Overhaul printing of orders, (fractional) ideals, and related objects #1840
Conversation
|
still missing fractional ideals |
|
I wonder if we should just do |
c1c4db8 to
59f7e09
Compare
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #1840 +/- ##
==========================================
- Coverage 76.61% 76.45% -0.17%
==========================================
Files 362 364 +2
Lines 115012 115635 +623
==========================================
+ Hits 88119 88407 +288
- Misses 26893 27228 +335
🚀 New features to boost your workflow:
|
|
Updated (went back to |
docs/src/howto/defineorder.md
Outdated
| Order of | ||
| number field with defining polynomial x^2 - 5 | ||
| over rational field | ||
| with Z-basis [1, a] |
There was a problem hiding this comment.
I am wondering if, according to OSCAR printing conventions, it would not be better to have something printed as:
Order
of number field with defining polynomial x^2 - 5
over rational field
with Z-basis [1, a]
All the of, over, from... should be on a new line just before what they refer to.
There was a problem hiding this comment.
Hm? Do you mean the following?
Order
of number field with defining polynomial x^2 - 5
over rational field
with Z-basis [1, a]
or else I don't understand the "rule".
There was a problem hiding this comment.
This is the same as when we define maps I guess
julia> F = hom(A, B, [x^2+z, y^2-1, z^3])
Ring homomorphism
from multivariate polynomial ring in 3 variables over QQ
to quotient of multivariate polynomial ring by ideal (x*y)
defined by
a -> x^2 + z
b -> y^2 - 1
c -> z^3
or what has been implemented with schemes also
julia> Y = affine_space(QQ,3)
Affine space of dimension 3
over rational field
with coordinates [x1, x2, x3]
But yeah, anyway, within OSCAR there are still inconsistencies... I was just giving my opinion. I guess the most important was to move the of in the indented part. Where you put the with Z-basis ... is up to taste, maybe. I would still argue to de-indent it. My reason is that the basis belong to the object, the number field does not.
There was a problem hiding this comment.
But the scheme example also has an of in the nonindented part. I really don't understand the rule (if there is actually one).
Compare with
julia> lie_algebra(QQ, :A, 5)
Abstract Lie algebra
of type A5
of dimension 35
over rational field
where all of things are indented, but where the indentation of over is the opposite of the scheme example.
There was a problem hiding this comment.
I guess in the case of scheme it is because the of dimension 3 is small enough to be put on the first line (which happens often in other printings), and it is not a standalone of, there is something after it.
I do not think there is a general rule, just a global guideline and suggestions. Some people might disagree with what I am telling you now, of course.
And for the Lie Algebra printing, well, I do not agree with it.
docs/src/howto/reduction.md
Outdated
| Ideal of order of cyclotomic field of order 7 | ||
| with 29-normal generators [29, z_7 + 22] | ||
| of norm 29 | ||
| of minimum 29 |
There was a problem hiding this comment.
And this printed as
Ideal
of order of cyclotomic field of order 7
with 29-normal generators [29, z_7 + 22]
of norm 29
of minimum 29
Probably here also the order of... could be more detailed but it really depends on how much information one wants to pass.
There was a problem hiding this comment.
This make it look like the second of are properties of the with 29-normal ... line.
There was a problem hiding this comment.
Yes that's my bad, I misunderstood it. Then why not keeping (non-indented) Norm: 29 and Minimum: 29 as it was before ?
There was a problem hiding this comment.
I find the rule of what is intended and what not quite arbitrary. So I intended to indent everything consistently.
Indentation should help parse things. But if indentation is not consistent, this does not help. For example printing many ideals would give something like
Ideal
of order of cyclotomic field of order 7
with 29-normal generators [29, z_7 + 22]
of norm 29
of minimum 29
Ideal
of order of cyclotomic field of order 7
with 29-normal generators [29, z_7 + 22]
of norm 29
of minimum 29
Ideal
of order of cyclotomic field of order 7
with 29-normal generators [29, z_7 + 22]
of norm 29
of minimum 29
It is now difficult to see where one thing starts/ends.
|
Hm, I don't like it kind of defeats the purpose of using indentation to mark where things start or end. |
162da85 to
05126e5
Compare
|
Added the relative stuff and also the pseudo-matrix. It went from to |
|
I am finished for now and I think it is a clear improvement. There might be some disagreement about minor details, but this would be the case for any change. |
|
@StevellM sorry for the confusion. You were right. I tried to adjust it accordingly. Can you check again? |
|
It looks nice, I particularly like the new printing for pseudo-matrices! You are right in the sense that it would be hard to please everybody. I think in general the new printing is good, and it is clearly an improvement, as you said. Would it make sense to indent more the numerator ideal, like or to just compactify it ? I have no preferences; the current printing looks fine to me. The last option would require maybe to, as in the case of integral ideals, add the maximal order to the printing, so more like (if I did not mess up whom the pseudo-basis belong to...) |
|
OK, I see. I think I will go with if there is no objection. |
|
Tests look good. Breakage is expected. I might use the chance to update some other printing as well in this PR here. |
93e873c to
b3754ee
Compare
No description provided.