I'm finding the downloads all quite complicated to work out which version
I want. I have a list of "Products", 5 of which are the 2007 CodeGear
product. How can I determine which I need? What's the difference in the
Unicode version? Given that they are all pretty much the same size, and
I'll need the Delphi 7 one too, is there not an installer which will do
any that I tell it to at run-time, having determined what I have on my PC?

/Matthew Jones/

I've downloaded a few, so I can get the manual, but could you adjust the
manual generator to not put each topic on a new page please? The majority
of the manual is little topics on a page of its own. Instead of 475 pages
to print, it could easily fit in about half that. Thanks!

/Matthew Jones/

Matthew,

<< I'm finding the downloads all quite complicated to work out which version
I want. I have a list of "Products", 5 of which are the 2007 CodeGear
product. How can I determine which I need? What's the difference in the
Unicode version? >>

Select which product/personality you want to use EDB with, and then
determine whether you want to use Unicode strings or not with the engine.
It's really that simple.

<< Given that they are all pretty much the same size, and I'll need the
Delphi 7 one too, is there not an installer which will do any that I tell it
to at run-time, having determined what I have on my PC? >>

It can't.  Unicode is simply a compiler option that determines whether EDB
uses WideStrings or AnsiStrings.

--
Tim Young
Elevate Software
www.elevatesoft.com

> then
> determine whether you want to use Unicode strings or not with the
> engine. It's really that simple.

This is going to sound silly, but how can I tell what I want? What are the
implications. I found a change note that said ElevateDB now supports
Unicode strings, but I don't know what that really means. If I use one, is
the database incompatible with a build without? Why would I choose one and
not the other? Is there a web page which will help me choose the right
option?

/Matthew Jones/

Matthew,

<< Why would I choose one and not the other? Is there a web page which will help me
choose the right option? >>

It depends what you are targetting. If your applications need to display more than
one alphabet side by side (as with our multilingual dictionary), then it has to be Unicode.
And Unicode is about as future-proof as you can get. But unless you are using .NET,
your Delphi visual controls won't yet handle Unicode and must be replaced by those
neat TMS (previously Tnt) Unicode components:  http://tmssoftware.com/
or by components from another third-party supplier of Unicode controls for Delphi.

On the other hand, if all your customers are in North America and Western Europe,
you can continue just using the default codepage Win1252 (West European Latin).
And in that case, you don't need the Unicode version of ElevateDB. But in that case,
forget about displaying Czech or Polish as well as French and German. Risky?

We have had big problems trying to support Russian as a dictionary language while
sticking with codepage Win1251 (Cyrillic). And adding Chinese and Japanese just
forced us down the Unicode route. It's just a shame that Borland dragged their feet for
so long with Unicode. At least CodeGear are now starting to address the problem!

So we are busy migrating from DBISAM to ElevateDB Unicode. And as we are sticking
with the Win32 version of Delphi 2007, we had to tackle the problem of Unicode visual
controls.
As an interim solution we found that replacing all the original controls (TForm, TEdit,
TDBEdit,
TMemo, TRadioGroup, TCheckbox etc etc) with their Tnt Unicode equivalents was a simple
search and replace. *That* part was not a big deal.  

HTH,
Dan

Lexicon Software Ltd, UK

Matthew


1. D2008 will be unicode, I'm not sure if you'll be able to turn it off
2. Do you need to write apps that show captions etc using unicode (eg Chinese, not sure which other languages, but there's lots more of them)

So if the answer to 2 is Yes or you intend to move to D2008 the answer is get unicode. If like me you only want to develop in English and you're going to avoid D2008 and later then go for the other one.

Roy Lambert

Dan


I doubt I'll ever need to use different alphabets, but how do you handle the variable length of a caption when using different ones?

Roy Lambert

Many thanks for that. We do have a lot of Chinese and similar users of one
package, and DBISAM seems to cope.

Does the database file format change at all? If we decided to go with the
Unicode version, would we notice any issues with things like
"field.AsString", or would it be automatic. And if we chose to go
ANSIString for now, would we have problems upgrading to Unicode later (in
v2 perhaps?)?

/Matthew Jones/

As you'll have seen, we have very international customers, so I'm looking
forward to using D2008 asap! Maybe that is indeed the answer, so long as
it isn't actually a problem for "DBISAM-like" work.

/Matthew Jones/

Roy,

<< how do you handle the variable length of a caption when using different ones? >>

By guesswork... Our screen layouts just leave plenty of white space for field captions,
so with luck one layout will work for all languages. As we allow "on-the-fly" change
of the interface language, it is fairly easy to test in our case.

Dan