Add SciLog ELN format description and example

[omkar-ethz]

Overview

In the 12th Consortium meeting, we presented SciLog's ELN export feature, which was still in development. Now that the feature is merged in SciLog, and we have a go-ahead from our other project collaborators in Switzerland, this PR aims to include SciLog as compatible with ELN File Format. Currently, only ELN export is implemented. ELN import will be implemented in the next couple of months.

Notes

• Added a SciLog example folder, with README describing SciLog's eln concepts
• Added an example .eln export (attached below is a screen recording showing the process in SciLog that was used to generate this example)
• Although there is no public test instance of SciLog, one can easily spin it up locally using the provided docker-compose file, and test the ELN and/or other features for themselves.

Screen recording

Screencast.from.01-26-2026.03.19.30.PM.webm

[NicolasCARPi]

I'll need to adapt my import code so it doesn't ignore a node with @type having multiple values. But I had a look and it looks alright. Once CI is green, please mark it ready for review and ask reviewers (typically me + @SteffenBrinckmann + @FlorianRhiem )

[SteffenBrinckmann]

Thank you for your PR.
@type has in the past caused issues (e.g. #103 and #104): the current way the implementations/specification work is that @type is a single string "File", "Datatype", "Comment" and that all the other categorization is going into the "genre":["Topic1","Topic2"]
Given that the issue keeps on popping up, we might have to discuss in the future meeting again.

[omkar-ethz]

@NicolasCARPi thank you for your feedback! I updated the example to pass the CI test, but also had to update the test code slightly to

1. accomodate for @type being array in the SciLog ELN
2. replace the ro-crate version to 1.1 (same as SampleDB) for roc-validator.

I think with this new example the PR is ready for review. I cannot add reviewers, so just mentioning @SteffenBrinckmann and @FlorianRhiem :)

[SteffenBrinckmann]

Looks great to me.
Since this changes the specification and we had discussed the @type=string or array issue a number of times, I do not want to just accept it without talking to the entire gang. Since the next meeting is coming up next month, lets wait for that meeting and re-discuss the issue again.

[omkar-ethz]

Thank you @SteffenBrinckmann ! Let's discuss it in the meeting next week.
In the meantime, could someone please approve the workflow run? Although pytest passes locally, I'd like to check it also passes in the CI

[FlorianRhiem]

I'm just working on supporting the nested Dataset nodes and now the lack of a contentSize for File nodes is causing errors for me. Do you think you'll be able to add contentSize (and possibly sha256) to File nodes?

[omkar-ethz]

I'm just working on supporting the nested Dataset nodes and now the lack of a contentSize for File nodes is causing errors for me. Do you think you'll be able to add contentSize (and possibly sha256) to File nodes?

We don't store these attributes currently, but i'm checking if we can add it to new logbooks, will get back to you in a couple of days.

[omkar-ethz]

I'm just working on supporting the nested Dataset nodes and now the lack of a contentSize for File nodes is causing errors for me. Do you think you'll be able to add contentSize (and possibly sha256) to File nodes?

Hi @FlorianRhiem , we have implemented the contentSize and sha256 for SciLog exports. I have updated the example .eln with a new export that contains these attributes.

@NicolasCARPi @SteffenBrinckmann any other blockers for merging this PR? :D