Not all words are converted

[turgic]

I have 2 files :
File has one row with : magn�sienne
Second file has one row with magn�sienne
The results for one document is :
magnésienne
For other :
magn?sienne
Implimentation :
$return[] = $this->utf8Encoding($row);

private function utf8Encoding($datas)
{
foreach ($datas as $key => $data) {
$datas[$key] = Encoding::fixUTF8($data);
}
return $datas;
}
Have you an idea ? Thx in advance

[millenniumtree]

Here's an explanation of that special character:

U+FFFD � REPLACEMENT CHARACTER used to replace an unknown, unrecognized or unrepresentable character

So if your data source has already replaced the original character with that dummy character, there may be no way to get it back.

[garrettw]

Although those two lines may appear the same in your text editor, the underlying bytes may not be the same. The latter character could be the literal U+FFFD question mark, while the first one could be some other thing that your editor doesn't recognize and thus replaces it visually with the question mark to indicate that fact.