In Python3 the behaviour of the inbuilt repr() function has changed when compared to Python2

Question

In Python3 the behaviour of the inbuilt repr() function has changed when compared to Python2.

In python2

repr("Дует Світязь")
"'\\xd0\\x94\\xd1\\x83\\xd0\\xb5\\xd1\\x82 \\xd0\\xa1\\xd0\\xb2\\xd1\\x96\\xd1\\x82\\xd1\\x8f\\xd0\\xb7\\xd1\\x8c'"

In Python3

repr("Дует Світязь")
"'Дует Світязь'"

How can I get repr() to work in Python3 the same as in Python2 or as an alternative another method to convert a string to escaped characters In my code. I use the escaped characters in lookup tables for my LCD display translation routines.

The platform is Raspberry Pi OS(Raspbian).

I have also tried using reprlib() without success. It gives the same output as repr(). Using string.encode goes some way to proving a solution, but the result is different to Python2 repr()

txt = "Дует Світязь"
txt.encode('utf8')
b'\xd0\x94\xd1\x83\xd0\xb5\xd1\x82 \xd0\xa1\xd0\xb2\xd1\x96\xd1\x82\xd1\x8f\xd0\xb7\xd1\x8c'

Answer 1

You can encode the string as bytes object:

print('Дует Світязь'.encode('utf-8'))

Prints:

b'\xd0\x94\xd1\x83\xd0\xb5\xd1\x82 \xd0\xa1\xd0\xb2\xd1\x96\xd1\x82\xd1\x8f\xd0\xb7\xd1\x8c'

---

EDIT:

s = str('Дует Світязь'.encode('utf-8')).lstrip('b').replace('\\', r'\\')
print(s)

Prints:

'\\xd0\\x94\\xd1\\x83\\xd0\\xb5\\xd1\\x82 \\xd0\\xa1\\xd0\\xb2\\xd1\\x96\\xd1\\x82\\xd1\\x8f\\xd0\\xb7\\xd1\\x8c'

Answer 2

My thanks to Andrej Kesely who put me on the right path for a solution

In Python2

s = repr(text)

In Python3

s = str(text.encode('utf-8')).lstrip('b')

I don't really understand why the authors of Python3 felt the need to change how a built in function such as repr() works. No doubt they have their reasons.