nimforum mirror - All chr() up to 127 returns correct unicode character, but chr([128++]) returns �

This single-line code:

echo chr(97) # a

Returns the letter "a" All Unicode decimal numbers up to 127 work.

But all Unicode numbers from 128 - 255 return � (replacement character)

This also happens to be the cutoff between Basic Latin and Latin-1 Supplement. https://en.wikipedia.org/wiki/List_of_Unicode_characters#Latin-1_Supplement

I've tried the corresponding function in Python on the terminal to see, but Python produces the correct character.

What am I missing?

Thank you,

--Peter

char is ASCII.

More specifically if you want to turn numbers into unicode characters you need to use the unicode module:

import unicode

for i in 120..127:
  echo i, ": ", i.Rune
for i in 162..168:
  echo i, ": ", i.Rune

With ASCII, the problem still exists: any number 127 or higher produces �.

This works! Now I understand -- thank you.

It seems that UTF-8 uses 2 bytes to store extended latin characters, and thus not (binary-) compatible with Latin-1.

echo len("\u00e6") # 2

Mirror of forum.nim-lang.org