Open Shenglai opened 6 years ago
The example above fails for me with the latest cwltool
.
kinow@ranma:/tmp/bunny-1.0.6$ mkdir /tmp/cwl
kinow@ranma:/tmp/bunny-1.0.6$ touch /tmp/cwl/a
kinow@ranma:/tmp/bunny-1.0.6$ touch /tmp/cwl/b
(venv) kinow@ranma:~/Development/python/workspace/cwl-v1.2$ cwltool /tmp/make_secondary.cwl --parent_file /tmp/cwl/a --children /tmp/cwl/b
INFO /home/kinow/Development/python/workspace/cwl-v1.2/venv/bin/cwltool 3.1.20220502060230
INFO Resolved '/tmp/make_secondary.cwl' to 'file:///tmp/make_secondary.cwl'
INFO [job make_secondary.cwl] /tmp/fk6y5rac$ docker \
run \
-i \
--mount=type=bind,source=/tmp/fk6y5rac,target=/jKWVxj \
--mount=type=bind,source=/tmp/v6sssram,target=/tmp \
--mount=type=bind,source=/tmp/cwl/a,target=/jKWVxj/a,readonly \
--mount=type=bind,source=/tmp/cwl/b,target=/jKWVxj/b,readonly \
--workdir=/jKWVxj \
--read-only=true \
--user=1000:1000 \
--rm \
--cidfile=/tmp/tg7_zeg5/20220613143313-613169.cid \
--env=TMPDIR=/tmp \
--env=HOME=/jKWVxj \
alpine \
true
INFO [job make_secondary.cwl] Max memory used: 0MiB
ERROR [job make_secondary.cwl] Job error:
("Error collecting output for parameter 'output': ../../../../../../tmp/make_secondary.cwl:33:5: 'path'", {})
WARNING [job make_secondary.cwl] completed permanentFail
{}
WARNING Final process status is permanentFail
It's something with the secondaryFiles
expression for the output
.
Hi all,
I have a workflow which requires downloading input files from s3 by uuid, (e.g. example.gz and example.gz.tbi are downloaded separately by its own uuid) and in later steps, these pre-downloaded files should be staged as a "file, secondary file" structure.
My intention is to use InitialWorkDirRequirement to avoid unnecessary copying of the input files.
Here is my cwl:
The output from cwltool engine is:
However, from the latest rabix 1.0.5:
I'm just wondering if it's noticed already and if there's a workaround for my case. Thank you very much in advance.