Valentina, Unicode, RB

[jda]

>I see on RB list many talks that RB strings on default are Unicode.
>So what ???
>
>In C++ on e.g. Windows, strings also are Unicode,
>But developer simply need convert Unicode to MacRoman string and then store
>in db that last string.
>
>Does all this talks means that RB do not provide easy way to convert strings
>between encoding ?
>
>Point here is that even Valentina will get support of Unicode strings, this
>is not fact that Western users may prefer it. Deal is that I think Unicode
>strings will be slower on indexing, search and sorting than MacRoman.
>I don't know exactly, but I assume this is true.
>I don't know yet how much slower, but if too much then you still may prefer
>for big dbs (with 100K - million records) to convert strings on READ/WRITE
>and get faster searches and sorts.
>
>Just a point.
>

Yes, you can convert to MacRoman rather easily in RB. But then of 
course you lose any characters specific for a given encoding! That 
means that, among other things, Greek characters are turned to mush. 
So do Japanese (for example), of course. My users hate that.

Just give us the choice, Ruslan. Search/sort speed for me is not an 
issue --Valentina is already more than fast enough for me. I'd 
*gladly* give up some speed to gain the benefits and flexibility of 
Unicode!

Jon