Refactor NIST

[jdnc]

This refactors the NIST module. Still need to update nist.rst, add remote tests

[jdnc]

I just realized the table doesn't read the last two columns well because of them being url links. Working on this.

 Observed     Ritz     Rel.    Aki    ... Type    TP       Line   
-------- ----------- ----- --------- ... ---- -------- ----------
  --    4020.871 (200)    5526.5 ... over "showBal oon('[',ev
  --  4052.18664    -- 1238100.0 ...   --   <a cla s='bib' ON
  --  4052.19376    --  737160.0 ...   --   <a cla s='bib' ON
  --  4052.22121    --  215030.0 ... over "showBal oon('[',ev
  --  4052.23222    --  737210.0 ...   --   <a cla s='bib' ON
  -- 4052.248747    -- 2412100.0 ... over "showBal oon('T',ev
  --  4052.24892    -- 1485800.0 ... over "showBal oon('[',ev
  --  4052.26147    --   18846.0 ...   --   <a cla s='bib' ON
 ...         ...   ...       ... ...  ...      ...        ...
  --     5525.19 (150)    2470.9 ... over "showBal oon('[',ev
  --    5711.464 (180)    3515.8 ... over "showBal oon('[',ev
  --     5908.22 (540)   70652.0 ...   --   <a cla s='bib' ON
  --    5956.845 (210)    5156.2 ... over "showBal oon('[',ev
  --    6291.918 (250)    7845.7 ... over "showBal oon('[',ev
  --    6771.993 (300)   12503.0 ...   --   <a cla s='bib' ON
  --    6946.756    --    688.58 ... over "showBal oon('[',ev

[keflavich]

Note that yesterday's version of process_asyncs had some errors in it; you should grab the latest. I'll merge the asyncs branch into master tomorrow, though, if you want to wait on that (we can discuss on hangout)

[jdnc]

There seems to be a slight issue in the rendering of the generated functions API via sphinx

[keflavich]

I've already used this successfully a few times:

>>> from astroquery import nist
>>> import astropy.units as u
>>> nist.Nist.query(minwav=1.63*u.um, maxwav=1.65*u.um, linename="Fe II")
<Table rows=1 names=('Observed','Ritz','Rel.','Aki','Acc.','Ei           Ek','Lower level','Upper level','Type','TP','Line')>
masked_array(data = [ (--, 1.643998, --, 0.006, 'C', '0.2321687    -    0.9863313', '3p6.3d7 | a 4F | 9/2', '3d6.(5D).4s | a 4D | 7/2', 'E2', 'T6677', --)],
             mask = [(True, False, True, False, False, False, False, False, False, False, True)],
       fill_value = (999999, 1e+20, 999999, 1e+20, 'N', 'N/A', 'N/A', 'N/A', 'N/', 'N/A', 999999),
            dtype = [('Observed', '<i8'), ('Ritz', '<f8'), ('Rel.', '<i8'), ('Aki', '<f8'), ('Acc.', 'S1'), ('Ei           Ek', 'S27'), ('Lower level', 'S20'), ('Upper level', 'S24'), ('Type', 'S2'), ('TP', 'S5'), ('Line', '<i8')])

>>> nist.Nist.query(minwav=1.59*u.um, maxwav=1.61*u.um, linename="Fe II")
<Table rows=1 names=('Observed','Ritz','Rel.','Aki','Acc.','Ei           Ek','Lower level','Upper level','Type','TP','Line')>
masked_array(data = [ (--, 1.599909, --, 0.00418, 'C', '0.3012936    -    1.076239', '3p6.3d7 | a 4F | 7/2', '3d6.(5D).4s | a 4D | 3/2', 'E2', 'T6677', --)],
             mask = [(True, False, True, False, False, False, False, False, False, False, True)],
       fill_value = (999999, 1e+20, 999999, 1e+20, 'N', 'N/A', 'N/A', 'N/A', 'N/', 'N/A', 999999),
            dtype = [('Observed', '<i8'), ('Ritz', '<f8'), ('Rel.', '<i8'), ('Aki', '<f8'), ('Acc.', 'S1'), ('Ei           Ek', 'S26'), ('Lower level', 'S20'), ('Upper level', 'S24'), ('Type', 'S2'), ('TP', 'S5'), ('Line', '<i8')])

>>> nist.Nist.query(minwav=2.16*u.um, maxwav=2.17*u.um, linename="H I")
<Table rows=1 names=('Observed','Ritz','Rel.','Aki','Acc.','Ei           Ek','Lower level','Upper level','Type','TP','Line')>
masked_array(data = [ (2.1661178, 2.166121, 8000, 304150.0, 'AAA', '12.7485385    -     13.32091736', 4, 7, --, 'T8637', 'L7421')],
             mask = [ (False, False, False, False, False, False, False, False, True, False, False)],
       fill_value = (1e+20, 1e+20, 999999, 1e+20, 'N/A', 'N/A', 999999, 999999, 999999, 'N/A', 'N/A'),
            dtype = [('Observed', '<f8'), ('Ritz', '<f8'), ('Rel.', '<i8'), ('Aki', '<f8'), ('Acc.', 'S3'), ('Ei           Ek', 'S31'), ('Lower level', '<i8'), ('Upper level', '<i8'), ('Type', '<i8'), ('TP', 'S5'), ('Line', 'S5')])

However, there ought to be a way to catch errors from empty tables:

>>> nist.Nist.query(minwav=2.17*u.um, maxwav=2.18*u.um, linename="H I")
ERROR: IndexError: list index out of range [astroquery.nist.core]
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/adam/repos/astroquery/astroquery/utils/class_or_instance.py", line 25, in <lambda>
    f = lambda *args, **kwds: self.fn(cls, *args, **kwds)
  File "/Users/adam/repos/astroquery/astroquery/nist/core.py", line 51, in newmethod
    result = self._parse_result(response)
  File "/Users/adam/repos/astroquery/astroquery/utils/class_or_instance.py", line 25, in <lambda>
    f = lambda *args, **kwds: self.fn(cls, *args, **kwds)
  File "/Users/adam/repos/astroquery/astroquery/nist/core.py", line 194, in _parse_result
    pre = pre_re.findall(response.content)[0]
IndexError: list index out of range

[jdnc]

Now this seems to handle this case correctly:

In [1]: import astropy.units as u

In [2]: from astroquery.nist import Nist

In [3]: Nist.query(minwav=2.17*u.um, maxwav=2.18*u.um, linename="H I")
ERROR: Exception: Result did not contain a table [astroquery.nist.core]
---------------------------------------------------------------------------

[keflavich]

@jdnc - rebase this whenever you have a chance (the process_asyncs code can now be imported)

[jdnc]

This is now rebased. There are issues in the rendering of query docs. Maybe we require a full-stop before the Parameters section begins ? e.g.

[astrofrog]

@jdnc - the first line of a docstring should always end with a period - I think that is the issue here.

[jdnc]

@astrofrog , since this was a minor change to process_asyncsI have pushed in the commit for a period at the end along with #158