Add alpha PHOENIX support, workflow recipes, instrument readers, and example notebook

.gitignore

@@ -1,3 +1,17 @@
-.idea/inspectionProfiles/
-.idea/modules.xml
-Spyctres/__pycache__/
+# Python bytecode
+__pycache__/
+*.py[cod]
+
+# Packaging / build artifacts
+*.egg-info/
+build/
+dist/
+
+# IDE files
+.idea/
+
+# Notebooks
+.ipynb_checkpoints/
+
+# Local development scripts
+scripts/phoenix_backend_compare.py

README.md

@@ -1,42 +1,117 @@
 # Spyctres
 
-Spyctres is a tool for analysing stellar spectra.  It can compare a measured spectrum with publicly-available 
-libraries of spectral templates in order to find the closest match.  
+Spyctres is a Python package for stellar spectral fitting and spectral typing from reduced spectra. It can compare a measured spectrum with publicly available spectral-template libraries to find the closest match.
 
 Developer: Etienne Bachelet
 
-Please note that Spyctres is still under development.  
+Spyctres is still under active development. It includes core fitting utilities, instrument I/O helpers, plotting tools, and example workflows. Recent additions include PHOENIX template-based fitting and a clearer separation between generic fitting code, workflow recipes, examples, and smoke tests.
+
+## Features
+
+- spectral fitting utilities in `Spyctres/Spyctres.py`
+- generic spectrum containers and reader dispatch in `Spyctres/io.py`
+- PHOENIX template support in `Spyctres/phoenix.py`
+- PHOENIX forward modelling in `Spyctres/phoenix_forward.py`
+- fitting helpers in `Spyctres/fitting.py`
+- workflow recipes in `Spyctres/recipes.py`
+- plotting helpers in `Spyctres/plotting.py`
 
 ## Installation
-For the time being, the best way to install Spyctres is to clone its [Github repository](https://github.com/ebachelet/Spyctres), 
-and run the setup.py.  It is recommended that you create and activate a virtual environment before proceeding with this
-installation. 
-
-```commandline
-venv> git clone https://github.com/ebachelet/Spyctres.git
-venv> cd Spyctres/
-venv> python setpy.py install
+
+Spyctres is currently intended for local editable installs during development. Creating and activating a virtual environment first is recommended.
+
+Spyctres requires Python 3.12 or later.
+
+```bash
+git clone https://github.com/ebachelet/Spyctres.git
+cd Spyctres
+pip install -e .
+```
+
+Some legacy workflows use `pysynphot` and its successor package `stsynphot`:
+
+```bash
+pip install pysynphot stsynphot
 ```
 
-Spyctres makes use of ```pysynphot``` and its successor package ```stsynphot``` as dependencies.  
-These can both be installed via pip:
+Those workflows also require the stellar template libraries linked from the [pysynphot installation documentation](https://pysynphot.readthedocs.io/en/latest/index.html#pysynphot-installation-setup). After downloading and unpacking them, set `PYSYN_CDBS` to their local root directory:
+
+```bash
+export PYSYN_CDBS=/path/to/cdbs
+```
+
+PHOENIX workflows require additional scientific Python dependencies and a local PHOENIX template directory.
+
+The PHOENIX templates may be downloaded from the Goettingen Spectral Library:
+
+- PHOENIX archive: `https://phoenix.astro.physik.uni-goettingen.de/`
+- PHOENIX v2 HiResFITS directory: `https://phoenix.astro.physik.uni-goettingen.de/data/v2.0/HiResFITS/PHOENIX-ACES-AGSS-COND-2011/`
+- PHOENIX v2 wavelength file: `https://phoenix.astro.physik.uni-goettingen.de/data/v2.0/HiResFITS/WAVE_PHOENIX-ACES-AGSS-COND-2011.fits`
+
+The wavelength file must be placed in the root directory of the PHOENIX v2 models.
+
+## PHOENIX template path
+
+The local PHOENIX path is resolved in this order:
 
-```commandline
-venv> pip install pysynphot
-venv> pip install stsynphot
+1. explicit command-line value
+2. environment variable `SPYCTRES_PHOENIX_DIR`
+3. config file `~/.config/spyctres/config.toml`
+
+Example config:
+
+```toml
+[paths]
+phoenix_dir = "/path/to/PHOENIXv2"
 ```
 
-Lastly, you will need to download the stellar spectrum library used as templates in Spyctres fitting 
-process.  Tarballs of these libraries are linked from the [```pysynphot``` ReadTheDocs page](https://pysynphot.readthedocs.io/en/latest/index.html#pysynphot-installation-setup)
- in the second table.
+## Quick start
+
+Useful entry points in the repository include:
+
+- `quick_example.py` for the legacy fitting workflow
+- `examples/full_spectrum_classification.ipynb` for PHOENIX classification
+- smoke tests under `scripts/`
 
-Download and unpack these files into a local directory.  In order to tell Spyctres where to find this library, 
-the path to it should be declared as an environment variable in any script or shell where the software is 
-run, e.g. 
+To open the PHOENIX example notebook:
 
-```python
-os.environ['PYSYN_CDBS'] =  '/my/path/cdbs/'
+```bash
+jupyter lab examples/full_spectrum_classification.ipynb
 ```
 
-## Running Spyctres
-The ```quick_example.py``` script in this repository provides a worked demonstration of the fitting process. 
+## Supported readers
+
+Current reader coverage includes:
+
+- X-SHOOTER 1D products
+- PEPSI `.dxt.nor`
+- FLOYDS ASCII/CSV exports
+- Gemini/GMOS ASCII exports
+
+Readers return a generic `SpectrumSegment` object so that fitting code can remain instrument-agnostic.
+
+## Project structure
+
+Spyctres is organized around four layers:
+
+- generic fitting core
+- workflow recipes
+- user-facing examples
+- developer smoke tests
+
+Notable files:
+
+- `Spyctres/recipes.py`
+- `examples/full_spectrum_classification.ipynb`
+- `scripts/xshooter_fit_smoketest.py`
+
+## Current limitations
+
+PHOENIX support should still be treated as alpha.
+
+In particular:
+
+- the example notebook is a first-pass classification workflow, not a final precision analysis
+- some workflows still require user judgment for wavelength windows, masking, resolving power, and continuum treatment
+- instrument-specific metadata quality varies across input formats
+- packaging and documentation are still minimal

Spyctres.egg-info/PKG-INFO

@@ -13,7 +13,8 @@ Classifier: Intended Audience :: Developers
 Classifier: Topic :: Software Development :: Build Tools
 Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
 Classifier: Programming Language :: Python :: 3
-Requires-Python: >=3.6,<4
+Classifier: Programming Language :: Python :: 3.12
+Requires-Python: >=3.12,<4
 License-File: LICENSE
 Requires-Dist: scipy
 Requires-Dist: numpy