What's Changed
Fixed Pydantic sphinx autodoc issues on Github deployment
Removed all Python pipeline item (processor, reader and writer) classes imports except for those required by the Pydantic sphinx autodoc bot from all (sub)package level __init__.py files. This benefits the overall setup speed of running any workflow, but does result in a change in how processors, readers and writers are called from the pipeline.yaml. Previously, they could be called from the packet level, going forward they should be called from the module level. For example:
⋮
- common.YAMLReader:
filename: energy_calibration_result.yaml
schema: edd.models.MCAEnergyCalibrationConfig
- edd.MCATthCalibrationProcessor:
⋮
- common.YAMLWriter:
filename: tth_calibration_result.yaml
force_overwrite: true
Should be modified to:
⋮
- common.reader.YAMLReader:
filename: energy_calibration_result.yaml
schema: edd.models.MCAEnergyCalibrationConfig
- edd.processor.MCATthCalibrationProcessor:
⋮
- common.writer.YAMLWriter:
filename: tth_calibration_result.yaml
force_overwrite: true
Note that for now the code will stay backward compatible with the old style pipeline.yaml files. However, it may become deprecated at a future time, so it is strongly suggested to update pipeline.yaml files to the new style going forward.
Pull request log
- Merge pull request #274 from CHESSComputing/dev by @rolfverberg in #275
- Merge main back into docs by @rolfverberg in #276
- perf: try removing imports from init's by @rolfverberg in #280
- build: add a manual pages deployment workflow by @rolfverberg in #281
- Docs: Fixing Pydantic sphinx issues on Github deployment by @rolfverberg in #282
- Merge pull request #275 from CHESSComputing/main by @rolfverberg in #283
Full Changelog: v2026.04.24...v2026.04.28
