…-raw-image-import # Conflicts: # sigima/io/__init__.py # sigima/io/image/__init__.py # sigima/io/image/base.py # sigima/locale/fr/LC_MESSAGES/sigima.po
The coordinated text writer imported the Sigima package root only to read its version, creating a cycle while the root loaded the I/O registry. Package metadata now lives in a dependency-free module and remains re-exported by Sigima. * [FIX] : Centralize package metadata in sigima._metadata * [FIX] : Read the version without importing the package root * [CHG] : Point setuptools version metadata to the new module
Contributor
|
@ThomasMalletCodra The remaining lint issues are fixed in eda56ab. The VS Code Ruff and Pylint tasks pass, and all required CI checks are green. |
There was a problem hiding this comment.
🟡 Changes recommended
There are confirmed runtime issues in the new text export options path (complex dtype handling) and an avoidable large-memory copy in the export write path.
Once you've addressed the issues Copilot identified, you can request another Copilot review.
Pull request overview
This PR extends Sigima’s headless I/O layer with (1) parameterized RAW image import and (2) format-aware image export preparation (normalization, dtype conversion, backend-specific writer options), while preserving legacy registry/read/write behavior where possible.
Changes:
- Added a dedicated RAW image reader (
RawImageFormat) with aRawImageImportParamdataset and comprehensive tests (dtype/endianness, layout validation, cancellation/progress). - Introduced an image export capability model +
ImageExportParamand wired export-option-aware writing throughwrite_imageand image formats, with extensive test coverage. - Updated registry/filter behavior and public exports (
sigima.io,sigima.params) and centralized package metadata intosigima._metadata.
File summaries
| File | Description |
|---|---|
| sigima/tests/io/raw_image_unit_test.py | Adds unit tests for RAW import parameters, layout validation, progress/cancel behavior, and public exports. |
| sigima/tests/io/ioregistry_unit_test.py | Updates registry tests for new filter grouping behavior and legacy reader signature compatibility. |
| sigima/tests/io/image_export_unit_test.py | Adds extensive tests for export normalization, dtype conversion, option validation/translation, and preview round-trips. |
| sigima/tests/io/coordinated_text_format_unit_test.py | Adds coverage for delimiter/precision options when exporting coordinated (non-uniform) text images. |
| sigima/params.py | Re-exports new I/O parameter datasets and updates parameter documentation listings. |
| sigima/locale/fr/LC_MESSAGES/sigima.po | Adds French translations for new export/import parameter UI strings. |
| sigima/io/image/raw.py | Implements parameterized RAW image import format + parameter dataset. |
| sigima/io/image/formats.py | Extends image formats to accept optional read params, adds export-option-aware writing, improves coordinated text writer, and switches version source. |
| sigima/io/image/export.py | Adds export capabilities model, option validation, safe conversion pipeline, and preview encoding/decoding helpers. |
| sigima/io/image/base.py | Extends image registry to accept optional read params and customizes save-filter generation; adds write_with_options. |
| sigima/io/image/init.py | Exposes new export/raw APIs from the image I/O package. |
| sigima/io/convenience.py | Extends convenience read APIs with optional read params; adds format-aware export path to write_image. |
| sigima/io/init.py | Re-exports new image export/import API surface at the top-level I/O module. |
| sigima/_metadata.py | Introduces centralized package metadata (version and URLs). |
| sigima/init.py | Imports and re-exports metadata symbols from sigima._metadata. |
| pyproject.toml | Points dynamic version resolution to sigima._metadata.__version__. |
Review details
- Files reviewed: 16/16 changed files
- Comments generated: 2
- Review effort level: Lite
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
Comment on lines
+896
to
+906
| if np.issubdtype(obj.data.dtype, np.integer): | ||
| fmt = "%d" | ||
| elif np.issubdtype(obj.data.dtype, np.floating) or np.issubdtype( | ||
| obj.data.dtype, np.complexfloating | ||
| ): | ||
| fmt = f"%.{values['precision']}e" | ||
| else: | ||
| raise NotImplementedError( | ||
| f"Writing data of type {obj.data.dtype} to text file is not supported." | ||
| ) | ||
| np.savetxt(filename, obj.data, fmt=fmt, delimiter=delimiter) |
Comment on lines
+165
to
+166
| exported_image = copy.deepcopy(image) | ||
| exported_image.data = prepare_image_for_export(image.data, filename, param) |
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
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.
No description provided.