Bengali Text-Processing: Combining Character Sequences:
Though in Bengali, the following character sequence is a valid and an expected one but we are not able to get that out of the existing rules on combining Bengali character sequences (made up of a base letter and one or more combining marks and are rendered around the base letter (above it, below it, etc.)) 
For example:

দিক্ টায়  (there should not be any space in between "ক্" and " টা")
09A6:   দ  BENGALI LETTER DA
09BF:   ি  BENGALI VOWEL SIGN I
0995:   ক  BENGALI LETTER KA
09CD:   ্  BENGALI SIGN VIRAMA
0020:      SPACE (adding this extra space is incorrect)
099F:   ট  BENGALI LETTER TTA
09BE:   া  BENGALI VOWEL SIGN AA
09DF:   য়  BENGALI LETTER YYA
0020:      SPACE
-----------
However, if we remove the unwanted space in between "ক্" "টা" then the above bengali word becomes:
দিক্টায় (wrong one)
09A6:   দ  BENGALI LETTER DA
09BF:   ি  BENGALI VOWEL SIGN I
0995:   ক  BENGALI LETTER KA
09CD:   ্  BENGALI SIGN VIRAMA
099F:   ট  BENGALI LETTER TTA
09BE:   া  BENGALI VOWEL SIGN AA
09DF:   য়  BENGALI LETTER YYA
0020:      SPACE
---------------------
But we have no problem to get the following valid character sequence using (09CD):
বাদ্লা 
09AC:   ব  BENGALI LETTER BA
09BE:   া  BENGALI VOWEL SIGN AA
09A6:   দ  BENGALI LETTER DA
09CD:   ্  BENGALI SIGN VIRAMA (here, correctly no space is needed after
                                                       09CD to get the above sequence) 
09B2:   ল  BENGALI LETTER LA
09BE:   া  BENGALI VOWEL SIGN AA
0020:      SPACE
--------
Such Rule for Combining Bengali character sequences (using 09CD) needs to corrected.

Thanks & Regards,
Goutam Kumar Saha