Re-word CI tests

This commit is contained in:
drpatelh 2021-02-07 23:40:26 +00:00
parent 6bbb71a022
commit 07f62a76c5

View file

@ -23,7 +23,7 @@ A repository for hosting [Nextflow DSL2](https://www.nextflow.io/docs/latest/dsl
- [Adding a new module file](#adding-a-new-module-file) - [Adding a new module file](#adding-a-new-module-file)
- [Module template](#module-template) - [Module template](#module-template)
- [Guidelines](#guidelines) - [Guidelines](#guidelines)
- [Testing](#testing) - [CI tests](#ci-tests)
- [Documentation](#documentation) - [Documentation](#documentation)
- [Uploading to `nf-core/modules`](#uploading-to-nf-coremodules) - [Uploading to `nf-core/modules`](#uploading-to-nf-coremodules)
- [Terminology](#terminology) - [Terminology](#terminology)
@ -224,19 +224,17 @@ The `saveFiles` function can be found in the [`functions.nf`](software/fastqc/fu
We also use a standardised parameter called `params.publish_dir_mode` that can be used to alter the file publishing method (default: `copy`). We also use a standardised parameter called `params.publish_dir_mode` that can be used to alter the file publishing method (default: `copy`).
### Testing ### CI tests
- All test data for `nf-core/modules` MUST be added to [`tests/data/`](tests/data/) and organised by filename extension. - All test data for `nf-core/modules` MUST be added to [`tests/data/`](tests/data/) and organised by filename extension.
- In order to keep the size of this repository as minimal as possible, pre-existing files from [`tests/data/`](tests/data/) MUST be reused if at all possible.
- Test files MUST be kept as tiny as possible. - Test files MUST be kept as tiny as possible.
- Every module MUST be tested by adding a test workflow with a toy dataset in the [`tests/`](tests/software/fastqc/main.nf) directory of the module. - Every module MUST have a test workflow utilising test data added to the appropriate directory e.g. [`tests/software/fastqc/main.nf`](tests/software/fastqc/main.nf)
- Generic files from [`tests/data/`](tests/data/) MUST be reused by importing them as `file(${launchDir}/tests/data/fastq/rna/test_single_end.fastq.gz)` - Any outputs produced by the test workflow MUST be included in the [pytest-workflow](https://pytest-workflow.readthedocs.io/en/stable) for that tool e.g. [`tests/software/fastqc/test.yml`](tests/software/fastqc/test.yml). `md5sum` checks are the preferable choice of test to determine file changes, however, this may not be possible for all outputs generated by some tools e.g. if they include time stamps or command-related headers. Please do your best to avoid just checking for the file being present e.g. it may still be possible to check that the file contains the appropriate text snippets.
- Any outputs produced by the test workflow MUST be included in the [pytest-workflow](https://pytest-workflow.readthedocs.io/en/stable) for that tool. md5sum is preferred, however it's acceptable to not have it on files that the hash changes due to various headers and timestamps (html). Please do your best to avoid just checking for the file being present.
- If the appropriate test data doesn't exist for your module then it MUST be added to [`tests/data/`](tests/data/).
- A filter for the module must be created in [`.github/filters.yml`](.github/filters.yml). Please include any paths specific for that tool or upstream of that tool (For example bowtie build is upstream of bowtie align). - A filter for the module must be created in [`.github/filters.yml`](.github/filters.yml). Please include any paths specific for that tool or upstream of that tool (For example bowtie build is upstream of bowtie align).