Sometimes, a pipeline can restart from intermediate processes with files produced by upstream processes or by other means. pipelines like sarek write the samplesheets with the produced files, so downstream processes can more easily be triggered later.
For the entry point for the restart we have a json schema that validates the samplesheet, but for the writing we are defining the columns manually without any validation. Having a write function that uses the json schema would ensure writing and reading functions are not diverging over time.
Sometimes, a pipeline can restart from intermediate processes with files produced by upstream processes or by other means. pipelines like sarek write the samplesheets with the produced files, so downstream processes can more easily be triggered later.
For the entry point for the restart we have a json schema that validates the samplesheet, but for the writing we are defining the columns manually without any validation. Having a
writefunction that uses the json schema would ensure writing and reading functions are not diverging over time.