Max Bachmann

In RapidFuzz I do the same for ninja/make. In addition I only install them when wheels are available, since the source build requires cmake to be available anyways. When installation...

> I'm hoping scikit-build's own backend will do this soon. An implementation inside scikit-build would likely always install cmake when it is not available yet. Note that this would break...

A similar error occurs on aarch64 + ppc64le when building for conda forge: https://dev.azure.com/conda-forge/feedstock-builds/_build/results?buildId=525876&view=logs&jobId=d1321064-f6c3-56b7-0172-6d994d01c836&j=696704cc-6fef-57a3-ea36-f27779b8cd5e

In [rapidfuzz](https://github.com/maxbachmann/RapidFuzz) I implement this metric using a sliding window approach, which skips comparisons only if it is really sure they can not occur. When searching for long needles (>...

Some things about this could be simplified: 1) only make this many x many. When users pass in a list with a single element this is good enough of a...

https://github.com/maxbachmann/RapidFuzz/commit/65fd6b0c7de9a4de121cdb0536256db2b19f80b1 implements the case of limit=1 to make it as performant es extractOne.

> Should you find it helpful, I would be glad to send you my implementation files. Seeing it properly cythonized in RapidFuzz (which I am technically unable to do myself)...

It has been quite a while since back then and a few things have changed in the meantime. One of the major changes is that going forward we will focus...

That's completely understandable. I just wanted to give you the option to continue working on this, since I know you did put quite a bit of work into this back...

This certainly makes sense. Internally we already run into this in the string versions. They use `StringWrapper` only to workaround this. I would consider this a breaking change though.