[CF-metadata] Invalid alias IDs in CF Standard Name Table v47

Murray Altheim Murray.Altheim at metservice.com
Mon Nov 20 19:08:54 MST 2017


My apologies, I spoke too soon. Upon realising that XML parsers always exit on validation

errors I should have waited until I'd fixed all of the errors. There were more to be found.


Upon making the aforementioned whitespace fixes I noted that the aliases were all

essentially meaningless, i.e., aliases to the same ID. These included the following IDs:


    <alias id="mole_fraction_of_chlorine monoxide_in_air">
    <alias id="mole_fraction_of_chlorine dioxide_in_air">
    <alias id="mole_fraction_of_dichlorine_peroxide_in_air">
    <alias id="mole_fraction_of_hypochlorous_acid_in_air">
    <alias id="rate_of_ hydroxyl_radical_destruction_due_to_reaction_with_nmvoc">
    <alias id="mole_fraction_of_hypochlorous_acid_in_air">


Following these changes I rigged up the XSD and tried validating the document. This
pointed out a duplicate of the alias:

 <alias id="surface_carbon_dioxide_mole_flux">
    <entry_id>surface_upward_mole_flux_of_carbon_dioxide</entry_id>
  </alias>

  <alias id="surface_carbon_dioxide_mole_flux">
    <entry_id>surface_downward_mole_flux_of_carbon_dioxide</entry_id>
  </alias>

which unless I'm mistaken is an attempt to alias an entry to two other different entries. This
would suggest that aliases are many-to-many. Is this intended? It seems odd, as then there
is no single resolution to an alias. [In my local file I've commented out the second element.]

What the above does suggest is that updates to the CF Standard Name Table XML document
are being posted without going through a validation process. Since I assume this document is
to be used in production environments (or at least it's our intention) I'm wondering who I
might discuss this issue with. And yes, I can offer assistance and/or forward a notated copy
of my fixed XML document if requested.

Cheers,

Murray

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cgd.ucar.edu/pipermail/cf-metadata/attachments/20171121/5de2ae16/attachment-0001.html>


More information about the CF-metadata mailing list