Skip to content

Conversation

@tsteenbe
Copy link
Member

See individual commits for details.

Update dependencies to latest available versions to benefit from
upstream improvements.

Signed-off-by: Thomas Steenbergen <opensource@steenbe.nl>
Include the missing plugin generation command and
add descriptions for each command for better clarity.

Signed-off-by: Thomas Steenbergen <opensource@steenbe.nl>
Move the docs directory from the website directory back to the root of the repository.
This change improves discoverability for users and aligns with
standard best practices used by many open source projects.

Signed-off-by: Thomas Steenbergen <opensource@steenbe.nl>
@tsteenbe tsteenbe added the documentation About end-user documentation label Dec 31, 2025
@tsteenbe tsteenbe requested review from a team as code owners December 31, 2025 05:47
Signed-off-by: Thomas Steenbergen <opensource@steenbe.nl>
@codecov
Copy link

codecov bot commented Dec 31, 2025

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 57.41%. Comparing base (1ae59c2) to head (1c2faab).

Additional details and impacted files
@@            Coverage Diff            @@
##               main   #11275   +/-   ##
=========================================
  Coverage     57.41%   57.41%           
  Complexity     1705     1705           
=========================================
  Files           346      346           
  Lines         12875    12875           
  Branches       1228     1228           
=========================================
  Hits           7392     7392           
  Misses         5005     5005           
  Partials        478      478           
Flag Coverage Δ
funTest-external-tools 13.67% <ø> (ø)
funTest-no-external-tools 30.93% <ø> (ø)
test-ubuntu-24.04 42.40% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

"@mdx-js/react": "^3.0.0",
"clsx": "^2.0.0",
"prism-react-renderer": "^2.3.0",
"@easyops-cn/docusaurus-search-local": "^0.52.2",
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Any idea why these updates were not covered by Renovate bot?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I was wondering the same when making these updates but couldn't find any wrong within renovate.json.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could it be, that it's configured to not update lockfiles?

Copy link
Member

@mnonnenmacher mnonnenmacher Jan 22, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As long as the version ranges defined in package.json still match the new version, Renovate will only update the package-lock.json. I think when using Renovate it's probably best to not use version ranges in the package.json at all because it makes it more obvious which versions are actually used.
Most updates to the package.json in this PR are just aligning with the versions already used in package-lock.json, for example, for clsx or prism-react-renderer. The React dependency was not upgraded by Renovate because #9565 was closed because Docusaurus was not compatible with React 19 back then.

Copy link
Member

@mnonnenmacher mnonnenmacher Jan 23, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I made a PR to let renovate pin the dependency versions (#11359) and will reopen #9565 afterwards so that the package.json updates will be done independently of the progress in this PR.


@OutputDirectory
val outputDirectory = project.layout.projectDirectory.file("website/docs/plugins").asFile
val outputDirectory = project.layout.projectDirectory.file("docs/plugins").asFile
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure about this change. There are website- / Docusaurus-specific files in docs, i.e. it's not generic Markdown files only. Also the argument about discoverability does not convince me, as users are not supposed to discover these from the repo, but to browse the website.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I planning to remove any website- / Docusaurus-specific files from docs so we just left with the Markdown files. Have done lots of ORT onboarding sessions and trainings - and often get asked this 'where are the docs' question so this one has been high on my over to-do lists. Working on bigger rewrite of the docs have been playing with exporting the docs a user manual as a PDF/ePub and importable wiki.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I planning to remove any website- / Docusaurus-specific files from docs so we just left with the Markdown files.

These files have a purpose, what will be the replacement?
The Markdown files also have Docusaurus specific content like frontmatter and Docusaurus specific extensions, what about those?

Have done lots of ORT onboarding sessions and trainings - and often get asked this 'where are the docs' question so this one has been high on my over to-do lists.

IMO the preferred way to consume the docs should be the website, not Markdown files rendered by GitHub. But if the issue is that people are not finding the docs on the website (despite the prominent link in the README), moving the folder will only promote reading them on GitHub. I would prefer to understand why people are not findings the docs and address that in a way that points them to the website.

playing with exporting the docs a user manual as a PDF/ePub and importable wiki

Is PDF export really something that anybody needs? And what is an importable wiki?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These files have a purpose, what will be the replacement?
The Markdown files also have Docusaurus specific content like frontmatter and > Docusaurus specific extensions, what about those?

I wrote only about removing Docusaurus specific files, but as part of the ongoing ORT docs rewrite effort I have moved features relying on Docusaurus file comments to Docusaurus configuration files.

IMO the preferred way to consume the docs should be the website, not Markdown files rendered by GitHub.

Enterprise users usually do not operate on the latest versions of tools; they are often two to nine months behind. This delay is understandable as thorough testing prior to rolling out a new version is needed to ensure there are no tool regressions and that the new tool versions works well within their established processes and any extensions.

ORT is no different - enterprise users that are very active in the ORT community usually are less than month in the version of ORT they use don't confuse them as being what standard among all ORT users. Past experiences from issues have shown that many users are typically three to six months behind the latest release.

The ORT users within enterprises need docs matching the "validated version" of ORT (fork) used within their organization. The ORT website only contains docs for the latest version of ORT which is why I have had requests to have ORT user guide PDFs to be include in each release or support formats that enable ORT docs to be imported into wiki such as Confluence. Discussed as a solution that they could host the ORT website within their organization but that is usually not feasible due to internal operational/resources/security challenges so in the end most up having their users read ORT docs on their organization's GitHub/GitLab server.

Creation of package curations and configurations within large enterprises is typically done by teams in Asia in for instance India or Philipines. People in these "compliance/clearance" teams typically work based on documented procedure which are underpinned by ORT docs. By making ORT docs in the future available in other formats we enable easier remixing of its contents into for instance organization specific ORT training materials and ORT operating procedure guidelines.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Enterprise users usually do not operate on the latest versions of tools; they are often two to nine months behind.

But that's a different problem statement than the original "docs discoverability" argument. To solve that problem, I'd prefer the website to have a version-dropdown (similar to Gradle's user manual) for user's to choose the rendered docs for a specific (past) release of ORT.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Our weekly release schedule would result in a very long version-dropdown on the website which isn't very user‑friendly. We should separate docs from the site so they can be remixed and published in other formats — including downloadable ORT docs, which this article recommends for the Japanese market.

Excerpt:

In the U.S., the typical flow is that a prospect visits a website, books a demo on their own, or drops off and is followed up aggressively by sales. A conversation happens quickly, a proof of concept or trial begins, and the deal either closes or the buyer moves on to a competitor. The cycle is fast and transactional. Buyers are comfortable testing tools, replacing them, and switching vendors if something does not work.
In Japan, the process is slower and more deliberate. A company will usually download product materials first.

I heard the same from Japanese and Korean companies at Open Source Summit Japan (I was in Tokyo last December for the Open Source Summit Japan / Open Compliance Summit) - they, along with Germany, care about open source compliance done properly. To attract users and contributors from those markets, we need to support their preferred evaluation flow by offering downloadable documentation.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have the feeling that this discussion is too broad for a PR review, I would suggest to put it on the agenda of one of the recurring meetings to get an alignment on the goals first before talking about technical details.

- pattern: "docs/**"
reason: "DOCUMENTATION_OF"
comment: >-
This directory contains project website with examples using licenses that do not apply to the OSS Review Toolkit.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The "examples [...] that do not apply to the OSS Review Toolkit" part sounds a bit odd to me, as if we were giving examples that are not workable.

Maybe better something like: "This directory contains documentation that is not distributed as part ORT releases."

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

documentation About end-user documentation

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants