Closed IanSudbery closed 5 years ago
ping @AndreasHeger
Hi both, CGATReport status at the moment is uncertain. The reports need updating and maintaining which I can't do as I currently don't have the bandwidth. My main objective is thus at the moment to ignore visualization but instead to clear up the pipelines to provide a set of few and simple tables to facilitate plotting in notebooks, Rmarkdown, etc.
(CGATReport itself is still well and alive as I use it for other projects)
Whats the way forward here then? Should we not try to fix bugs in the cgat-flow reports. Should they be removed?
I guess the action depends on what we want to do long-term. We need to make a decision if the reports are useful and we want to keep them. If we keep them, we need to come up with a way to maintain them.
I think we should start to remove all references to cgat-report in the production pipelines, considering that they are broken and I dont think any of us has time to fix them. Whats the consensus?
I am happy with this.
The readqc pipeline reads gets the PARAMS dictionary from CGATReport.Utils. This dictionary should have been filled in conf.py, but seems to be empty in
pipeline_docs/pipeline_readqc/tracks/ReadqcReport.py
.For example, create a readqc pipeline with the following:
Edit the resulting
pipeline.yml
so that it contains the following line:Now add the following code to
cgat-flow/CGATPipelines/pipeline_docs/pipeline_readqc/trackers/ReadqcReport.py
, below the parametrization block:now run the report with:
You'll find the resulting
test
file contains:What is the status of the CGATReport reports at the moment? Are they still supported? If so they need updating to cope with changes in CGATReport and the move from
ini
files toyml
files. If not, then they should probably be removed before publication.