How to print accented characters with U8g2lib in character mode ? about u8g2 HOT 4 CLOSED

You need to check, whether the required character is available in the selected font. In your example the selected font doesn't contain the desired character: https://github.com/olikraus/u8g2/wiki/fntgrpopengameart#chroma48medium8

I suggest to use this font instead: https://github.com/olikraus/u8g2/wiki/fntgrpcodeman38#pressstart2p

from u8g2.

OK, in the code snippet above, I changed the font, as folllows:

    //u8x8_i2c.setFont(u8x8_font_chroma48medium8_r);                              
    u8x8_i2c.setFont(u8x8_font_pressstart2p_f);                                     

Looking at the character map, I can verify that the characters printed are according to the character codes:
example: when printing 'char(224)', it prints 'à', which is correct.

But when printing "à", it prints "Ä " (wrong character, followed by a space), which is not correct.
Check: printing asc('à) returns 195. The character map of the font indeed indicates Ä for character code 195.

So, Arduino does not translate a character to the character code expected by the OLED.
Example: it sends character code 195, and not 224, when printing 'à'.

Does that mean I'm again using a wrong font ? I would again appreciate your help.

Best regards
Herwig

from u8g2.

The short suggestion is: Read out the hex code from the table. For Ä this would be 0xc4. Hex codes can be embedded into strings, by using \x escape sequence: "\xc4pfel" would print the German word "Äpfel" (apples).

Long answer: If you enter "Ä" (or any other char) into your keyboard, then:

1. your text editor translates the key into an encoded which is used / configured in your text editor. This might be UTF-8 in case of Arduino IDE. The UTF8 encoding for Ä is "0xC3 0xA4", which as nothing todo with 0xC4
2. Once your c-code contains the byte sequence "0xC3 0xA4" (remember you will always see Ä because your editor is configured for UTF-8), then the C-Compiler has to understand what exactly is "0xC3 0xA4". Usually it doesn't do much, but some remarks are here: https://gcc.gnu.org/onlinedocs/gcc-4.0.4/cpp/Character-sets.html
3. Finally the software (in this case U8g2 / U8x8) has to decode the encoded character and fetch the correct glyph from the font. In this case, u8x8 and u8g2 have to assume, what input encoded you might have, because there is no way to tell u8g2 / u8x8 how your editor is configured.

Now, this is what u8g2 and u8x8 will do: Both will assume ISO/IEC 8859 encoding.
U8g2 (as a special feature) also supports UTF8 if enabled (https://github.com/olikraus/u8g2/wiki/u8g2reference#enableutf8print) or if you use the drawUTF8 function.

U8x8 has a similar function, but my suggestion is not to trust that u8x8 function.

Coming back to your problem: Your IDE will probably use UTF8 encoding, however U8x8 assumes ISO/IEC 8859 encoding... so no wonder the wrong chars will appear. So better use the \xYY number code from the u8x8 font table.

from u8g2.

OK, thanks for the quick reply.

from u8g2.