Print

Print


Hi Jonathan,

On 4/29/11 1:04 PM, Jonathan wrote:
> Not only does the first proposal make it hard sort the entries but it 
> will also be harder to parse because the symbol with sometimes be 1 
> character long, sometimes 2 and sometimes 3.  So extra checking has do 
> be done to verify if the next character is the start of a new symbol 
> or a fill and or rotation modifier.

I agree; however, removing these characters from the initial Unicode 
proposal may be required to get approval from the Unicode committees.

I understand the potential annoyances and problems for programmers who 
do not use these codepoints.

My previous email was unclear on the new codepoint assignments.  They 
should be as follows.

The first proposal will be as follows:
Fill-1 [removed from code chart]
Fill-2 U+1DA9B
Fill-3 U+1DA9C
Fill-4 U+1DA9D
Fill-5 U+1DA9E
Fill-6 U+1DA9F

Rotation-1 [removed from code chart]
Rotation-2  U+IDAA1
Rotation-3  U+IDAA2
Rotation-4  U+IDAA3
Rotation-5 U+IDAA4
Rotation-6 U+IDAA5
...

A second proposal would include the 2 remaining control characters to 
complete the set.
Fill-1 U+1DA9A
Rotation-1 U+1DAA0

With the first proposal, we may be able to move forward with wide spread 
consensus.

I will be working off of the second proposal, so I will continue to use 
the Fill-1 and Rotation-1 characters in my data and code libraries.  The 
only change for my data will be to advance the fill and rotation 
codepoints by 14.

Regards,
-Steve