I have been struggling with this problem for a few years. I have
written a basic relational database design tool:
http://www.hardgeus.com/pgdesigner/
I have completely rewritten this program 3 times. I initially
implemented it in Fltk, but ran into a bunch of limitations in the API.
I rewrote it in WxGTK and was pretty happy with the results -- until I
ported to Windows. My program was so abysmally broken in Windows that I
threw the code away and vowed never again to use WxWidgets. I recently
ported the entire codebase over to QT4 and have been VERY happy with it.
Unfortunately, my love for QT4 has put my project into a bit of a
limbo...Most versions of Linux don't "play nice" with multiple versions
of the same API on the machine, and most Linux distros use QT3 since
that's the underlying API of KDE. I already get a ton of mail from
people trying to compile the old versions of PGDesigner, I'd hate to
think what would happen if I tried to release my QT4 version now...My
build setup is pretty nuts, to say the least:
http://www.hardgeus.com/index.php?ndailyupdateid=685
It sucks, because I use pgDesigner almost every day to visualize my
Postgres databases, but QT4 just isn't "there" enough for me to support it.
Anyway, you didn't ask for a novel...I would recommend QT3 for your
project. I am not entirely certain how much I am using is QT4 specific,
but I have been very happy with the signal/slot architecture, clean
database handling, and very robust variant-like variable handling (i.e.
I don't have to have giant bloated type-checking when copying data out
of my database into a local variable).
Ritesh Nadhani wrote:
Hey
Sorry for the critical mistake. Pressed the SEND button too early. In
the first para I meant:
The reason I *want* to develop the project in wxWindows or a C/C++
based toolkit is that in the end I would be able to compile a binary
which will have least dependency and can be bundled for downloaded in
a single binary. With my experience while developing and selling
SQLyog, I came across many customers who were working on a slow dial
up connection for whom downloaded a 10MB package was also a pain. I
have had customers who just had plain vanilla Win98 machines and
SQLyog used to run great on it.
instead of:
The reason I *don't want* to develop the project in wxWindows or a C/C++
based toolkit is that in the end I would be able to compile a binary
which will have least dependency and can be bundled for downloaded in
a single binary. With my experience while developing and selling
SQLyog, I came across many customers who were working on a slow dial
up connection for whom downloaded a 10MB package was also a pain. I
have had customers who just had plain vanilla Win98 machines and
SQLyog used to run great on it.
Ritesh
On 11/28/06, Ritesh Nadhani <rn.mailing@xxxxxxxxx> wrote:
Hello All
Sorry for the late reply. Been a little busy with my assignments.
I will try to answer all the queries in this mail.
The reason I don't want to develop the project in wxWindows or a C/C++
based toolkit is that in the end I would be able to compile a binary
which will have least dependency and can be bundled for downloaded in
a single binary. With my experience while developing and selling
SQLyog, I came across many customers who were working on a slow dial
up connection for whom downloaded a 10MB package was also a pain. I
have had customers who just had plain vanilla Win98 machines and
SQLyog used to run great on it.
Another reason why suggested wxWidgets is because I have worked with
it before and I am comfortable with the system. As somebody suggested,
even wxPython looks good as Python greatly increases the speed of
implementation.
Also, IDEs like Delphi etc. are out of question as I cant afford to
buy the licenses.
I have no experience with XUL but it looks good. I am not sure, how
easy is to design GUI with lots of forms etc, with XUL. Writing an
initial prototype in XUL will make things more clear.
Even though I am not related to Webyog (developer of SQLyog) anymore,
I had started a project up there called SQLyog Max (which didn't work
due to time constraints rather then technological constraints) and we
even released one BETA release that had support for both MySQL and
PostgreSQL.
I believe developing an actual prototype would help our cause more
then just deciding upon which tool kit to use. My semester gets over
on 15th December. I plan to sit with it after that. Right now I am
thinking of an architecture which can support something like this.
If you have any idea how something like should be designed or
architectured, please provide me with your invaluable suggestions.
In other related question, my primary desktop of usage is Mac OS X but
I will be buying a Ubuntu box soon so expect decent development
parallely in Mac and Linux. I am not sure about Windows but if we
write correct wxWidgets/wxPython code, it should be a 0 issue to get
it compiled in Windows.
Ritesh
On 11/28/06, Rich Shepard <rshepard@xxxxxxxxxxxxxxx> wrote:
> On Tue, 28 Nov 2006, Merlin Moncure wrote:
>
> > For a general purpose language, lately I've been taking a really
good look
> > at 'D', which looks to be an amazing language. Has anybody tried
to hook
> > up postgresql to D?
>
> No, I haven't. But, if you want a cross-platform language and
GUI toolkit,
> consider Python and wxPython.
>
> --
> Richard B. Shepard, Ph.D. | The Environmental
Permitting
> Applied Ecosystem Services, Inc.(TM) | Accelerator
> <http://www.appl-ecosys.com> Voice: 503-667-4517 Fax:
503-667-8863
>
> ---------------------------(end of
broadcast)---------------------------
> TIP 3: Have you checked our extensive FAQ?
>
> http://www.postgresql.org/docs/faq
>
---------------------------(end of broadcast)---------------------------
TIP 3: Have you checked our extensive FAQ?
http://www.postgresql.org/docs/faq
