EDB is built to compile Delphi code into Windows applications.

I am not a Delphi-hater, and I really don't want this thread to go in that direction (

I feel there has been a gradual withering of Delphi, whilst it is still in development and has actually improved a lot in recent years I see the ramping up of the price of licences and the ramping down of development as negatives that may ultimately kill the product.

With 100,000+ lines of code and the need to be on a "live" platform, I am always a bit curious about other options. I looked long & hard at Visual Studio, before deciding I didn't want to be that closely locked into Windows or C#.

Now RemObjects seem to have gone quite a distance developing a replacement development tool-chain for Object Pascal developers.

Elements seems to be the IDE, and Hydra seems to be a tool to allow Delphi Object Pascal source files to be integrated into Elements projects.

Elements incorporates a whole load of modernizations compared to Delphi, and works within the Visual Studio environment.

* Is anyone on these groups using Elements?
* Any experiences of working with it, or the companion tools?
* Is it possible / easy to access and work with EDB databases from within Elements?
* Any sense of whether RemObjects are likely to do a better job of developing their product
  than Idera over the next few years?

* TIM (if you're reading) might we get Elements EDB components at any point in the future?

Adam,

<< * TIM (if you're reading) might we get Elements EDB components at any point in the future? >>

What you need to use EDB with any .NET code is the ElevateDB .NET Data Provider, which is included in the ElevateDB DAC products.  You can also use the ElevateDB ODBC Driver, but that isn't a "built-in" solution.

Tim Young
Elevate Software
www.elevatesoft.com

Adam Brett wrote:

> Elements seems to be the IDE, and Hydra seems to be a tool to allow
> Delphi Object Pascal source files to be integrated into Elements
> projects.

Not sure that is quite right, but the idea is pretty much right. Hydra
allows mixing of Delphi and .Net code.

I think the key is that if you have Delphi code, and want to take
advantage of .Net, the RemObjects systems are a good option. RO seem to
be doing good stuff, and they keep moving it forward in interesting
ways. But you are essentially moving from one island to another island,
though this one has a bridge to, err, a country where you can migrate
to easily in the future. Okay, analogy failure, but once your code is
working in .Net, you can mix C# easily, and/or go to mobile etc with
their other compilers. It is a fair way to move if that's where you
want to go with existing source.

(Personally, I've started using C# in VisualStudio, and it is working
fine, but I started programming in C++ before I went to Delphi 1.)


--

Matthew Jones

Thanks Matthew

>>you are essentially moving from one island to another island,
>>though this one has a bridge to, err, a country where you can migrate
>>to easily in the future.

Yes, I think I got the "island to island" analogy. Essentially it is a switch of IDE from Idera Delphi to RemObjects "Elements", where Elements is actually housed in Visual Studio.

The advantage is that Visual Studio is unlikely to cease to exist, and that if ObjectPascal really dies a transition to C# is possible.

I guess that I wanted a clearer picture of the situation in terms of to what extent code is really portable, i.e. have RO recreated the VCL, or would I have to re-code masses of stuff.

I mostly have source code, but I use a fair number of 3rd Party components ... TMS, DevEx etc., I guess I am also curious about how that might work, or not.

I am curious to ask here, as I think the users on this forum are relatively unbiased. If I try to ask these questions on other forums I might get a lot of trolling.

Adam Brett wrote:

> have RO recreated the VCL, or would I have to re-code masses of stuff.

My understanding is no, you are on your own with the UI. Or rather you
now have to adopt a .Net UI framework, and can take advantage of things
like the DevExpress components for that sort of thing. But it isn't an
"open project, recompile, and done" at all. That's where Hydra would
help by allowing modules to be written in .Net and used in the existing
Delphi applications. I have no idea how sensible that is though.

You are right though that it is a better island to be on, with more
prospects. Let us know how you get on!

--

Matthew Jones

>> Let us know how you get on!

Not sure I am ready to jump just yet. I actually felt pretty positive about Delphi during the Embarcadero excursion, the product was improving and I actually quite liked the fact that it wasn't closely linked to Microsoft.

I don't need Delphi to be bleeding edge, if it adopts useful language features slowly (i.e. generics) that's fine, provided that the product continues to develop and there are enough tools around to make for a really good IDE.

