[asaibx]

As others have commented, love the ambitiousness of this! However, Unicode searches do not seem to work at all -- not just "中文" but also even "français" gives an error. Unicode support is something you definitely want to build in from the very beginning in order to avoid headaches (for you and users) in the future. Even if there is no content in the index, the presence of non-ASCII characters in the search term should not lead to a server error. Suggest you make Unicode the default encoding for everything even if you are not planning on supporting non-English search results for the moment, just to avoid unexpected errors when people search for things like "café" for example.

[misterman0]

The database has Unicode support but apparently my normalization and tokenization does not. I will have a look at that ASAP. Thanks!