Zulip Chat Archive
Stream: maths
Topic: New folder analysis/inner_product_space
Heather Macbeth (Sep 24 2021 at 17:52):
The file analysis.normed_space.inner_product
is 2900 lines long, and now that conjugate-linear maps are in mathlib it's going to grow more. So I propose making a new folder analysis.inner_product_space.*
and splitting the big file into several pieces. Here's a draft: branch#split-inner-product. Comments are welcome!
One benefit is that the main file defining inner product spaces no longer imports deriv
and times_cont_diff
. (Previously these were imported, for the sake of 200 lines about the derivative of the norm, etc., but now these are in a separate file from the definitions.)
I've done my best sorting out authorship of the split files but feel free to suggest people/contributions I've missed. I currently have:
analysis.inner_product_space.basic
: @Joe for #1248, @Sebastien Gouezel for #3060, @Frédéric Dupuis for #4057analysis.inner_product_space.calculus
: @Yury G. Kudryashov for #5089, #5600, #6275analysis.inner_product_space.projection
: @Joe for #1248, @Frédéric Dupuis for #4057, myself for a series of smaller PRs (#5408, #5474, #5543, #5558, #5734, #5876, #8884)
Heather Macbeth (Sep 24 2021 at 17:56):
I have a few TODOs if people like the main refactor:
- deduce continuity of the inner product directly rather than from differentiability (so it can live in
basic
rather thancalculus
) - split
analysis.normed_space.pi_Lp
andanalysis.normed_space.dual
into normed-space and inner-product-space parts.
Scott Morrison (Sep 25 2021 at 01:57):
This is great! I agree splitting pi_Lp
and dual
later would be good, but you might just want to get this refactor through asap before conflicts start hitting it.
Heather Macbeth (Sep 25 2021 at 02:36):
Good point! OK, #9382.
Heather Macbeth (Sep 25 2021 at 14:37):
Second PR #9388 is now live, to split analysis.normed_space.pi_Lp
and analysis.normed_space.dual
. Everything was clearly the work of one person except the new analysis.inner_product_space.pi_L2
, which I have credited to @Joseph Myers for #2852, @Sebastien Gouezel for #3060, and myself for #5949, #6877.
Last updated: Dec 20 2023 at 11:08 UTC