bug#36070: 27; feature request '(Describe Char Unidata List) to include 'kDefinition' value

[Van L]

> On 4 Jun 2019, at 01:06, Eli Zaretskii <eliz@gnu.org> wrote:
> 
>> --8<---------------cut here---------------start------------->8---
>> Character code properties: customize what to show
>>  name: CJK IDEOGRAPH-5165
>>  general-category: Lo (Letter, Other)
>>  decomposition: (20837) ('入')
>> --8<---------------cut here---------------end--------------->8---
> 
> This comes from UnicodeData.txt, our source for the Unicode properties
> of all the characters.  We parse it into uni-*.el files as part of the
> build.
> 
>> The Readings table, in particular, is nice to have for the 'kDefinition'.
>> 
>> --8<---------------cut here---------------start------------->8---
>> | Data type   | Value                    |
>> |-------------+--------------------------|
>> | kDefinition | enter, come in(to), join |
>> |             |                          |
>> --8<---------------cut here---------------end--------------->8---
> 
> This comes from Unihan_Reading.txt, a different file that is part of
> the Unihan database.
> 
> We don't currently have a property where to put this value, so we need
> first to extend the properties.  And then we will need to parse the
> above file and populate the property.  Patches welcome.  Bonus points
> for reviewing other properties of the Unihan DB and adding whatever is
> useful.  See UAX#38 (http://www.unicode.org/reports/tr38/), for the
> description of the properties.

Thanks for pointing this out. I definitely want to know more about the Unihan 
DB and extend the handling of this information.

-- Van