Opened 6 years ago

Last modified 5 years ago

#18298 new enhancement

prime_range vs primes

Reported by: vdelecroix Owned by:
Priority: major Milestone: sage-6.7
Component: number theory Keywords:
Cc: leif Merged in:
Authors: Reviewers:
Report Upstream: N/A Work issues:
Branch: Commit:
Dependencies: Stopgaps:

Status badges


There is currently a prime_range function in sage.rings.fast_arith which return a list of prime in a given range {0, ..., n-1} and a function primes in sage.rings.arith which return an iterator.

We should have only one entry point and the answer would better be an iterator.

Change History (7)

comment:1 Changed 6 years ago by kcrisman

Make sure that it's still easy to use for newcomers - my concern is that it might be non-obvious what to do with an iterator as opposed to a list, which is very clear. This is the sort of thing that someone with zero programming experience is most likely to use - hey, get a list of the primes less than 1000! Maybe that's even an argument for keeping both, or maybe not. But at any rate it should be clear what to do, plus deprecation issues.

comment:2 Changed 6 years ago by vdelecroix

I do care about newcomers and this is one motivation for this ticket. But currently primes return an iterator while prime_powers return a list! On the other hand, my choice was motivated by Python 3 where even range would return an iterator. If you feel like we should have both functions (one returning a list and one returning an iterator) please tell me.

comment:3 Changed 6 years ago by kcrisman

Hmm, not sure. My personal preference would be to have default return a list, with a flag to return an iterator, but that will probably not be popular for most others.

comment:4 Changed 6 years ago by leif

  • Cc leif added

comment:5 Changed 5 years ago by cremona

I found this ticket and the next one, and wonder if they are going to be worked on?

My current beef is that the docstring for primes is wrong. The only argument with no default is "start" and it says that this is a lower bound (as the name suggests) but in fact it is an upper bound unless "stop" is also set! ALso there is no way at present (I think) to get an open-ended iterator with only one parameter, you have to use primes(2,infinity). You can also use Primes() as in "for p in Primes(): ..." but I have no idea how efficient that is.

Last edited 5 years ago by cremona (previous) (diff)

comment:6 follow-up: Changed 5 years ago by leif

I don't get what Vincent intends to do. Merge both functions into a single one (with an even more complicated interface)?

Both have pretty different applications, and the functions' docstrings refer to each other.

In the long run, we could of course implement something better...

comment:7 in reply to: ↑ 6 Changed 5 years ago by leif

Replying to leif:

In the long run, we could of course implement something better...

See also #9451... ;-)

Note: See TracTickets for help on using tickets.