Both these components have a a compile_results processes that combines the results and appends them into a single report.tsv file. This file is then symlinked to the publishDir folder. Upon re-execution of the pipeline with the resume flag, the combine_reports is executed, creating a new report.tsv file that in then symlinked to the publishDir, overwriting the original one and all previous results are lost.
So far I can't figure out a way to prevent this from happening and still keep the results in a single tsv file. A solution is to stop this aggregating behavior and keep the results in a file per sample. Do you agree with this solution?
jacarrico
commented
5 years ago
Both these components have a a compile_results processes that combines the results and appends them into a single report.tsv file. This file is then symlinked to the publishDir folder. Upon re-execution of the pipeline with the resume flag, the combine_reports is executed, creating a new report.tsv file that in then symlinked to the publishDir, overwriting the original one and all previous results are lost.
So far I can't figure out a way to prevent this from happening and still keep the results in a single tsv file. A solution is to stop this aggregating behavior and keep the results in a file per sample. Do you agree with this solution?