Re: UnicodeData.txt is invalid, flawed, broken, corrupt and wrong

From: Theodore H. Smith (
Date: Sat Jun 11 2005 - 15:26:29 CDT

  • Next message: fantasai: "Re: Arabic letters separated by markup"

    On 11 Jun 2005, at 21:09, Aki Inoue wrote:

    > Theodore,
    > According to the Unicode code chart KELVIN SIGN U212A has
    > canonically decomposition mapping to Latin K U004B so the Unicode
    > database is correct.

    Doesn't that imply that K composes to U212A? If not, then where do we
    collect the composition data from?

    > Note, in Normalization Form C processing, you don't map single
    > character canonical mappings such as KELVIN SIGN or ANGSTROM SIGN.


    I probably missed that part. Now I must scan through these long
    normalisation documents again to find that sentance I missed.

    This archive was generated by hypermail 2.1.5 : Sat Jun 11 2005 - 15:27:51 CDT