Release 3 0

CLAUDE.md

@@ -0,0 +1,39 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Repository overview
+
+Static website for the PhotoDissociation Region Toolbox (dustem.astro.umd.edu), served via Apache SSI (Server Side Includes — `<!--#include virtual="..." -->`). Describes PDR modeling, hosts model images/FITS files, and documents the `pdrtpy` Python package (https://pdrtpy.readthedocs.io). There is no build step, bundler, or JS framework for the static pages — HTML is either hand-edited or generated by the Python script in `autogen/`.
+
+## Regenerating the model pages
+
+The `models/` directory (one subdirectory per ModelSet, e.g. `wk2006_Z1.0_constant_density`, `smc_Z0.2_L45_constant_density`) is generated output, not hand-edited. Each ModelSet directory contains generated `.fits`, `.png`, and `.html` files for every model ratio/quantity, plus an `index.html`.
+
+To regenerate:
+
+```
+cd autogen
+python make_model_webpages.py            # generate all non-KOSMA-tau ModelSets
+python make_model_webpages.py -k         # also include KOSMA-tau (kt*) models
+python make_model_webpages.py -m <name>  # only a given modelset
+python make_model_webpages.py -q         # quick: skip plot generation, only rebuild all_models index
+```
+
+This requires `pdrtpy` installed (`ModelSet`, `ModelPlot` from `pdrtpy.modelset` / `pdrtpy.plot.modelplot`) and writes output to `../models.new/` (a sibling directory, NOT `../models/` — output must be reviewed and manually promoted/renamed into `models/` before it takes effect on the site). The script pools across `os.cpu_count() - 2` processes via `multiprocessing.Pool`.
+
+Key behaviors to know when touching `autogen/make_model_webpages.py`:
+
+- `lmc` and `wk2006` model name prefixes are currently skipped entirely (line ~65), as is any `kt*` set when `-k`/`--kosmatau` is not passed.
+- Per-ModelSet HTML "explain" text and display titles are hardcoded in `explain` and `model_title` dicts inside `process_modelset()` — add entries there when introducing a new ModelSet family (matched by prefix of `ms.keyname`).
+- Model titles/units arrive from `pdrtpy` containing LaTeX (`$...$`); the script does manual `.replace()` cleanup to turn them into HTML/entities before rendering. If `$` still appears in a title after cleanup, the script prints `############ OOPS missed some latex ...` — check for this in output when adding new model quantities.
+- Ionized-gas line-ratio models (`model.wcs.wcs.ctype[0] == "T_e"`) are plotted/labeled differently (electron temperature/density axes) than standard PDR density/FUV-flux models — this drives `ms.keyname = "hii"` and different axis labels.
+- Each ModelSet directory also gets a `<keyname>_models.tgz` tarball referenced from its page; this is not created by the script and must exist separately in `models/`.   It is not tracked by github since it is a large binary file.  Marc ensures these files are up to date. 
+
+Jinja2 templates live in `autogen/`: `index_page_jinja_template.html` (per-ModelSet index), `model_page_jinja_template.html` (per-model-quantity page), `all_models_page_jinja_template.html` (top-level `models/index.html` listing all ModelSets). These duplicate the nav/head markup rather than sharing it with the hand-written pages — when nav changes, update it in the templates *and* in every top-level `.html` file and `includes/nav.inc`.
+
+## Page structure conventions
+
+- `includes/head.inc`, `includes/nav.inc`, `includes/footer.inc` are SSI includes intended to be shared across hand-written pages, but note the top-level pages (`index.html`, `models.html`, `tools.html`, `docs.html`) inline their own copy of the nav markup rather than including `nav.inc` — check which pattern a given page already uses before editing.
+- CSS: `css/pdrt.css` is site-specific; `freelancer/css/*` is the vendored Start Bootstrap "Freelancer" theme (https://startbootstrap.com/themes/freelancer/) — don't hand-edit vendored files, override in `pdrt.css` instead.
+- Root-relative paths (`/css/...`, `/includes/...`, `/models/...`) are used throughout — pages assume they are served from the site root, not from a subdirectory.

autogen/all_models_page_jinja_template.html

@@ -35,7 +35,8 @@
                 <h1 class="pre-wrap mymasthead-heading font-weight-bold mb-0">
 Links to All Available ModelSets</h1>
                 <p class="myp">
-Click on the description to explore a ModelSet (see plots, download FITS files).
+To see plots and download individual FITS files for a given ModelSet, click on its ModelSet Description.
+
 <table  class="table mytable table-striped table-striped table-bordered" bgcolor="white" >
 <tr><th>ModelSet name in <tt><a href="https://pdrtpy.readthedocs.io">pdrtpy</a></tt></th><th>ModelSet Description</th></tr>
 {% for key,value in all_models.items() %}

autogen/index_page_jinja_template.html

@@ -36,7 +36,7 @@ <h1 class="pre-wrap mymasthead-heading font-weight-bold mb-0">
 {{modelset.header}}</h1>
                 <p class="myp">
 Below are links to the individual FITS files
-and plots in the <i>{{modelset.name}}</i> <a class="mya"
+and plots in the <i>{{modelset.name}} z={{modelset.z}} losangle={{modelset.losangle}}</i>  <a class="mya"
 href="https://pdrtpy.readthedocs.io/en/latest/pdrtpy.modelset.html">ModelSet</a>. We also provide a <a class="mya" href="{{modelset.tarball}}">gzipped tarball</a> of all the <i>{{modelset.keyname}}</i> FITS and PNG files.
 
 <table  class="table mytable table-striped table-striped table-bordered" bgcolor="white" >