Vinay Jain <vinayj(at)sarathi(dot)ncst(dot)ernet(dot)in> writes:
> Which takes a lot of time the reason is that in every indchar_lt
> function call SPI connection is made and destroyed...
I cannot imagine how you'd think that that would be practical from a
performance standpoint.
> Here is the problem which i found in postgresql...
> after setting collating sequence in proper way(i.e. C) the order was
> on basis of unicode values...but in Hindi Language some of combined
> unicode values makes a single character
> similarly length is not appropriate for these reasons & hence
> substring operations
Why don't you just use a Hindi locale? The infrastructure should all be
there already for you.
regards, tom lane