I want to use the snakemake utility (5.6.0) to use files stored on the EGA. First I wanted to try the code written in the official documentation, so I tried this:
import snakemake.remote.EGA as EGA
ega = EGA.RemoteProvider()
rule get_remote_file_ega:
input:
ega.remote("ega/dataset_id/foo.bam")
output:
"data/foo.bam"
shell:
"cp {input} {output}"
Before executing the script I created environment variables as necessary (EGA_USERNAME and EGA_PASSWORD).
Then I get the following error:
me:~/scripts$ snakemake -s test_ega.smk
Building DAG of jobs...
Traceback (most recent call last):
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/__init__.py", line 551, in snakemake
export_cwl=export_cwl)
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/workflow.py", line 433, in execute
dag.init()
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/dag.py", line 122, in init
job = self.update([job], progress=progress)
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/dag.py", line 603, in update
progress=progress)
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/dag.py", line 655, in update_
missing_input = job.missing_input
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/jobs.py", line 396, in missing_input
for f in self.input
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/jobs.py", line 397, in <genexpr>
if not f.exists and not f in self.subworkflow_input)
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/io.py", line 208, in exists
return self.exists_remote
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/io.py", line 119, in wrapper
v = func(self, *args, **kwargs)
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/io.py", line 258, in exists_remote
return self.remote_object.exists()
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/remote/EGA.py", line 173, in exists
return self.parts.path in self.provider.get_files(self.parts.dataset)
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/remote/EGA.py", line 126, in get_files
"data/metadata/datasets/{dataset}/files".format(dataset=dataset))
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/remote/EGA.py", line 96, in api_request
headers["Authorization"] = "Bearer {}".format(self.token)
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/remote/EGA.py", line 77, in token
self._login()
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/remote/EGA.py", line 45, in _login
"client_id" : self._client_id(),
File "/home/puissant/miniconda3/lib/python3.7/site-packages/snakemake/remote/EGA.py", line 151, in _client_id
return self._credentials("EGA_CLIENT_ID")
NameError: name 'self' is not defined
The part of the code involved is there (EGA.py line 151):
1 @classmethod
2 def _client_id(cls):
3 return self._credentials("EGA_CLIENT_ID")
Could the error come from a "self" instead of a "cls" on line 3? Because after changing it to "cls" the error moved to the next block, built in the same way. My understanding of python objects being limited, I hope I don't say great absurdities.
Have I forgotten any steps or misunderstood any of them?
You are correct, you should be using cls (which presumably stands for 'class' here) rather than self. self is generally the name used for instances of classes i.e. objects. If you are using self elsewhere in the function, you need to switch them to cls.