mcviewtest: spawn the new mcplotdiff-html plotter

[willend]

Free-form text area

Please describe what your PR is adding in terms of features or bugfixes:

---

Declaration of use of AI-tools

• Please add a checkmark here if you used AI-tools during the work for this contribution
• Furter, please describe how / where and for what the tools were used:

Claude, prompt:

"Now I'd like you to focus on the mcviewtest tool which is part of tools/Python/mctest/.

The mctest tool can generate a set of test data for McStas, based on %Example lines in the instr files. As part of this, it calls the mcplot-html plotter per test result.

The folder structure from an mctest can be found at e.g. http://new-nightly.mcstas.org/todays-datafiles/mcstas-3.99.99_mpi_x_20_1e7_Linux_20260709_0902_23/

Afterwards, mcviewtest is called to generate an overview table - including columns that correspond e.g. to CPU and GPU execution - or different operating systems, see the html file at http://new-nightly.mcstas.org/todays-datafiles/2026-07-09_output.html (spans the above mctest output + 2 other mctest's.

Consider the first column in the html file the 'reference' test. Comparing each of the OTHER columns to the first one, I'd like you to add 'difference' cells corresponding to a mcplotdiff-html comparion between the dataset outputs, e.g. McStas_8GPU_openacc_mpi_x_7_1e7_Linux_20260709_0001_06_1e7/BNL_H8/1/ and mcstas-3.99.99_mpi_x_20_1e7_Linux_20260709_0902_23/BNL_H8/1/ - adding a link the the mcviewtest template."

Various local adaptation added

---

Development OS / boundary conditions

Please describe what OS you developed and tested your additions on, and if any special dependencies are required:

---

PR Checklist for contributing to McStas/McXtrace

For a coherent and useful contribution to McStas/McXtrace, please fill in relevant parts of the checklist:

• My contribution contains something else

  • Explanation is added in free form text above or below the checklist

---

[willend]

A first demo of what this can do is available through http://new-nightly.mcstas.org/demo-diffplots/2026-07-09_output.html. Where possible (simulation worked and produced ascii output), the other datasets are compared against the left-most column (“ref”).

While this still has rough edges (e.g. very long legends on links etc.) I think it will be very useful for instantly getting a good idea of what is up when a test starts to fail numerically.

[willend]

With the latest few commits (at 5ac3e2c) the output got a couple of stylistic changes.

Overview:

(DIFF separated from rest of cell with a <hr>)

Diffplot:

(Datasets labeled ‘A’ and ‘B’ in header, referral to these in link labels)

[willend]

After last commit, individual ‘monitor’ plots look like this:


(with link to the underlying ascii datafiles)

where as the ‘diff-monitor’ plots look like this:


(the ascii data do not exist in ‘physical form’)

[willend]

Here is a link to the output from mcstas-conda-testsuite as run from the PR bramch
http://new-nightly.mcstas.org/combined/combined_output.html