gh-109653: Reduce import overhead of uuid module on Linux

[hauntsaninja]

This follows in the footsteps of #21586
This speeds up import uuid by about 6ms on Linux.

Before:

λ hyperfine -w 4 "./python -c 'import uuid'"
Benchmark 1: ./python -c 'import uuid'
  Time (mean ± σ):      20.4 ms ±   0.4 ms    [User: 16.7 ms, System: 3.8 ms]
  Range (min … max):    19.6 ms …  21.8 ms    136 runs

After:

λ hyperfine -w 4 "./python -c 'import uuid'"
Benchmark 1: ./python -c 'import uuid'
  Time (mean ± σ):      14.5 ms ±   0.3 ms    [User: 11.5 ms, System: 3.2 ms]
  Range (min … max):    13.9 ms …  16.0 ms    175 runs

• Issue: Improve import time of various stdlib modules #109653

[zooba]

I'm not the right person to review this.

You need someone who knows whether sys.platform == "linux" implies the same things as the platform module. In general, they don't, so if you want my review I would reject this change as being too risky.

[hauntsaninja]

Thanks for the quick response! I'll wait for someone else to review.

The value of sys.platform comes from

cpython/configure

Line 4296 in 1032326

case $MACHDEP in

while the value of platform.system comes from os.uname + munging and falling back to sys.platform

Note we don't need a one-to-one correspondence here (and I'm not sure there is one, e.g. on ancient Python versions it was possible to have sys.platform == "linux2" while presumably platform.system() was still Linux). But we just need implication in one direction since we preserve the else branch.

[ofek]

This seems correct to me; I don't see any risk to this change.