I need some help with Krakenuniq_build and understanding inputs: The input directive is:

input:
    tuple val(meta), path(custom_library_dir, stageAs: "library/*"), path(custom_taxonomy_dir, stageAs: "taxonomy/*"), path(custom_seqid2taxid, stageAs: 'seqid2taxid.map')
    val keep_intermediate

The

meta.yml

describes each input as:

input:
  - - meta:
        type: map
        description: |
          Groovy Map containing sample information
          e.g. [ id:'test', single_end:false ]
    - custom_library_dir:
        type: directory
        description: Optional directory containing custom fasta files for database
          build
        pattern: "*"
    - custom_taxonomy_dir:
        type: directory
        description: Optional directory containing custom taxonomy files for
          database build
        pattern: "*"
    - custom_seqid2taxid:
        type: file
        description: custom seqid2taxid
        ontologies: []
  - keep_intermediate:
      type: boolean
      description: "Keep intermediate files that are not used by the database itself"
      pattern: true|false

The test supplies

custom_library_dir

as a

file

, and

custom_taxonomy_dir

as

directory

. This means though that when the inputs are staged in the process, it looks like

library 
  \- *.fasta
taxonomy
  \- folder_name (usually also taxonomy)
      |- names.dmp
      \- nodes.dmp

These are then `mv`ed into the database directory, where Krakenuniq recreates the

names.dmp

and

nodes.dmp

again in the

taxonomy

folder ( the parent one ). My problem is, I don't actually understand whether the inputs should be directories as listed in the

meta.yml

or if they should be files as suggested by the

input:

.

OK I have 10 minutes