There is an issue regarding to the searching in ANSI mode under Multi-Byte
locales (for example, Chinese or Korean). MB locales allow to show Unicode characters
so that 2 bytes can form the single Multi-Byte character.
But this is not taken into account by ATStreamSearch component, which
always treats 1 byte as 1 single ANSI character. This leads to the incorrect
search results comparing to the standard Windows controls (which support MB locales correctly).
This issue can be seen in the ATStreamSearch demo. You can open ANSI file
containing some Multi-Byte characters, and it will be displayed correctly in the upper ATBinHex control.
But the search results for MB characters will be highlighted incorrectly.