However, the RO interpretation of ObjectPascal "Oyxgene", seems interesting & powerful, and obviously being able to transition to other languages gives a feeling of safety.

If I had all the time in the world I would download Visual Studio + Elements & have a play ... unfortunately writing this post has used up most of my free time for the day! I have to get back to actually fixing stuff (using Delphi) more or less now!

Adam Brett wrote:

> If I had all the time in the world I would

Likewise. Which is why I took a contract at a lower rate to get into C#
and the .Net world. I have a friend who has been using C# since the
start and he has been able to apply the same language to all areas,
web, mobile, services. Delphi is still good to me, but I think I have
to take the plunge and move on. Mind, I was saying that all the time he
was using it, and it took me this long to really get moving. And Delphi
is still a big thing for me - I just won't start anything new in it...

--

Matthew Jones

>>it took me this long to really get moving. And Delphi
>>is still a big thing for me - I just won't start anything new in it...

What was the journey like Matthew, which parts were hard / easy & were there useful places you found information to help the transition?

Also, it sounds as though you were doing a series of projects rather than managing 1 v. large code-base, as I am, making it easier for you to switch. Do you think it would be possible to transition a large code-base, or would it just be too much work?

Adam Brett wrote:

> > > it took me this long to really get moving. And Delphi
> > > is still a big thing for me - I just won't start anything new in
> > > it...
>
> What was the journey like Matthew, which parts were hard / easy &
> were there useful places you found information to help the transition?
>
> Also, it sounds as though you were doing a series of projects rather
> than managing 1 v. large code-base, as I am, making it easier for you
> to switch. Do you think it would be possible to transition a large
> code-base, or would it just be too much work?

Hard to generalise from me I suspect, but one thing that I found was
that there seemed to be a lot of "magic" involved in C#/.Net. I'd think
"how do I do something" and google it, and there would be various
variations some old and some new, some workable for me and some not. I
spent a day trying to do some thing and in the end found 5 lines of
code that did it perfectly but I had no clue how. I don't like that!
However, I can recommend the PluralSight video courses - they seem to
be the leader in development stuff, and the course on LINQ in
particular goes through the evolution of it and that explained a lot of
the magic. There is a lot to take in that is not Delphi, and of course
that affects productivity. In Delphi, if I needed to do something, I'd
probably done it before and can grab that code. Not so with C#, and
there are new ways to do things.

As to whether a project is worth converting, that depends on the
project. The one I am doing is an old Delphi application that has been
maintained by a number of programmers and really needed some polish.
The customers want more though, and Mac and web and whatever beckon.
Hence it made sense to start over - it isn't the worlds most difficult
application, but it has some nuances that mean it has been good to be
able to take known good Delphi calculation code and use it in EWB, and
convert it to C# too. I reckon this first project is probably double
the usual time, but I now feel a lot more confident about C# than when
I'd helped out on another existing project and had little clue how
things fit together.

But this is also part of the modern way of doing things - making things
modular with REST or similar APIs, so you can start with a load of
modules written one way, and then implement them with something else
later.

--

Matthew Jones

Hello Adam,
I see that the Dexit movement is gaining strength. Dexit is to Delphi as Brexit
is to Britain. Welcome aboard.

You mentioned RemObjects. There are other alternatives.

In the last two (2) weeks some developments have taken place:-

1) launch of the Free Pascal and Lazarus Foundation.
http://foundation.freepascal.org/

2) launch of NewPascal.
www.newpascal.org
They state "NewPascal exists to help both FPC/Lazarus and mORMot Open Source projects. This distribution is similar to what Ubuntu is vs Debian. Less platforms supported, but with latest features on main Linux/Windows targets.".

The NewPascal roadmap can be seen at the Synopse forum under 'NewPascal Distribution'.

Only a few months back TMS Software launched a new component set. This is called FNC.
See their 'TMS FNC UI Pack' which supports 3 frameworks and 5+ operating systems. That includes
the LCL (Lazarus Component Library).

You may be interested to read in the http://www.freepascal.org/ third party forum about
people moving away from RemObjects.

Regards,
 Peter Evans