Boop application misinterprets UTF-8 text as the native codepage of the operating system
DragRedSim opened this issue · comments
As per the title.
Example: attempting to boop from the game https://www.steamgriddb.com/game/5262323 gives the following:
Note the character in the title which maps to hex C2, which in Windows-1252 produces a Latin Capital letter A with circumflex , but which in UTF-8 is a pointer to Latin-1 Supplement, with the following character, hex AE, being shown in its Windows-1252 representation as the correct character, the registered trademark symbol. The string displays correctly on the SteamGridDB site, since pages there declare the UTF-8 encoding
Function reference:
Line 892 in 3fb75d2
Since the string is passed in as a char* pointer, it will be treated as a set of bytes without care given to the encoding within.
As a workaround, if the user enables UTF-8 support in Windows 11, it should parse the characters correctly.
I suspect this will also extend to the qsort() call within the function, causing certain cases not to match where they may otherwise appear to do so.