Re: PATH duplications

[voidlinux-github@inbox.vuxu.org]

[-- Attachment #1: Type: text/plain, Size: 2883 bytes --]

New comment by loreb on void-packages repository

https://github.com/void-linux/void-packages/issues/18777#issuecomment-585884449

Comment:
TLDR: imho the speedup is not worth the confusion, but worth adding a comment ("not having these in $PATH would confuse most people" or "these are not needed in void") either way just in case.

A script that runs a command many times - successfully:
```
#! /bin/sh
for i in 1 2 3 4 5 6 7 8 9 0 a b c d e f g h i j k l m n o p q r s t u v w x y z ; do
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
	cat /dev/null; cat /dev/null; cat /dev/null; cat /dev/null
done
```

To test non existing commands:
```
#! /bin/sh
foo&&exit 123 # must NOT be in $PATH
for i in 1 2 3 4 5 6 7 8 9 0 a b c d e f g h i j k l m n o p q r s t u v w x y z ; do
	foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo
	foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo
	foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo
	foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo
	foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo
	foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo
	foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo; foo
done
```

On my computer the first is just as fast even if we replace the first few entries of path with something bogus - PATH=/usr/local/nibs:/usr/local/nib/...:/usr/bin
The second script is indeed almost 3x slower when run with the full path instead of PATH=/usr/bin, but we are talking 0.05s vs 0.13s when we call a nonexisting command 3000 times.

All of that on a cheap laptop from 2015 without an SSD btw - does anyone have different numbers?