[ronnodas]

Right now the methods position_{min,max,minmax}_by_key evaluate the user provided key function each time each item is compared with another, which seems unnecessary, especially in case computing the key is expensive. Note that in any correct implementation, every item must be compared at least once, so each key must be computed at least once.

This does theoretically change the behavior since the key function is allowed to be stateful, in particular if it is non-deterministic.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.72%. Comparing base (6814180) to head (a593dd5).
⚠️ Report is 223 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1136      +/-   ##
==========================================
- Coverage   94.38%   93.72%   -0.67%     
==========================================
  Files          48       52       +4     
  Lines        6665     6484     -181     
==========================================
- Hits         6291     6077     -214     
- Misses        374      407      +33     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[phimuemue]

Hi @ronnodas, thanks, this looks reasonable.

[phimuemue]

And I guess this also opens the door for your idea about making the function FnMut(Self::Item)->K instead of FnMut(&Self::Item)->K, right?

[ronnodas]

And I guess this also opens the door for your idea about making the function FnMut(Self::Item)->K instead of FnMut(&Self::Item)->K, right?

Yes, I can file a PR for that once this one is merged.