正确的解决方案，包括空格：

Correct solution, including spaces:

use open ':std', ':encoding(UTF-8)';
use Encode;

my $escaped = '%41F%2F%424+%41F%41E%414%416%410%420%41A%410+%418%417+%421%412%418%41D';
(my $unescaped = $escaped) =~ s/\+/ /g;
$unescaped =~ s/%([[:xdigit:]]+)/chr hex $1/eg;
print $unescaped;
# П/Ф ПОДЖАРКА ИЗ СВИН

Credit转到 Renaud Bompuis 首先识别这些是以％为前缀的Unicode代码点。

Credit goes to Renaud Bompuis for recognising as the first that these are Unicode code-points prefixed with %.

我想补充一点，这个问题的编码方案非常不寻常，我以前没有见过。通常，人们期望字符串П/ФПОДЖАРКАИЗСВИН被编码为％D0％9F％2F％D0％A4 +％D0％9F ％D0％9E％D0％94％D0％96％D0％90％D0％A0％D0％9A％D0％90 +％D0％98％D0％97 +％D0％A1％D0％92％D0％98 ％D0％9D ，也就是说，首先将字符编码为UTF-8，然后八位字节进行百分比转义。此方案适用于 Dr.Kameleon 的答案。

I wish to add that the encoding scheme from the question is very unusual, I haven't seen it before. Normally one would expect the characters string П/Ф ПОДЖАРКА ИЗ СВИН to be encoded as %D0%9F%2F%D0%A4+%D0%9F%D0%9E%D0%94%D0%96%D0%90%D0%A0%D0%9A%D0%90+%D0%98%D0%97+%D0%A1%D0%92%D0%98%D0%9D, that is to say, first the characters are encoded into UTF-8, then the octets are percent-escaped. This scheme works with the answer from Dr.Kameleon.