Merge pull request #616 from keflavich/issue616

Astroquery.splatalogue query not respecting "exclude" keyword(?)

Author: keflavich

astroquery/splatalogue/__init__.py

@@ -7,7 +7,7 @@
 
 :Originally contributed by:
 
-     Magnus Vilehlm Persson ([email protected])
+     Magnus Vilhelm Persson ([email protected])
 """
 from astropy import config as _config
 

astroquery/splatalogue/core.py

@@ -8,6 +8,7 @@
 import warnings
 from astropy.io import ascii
 from astropy import units as u
+from astropy import log
 from ..query import BaseQuery
 from ..utils import async_to_sync
 from ..utils.docstr_chompers import prepend_docstr_noreturns
@@ -129,7 +130,8 @@ def _parse_kwargs(self, min_frequency=None, max_frequency=None,
                       show_upper_degeneracy=None, show_molecule_tag=None,
                       show_qn_code=None, show_lovas_labref=None,
                       show_lovas_obsref=None, show_orderedfreq_only=None,
-                      show_nrao_recommended=None):
+                      show_nrao_recommended=None,
+                      parse_chemistry_locally=False):
         """
         The Splatalogue service returns lines with rest frequencies in the
         range [min_frequency, max_frequency].
@@ -164,7 +166,9 @@ def _parse_kwargs(self, min_frequency=None, max_frequency=None,
 
             ``' H2CO '`` - Just 1 species, H2CO. The spaces prevent including
                            others.
-
+        parse_chemistry_locally : bool
+            Attempt to determine the species ID #'s locally before sending the
+            query?  This will prevent queries that have no matching species.
         chem_re_flags : int
             See the `re` module
         energy_min : `None` or float
@@ -185,7 +189,9 @@ def _parse_kwargs(self, min_frequency=None, max_frequency=None,
         exclude : list
             Types of lines to exclude.  Default is:
             (``'potential'``, ``'atmospheric'``, ``'probable'``)
-            Can also exclude ``'known'``
+            Can also exclude ``'known'``.
+            To exclude nothing, use 'none', not the python object None, since
+            the latter is meant to indicate 'leave as default'
         only_NRAO_recommended : bool
             Show only NRAO recommended species?
         line_lists : list
@@ -264,10 +270,13 @@ def _parse_kwargs(self, min_frequency=None, max_frequency=None,
             # include all
             payload['sid[]'] = []
         elif chemical_name is not None:
-            species_ids = self.get_species_ids(chemical_name, chem_re_flags)
-            if len(species_ids) == 0:
-                raise ValueError("No matching chemical species found.")
-            payload['sid[]'] = list(species_ids.values())
+            if parse_chemistry_locally:
+                species_ids = self.get_species_ids(chemical_name, chem_re_flags)
+                if len(species_ids) == 0:
+                    raise ValueError("No matching chemical species found.")
+                payload['sid[]'] = list(species_ids.values())
+            else:
+                payload['chemical_name'] = chemical_name
 
         if energy_min is not None:
             payload['energy_range_from'] = float(energy_min)
@@ -291,7 +300,12 @@ def _parse_kwargs(self, min_frequency=None, max_frequency=None,
             raise ValueError("Invalid version specified.  Allowed versions "
                              "are {vers}".format(vers=str(self.versions)))
 
-        if exclude is not None:
+        if exclude == 'none':
+            for e in ('potential', 'atmospheric', 'probable', 'known'):
+                # Setting a keyword value to 'None' removes it (see query_lines_async)
+                log.debug("Setting no_{0} to None".format(e))
+                payload['no_' + e] = None
+        elif exclude is not None:
             for e in exclude:
                 payload['no_' + e] = 'no_' + e
 
@@ -383,6 +397,9 @@ def query_lines_async(self, min_frequency=None, max_frequency=None,
                                                    max_frequency=max_frequency,
                                                    **kwargs))
 
+        # Add an extra step: sometimes, need to REMOVE keywords
+        data_payload = {k:v for k,v in data_payload.items() if v is not None}
+
         if get_query_payload:
             return data_payload
 

astroquery/splatalogue/tests/test_splatalogue.py

@@ -3,6 +3,7 @@
 from ...utils.testing_tools import MockResponse
 from astropy import units as u
 from astropy.tests.helper import pytest, remote_data
+from astropy.extern.six.moves import urllib_parse
 import requests
 import os
 
@@ -106,13 +107,53 @@ def test_band_crashorno():
     assert exc.value.args[0] == "Invalid frequency band."
 
 
-# regression test : version selection should work
-# Unfortunately, it looks like version1 = version2 on the web page now, so this
-# may no longer be a valid test
+#Upstream changed: there is no distinction between versions for this molecule
+# # regression test : version selection should work
+# # Unfortunately, it looks like version1 = version2 on the web page now, so this
+# # may no longer be a valid test
+# @remote_data
+# def test_version_selection():
+#     results = splatalogue.Splatalogue.query_lines(
+# 			    min_frequency= 703*u.GHz,
+# 			    max_frequency=706*u.GHz,
+# 			    chemical_name='Acetaldehyde',
+# 			    version='v1.0'
+# 			    )
+#     assert len(results)==1
+
+def test_exclude(patch_post):
+    # regression test for issue 616
+    d = splatalogue.Splatalogue.query_lines_async(114 * u.GHz, 116 * u.GHz,
+                                                  chemical_name=' CO ',
+                                                  exclude=None,
+                                                  get_query_payload=True)
+
+    exclusions = {'no_atmospheric': 'no_atmospheric',
+                  'no_potential': 'no_potential',
+                  'no_probable': 'no_probable',}
+
+    for k,v in exclusions.items():
+        assert d[k] == v
+
+    d = splatalogue.Splatalogue.query_lines_async(114 * u.GHz, 116 * u.GHz,
+                                                  chemical_name=' CO ',
+                                                  exclude='none',
+                                                  get_query_payload=True)
+
+    for k,v in exclusions.items():
+        assert k not in d
+
+    for k in d:
+        assert k[:3] != 'no_'
+
 @remote_data
-def test_version_selection():
+def test_exclude_remote():
+    # regression test for issue 616
+    # only entry should be  D213CO  Formaldehyde 351.96064  3.9e-06            --            -- 6(2,4)-5(2,3)            -2.0065 ...            0.0                          -2.94058                  --      44.097 63.44616    55.83714 80.33772     CDMS
     results = splatalogue.Splatalogue.query_lines(
-        min_frequency=703 * u.GHz, max_frequency=706 * u.GHz,
-        chemical_name='Acetaldehyde', version='v1.0')
-
-    assert len(results) == 133
+        min_frequency=351.9*u.GHz,
+        max_frequency=352.*u.GHz,
+        chemical_name='Formaldehyde',
+        exclude='none',
+        )
+    assert len(results) >= 1