Opened 4 years ago
Closed 4 years ago
#20254 closed defect (fixed)
Improvements on padic Lfunctions for elliptic curves
Reported by:  wuthrich  Owned by:  

Priority:  minor  Milestone:  sage7.2 
Component:  elliptic curves  Keywords:  days71, padic Lfunctions 
Cc:  roed  Merged in:  
Authors:  Chris Wuthrich  Reviewers:  Julian Rüth 
Report Upstream:  N/A  Work issues:  
Branch:  f29d685 (Commits)  Commit:  f29d68508ffcaa40aadbfa1a6594633fdf388094 
Dependencies:  Stopgaps: 
Description
There are a few issues about padic Lfunction for elliptic curves. One is that the coefficient ring of supersingular series should be an padic Eisenstein field rather than a quotient field.
Change History (27)
comment:1 Changed 4 years ago by
 Keywords days71 added
comment:2 Changed 4 years ago by
 Cc roed added
comment:3 Changed 4 years ago by
 Summary changed from Improvments on padic Lfunctions for elliptic curves to Improvements on padic Lfunctions for elliptic curves
comment:4 Changed 4 years ago by
Oh, there is a lot to do, I discovered.
The #20244 will be independent as I will call extension directly rather than root_field
.
The main issue is the normalisation for padic Lfunction involving negative modular symbols. They are used when a quadratic twist is used and that happens even for E.sha().an_padic(p)
. I am close to resolving this.
The major outstanding issue that I will not change here is #4656 which makes some coefficients (with negative precision sometimes) disappear from the series.
comment:5 Changed 4 years ago by
 Branch set to /u/wuthrich/ticket/20254
 Commit set to 45297356bc88459869d18fc16f2acee7bff60131
comment:6 Changed 4 years ago by
 Commit changed from 45297356bc88459869d18fc16f2acee7bff60131 to a265fc0f458017a32e6a923b836901a8baaaf3c0
 Status changed from new to needs_review
This is ready for review.
Apart from cosmetic changes in documentation and code, this does the following
 The negative modular symbols are now correctly normalised.
[r]^
is the quotient of the imaginary part of2 pi i int_oo^0 f(z)dz
by the imaginary part of the period on the imaginary axis with smallest positive imaginary part.
 The output of a
series
for a supersingular padic Lfunction is now an element in the Eisenstein ring generated by the root of the Frobenius polynomial. In particular, they are printed alphaadically.
New commits:
6677b7b  trac 20254: further small adjustments to twists in padic lseries

139514f  Merge branch 'develop' into oxford

1a7a261  trac 20254: correcting small errors

a265fc0  trac 20254: final doctests adjustments

comment:7 Changed 4 years ago by
Some minor remarks:
 You added a few chunks of commented out code. These should be removed I guess. If they can't then there should at least be a comment what they are about.
_poly()
has no complete docstring (in particular no doctests) You need to set your author in the ticket
Without understanding all the details, everything else seems alright. Feel free to set to postitive review yourself once you adressed the above.
comment:8 Changed 4 years ago by
 Reviewers set to Julian Rüth
comment:9 Changed 4 years ago by
 Status changed from needs_review to needs_work
comment:10 Changed 4 years ago by
 Commit changed from a265fc0f458017a32e6a923b836901a8baaaf3c0 to 8b7674311465d1a604ae44fc56aceef16687c0f0
 Status changed from needs_work to positive_review
comment:11 Changed 4 years ago by
 Status changed from positive_review to needs_work
$ git fetch trac /u/wuthrich/ticket/20254 fatal: Invalid refspec '/u/wuthrich/ticket/20254'
comment:12 Changed 4 years ago by
 Branch changed from /u/wuthrich/ticket/20254 to u/wuthrich/ticket/20254
git fetch trac u/wuthrich/ticket/20254
seems to work.
comment:13 Changed 4 years ago by
 Status changed from needs_work to positive_review
comment:14 Changed 4 years ago by
Sorry if I did something wrong.
comment:15 Changed 4 years ago by
Thanks!
comment:16 Changed 4 years ago by
 Status changed from positive_review to needs_work
Doctests fail (just as in the patchbot, click on the yellow questionmark in the ticket description)
comment:17 Changed 4 years ago by
 Commit changed from 8b7674311465d1a604ae44fc56aceef16687c0f0 to a13fdf4122ddb95f67d341dd3b9acad09d890d70
Branch pushed to git repo; I updated commit sha1. New commits:
a13fdf4  trac 20254: one more doctest

comment:18 Changed 4 years ago by
 Status changed from needs_work to needs_review
Oi, I am sorry for overlooking this one. Not my day.
comment:19 Changed 4 years ago by
 Status changed from needs_review to positive_review
comment:20 Changed 4 years ago by
Merge conflict with #14828
comment:21 Changed 4 years ago by
 Status changed from positive_review to needs_work
comment:22 Changed 4 years ago by
I can't work on this for a week at least. The resolution is that the file padic_lseries.py
should be the one in this branch.
comment:23 Changed 4 years ago by
 Commit changed from a13fdf4122ddb95f67d341dd3b9acad09d890d70 to 8e53c1c85f28bd6ab10e3e2e998127b5ea8e9a67
comment:24 Changed 4 years ago by
 Status changed from needs_work to needs_review
comment:25 Changed 4 years ago by
 Branch changed from u/wuthrich/ticket/20254 to u/mmasdeu/20254
 Commit changed from 8e53c1c85f28bd6ab10e3e2e998127b5ea8e9a67 to f29d68508ffcaa40aadbfa1a6594633fdf388094
I have changed old style doctests ("...") in ell_modular_symbols.py to the new style ("....:"). This should turn the patchbot greener.
comment:26 Changed 4 years ago by
 Status changed from needs_review to positive_review
comment:27 Changed 4 years ago by
 Branch changed from u/mmasdeu/20254 to f29d68508ffcaa40aadbfa1a6594633fdf388094
 Resolution set to fixed
 Status changed from positive_review to closed
Is the coefficient ring issue resolved by #20244? What other changes should be made for padic Lfunctions of elliptic curves?