Skip to content

getting-started.md

Latest commit

 

History

History
27 lines (20 loc) · 2.08 KB

getting-started.md

File metadata and controls

27 lines (20 loc) · 2.08 KB
Raw

Getting started

  1. Set the site identity and canonical public URL in pyproject.toml (tool.orinoco.site).
  2. Run pixi run build and pixi run serve to preview the starter site.
  3. Replace the records in site-specific/metadata/records/ with your own metadata and rebuild.
  4. Add optional pages, assets, and appearance overrides under site-specific/; see file ownership.
  5. Configure repository Pages and curation settings before enabling hosted editing.

The site record's associated_with selects people; a project's part_of places it within the site. Projects name contributors through associated_with, and publications name authors through attributed_to. When changing a record's pid, update references to it in other records too.

To test local package edits, run pixi run dev-enable ../orinoco-lite-dev; restore the preceding package selection with pixi run dev-disable. Both tasks record the package selection, lockfile, and development link with DataLad, leaving unrelated site edits uncommitted. The recorded editable connection still depends on that local checkout.

Use template updates to update the scaffold and package through a GitHub draft pull request or the local CLI.

Use pixi run orinoco-lite validate to check inputs without generating a site or projection, and pixi run orinoco-lite verify-site build/site to check an existing local build without rebuilding it. Builds reuse unchanged metadata projections; pixi run build --no-cache repeats projection and semantic checks without fetching new source data.

For GitHub Pages, pixi run build-pages builds the website and emits its publication bundle from clean committed inputs. The workflow deploys the site, then runs publication record to retain that successful build outside the source branch. There is no separate preparation step and no rebuild after deployment.

Metadata acquisition and curation programs may live under extensions/ and run through explicit adapter tasks. They must write proposals or reviewed metadata inputs; the website build never imports or executes them.