← 返回命令列表

Linux command

cwl-runner 命令

文本

复制后可按需替换文件名、目录或参数。

常用示例

Run a CWL workflow

cwl-runner [workflow.cwl] [inputs.yml]

Run with specific input values

cwl-runner [workflow.cwl] --[input_name] [value]

Run with Docker disabled

cwl-runner --no-container [workflow.cwl] [inputs.yml]

Run with custom output directory

cwl-runner --outdir [/path/to/output] [workflow.cwl] [inputs.yml]

Validate a CWL document

cwl-runner --validate [workflow.cwl]

Run in parallel

cwl-runner --parallel [workflow.cwl] [inputs.yml]

说明

cwl-runner executes workflows defined in Common Workflow Language (CWL), a specification for describing analysis workflows in data-intensive sciences. CWL workflows are portable across different execution environments. The runner takes a CWL document and input parameters, resolves dependencies, and executes tools in the correct order. Steps can be run in containers (Docker/Singularity) for reproducibility, or natively if tools are installed locally. Input parameters can be specified in a YAML/JSON file or as command-line arguments. Outputs are placed in the specified output directory with a manifest describing the results.

参数

--validate
Validate the CWL document without running.
--outdir _path_
Output directory for results.
--no-container
Disable Docker/container execution.
--parallel
Run workflow steps in parallel.
--cachedir _path_
Directory for caching intermediate results.
--tmpdir-prefix _path_
Prefix for temporary directories.
--debug
Enable debug output.
--quiet
Only print warnings and errors.
--validate
Validate the CWL document without running.
--leave-tmpdir
Don't delete temporary directories.
--version
Print version and exit.
--make-template
Generate template input object.
--on-error _action_
Workflow behavior on failure (stop or continue).
--singularity
Use Singularity container runtime instead of Docker.
--podman
Use Podman container runtime instead of Docker.

FAQ

What is the cwl-runner command used for?

cwl-runner executes workflows defined in Common Workflow Language (CWL), a specification for describing analysis workflows in data-intensive sciences. CWL workflows are portable across different execution environments. The runner takes a CWL document and input parameters, resolves dependencies, and executes tools in the correct order. Steps can be run in containers (Docker/Singularity) for reproducibility, or natively if tools are installed locally. Input parameters can be specified in a YAML/JSON file or as command-line arguments. Outputs are placed in the specified output directory with a manifest describing the results.

How do I run a basic cwl-runner example?

Run `cwl-runner [workflow.cwl] [inputs.yml]` in a terminal, then adjust file names, paths, flags, or remote targets for your system.

What does --validate do in cwl-runner?

Validate the CWL document without running.