-
-
Notifications
You must be signed in to change notification settings - Fork 52
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Survey appends pyvo requested format on top of assumed format #621
Comments
If you do sia_service.baseurl
After your example you see that in the service URL there are already some parameters. This is not what is expected, and this is why on the server side it looks like you give two times the FORMAT parameter when you try to give one in pyvo. If the others confirm that this is the root of the problem, then I guess the solution is to raise the issue with the contact address: res = registry.search(registry.Servicetype("sia"), registry.Waveband("optical"))['SDSSDR9']
res.get_contact()
|
I recall a related issue from a few years ago, but only have limited GH access at the moment so cannot dig it up. At that time we concluded it was a server side issue. |
On Thu, Nov 21, 2024 at 01:31:55AM -0800, Brigitta Sipőcz wrote:
I recall a related issue from a few years ago, but only have
limited GH access at the moment so cannot dig it up. At that time
we concluded it was a server side issue.
It is, in at least two separate senses:
(a) it's unfriendly to bake in protocol parameters into access URLs,
because that tries to override choices the user or the client may
want to make (what if I'm not interested in FTIS files at all in a
global dataset discovery?)
(b) While I admit that SIAv1 is not really clear on the behaviour
when a parameter is repeated, for FORMAT it is fairly obvious that
ideally, you would interpret that as a logical or, or, failing that,
you overwrite a previous setting. Concatenating with a comma and
interpreting the result as a new media type certainly is nothing a
user can expect or for which any plausible use case could be
conceived.
I was briefly thinking about whether pyVO should do anything about
this in the spirit of the golden rule of interoperability ("be strict
in what you produce, but lenient in what you accept"); but whatever I
could come up with would only make the situation worse by introducing
even more unintuitive behaviour.
So, it's really up to the operators to fix this, starting with an
answer to: Why do they preset FORMAT in the first place?
|
Then we agree that it is on the service provider side. @duytnguyendtn : looks like you'll have to contact skyserver with the email in my precedent answer if you want this fixed |
Hi PyVO maintainers!
A collaborator of mine found a bug that I can't tell if it originates from PyVO or is a faulty survey. They found an example where seems if I request a specific format in my SIA search query, the underlying service receives an amalgamation of what seems to be an assumed format, that's then quashed together with my requested format.
The below queries the registry with the optical SDSSDR9 survey on the source M32, with a radius of 1 degree and specifically filters "image/fits" format only:
results in the following traceback:
Full Traceback
if I remove the
format="image/fits"
arg, the query returns successfully.Thanks in advance!
The text was updated successfully, but these errors were encountered: