Merge pull request pyne#8 from paulromano/add-jeff33

Add script to convert JEFF 3.3 data from OECD/NEA

convert_jeff32.py

@@ -3,11 +3,11 @@
 import argparse
 import glob
 import os
-import sys
 import tarfile
 import zipfile
 from collections import defaultdict
 from string import digits
+from urllib.parse import urljoin
 
 import openmc.data
 from openmc._utils import download
@@ -23,8 +23,6 @@
 processing the data may require as much as 40 GB of additional free disk
 space. Note that if you don't need all 11 temperatures, you can modify the
 'files' list in the script to download only the data you want.
-
-Are you sure you want to continue? ([y]/n)
 """
 
 class CustomFormatter(argparse.ArgumentDefaultsHelpFormatter,
@@ -35,19 +33,24 @@ class CustomFormatter(argparse.ArgumentDefaultsHelpFormatter,
     description=description,
     formatter_class=CustomFormatter
 )
-parser.add_argument('-b', '--batch', action='store_true',
-                    help='supresses standard in')
-parser.add_argument('-d', '--destination', default='jeff-3.2-hdf5',
+parser.add_argument('-d', '--destination', default='jeff32_hdf5',
                     help='Directory to create new library in')
+parser.add_argument('--download', action='store_true',
+                    help='Download files from OECD-NEA')
+parser.add_argument('--no-download', dest='download', action='store_false',
+                    help='Do not download files from OECD-NEA')
+parser.add_argument('--extract', action='store_true',
+                    help='Extract tar/zip files')
+parser.add_argument('--no-extract', dest='extract', action='store_false',
+                    help='Do not extract tar/zip files')
 parser.add_argument('--libver', choices=['earliest', 'latest'],
                     default='latest', help="Output HDF5 versioning. Use "
                     "'earliest' for backwards compatibility or 'latest' for "
                     "performance")
+parser.set_defaults(download=True, extract=True)
 args = parser.parse_args()
 
-response = input(download_warning) if not args.batch else 'y'
-if response.lower().startswith('n'):
-    sys.exit()
+print(download_warning)
 
 base_url = 'https://www.oecd-nea.org/dbforms/data/eva/evatapes/jeff_32/Processed/'
 files = ['JEFF32-ACE-293K.tar.gz',
@@ -66,37 +69,32 @@ class CustomFormatter(argparse.ArgumentDefaultsHelpFormatter,
 # ==============================================================================
 # DOWNLOAD FILES FROM OECD SITE
 
-files_complete = []
-for f in files:
-    # Establish connection to URL
-    url = base_url + f
-    downloaded_file = download(url)
-    files_complete.append(f)
+if args.download:
+    for f in files:
+        download(urljoin(base_url, f))
 
 # ==============================================================================
 # EXTRACT FILES FROM TGZ
 
-for f in files:
-    if f not in files_complete:
-        continue
-
-    # Extract files
-    if f.endswith('.zip'):
-        with zipfile.ZipFile(f, 'r') as zipf:
-            print('Extracting {}...'.format(f))
-            zipf.extractall('jeff-3.2')
-
-    else:
-        suffix = 'ACEs_293K' if '293' in f else ''
-        with tarfile.open(f, 'r') as tgz:
-            print('Extracting {}...'.format(f))
-            tgz.extractall(os.path.join('jeff-3.2', suffix))
-
-        # Remove thermal scattering tables from 293K data since they are
-        # redundant
-        if '293' in f:
-            for path in glob.glob(os.path.join('jeff-3.2', 'ACEs_293K', '*-293.ACE')):
-                os.remove(path)
+if args.extract:
+    for f in files:
+        # Extract files
+        if f.endswith('.zip'):
+            with zipfile.ZipFile(f, 'r') as zipf:
+                print('Extracting {}...'.format(f))
+                zipf.extractall('jeff-3.2')
+
+        else:
+            suffix = 'ACEs_293K' if '293' in f else ''
+            with tarfile.open(f, 'r') as tgz:
+                print('Extracting {}...'.format(f))
+                tgz.extractall(os.path.join('jeff-3.2', suffix))
+
+            # Remove thermal scattering tables from 293K data since they are
+            # redundant
+            if '293' in f:
+                for path in glob.glob(os.path.join('jeff-3.2', 'ACEs_293K', '*-293.ACE')):
+                    os.remove(path)
 
 # ==============================================================================
 # CHANGE ZAID FOR METASTABLES

convert_jeff33.py

@@ -0,0 +1,83 @@
+#!/usr/bin/env python
+
+import argparse
+from pathlib import Path
+import tarfile
+import sys
+
+import openmc.data
+from openmc._utils import download
+
+
+# Make sure Python version is sufficient
+assert sys.version_info >= (3, 6), "Python 3.6+ is required"
+
+description = """
+Convert JEFF 3.3 ACE data distributed by OECD/NEA into an HDF5 library
+that can be used by OpenMC. It will download an 1.3 GB archive containing
+all the ACE files, extract it, convert them, and write HDF5 files into a
+destination directory.
+
+"""
+
+class CustomFormatter(argparse.ArgumentDefaultsHelpFormatter,
+                      argparse.RawDescriptionHelpFormatter):
+    pass
+
+
+parser = argparse.ArgumentParser(
+    description=description,
+    formatter_class=CustomFormatter
+)
+parser.add_argument('-d', '--destination', type=Path, default=Path('jeff33_hdf5'),
+                    help='Directory to create new library in')
+parser.add_argument('--download', action='store_true',
+                    help='Download tarball from OECD-NEA')
+parser.add_argument('--no-download', dest='download', action='store_false',
+                    help='Do not download tarball from OECD-NEA')
+parser.add_argument('--extract', action='store_true',
+                    help='Extract zip files')
+parser.add_argument('--no-extract', dest='extract', action='store_false',
+                    help='Do not extract .tgz file if it has already been extracted')
+parser.add_argument('--libver', choices=['earliest', 'latest'],
+                    default='earliest', help="Output HDF5 versioning. Use "
+                    "'earliest' for backwards compatibility or 'latest' for "
+                    "performance")
+parser.set_defaults(download=True, extract=True)
+args = parser.parse_args()
+
+# Download JEFF 3.3 library
+filename = 'JEFF33-n_tsl-ace.tgz'
+url = f'http://www.oecd-nea.org/dbdata/jeff/jeff33/downloads/{filename}'
+if args.download:
+    download(url)
+
+# Extract tar file
+if args.extract:
+    with tarfile.open(filename, 'r') as tgz:
+        print(f'Extracting {filename}...')
+        tgz.extractall()
+
+# Create output directory if it doesn't exist
+args.destination.mkdir(parents=True, exist_ok=True)
+
+# Get a list of all ACE files
+paths = sorted(Path('ace').glob('*.[Aa][Cc][Ee]'))
+
+lib = openmc.data.DataLibrary()
+for p in sorted(paths):
+    print(f'Converting: {p}')
+    if 'jeff33' in str(p):
+        data = openmc.data.IncidentNeutron.from_ace(p)
+        if 'm.' in str(p):
+            # Correct metastable
+            data.metastable = 1
+            data.name += '_m1'
+    else:
+        data = openmc.data.ThermalScattering.from_ace(p)
+
+    h5_file = args.destination / f'{data.name}.h5'
+    data.export_to_hdf5(h5_file, 'w', libver=args.libver)
+    lib.register_file(h5_file)
+
+lib.export_to_xml(args.destination / 'cross_sections.xml')