Previously callers of the binary classifier were confusing the length field to refer to the bytes in the buffer instead of characters to decode. This corrects that. It also changes the API to be aware of the file extent of the buffer to avoid scanning data that might not be initialized.
Previously callers of the binary classifier were confusing the length field to refer to the bytes in the buffer instead of characters to decode. This corrects that. It also changes the API to be aware of the file extent of the buffer to avoid scanning data that might not be initialized.