Reimplement elements of Zp and Qp using templates

[roed]

The goal of this ticket is to reimplement elements of Zp and Qp using the ideas from sage/rings/polynomial/polynomial_template.pxi. This should make p-adics in Sage both easier to maintain and easier to add new classes.

The final patch should not affect the external interface to Zp and Qp.

---

Apply

• trac_12555.patch
• trac_12555_review.patch
• 12555_rebase_6223.patch

[sydahmad]

Replying to roed:

The goal of this ticket is to reimplement elements of Zp and Qp using the ideas from sage/rings/polynomial/polynomial_template.pxi. This should make p-adics in Sage both easier to maintain and easier to add new classes.

[roed]

Now includes capped absolute and fixed modulus elements. There are a few problems remaining:

• teichmuller_list is disabled since it had a bug and I didn't want to fix it at the time. I'll track this down later.
• Some coercions aren't working since the element class has changed.

This version should be enough for Justin to compare the performance of the templated and non-templated versions. I'm now going to create another patch on top of this which removes the untemplated versions.

[roed]

Still some bugs, but it should be useable to compare timings

[roed]

I've fixed all known bugs; currently doctesting

[was]

this is my REFEREE patch after reading through half the code. I stopped at padic_ZZ_pX_FM_element.pyx. This is the ultimate patch bomb. Very hard to referee...

[saraedum]

A review of this patch is at https://github.com/saraedum/padic-review/commit. We agreed to work on the remaining issues there.

[roed]

I've made a pull request to https://github.com/saraedum/padic-review/commit, addressing the comments. Tests all pass for me.

We had discussed having a document that runs some overall tests. I haven't made that yet, but can this week. If someone has time to take a look at this this week, I'll be available to make changes....

[saraedum]

Replying to roed:

I've made a pull request to https://github.com/saraedum/padic-review/commit, addressing the comments. Tests all pass for me.

Great.

We had discussed having a document that runs some overall tests. I haven't made that yet, but can this week. If someone has time to take a look at this this week, I'll be available to make changes....

You haven't started to work on this already, have you? I just started to work on these tests. I'll push to github as soon as I have something working.

[roed]

We had discussed having a document that runs some overall tests. I haven't made that yet, but can this week. If someone has time to take a look at this this week, I'll be available to make changes....

You haven't started to work on this already, have you? I just started to work on these tests. I'll push to github as soon as I have something working.

No, I haven't started. I'll wait to see what you come up with.

Do we have a list of things that need to be done before this gets a positive review?

[saraedum]

Replying to roed:

Do we have a list of things that need to be done before this gets a positive review?

Not yet. Should we use github issues for this?

[roed]

Replying to saraedum:

Replying to roed:

Do we have a list of things that need to be done before this gets a positive review?

Not yet. Should we use github issues for this?

Sure, that sounds good.

[saraedum]

apply trac_12555.patch

[saraedum]

apply trac_12555.patch

[saraedum]

apply trac_12555.patch

[saraedum]

apply trac_12555.patch

[saraedum]

apply trac_12555.patch

[saraedum]

review patch to match changes done in 5.9.beta0

[jdemeyer]

Please rebase to #6223.

[jdemeyer]

combined github patch

[jdemeyer]

This still needs work rebasing to #6223. Also PyErr_Occured() should not be used anymore in favour of standard Python except handling.

[jdemeyer]

12555_rebase_6223.patch needs review.

[roed]

Those changes look fine to me.

[jdemeyer]

Please fix this documentation problem (I won't):

[139 [690]] [130
! Missing $ inserted.
<inserted text> 
                $
l.10592 $i^{\mbox{prec_cap}
                           }$ if the element was defined by
? ]
! Emergency stop.
<inserted text> 
                $
l.10592 $i^{\mbox{prec_cap}
                           }$ if the element was defined by
!  ==> Fatal error occurred, no output PDF file produced!
Transcript written on padics.log.
 [17make[1]: *** [padics.pdf] Error 1
make[1]: Leaving directory `/mazur/release/merger/sage-5.10.beta1/devel/sage-main/doc/output/latex/en/reference/padics'

[jdemeyer]

* ping *

[roed]

Thanks for the ping. I had a visitor this week, but he just left so I'll take a look.