Download failure

Question

I am trying to match a sample of quasars (4200 objects) to DESI DR1 with the following query;

SELECT a.*, b.flux_r, b.flux_ivar_r, b.flux_z, b.flux_ivar_z
FROM mydb://desi_highz AS a, desi_dr1.photometry AS b
WHERE q3c_join(a.ra, a.declination, b.ra, b.dec, 1./3600.);

The query executes without error and displays the first 100 rows of the table, however, when I try to save the output to mydb I get an error and when I try to download the table as a .csv I get a load of junk. If I open the error message in the jobs tab I don't get anything meaningful - does anyone have any ideas?

Answer 1

I assume you are using the query interface on our Data Explorer page? If so, that interface uses ADQL, and mydb:// tables cannot be aliased in ADQL (the "FROM mydb://desi_highz AS a" part of your query). Instead, I would recommend using our Jupyter notebook server and submitting the query in SQL like:

from dl import queryClient as qc

q = """
SELECT a.*, b.flux_r, b.flux_ivar_r, b.flux_z, b.flux_ivar_z
FROM mydb://desi_highz AS a, desi_dr1.photometry AS b
WHERE q3c_join(a.ra, a.declination, b.ra, b.dec, 1./3600.)
"""

res = qc.query(sql=q, out='mydb://desi_highz_results')

Or if you want to download the table as a .csv file:

res = qc.query(sql=q, out='./desi_highz_results.csv')

Comments

FIXED: I needed to authClient.login() before accessing mydb

Thank you - I was using the Data Explorer. When I navigate to "Launch notebook" and try the above I get the following error: "queryClientError: Error: QM: MyDB table desi_highz does not exist'". I uploaded it to mydb on the data explorer page, is there an additional step required to get access to the database in the Jupyter lab?

---

Oh yes sorry I forgot to include that in my original reply but yep you got it, you need to be logged in with authClient first