mbstring 是否规范化 utf-8 字符串？
Does mbstring normalize utf-8 strings?
我注意到 mb_convert_encoding always returns valid utf-8, which is great for cleaning up user input, but I wonder whether it will return a normalized form of utf-8 or whether I would have to do that manually after with the \Normalizer class?
阅读source，mb_convert_encoding似乎没有正常化。它似乎在编码之间进行转换，然后替换非法字符，仅此而已。
我注意到 mb_convert_encoding always returns valid utf-8, which is great for cleaning up user input, but I wonder whether it will return a normalized form of utf-8 or whether I would have to do that manually after with the \Normalizer class?
阅读source，mb_convert_encoding似乎没有正常化。它似乎在编码之间进行转换，然后替换非法字符，仅此而已。