Search provider - NASA ADS

[vilhelmp]

Rewrote the NASA ADS search I had lying around for inclusion into astroquery. Instead of parsing the HTML results with Beautiful Soup, I request the results in XML, and use the xml.dom.minidom inspector to parse the results.

Simple search:

    import astroquery.nasa_ads as ads
    result  = ads.ADS.query_simple('^Ginsburg')
    # result is now a Astropy.Table table. eg:
    result['authors'] # to see the list of lists of authors

However, print result raises an error, I think it's because of the encoding, in the function astroquery.nasa_ads.utils.get_data_from_xml the parsing of the individual elements happens (.encode("utf-8") tries to encode the text field into utf-8).

[keflavich]

For tables, you can use the function astropy.table.Table.convert_bytestring_to_unicode to fix some unicode issues, perhaps - I can't promise this is relevant to the problem you've noted, I'm just trying to pass on something I recently learned.

Thanks for the PR. I'll try to review it in more detail when I'm back from travel

[vilhelmp]

I just discovered this: https://github.com/andycasey/ads

[keflavich]

It is nice but we need an API key to use that. Maybe we can get one for
astroquery? I think a python interface to the simpler, old ADS form is
still useful

On Tue, Mar 17, 2015 at 11:18 AM, Magnus Vilhelm Persson <
notifications@github.com> wrote:

I just discovered this: https://github.com/andycasey/ads

—
Reply to this email directly or view it on GitHub
#499 (comment).

Adam

[keflavich]

Can you rebase or merge against master? we fixed the astropy build issue in #503

[vilhelmp]

Yeah I don't know what I just did. I tried to rebase again, before adding the doc to index.rst. Got an error, that there were conflicts, from the first commit. Then I thought, I clone my fork again locally (to a new destination), and then work on that instead, that seem to have screwed up things on my machine, now I don't know what to do.

[keflavich]

I'll investigate tomorrow.

[keflavich]

@vilhelmp if you want some live help dealing with the git-related stuff, join gitter (https://gitter.im/astropy/astropy) and I can help walk you through the issues. Re-cloning should never be necessary, though I did it a few times before I understood rebasing.

As for the error, it's still a simple one:
https://travis-ci.org/astropy/astroquery/jobs/58165454#L1473
just need to add the NASA ADS entry in the index.

[coveralls]

Coverage decreased (-0.33%) to 65.88% when pulling 1a659ec on vilhelmp:master into 909fc9a on astropy:master.

[vilhelmp]

I get some errors in other modules, I think, probably because I updated my fork from upstream, and then pushed my changes. Since I am organizing a Software Carpentry boot camp today and tomorrow, I don't have time during daytime for gitter.

[keflavich]

Travis is passing. Coverage is complaining because you don't have tests, but that isn't a blocker - tests would be really great, but I'm willing to incorporate un-locally-tested tools for the time being at least. So, is this ready to merge, from your side?

[keflavich]

Actually, on further inspection, we need some docs and at least a remote test/example...

[coveralls]

Coverage decreased (-0.33%) to 65.88% when pulling 9f3e83c on vilhelmp:master into 8540277 on astropy:master.

[keflavich]

this stuff should be in an indented code block:

.. code-block:: python

    from astro....

[keflavich]

the failure is just sphinx

[coveralls]

Coverage increased (+0.0%) to 66.21% when pulling bd34224 on vilhelmp:master into 8540277 on astropy:master.

[keflavich]

you're green now! ready to merge?

[keflavich]

all these commented out things should be removed

[vilhelmp]

Think I fixed everything!

[coveralls]

Coverage decreased (-0.0%) to 66.2% when pulling 2aae04f on vilhelmp:master into 8540277 on astropy:master.

[coveralls]

Coverage decreased (-0.0%) to 66.2% when pulling 2aae04f on vilhelmp:master into 8540277 on astropy:master.