#16127: Fix comparison of PARI objects
-------------------------------------+-------------------------------------
Reporter: pbruin | Owner:
Type: defect | Status: new
Priority: critical | Milestone: sage-6.4
Component: interfaces | Resolution:
Keywords: pari comparison | Merged in:
Authors: | Reviewers:
Report Upstream: N/A | Work issues:
Branch: | Commit:
u/pbruin/16127-gcmp | a119b00099787f5cb8c3cc8477119184e16806ee
Dependencies: | Stopgaps:
-------------------------------------+-------------------------------------
Comment (by pbruin):
Replying to [comment:7 jdemeyer]:
> Replying to [comment:5 pbruin]:
> > Maybe it is better to write a variant of `gcmp_sage()` that calls
`gcmp()` and catches errors using PARI's `pari_CATCH` and `pari_TRY`
macros so that we don't have to raise Python exceptions.
> If you do this, I would prefer to see this function in the PARI library
(call it `gcmp_any()` for example). Hopefully, this can be done in a way
which is acceptable for upstream.
I ended up practically copying the existing (static) `gen2.c:gequal_try()`
and calling it `gcmp_try()`. PARI already has another (somewhat
different) universal comparison function `cmp_universal()`; I don't think
the PARI library currently has a use for this `gcmp_try()`.
New branch coming soon, but here are the new timings (same notation as in
comment:5), which leave nothing more to wish for in my opinion:
{{{
cmp cmp == == < <
before after before after before after
(a, a) 189 ns 176 ns 151 ns 142 ns 156 ns 156 ns
(a, b) 1.38 µs 324 ns 568 ns 520 ns 608 ns 212 ns
(a, c) 328 ns 276 ns 171 ns 272 ns 166 ns 178 ns
(a, d) 1.06 µs 984 ns 500 ns 284 ns 572 ns 300 ns
(a, e) 968 ns 1.27 µs 512 ns 224 ns 456 ns 332 ns
(a, f) 1.23 µs 1.24 µs 580 ns 408 ns 532 ns 336 ns
(b, b) 175 ns 164 ns 748 ns 157 ns 864 ns 216 ns
(b, c) 3.96 µs 348 ns 1.59 µs 348 ns 1.42 µs 232 ns
(b, d) 1.44 µs 1.38 µs 748 ns 336 ns 696 ns 352 ns
(b, e) 1.47 µs 1 µs 796 ns 264 ns 624 ns 336 ns
(b, f) 1.61 µs 1.24 µs 824 ns 404 ns 696 ns 480 ns
(c, c) 176 ns 180 ns 153 ns 156 ns 140 ns 186 ns
(c, d) 3.56 µs 2.52 µs 1.19 µs 288 ns 1.33 µs 348 ns
(c, e) 3.04 µs 2.72 µs 1.24 µs 216 ns 1.18 µs 324 ns
(c, f) 3.12 µs 2.52 µs 1.83 µs 288 ns 1.5 µs 292 ns
(d, d) 182 ns 175 ns 556 ns 156 ns 492 ns 296 ns
(d, e) 1.96 µs 1 µs 548 ns 232 ns 540 ns 328 ns
(d, f) 2.28 µs 1.29 µs 672 ns 288 ns 640 ns 312 ns
(e, e) 169 ns 176 ns 496 ns 153 ns 532 ns 328 ns
(e, f) 2.16 µs 992 ns 620 ns 296 ns 548 ns 296 ns
(f, f) 179 ns 181 ns 656 ns 158 ns 652 ns 328 ns
(t, t) 166 ns 180 ns 568 ns 688 ns 708 ns 752 ns
(t, u) 1.48 µs 1.48 µs 648 ns 788 ns 648 ns 772 ns
}}}
--
Ticket URL: <http://trac.sagemath.org/ticket/16127#comment:8>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/d/optout.
