- Notifications
You must be signed in to change notification settings - Fork15
Add MOC vs. other geometries casts#6
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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.
Already on GitHub?Sign in to your account
Uh oh!
There was an error while loading.Please reload this page.
Conversation
in contains and intersects constraints.I'm also adding a little file on how to hack pg_sphere; yes, that'smainly repeating stuff from the pg docs, but I guess it'll helpa few folks; it would have helped me.
I hadn't done this before because the warnings spoiled the regressionexpectation. I'm now shutting those up for while I'm doing the dropping.
msdemlei commentedMar 14, 2022
Can I bump this? The CI failures, I think, are unrelated to the PR, as pgsphere probably doesn't work with 9.x anyway (or does it? And if so, is there a strong case to not drop 9.x support?). I give you this isn't perfect and rather certainly won't be terribly fast, but then itwill help for tables in the ~1e4 rows range, such as rr.stc_spatial in future RegTAP (the VO registry discovery protocol,https://github.com/ivoa-std/RegTAP). |
msdemlei commentedAug 17, 2022
Since it'd be great to have these operators for RegTAP 1.2,https://ivoa.net/documents/RegTAP/20220519/, can I nag again about this? I realise we'll not win beauty contests with this, but I'd say it's reasonably harmless and cheap while giving a rather real benefit... |
gmantele left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
It seems all right to me.
It passes all CI for Postgres 10 and more.
This PR will enable search conditions such as scircle <@ moc. It is implemented as syntactic sugar, turning this, in effect, into smoc(max_order(moc), scircle) <@ moc.
While this has several drawbacks against a strict implementation of contains and overlap between spherical geometries and mocs – most notably, it will be a lot slower when comparing a constant geometry against moc-valued columns, because the conversion will be re-done in every row –, implementing these exact operations is at least hard.
For casual users, I argue, this will provide non-surprising behaviour.
This PR also bumps the version to 1.2.1. Piggyback on it rides a brief hacking guide I came up with while writing this.