-
Notifications
You must be signed in to change notification settings - Fork 15
Add function to create spoly from an array of numbers in radians #96
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Conversation
I believe the answer to both question is "yes". Mind that besides the parameter being NULL, individual array elements might also be NULL. |
@esabol I think, everything ok except of some trivial formatting changes. Thank you! |
@esabol I'm ok to accept the patch. Could you please squash the commits? |
2cbb5b0
to
0fcfa36
Compare
Rebased and squashed. Ready to merge whenever you are. Thank you for your reviews, @vitcpp ! |
It seems I'm the only reviewer. Lets merge it, since the risks to break the pgpshere are minimal. |
As discussed in issue #95, this pull request adds a constructor function to create an spoly from an array of numbers in radians. It is very similar to spoly_deg() except without the calls to
deg_to_rad()
.Question 1: Can the spoly(float8[]) function be marked PARALLEL SAFE? My reading of the documentation for PARALLEL SAFE doesn't mention anything about accessing arrays or allocating memory, so I think it should be.
Question 2: Can it be marked STRICT?
select spoly(NULL::float8[])
returns null, I believe.If the answer to either or both questions, I would like to make the same change(s) to
spoly_deg
also.