Ed Leafe wrote:
On Jun 5, 2006, at 9:24 PM, Alex Tweedly wrote:
I know I'm not going to write my UIs in code - but I also know I'm
not going to give up the strengths of the various tools I've
developed expertise in without an almighty struggle.
I'm glad that Norman Winn posted his message; he summed up
exactly what I was going to say to you. Just as you're reluctant to
give up your tools, there are many, many more folks out there who
are reluctant to give up their GUI development tools. I'll guarantee
that if you take a Visual Studio developer and show him how great it
is to develop GUIs in Python, he'll be shocked that you're using text
files to create GUIs. He'll then go and tell all his MSFT buddies
that Python sucks.
That's really not the direction I'm coming from (far less trying to head
towards). I've been a Pythoncard user for the last couple of years, as
well as an occasional Pythoncard developer. I'm hoping Dabo will some
day be as easy to use as PythonCard, and trying to help make that
happen. (Actually, I'm *sure* it will some day be as easy to use - I'm
hoping to help make that happen sooner).
Today, Dabo (really Dabo.ui) can be approached in two quite distinct
ways - via Class Designer and as a wx wrapper.
CD will, some day, be a complete IDE - but as you said a few days ago,
today it is at a very early stage, and it would be hard to develop a
large app using it.
Using dabo.ui simply as a wx wrapper, it would be perfectly feasible to
develop an arbitrarily complex app - but you'd have to give up all the
simplicity and pleasure of using a GUI editor to create your UI controls
and layout.
I was hoping to find an "in-between" solution that allowed the use if CD
to keep the ability to use a GUI layout editor without having to wait
until the IDE aspect was more complete. Or to put it another way -
aiming to get 80% of the simplicity with only 20% of CD.
I know this sounds grandiose, but our eventual goal is to be the
Visual Studio of the Python world. We want to make it as easy to
create GUI applications in Python as it is in any proprietary
language. We chose wxPython as our initial UI toolkit because it
looked the best on every platform out there. Its API sucks, though,
at least from a Python POV: it's too much like C++, with getters/
setters and those god-awful
wx.UPPER_CASE_CONSTANTS_WITH_LOTS_OF_UNDERSCORES. That's why we
created our own API: not because wxPython was too hard for us; hell,
wrapping an API is much more difficult than using it! ;-) But
because we want to make it easy for everyone to use Dabo's UI. E.g.,
if I want to set the upper position of a control, it makes sense to
use a syntax like control.Top = 33; the wxPython method is pretty
ugly. Now imagine you want to bottom-align a series of controls in
wxPython! In Dabo, you just set their Bottom property to the same
value, and you're done!
That's a great long term goal, I'm very happy you have it as a goal. I
was only trying to find an intermediate goal that would get as many of
the advantages as possible but sooner (and hopefully without diverting
far off the path you need to follow anyway). It turns out that the
initial idea of an intermediate goal was too far off to the side - so
let's forget that one. I have a couple of other ideas for potential
"intermediate goals" - I'll explore them, and (if necessary) come back
and ask more questions.
But I'm not trying to distract the authors from their goal, or change
the (long term) target - merely identify a suitable subset of the long
term goal that would allow development to proceed sooner.
I think it would be difficult to write (for instance) some of the
wxPython or PythonCard sample (demo) programs using CD today. I know I
struggled with a couple of them (while a couple of others were a
breeze). Without CD they were all easy-ish - but I don't want to write
code to create and layout the UI.
I still use BBEdit to write Dabo; I much prefer that editor over
Dabo's. That means that at some point, more work will have to be put
into Dabo's editor to add the kinds of things that make life simple
in BBEdit. I'm hoping that people who love other editors will be
similarly inspired to contribute features that will make Dabo's
editor work better for them.
--
Alex Tweedly http://www.tweedly.net
--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.8.2/356 - Release Date: 05/06/2006
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-users
