Add marimo example

.gitignore

@@ -9,3 +9,6 @@ SolarStationsOrg-station-catalog.csv
 
 # MyST build outputs
 _build
+
+# Jupyter book
+docs/.jupyter-book-marimo

.readthedocs.yaml

@@ -8,15 +8,20 @@ build:
     python: "3.13"
     nodejs: "22"
   commands:
-    # Install Python dependencies
-    - pip install -r docs/requirements.txt
+    # Install Python dependencies via uv, creating .venv at the repo root
+    # (myst.yml's marimo plugin path, ../.venv/bin/jupyter-book-marimo, expects it there)
+    - UV_PROJECT_ENVIRONMENT=.venv uv sync
     # Install mystmd
     - npm install -g mystmd
     # Build the site (myst.yml lives in docs/, so build from there)
-    # --execute re-runs notebook code cells instead of using their stored outputs.
-    - cd docs && myst build --html --execute
+    - cd docs && UV_PROJECT_ENVIRONMENT=.venv uv run jupyter-book build --html
     # Copy the output to Read the Docs expected location
-    - mkdir -p $READTHEDOCS_OUTPUT/html/
-    - cp -r docs/_build/html/. "$READTHEDOCS_OUTPUT/html"
+    - mkdir -p "$READTHEDOCS_OUTPUT/html/"
+    - cd docs && cp -r _build/html/. "$READTHEDOCS_OUTPUT/html/"
     # Clean up build artifacts
-    - rm -rf docs/_build
+    - cd docs && rm -rf _build
+
+python:
+  install:
+    - method: uv
+      command: sync

docs/myst.yml

@@ -20,9 +20,14 @@ project:
   settings:
     output_matplotlib_strings: remove
 
+  plugins:
+    - type: executable
+      path: ../.venv/bin/jupyter-book-marimo
+
   toc:
     - file: intro.ipynb
     - file: station_catalog.ipynb
+    - file: station_catalog_marimo.md
     - file: station_requirements.md
       title: Station requirements
     - file: station_metadata.md

docs/requirements.txt

@@ -10,3 +10,4 @@ unidecode==1.4.0
 openpyxl==3.1.5
 xarray==2026.7.0
 h5netcdf==1.8.1
+jupyter-book-marimo==0.0.1

docs/station_catalog_marimo.md

@@ -0,0 +1,72 @@
+```{marimo} python
+import marimo as mo
+import pandas as pd
+import numpy as np
+#import kgcpy
+```
+
+
+```{marimo} python
+main_url = "https://raw.githubusercontent.com/AssessingSolar/solarstations/refs/heads/main/solarstations.csv"
+esmap_url = "https://raw.githubusercontent.com/AssessingSolar/solarstations/refs/heads/main/esmap_stations.csv"
+country_by_continent_url = "https://raw.githubusercontent.com/AssessingSolar/solarstations/refs/heads/main/data/country_by_continent.json"
+nasa_power_url = "https://raw.githubusercontent.com/AssessingSolar/solarstations/refs/heads/main/data/nasa_power_annual_irradiance_global.csv"
+
+main_stations = pd.read_csv(main_url, dtype={'Tier': str}).fillna('')
+esmap_stations = pd.read_csv(esmap_url, dtype={'Tier': str}).fillna('')
+stations = pd.concat([main_stations, esmap_stations], axis='rows', ignore_index=True)
+
+country_by_continent = pd.read_json(country_by_continent_url, typ="series").to_dict()
+
+stations['Continent'] = stations['Country'].map(country_by_continent)
+
+#stations['Koeppen Geiger classification'] = stations.apply(lambda x: kgcpy.lookupCZ(x['Latitude'], x['Longitude']), axis=1)
+#kg_climates = {'A': 'Tropical', 'B': 'Dry', 'C': 'Temperate', 'D': 'Continental', 'E': 'Polar', 'O': 'Ocean'}
+#stations['Koeppen Geiger climate zone'] = stations['Koeppen Geiger classification'].apply(lambda x: kg_climates[x[0]])
+
+instrumentation = stations['Instrumentation'].str.split(';', expand=True)
+is_tier_1 = (instrumentation=='G').any(axis=1) & (instrumentation=='B').any(axis=1) & (instrumentation=='D').any(axis=1)
+stations['Tier'] = 2 - is_tier_1.astype(int)
+
+annual_irradiance = pd.read_csv(nasa_power_url, index_col=[0, 1])
+for index, row in stations.iterrows():
+    lat_round = round(row['Latitude']*2-0.5, 0)/2 + 0.25
+    lon_round = round(row['Longitude']*2-0.5, 0)/2 + 0.25
+    try:
+        stations.loc[index, ['GHI_typical_kWh_m2', 'DHI_typical_kWh_m2', 'DNI_typical_kWh_m2']] = \
+            annual_irradiance.loc[(lat_round, lon_round), :]
+    except KeyError as e:
+        pass
+    # Manual add data missing from the climatological file (data retrieved from NASA's webinterface)
+    if row['Station name'] == 'Funafuti':
+        stations.loc[index, ['GHI_typical_kWh_m2', 'DHI_typical_kWh_m2', 'DNI_typical_kWh_m2']] = \
+            np.array([5.33*365, 2.12*365, 4.37*365]).astype(int)
+    if row['Station name'] == 'South Pole':
+        stations.loc[index, ['GHI_typical_kWh_m2', 'DHI_typical_kWh_m2', 'DNI_typical_kWh_m2']] = \
+            np.array([3.0*365, 1.38*365, 5.13*365]).astype(int)
+
+stations = stations[~stations['Instrumentation'].str.contains('G;Ds')]  # remove Tier 3 stations
+
+stations = stations.sort_values('Station name', ignore_index=True)
+
+## Write file containing all columns, linked to above
+#stations.to_csv('./SolarStationsOrg-station-catalog.csv', index=False)
+
+## Format station name as hyperlinks if URL is available.
+## Do this after the new CSV is written, since we don't want HTML in that.
+#for index, row in stations.iterrows():
+#    if row['URL'].startswith('http'):
+#        stations.loc[index, 'Station name'] = f'<a href="{row["URL"]}">{row["Station name"]}</a>'
+
+
+table = mo.ui.table(
+    stations,
+    pagination=True,
+    freeze_columns_left=["Station name"],
+    hidden_columns=["URL"],
+)
+```
+
+```{marimo} python
+table
+```

pyproject.toml

@@ -0,0 +1,23 @@
+[project]
+name = "SolarStations.org"
+version = "0.1.0"
+description = "A catalog of multi-component solar irradiance monitoring stations."
+readme = "README.md"
+requires-python = ">=3.13"
+dependencies = [
+    "folium==0.20.0",
+    "itables==2.2.2",
+    "jupyter-book==2.1.6",
+    "kgcpy==1.1.8",
+    "matplotlib==3.11.1",
+    "numpy==2.5.1",
+    "pandas==3.0.3",
+    "pvlib==0.15.2",
+    "unidecode==1.4.0",
+    "openpyxl==3.1.5",
+    "xarray==2026.7.0",
+    "h5netcdf==1.8.1",
+    "jupyter-book==2.1.6",
+    "jupyter-book-marimo>=0.0.2",
+    "marimo",
+]