[GSoC] Add programmatic parser for Seaton opacity tables

carsus/io/literature/__init__.py

@@ -0,0 +1,9 @@
+"""
+Literature data readers for the TARDIS/Carsus codebase.
+"""
+
+from carsus.io.literature.seaton import get_seaton_opacity_df
+
+__all__ = [
+    'get_seaton_opacity_df',
+]

carsus/io/literature/seaton.py

@@ -0,0 +1,70 @@
+"""
+Reader for Seaton opacity data (Seaton 1992).
+
+This module provides functionality to read continuum opacity data from 
+the Seaton archive (VI/80 at CDS).
+"""
+
+import pandas as pd
+import requests
+import io
+import gzip
+
+
+def get_seaton_opacity_df(file_name):
+    """
+    Load and parse a Seaton opacity table from the CDS archive.
+
+    Parameters
+    ----------
+    file_name : str
+        The name of the gzip-compressed file to load from the Seaton 
+        archive (e.g., 's92.201.gz').
+
+    Returns
+    -------
+    pandas.DataFrame
+        A DataFrame with columns:
+        - logT : float
+            Log10 of temperature in Kelvin
+        - logNe : float
+            Log10 of electron density
+        - kappa_planck : float
+            Planck-weighted mean opacity
+        - kappa_rosseland : float
+            Rosseland-weighted mean opacity
+    """
+    url = f"https://cdsarc.cds.unistra.fr/ftp/VI/80/{file_name}"
+    response = requests.get(url)
+
+    with gzip.GzipFile(fileobj=io.BytesIO(response.content)) as f:
+        lines = [line.decode('utf-8').strip() for line in f.readlines() if line.strip()]
+
+    data_rows = []
+    current_log_t = None
+
+    # Standard Seaton format: logT = Index / 40
+    for line in lines[1:]:  # Skip metadata line
+        parts = line.split()
+        if not parts:
+            continue
+
+        # Detect Block Header: (e.g., 140 14 68 2)
+        # These define the constant log(T) for the following rows
+        if len(parts) == 4 and float(parts[0]) >= 140 and '.' not in parts[0]:
+            current_log_t = float(parts[0]) / 40.0
+            continue
+
+        # Parse Data Rows: [Index] [logNe] [Planck_Opacity] [Rosseland_Opacity]
+        if current_log_t is not None and len(parts) >= 4:
+            try:
+                data_rows.append({
+                    'logT': current_log_t,
+                    'logNe': float(parts[1]),
+                    'kappa_planck': float(parts[2]),
+                    'kappa_rosseland': float(parts[3])
+                })
+            except ValueError:
+                continue
+
+    return pd.DataFrame(data_rows)

carsus/io/literature/tests/test_seaton.py

@@ -0,0 +1,42 @@
+"""
+Tests for Seaton opacity reader.
+"""
+
+import gzip
+import io
+from unittest import mock
+
+import pandas as pd
+import pytest
+
+from carsus.io.literature import get_seaton_opacity_df
+
+
+def create_sample_data():
+    """Create gzip-compressed sample Seaton data."""
+    sample_lines = [
+        "OP Version 2.0 S92 201",
+        "140        14   68    2",
+        "14    -4.0 1.23e-1 1.15e-1",
+        "16    -3.5 1.45e-1 1.32e-1",
+    ]
+    buffer = io.BytesIO()
+    with gzip.GzipFile(fileobj=buffer, mode='wb') as f:
+        f.write("\n".join(sample_lines).encode('utf-8'))
+    return buffer.getvalue()
+
+
+@mock.patch('carsus.io.literature.seaton.requests.get')
+def test_get_seaton_opacity_df(mock_get):
+    """Test reading Seaton opacity data."""
+    mock_get.return_value.content = create_sample_data()
+
+    df = get_seaton_opacity_df("s92.201.gz")
+
+    assert isinstance(df, pd.DataFrame)
+    assert len(df) == 2
+    assert list(df.columns) == ['logT', 'logNe', 'kappa_planck', 'kappa_rosseland']
+    assert (df['kappa_planck'] > 0).all()
+    assert (df['kappa_rosseland'] > 0).all()
+
+