Preview this book locally

Use these commands from the repository root (the folder that contains myst.yml). After the server starts, open one book URL: http://localhost:3000/.

Jupyter Book 2 also starts a file/content server (usually port 3100). That is part of the same preview, not a second book. Prefer Chrome or Edge for the navbar PDF link (editor-embedded browsers often show a blank PDF). Or open exports/sciml_ps_notes.pdf on disk after scripts/build_pdf.ps1.

1. One-time environment setup¶

The course environment installs Python, Jupyter Book 2, and Node.js (required by JB2’s MyST site server).

conda env create -f environment.yml
conda activate sciml_ps

On Windows PowerShell you can instead run:

.\scripts\setup_env.ps1
conda activate sciml_ps

If the environment already exists and dependencies changed (for example after Node.js was added):

conda env update -f environment.yml --prune
conda activate sciml_ps

Confirm both tools:

jupyter book --help
node --version

2. Live preview (recommended while reading or editing)¶

Option A — helper script (Windows)¶

conda activate sciml_ps
.\scripts\preview.ps1

Optional port or notebook execution at build time:

.\scripts\preview.ps1 -Port 3001
.\scripts\preview.ps1 -Execute

Option B — helper script (macOS / Linux)¶

conda activate sciml_ps
chmod +x scripts/preview.sh scripts/build.sh   # once
./scripts/preview.sh

./scripts/preview.sh --port 3001
./scripts/preview.sh --execute

Option C — raw Jupyter Book commands¶

Use HOST=localhost with --keep-host so the theme UI and the content/PDF server agree on hostnames (PDF links redirect to http://localhost:3100/...). Helper scripts already do this.

conda activate sciml_ps
$env:PYTHONUTF8 = "1"
$env:HOST = "localhost"
jupyter book start --port 3000 --keep-host

conda activate sciml_ps
export HOST=localhost
jupyter book start --port 3000 --keep-host

jupyter book start --port 3001 --keep-host   # only if 3000 is busy
jupyter book start --execute --keep-host

Edit any page under book/, save, and refresh the browser.

Open in the browser: http://localhost:3000/
The preview process must stay running in that terminal (do not close it). Do not start a second preview while one is running.

3. Static HTML build (no live server)¶

Windows¶

conda activate sciml_ps
.\scripts\build.ps1

macOS / Linux¶

conda activate sciml_ps
./scripts/build.sh

Raw command¶

jupyter book build

Open the index.html path printed by the CLI (often under _build/).

4. What you should see¶

• Sidebar parts: Front matter, Preparation, Foundations, Dynamics, Stability, Identification, Advanced, Laboratories, Reference

• This page under Front matter

• Stub chapters until prose is filled

5. Troubleshooting¶

Maintainer detail (PDF engine, cover provenance, hosting): docs/guides/site_hosting.md. Editing / long-term growth: docs/guides/ (not book pages).