search

ivoa-std / ADQL

Astronomical Data Query Language Standard
https://wiki.ivoa.net/twiki/bin/view/IVOA/ADQL
Creative Commons Attribution Share Alike 4.0 International
7 stars 7 forks source link

[RFC] DCP suggested modifications #88

Closed gmantele closed 1 year ago

gmantele commented 1 year ago

All these commits are modifications suggested by DCP WG during ADQL-2.1's RFC.

gmantele commented 1 year ago

I agree with you. I added this definition in here because the paragraph was mentionning the second version of the geometrical function. Knowing that, according to me, it makes sense to explain what parameters roughly are: that's to say a coord_value is a pair of coordinates. But I'd definitely not go further than that. This simple addition does not hurt and may be useful for someone that is not familiar with ADQL-2.1 changes and wants to know more about the number and rough category of parameters. But I agree with you, I do not intend to do that for all terms...otherwise it would multiple copy-pastes of the BNF (which is already terrible to maintain).

msdemlei commented 1 year ago

On Thu, Apr 27, 2023 at 04:40:20AM -0700, Grégory Mantelet wrote:

function. Knowing that, according to me, it makes sense to explain what parameters roughly are: that's to say a coord_value is a pair of coordinates. But I'd definitely not go further than that.

Well, I won't stand in your way, but as I said you shouldn't add this rule to the "and... and... and" concatenation, because that would imply that coord_value is a numeric_value_function, which it is not.

So, if you want to keep it, just make the "and" in front of the role a "where" -- that ought to avoid this ambiguity at least for attentive readers. And the others won't notice the numeric_value_function implication in the first place :-)

gmantele commented 1 year ago

@msdemlei , is it better like this?