Factor out the HITRAN reader so it can be used on pre-downloaded files

astroquery/hitran/__init__.py

@@ -24,6 +24,6 @@ class Conf(_config.ConfigNamespace):
 
 conf = Conf()
 
-from .core import Hitran, HitranClass
+from .core import Hitran, HitranClass, parse_hitran_text
 
-__all__ = ['Hitran', 'HitranClass', 'conf']
+__all__ = ['Hitran', 'HitranClass', 'parse_hitran_text', 'conf']

astroquery/hitran/core.py

@@ -6,7 +6,7 @@
 from . import conf
 from .utils import parse_readme
 
-__all__ = ['Hitran', 'HitranClass']
+__all__ = ['Hitran', 'HitranClass', 'parse_hitran_text']
 
 
 @async_to_sync
@@ -227,26 +227,71 @@ def _parse_result(self, response, *, verbose=False):
         """
         Parse a response into an `~astropy.table.Table`
         """
-        formats = parse_readme(self.FORMATFILE)
-
-        dtypes = [entry['dtype'] for entry in formats.values()]
-
-        rows = []
-        for line in response.text.split('\n'):
-            if line.strip():
-                row = []
-                start = 0
-                for key, entry in formats.items():
-                    formatter = entry['formatter']
-                    length = entry['length']
-                    value = formatter(line[start:start+length])
-                    row.append(value)
-                    start = start + length
-                rows.append(row)
-
-        result = Table(rows=rows, names=formats.keys(), dtype=dtypes)
-
-        return result
+        return parse_hitran_text(response.text, formatfile=self.FORMATFILE)
+
+    @staticmethod
+    def read(filename, *, formatfile=None):
+        """
+        Read a HITRAN ``.par`` file from disk into an `~astropy.table.Table`.
+
+        Parameters
+        ----------
+        filename : str
+            Path to a HITRAN ``.par`` (160-column fixed-width) file.
+        formatfile : str, optional
+            Path to a HITRAN format ``readme.txt`` file describing the
+            column layout. Defaults to the bundled HITRAN 2004 format.
+
+        Returns
+        -------
+        result : `~astropy.table.Table`
+            Parsed line list.
+        """
+        if formatfile is None:
+            formatfile = HitranClass.FORMATFILE
+        with open(filename, 'r') as fh:
+            text = fh.read()
+        return parse_hitran_text(text, formatfile=formatfile)
+
+
+def parse_hitran_text(text, *, formatfile=HitranClass.FORMATFILE):
+    """
+    Parse the text of a HITRAN ``.par`` line list into an `~astropy.table.Table`.
+
+    Parameters
+    ----------
+    text : str
+        Contents of a HITRAN ``.par`` file (160-column fixed-width records,
+        one per line).
+    formatfile : str, optional
+        Path to a HITRAN format ``readme.txt`` file describing the column
+        layout. Defaults to the bundled HITRAN 2004 format.
+
+    Returns
+    -------
+    result : `~astropy.table.Table`
+        Parsed line list.
+    """
+    formats = parse_readme(formatfile)
+
+    dtypes = [entry['dtype'] for entry in formats.values()]
+
+    rows = []
+    for line in text.split('\n'):
+        if line.strip():
+            row = []
+            start = 0
+            for key, entry in formats.items():
+                formatter = entry['formatter']
+                length = entry['length']
+                value = formatter(line[start:start+length])
+                row.append(value)
+                start = start + length
+            rows.append(row)
+
+    result = Table(rows=rows, names=formats.keys(), dtype=dtypes)
+
+    return result
 
 
 Hitran = HitranClass()

astroquery/hitran/tests/test_hitran.py

@@ -5,7 +5,7 @@
 from astropy import units as u
 from astropy.table import Table
 
-from ...hitran import Hitran
+from ...hitran import Hitran, parse_hitran_text
 
 HITRAN_DATA = 'H2O.data'
 
@@ -21,8 +21,8 @@ def __init__(self):
 
     @property
     def text(self):
-        with open(self.filename) as f:
-            return f.read()
+        with open(self.filename) as fh:
+            return fh.read()
 
 
 def test_query_async():
@@ -36,19 +36,54 @@ def test_query_async():
     np.testing.assert_almost_equal(response['numax'], 10.)
 
 
+EXPECTED_KEYS = {'molec_id', 'local_iso_id', 'nu', 'sw', 'a',
+                 'gamma_air', 'gamma_self', 'elower',
+                 'n_air', 'delta_air', 'global_upper_quanta',
+                 'global_lower_quanta', 'local_upper_quanta',
+                 'local_lower_quanta', 'ierr1', 'ierr2',
+                 'ierr3', 'ierr4', 'ierr5', 'ierr6', 'iref1',
+                 'iref2', 'iref3', 'iref4', 'iref5', 'iref6',
+                 'line_mixing_flag', 'gp', 'gpp'}
+
+
 def test_query():
     hitran = Hitran()
     response = MockResponseHitran()
     tbl = hitran._parse_result(response)
     assert isinstance(tbl, Table)
     assert len(tbl) == 122
-    assert set(tbl.keys()) == set(['molec_id', 'local_iso_id', 'nu', 'sw', 'a',
-                                   'gamma_air', 'gamma_self', 'elower',
-                                   'n_air', 'delta_air', 'global_upper_quanta',
-                                   'global_lower_quanta', 'local_upper_quanta',
-                                   'local_lower_quanta', 'ierr1', 'ierr2',
-                                   'ierr3', 'ierr4', 'ierr5', 'ierr6', 'iref1',
-                                   'iref2', 'iref3', 'iref4', 'iref5', 'iref6',
-                                   'line_mixing_flag', 'gp', 'gpp'])
+    assert set(tbl.keys()) == EXPECTED_KEYS
+    assert tbl['molec_id'][0] == 1
+    np.testing.assert_almost_equal(tbl['nu'][0], 0.072059)
+
+
+def test_parse_hitran_text():
+    with open(data_path(HITRAN_DATA)) as fh:
+        text = fh.read()
+    tbl = parse_hitran_text(text)
+    assert isinstance(tbl, Table)
+    assert len(tbl) == 122
+    assert set(tbl.keys()) == EXPECTED_KEYS
+    assert tbl['molec_id'][0] == 1
+    np.testing.assert_almost_equal(tbl['nu'][0], 0.072059)
+
+
+def test_parse_hitran_text_matches_parse_result():
+    """parse_hitran_text and _parse_result must produce identical tables."""
+    response = MockResponseHitran()
+    tbl_from_response = Hitran._parse_result(response)
+    tbl_from_text = parse_hitran_text(response.text)
+    assert len(tbl_from_response) == len(tbl_from_text)
+    assert tbl_from_response.keys() == tbl_from_text.keys()
+    for key in tbl_from_response.keys():
+        np.testing.assert_array_equal(tbl_from_response[key],
+                                      tbl_from_text[key])
+
+
+def test_read_par_file():
+    tbl = Hitran.read(data_path(HITRAN_DATA))
+    assert isinstance(tbl, Table)
+    assert len(tbl) == 122
+    assert set(tbl.keys()) == EXPECTED_KEYS
     assert tbl['molec_id'][0] == 1
     np.testing.assert_almost_equal(tbl['nu'][0], 0.072059)