[BREAKING] KeyError: '_samples' Cant validate schema about eido HOT 11 CLOSED

Ok so... New Year and I have finally taken enough of a break and reset to fix this issue once and for all and it's a stupid mistake XD.

So basically my issue was an indentation (or lack thereof) in my schema reference file. Basically, I messed up the indentation, so I had:

name: "My Pipeline"
description: My Description
imports:
  - http://schema.databio.org/pep/2.1.0.yaml
properties:
  samples:
  type: array
  items:
    ...<insert rest here>

when I needed to add indents like so:

name: "My Pipeline"
description: My Description
imports:
  - http://schema.databio.org/pep/2.1.0.yaml
properties:
  samples:
    type: array
    items:
      ...<insert rest here>

The moral of the story, if you can't fix it and you're getting frustrated, take a break 😂

from eido.

Could you post your PEP and the schema here? Also, what versions of peppy and eido are you using?

from eido.

Peppy: v0.31.1
Eido: v0.1.5

PEP:

description: "Schema for PEP sample declaration information for this workflow"
pep_version: "2.1.0"
sample_table: "samples.csv"
subsample_table: "partitions.csv"

**PEP Schema: **

name: "My Pipeline"
description: My Description
imports:
  - http://schema.databio.org/pep/2.0.0.yaml
properties:
  sample_name:
    type: string
    description: Identifier
  dataset:
    type: string
    description: Dataset Name
  file:
    type: string
    description: Filename
  reference_genome:
    type: string
    description: reference ID
required:
  - sample_name
  - dataset
  - file
  - reference_genome
files:
  - file

from eido.

I can confirm this also happens on eido v0.1.6-dev

from eido.

Hi @G-kodes your schema isn't reflecting the structure of the PEP correctly, which has both config and samples sections. Update your schema to this:

name: "My Pipeline"
description: My Description
imports:
  - http://schema.databio.org/pep/2.0.0.yaml
properties:
  samples:
    type: array
    items:
      type: object
      properties:
        sample_name:
          type: string
          description: Identifier
        dataset:
          type: string
          description: Dataset Name
        file:
          type: string
          description: Filename
        reference_genome:
          type: string
          description: reference ID
      required:
        - sample_name
        - dataset
        - file
        - reference_genome
      files:
        - file
required: 
  - samples

It should work. You can see some more example schemas here: https://github.com/databio/schema.databio.org/tree/master/pipelines

from eido.

Your schema wasn't what you intended, but also, eido wasn't correctly interpreting this schema (which should have been asking for those properties on the main item). So, the error should have been different than what it showed, which may have pointed you down the path to fix your schema.

So, I've fixed the error message, so this will be more informative if a future user tries to do the same thing. This will show up on the next release (0.1.6).

from eido.

@nsheff thank you very much for the help! I guess this is chalked up to me not understanding the documentation properly so feeling a bit dumb this side XD.

I have updated my schema as you have directed, however unfortunately I am still getting an error:

Traceback (most recent call last):
  File "/opt/homebrew/lib/python3.9/site-packages/snakemake/__init__.py", line 593, in snakemake
    workflow.include(
  File "/opt/homebrew/lib/python3.9/site-packages/snakemake/workflow.py", line 1182, in include
    exec(compile(code, snakefile.get_path_or_uri(), "exec"), self.globals)
  File "/Users/g-kodes/Documents/Pharmacogenetic-Analysis-Pipeline/workflow/Snakefile", line 37, in <module>
  File "/opt/homebrew/lib/python3.9/site-packages/snakemake/workflow.py", line 1267, in pepschema
    eido.validate_project(project=pep, schema=schema, exclude_case=True)
  File "/opt/homebrew/lib/python3.9/site-packages/eido/validation.py", line 45, in validate_project
    _validate_object(project_dict, preprocess_schema(schema_dict), exclude_case)
  File "/opt/homebrew/lib/python3.9/site-packages/eido/schema.py", line 32, in preprocess_schema
    "items" in schema_dict[PROP_KEY]["_samples"]
TypeError: argument of type 'NoneType' is not iterable

Am I just missing something basic here?

from eido.

Hmm. I am not reproducing this, using either eido 0.1.5 or 0.1.6. Can you provide your complete PEP (you posted the yaml, I need the csvs) by chance? This error appears to be indicating that you don't have any samples in your PEP, but that's confusing to me. It looks like the samples array in your schema isn't being recognized.

Are you sure you copied the schema correctly?

from eido.

Also, you might try with the latest dev version which is almost released, (0.1.6)

pip install https://github.com/pepkit/eido/archive/refs/heads/dev.zip

You can check version with eido --version.

from eido.

Hi! My apologies, this is turning into a support ticket!

My edit version output is eido 0.1.5

and my complete PEP file is:

description: "Schema for PEP sample declaration information for this workflow"
pep_version: "2.1.0"
sample_table: "samples.csv"

I have attempted to use the newer Eido version you indicated above, however it appears to be dependant on an un-released version of peppy, version >=0.32.0. I can only find up till version 0.31.2 online?

from eido.

If samples.csv doesn't exist, you should be getting this exception:

peppy.exceptions.SampleTableFileException: Could not read table: samples.csv. Caught exception: FileNotFoundError(2, 'No such file or directory')

If I do have a CSV file with at least 1 sample, I can validate that PEP with eido 0.1.5:

cat samples.csv

sample_name,attr,dataset,file,reference_genome
test1,test2,test,test,test

eido --version
> eido 0.1.5
eido validate pep.yaml -s pep_schema.yaml 
> Validation successful
cat pep.yaml

Result:

description: "Schema for PEP sample declaration information for this workflow"
pep_version: "2.1.0"
sample_table: "samples.csv"

And for the schema:

cat pep_schema.yaml

Result:

name: "My Pipeline"
description: My Description
imports:
  - http://schema.databio.org/pep/2.0.0.yaml
properties:
  samples:
    type: array
    items:
      type: object
      properties:
        sample_name:
          type: string
          description: Identifier
        dataset:
          type: string
          description: Dataset Name
        file:
          type: string
          description: Filename
        reference_genome:
          type: string
          description: reference ID
      required:
        - sample_name
        - dataset
        - file
        - reference_genome
      files:
        - file
required:
  - samples

Really, eido 0.1.5 should be working for you. But anyway, if you want to upgrade, peppy 0.32.0 is pending release. You can install 0.32.0 of peppy in the same way, sorry about that:

pip install https://github.com/pepkit/eido/archive/refs/heads/dev.zip

from eido.