bjorn-martinsson

Can you be more specific?

As a small note on the fast mod mul. Locally on my windows computer using PyPy the code runs around a factor of 10 faster with modmul compared to doing...

> With the new version of PyPy on Windows, what are your thoughts on this moving forward? Once CF updates (assuming it ever updates) then we can remove all the...

About long_to_decimal, here are some things that I've noticed/thoughts. 1. Increasing the size of BITLIM to something like 32768 gives a noticable performance boost for me. 2. I think it...

I agree that persistent dictionaries are bad, but I never intended `mem` to be one. About the number of keys used. I fixed my implementation to use fewer keys. Now...

> I'm not going to argue about this. As I said, I didn't find the code hard to understand to begin with, so this is just random thrashing to me....

I've complied the three different styles for `int -> str` conversion and `str -> int`. The first style is the one I used in the comment in gh-90716. It is...

As I discussed before, I think it is important to be consistent with the style. Currently `int_to_decimal` and `_str_to_int_inner` are coded in different ways even though they fundamentally do the...

Here is a `_str_to_long_inner` implementation in the same exact style as the one Tim just posted. ```py def _str_to_long_inner(s): """Asymptotically fast conversion of a 'str' to an 'int'.""" # Function...

On the subject on codegolfing the z function. You can save one more character by changing ```cpp z[i] = i >= r ? 0 : min(r - i, z[i -...