Support for UNICODE characters is available in Tally 9. This post deals with some issues relating to UNICODE.
1) When we programmatically import / export data from Tally 9, the default character-set used by Tally Software is ASCII.
2) However, in case if NO Company is opened in Tally 9 and we programmatically send a SQL-Request to Tally (like to check if any company is open or not), the RESPONSE received from Tally is always in UNICODE. This may cause error as generally the Tally Integration programs written now-a-days do not provide support for handling UNICODE characters. This prima-facia appears to be a BUG in Tally 9, but has not been confirmed by Tally Solutions, Bangalore as of now.
BUG
Tally 9 returns data in UNICODE even though the character-set is set as ASCII (i.e. UNICODE="No"). This occurs when we send a SQL-Request to Tally when NO company is open in Tally 9. It does not occur in any other situations tested by us till now.