From 3665487733f991777e4b8e6d0cfb13358a9285c8 Mon Sep 17 00:00:00 2001
From: Nicolas Kruse <nicolas@nonan.net>
Date: Mon, 2 Jun 2025 16:25:17 +0200
Subject: [PATCH] helper scripts for compile db added

---
 thermo_data/combine_data.py   | 80 +++++++++++++++++++++++++++++++++++
 thermo_data/compile_to_bin.py | 67 +++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 thermo_data/combine_data.py
 create mode 100644 thermo_data/compile_to_bin.py

diff --git a/thermo_data/combine_data.py b/thermo_data/combine_data.py
new file mode 100644
index 0000000..b713dd3
--- /dev/null
+++ b/thermo_data/combine_data.py
@@ -0,0 +1,80 @@
+import yaml
+import xml.etree.ElementTree as ET
+import glob
+import sys
+
+
+def main():
+    output_file = sys.argv[1]
+    input_files = sys.argv[2:]
+
+    inp_therm_prop: list[dict] = []
+
+    for glop_filter in input_files:
+        for file_name in glob.glob(glop_filter):
+            print(f'Processing file: {file_name}...')
+            if file_name.endswith('.xml'):
+                inp_therm_prop += get_xml_data(file_name)
+            else:
+                with open(file_name, 'r') as f:
+                    inp_therm_prop += yaml.safe_load(f)['species']
+
+    therm_prop = []
+    added_species_names: set[str] = set()
+
+    for species in inp_therm_prop:
+        name = species['name']
+        assert isinstance(name, str)
+        for element in species['composition']:
+            name = name.replace(element.upper(), element)
+        if name not in added_species_names and 'E' not in species['composition']:
+            species['name'] = name
+            therm_prop.append(species)
+            added_species_names.add(name)
+
+    with open(output_file, 'w') as f:
+        f.write('species:\n')
+        for species in sorted(therm_prop, key=lambda s: s['name']):
+            f.write('\n- ' + yaml.dump({'name': species['name']}, default_flow_style=False))
+            f.write(f'  composition: {yaml.dump(species["composition"],default_flow_style=True)}')
+            f.write('  thermo:\n')
+            f.write(f'    model: {species["thermo"]["model"]}\n')
+            f.write(f'    temperature-ranges: {yaml.dump(species["thermo"]["temperature-ranges"],default_flow_style=True)}')
+            f.write('    data:\n')
+            for d in species['thermo']['data']:
+                f.write(f'    - {d}\n')
+            f.write('    ' + yaml.dump({'note': species['thermo']['note']}, default_flow_style=False))
+
+    print('Added: ', ', '.join(sorted(added_species_names)))
+
+
+def get_xml_data(file_name: str) -> list[dict]:
+    xml_data_list: list[dict] = []
+    tree = ET.parse(file_name)
+    root = tree.getroot()
+    for i, child in enumerate(root):
+
+        elements = {el[0]: int(el[1]) for c in child.find('elements').findall('element') for el in c.items()}
+
+        values_temp = [tr for tr in child.findall('T_range')]
+        t_ranges = [float(v.attrib['Tlow']) for v in values_temp] + [float(values_temp[-1].attrib['Thigh'])]
+        data = [[float(v.text) for v in tr] for tr in values_temp]
+
+        is_gas = child.find('condensed').text == 'False'
+
+        if is_gas:
+            xml_data_list.append({
+                'name': child.attrib['inp_file_name'],
+                'composition': elements,
+                'thermo': {
+                    'model': 'NASA9',
+                    'temperature-ranges': t_ranges,
+                    'data': data,
+                    'note': child.find('comment').text
+                }
+            })
+    return xml_data_list
+
+
+if __name__ == "__main__":
+    main()
diff --git a/thermo_data/compile_to_bin.py b/thermo_data/compile_to_bin.py
new file mode 100644
index 0000000..a50ff62
--- /dev/null
+++ b/thermo_data/compile_to_bin.py
@@ -0,0 +1,67 @@
+import yaml
+import struct
+import sys
+
+
+def main():
+    input_file = sys.argv[1]
+    output_file = sys.argv[2]
+
+    assert not output_file.endswith('.yml') and not output_file.endswith('.yaml'), 'Binary output file should not have yaml-extension'
+
+    with open(input_file) as f:
+        data = yaml.safe_load(f)
+
+    data = list({v['name']: v for v in data['species']}.values())
+
+    species_names = ' '.join(s['name'] for s in data)
+
+    header_len = 8
+
+    offset = 8 + len(species_names) + len(data) * header_len
+
+    header_list = []
+    body_list = []
+    added_list = set()
+
+    for dat in data:
+        if dat['name'] not in added_list:
+
+            composition_count = len(dat['composition'])
+            model = {'NASA9': 9}[dat['thermo']['model']]
+            temperatures = dat['thermo']['temperature-ranges']
+            ref_string = dat['thermo']['note'].encode('utf-8')
+            header = struct.pack('<I4B', offset, composition_count, model, len(temperatures), len(ref_string))
+
+            assert len(header) == header_len
+
+            header_list.append(header)
+
+            composition = [el for k, v in dat['composition'].items() for el in [k.ljust(2).encode('ASCII'), v]]
+            data_vals = [d for darr in dat['thermo']['data'] for d in darr[:model]]
+
+            assert len(dat['thermo']['data']) == len(temperatures) - 1, f"Temperature data length mismatch for {dat['name']}."
+
+            if any(len(d) != model for d in dat['thermo']['data']):
+                print(f"Warning: Data length mismatch for {dat['name']}. Expected {model} coefficients, got {len(dat['thermo']['data'][0])}.")
+
+            format_string = '<' + '2sB' * composition_count + f'{len(temperatures)}f{len(data_vals)}f{len(ref_string)}s'
+
+            body = struct.pack(format_string, *(composition + temperatures + data_vals + [ref_string]))
+            body_list.append(body)
+
+            offset += len(body)
+            added_list.add(dat['name'])
+
+    with open(output_file, 'wb') as f:
+        f.write(b'gapy')
+        f.write(struct.pack('<I', len(species_names)))
+        f.write(species_names.encode('ASCII'))
+        for dat in header_list:
+            f.write(dat)
+        for dat in body_list:
+            f.write(dat)
+
+
+if __name__ == '__main__':
+    main()