Dear members， Thank Wang Wei. Here are two excels : zh_in_cdnc_repo.xlsx , 19537 Hanzi which includes all cdnc Hanzi(19520) and HKIRC submit 15 Hanzi (58DC、52A4、52C5、53DA、57D7、5BD7、5FDF、617D、6407、64E1、64E5、661E、6900、6AC8、6C39) , (58B5, 3A5C NOT in MSR). zh_in_reduced_repo.xlsx, 13237 Hanzi. This version is not complete. Not all Hanzi in MSR and Hanzi need to be revised for generating variant groups. Best Regards. Qi chao 发件人： 王伟 发送时间： 2015年4月14日(星期二) 下午10:07 收件人： ChineseGP 主题： [ChineseGP] Thursday meeting on 16th April Dear All I'd like to hold CGP meeting 3PM~4PM Thursday With the help from Mr. QI Chao, we have generated two different repertoire, the original CDNC table (18000+ code points), and a reduced smaller one (13000+ code points) However, compared with JPRS table, the TWO overlapping sets are quite similar. We really need to discuss pros and cons! of two different table, and which one should be adopted. Mr. QI, please send out the two EXCEL files of code points to everyone. Thank you! Regards Wang Wei