New module: GRIMER

[Jonahnki]

Closes #11566

Description

Adds a new nf-core module for GRIMER (v1.1.0), a tool that generates interactive HTML dashboards for contamination detection in metagenomics and viromics datasets.

Developed during the May 2026 nf-core x VirJenDB Virus Bioinformatics Hackathon.

Checklist

• environment.yml matches bioconda package
• main.nf passes lint (47/47 tests passed)
• meta.yml complete and schema-valid
• nf-test dry run passes (2/2 tests discovered)
• All pre-commit hooks pass

[Jonahnki]

The stub test now uses an existing nf-core/test-datasets file to pass CI. I hope to add in a proper GRIMER-specific count table TSV to nf-core/test-datasets in a follow-up post-PR after review

[SPPearce]

Can you create this file on-the-fly within the nf.test please.

[Jonahnki]

Thanks for the latest comments @SPPearce! I've gone through everything and pushed some fixe

*The metadata input in main.nf was tripping the linter since anything prefixed with meta gets treated as a Groovy map and renamed it to sample_metadata and that cleared it up
*Synced meta.yml to match while I was at it; --fix had silently collapsed the sample_metadata block to {} so I restored the full field descriptions manually. All 51 lint checks are green now
*For the test part, I swaped out the bare assert calls for snapshot(process.out).match() and added a -stub block. Test data is now generated on-the-fly in setup {} so the static tests/testdata/test_count_table.tsv wasn't needed anymore and I've removed it
*.nf-core.yml is also back to upstream state

[Jonahnki]

Managed to fix the two GRIMER-specific CI failures in the latest push:

*Version eval: grimer --version outputs a multi-line ASCII art banner rather than a clean version string, which caused the snapshot mismatch on CI. Changed the eval to grimer --version 2>&1 | tail -1 which reliably extracts just the bare version number from the last line of the output
*meta.yml sync: ran nf-core modules lint --fix to update the version eval string in both the outputs and topics blocks to match the updated main.nf. All 51 lint checks now passing and the stub snapshot has been regenerated to match

The remaining CI failures (SENTIEON, HMMER, WFMASH, SALSA2, KRAKENUNIQ etc.) are infrastructure-level flakes — runner disk full (no space left on device) and conda HTTP 403 errors and unrelated to this PR

[Jonahnki]

The confirm-pass-lint gate failure is downstream of the multiqc and multiqcsav lint failures, which seems like pre-existing issues in those modules unrelated to this PR. The GRIMER module lint passes cleanly. I'm happy to have a reviewer confirm it

[SPPearce]

The remaining CI failures (SENTIEON, HMMER, WFMASH, SALSA2, KRAKENUNIQ etc.) are infrastructure-level flakes — runner disk full (no space left on device) and conda HTTP 403 errors and unrelated to this PR

You need to semi-regularly update the branch, so it only tests your changes, rather than unrelated other modules.

[SPPearce]

Can you use the test file you are generating in the test currently with an actual run, not just a stub test, to get some output?

[SPPearce]

This file is not required (and hasn't been for some time)

[SPPearce]

Suggested change

	def m_arg = sample_metadata ? "-m ${sample_metadata}" : ''
	def m_arg = sample_metadata ? "-m ${sample_metadata}" : ''

[SPPearce]

This isn't correct ;)
Need to run the snapshot with a profile (docker/singularity/conda) that has the tool.