I have an ongoing problem (2 years or thereabouts) with Excel 2010 incorrectly sorting certain Unicode characters.
I use Excel to manage language-data (text) in Greek and Coptic Unicode. One of the essential functions I need is to be able to correctly sort the words in alphabetic order.
For Greek, sorting works correctly. For Coptic, which has 30 letters in the alphabet, the last five letters of the alphabet are always sorted at the beginning of the list.
My suspicion is that the problem is related to the Unicode “code points”. The hex values of the Coptic letters, followed by the letter name and the correct position in the alphabet, of these last 5 letters are as follows:
03E2 (Shei) #25
03E5 (Fei) #26
03E9 (Hori) #27
03EB (Gangia) #28
03ED (Shima) #29
03EF (Dei) #30
The letters that precede this group are in a different code-point range (the first 4 of which are listed here):
2C81 (Alfa) #1
2C83 (Vida) #2
2C86 (Gamma) #3
2C88 (Dalda etc.) #4 through # 24.
When sorting these glyphs the correct sort order SHOULD be:
2C81 (Alfa) #1
2C83 (Vida) #2
2C86 (Gamma) #3
2C88 (Dalda etc.) #4 through # 24.
03E2 (Shei) #25
03E5 (Fei) #26
03E9 (Hori) #27
03EB (Gangia) #28
03ED (Shima) #29
03EF (Dei) #30
Instead it is:
#25
#26
#27
#28
#29
#30
#1
#2
#3
#4 etc.
Can anyone confirm that Excel 2010 sorts Unicode based on these code points?
If so, do you have any idea how to work around this problem?
I might mention that when I use the spreadsheet included in Open Office (which I passionately dislike) that the sort works correctly. Additionally, I have colleagues who report that sorting works correctly in Excel 2003, an even greater mystery.
Thank you for your help.
Bookmarks