Re: Patch to add Sinhala (Sri Lanka) collation to MySQL

[Harshula]

Hi Alexander,

On Tue, 2007-02-13 at 19:55 +0400, Alexander Barkov wrote:

>
> http://cvs.savannah.nongnu.org/viewcvs/*checkout*/sinhala/sinhala/patches/mysql-output-utf8_sinhala_ci.txt
> 
> Shouldn't the pair of characters "0D9A 0DCA" (N84 in the above chart)
> be sorted before single character "0D9A" (N21 in the above chart)?

No.

> 2. According to the University of Colombo School of Computing,
> there is a number of different sorting orders for Sinhala:
> 
> http://www.ucsc.cmb.ac.lk/ltrl/public/collationDocs.html
> 
> - Alphabetical Order at the Election Commissioner's Office.pdf
> - Alphabetical Order at National Library.pdf
> - Order of Alphabet and System of Transliteration followed by the
> University at University Libraries.pdf
> - Sinhala_Collation_&_Encoding_2[1].0.pdf
> 
> 
> What is the difference between them, and which sorting order
> did you follow when implementing the collation?

I think those are the documents collected for the literature survey
prior to the collation sequence being standardised by the Sri Lankan
Standards Institute (SLSI) as part of SLS1134. I suspect none of the
above documents resemble the final collation sequence.

SLS1134 Part 1 (Collation Sequence) completed its public comment phase
in 2006 and the final document should be finalised. I'll find out about
its status next week and see if I can get hold of an official copy.

cya,
#