pythoncard-users — PythonCard Users and Developers

Re: [Pythoncard-users] sending parameter from child window to parent?

[Phil E. <ph...@li...>]

Alec Bennett wrote:
> I'm wondering if anyone might give me a clue about how to send a 
> parameter from a child window to the parent? I have the child window 
> launched, it can set gui components in the parent (for example 
> self.parent.components.whatever.text = "hello"), and the parent can set 
> gui components in the child, but I'm unable to set variables directly.
> 
> I'm building the child/parent relationship the way David does in this 
> tutorial:
> 
> http://pythoncard.sourceforge.net/walkthrough3.html
> 
> I could use some duct tape and have a thread checking the contents of a 
> text field that I set from the other window, but that's just so ugly.
> 
> I've gone through all the samples and all the documentation I could find 
> and can't figure it out.
> 
> Anyone?
> 

The trick is in the fact that your child window is always active, but is 
(usually anyway, YMMV) not actually made visible on screen until needed, 
by calling:

self.myChildWindow.show()

You should be able to just set a class member variable in your child 
window code and have the parent window pick it up later. So in your 
child window, you'd want:

self.someVariable = ['spam', 'chips', 'beans']

Then in your main window, you can access this as 
self.myChildwindow.someVariable. As you've already discovered, it works 
the other way round, too, i.e. you could do this in your parent window:

self.otherVariable = ['spam', 'eggs', 'spam']

Then have the child window reference this as self.parent.otherVariable.

-- 

Regards

Phil Edwards
Brighton, UK

[Pythoncard-users] sending parameter from child window to parent?

[Alec B. <wry...@gm...>]

I'm wondering if anyone might give me a clue about how to send a parameter
from a child window to the parent? I have the child window launched, it can
set gui components in the parent (for example
self.parent.components.whatever.text = "hello"), and the parent can set gui
components in the child, but I'm unable to set variables directly.

I'm building the child/parent relationship the way David does in this
tutorial:

http://pythoncard.sourceforge.net/walkthrough3.html

I could use some duct tape and have a thread checking the contents of a text
field that I set from the other window, but that's just so ugly.

I've gone through all the samples and all the documentation I could find and
can't figure it out.

Anyone?

Re: [Pythoncard-users] PythonCard app not closing cleanly

[Phil E. <ph...@li...>]

Phil Edwards wrote:
> 
> break things on Linux. I've checked an earlier version of 
> standaloneBuilder and this has the same bug, so I'm reasonably sure this 
> isn't something that I've introduced into the code by my recent changes.
> 

Whatever the problem is, it doesn't appear to be something caused by my 
code. The codeEditor sample displays exactly the same behaviour. I'll 
continue to investigate.


-- 

Regards

Phil Edwards
Brighton, UK

[Pythoncard-users] PythonCard app not closing cleanly

[Phil E. <ph...@li...>]

Hi All:

Long time no post and all that, hope everyone is well! :-)

I'm having an odd problem with the latest round of enhancements for the 
standaloneBuilder tool which I'd appreciate any input on that may be 
forthcoming.

Platform details first...I'm running Mandriva 2007.0, Python 2.4.3, 
wxPython 2.6.3.3 and PythonCard 0.8.2 on my Linux machine and XP Pro, 
Python 2.4.3, wxPython 2.8.4.0 and PythonCard 0.8.2 on my Windows box.

If you change some aspect of a standaloneBuilder project and then try to 
close the program without saving the project, it throws up a 'Project 
not saved, save now?' dialog with yes, no and cancel buttons. Within the 
code, this is handled by the 'on_close' event:

     def on_close(self, event):
         if self.doExit():
             print 'on_close: calling event.Skip()'
             event.Skip()
             print 'on_close: back from event.Skip()'

the doExit() routine looks like this:

     def doExit(self):
         bull = True
         if self.documentChanged:
             save = self.saveChanges()
             if save == wx.ID_CANCEL:
                 bull = False
             elif save == wx.ID_NO:
                 bull = True
             else:
                 if self.documentPath is None:
                     bull = self.on_menuFileSaveAs_select(None)
                 else:
                     self.saveFile(self.documentPath)
                     bull = True

         print 'doExit: returning [%s]' % bull
         return bull

On my Linux box, I make a change to the project file then try to exit 
the program and select 'No' when it asks me if I want to save the 
changes. The main window then disappears but the program is still 
running. From the print statements I've put into the code, at the point 
where the on_close event fires, the event.Skip() appears to be getting 
ignored, and in my terminal window, I see both print lines, i.e

on_close: calling event.Skip()
on_close: back from event.Skip()

Under Windows, everything works as it is supposed to and the program 
exits cleanly back to a command prompt.

On both Linux and Windows, clicking the 'yes' and 'cancel' buttons 
results in the expected behaviour, it's just clicking 'no' that seems to 
break things on Linux. I've checked an earlier version of 
standaloneBuilder and this has the same bug, so I'm reasonably sure this 
isn't something that I've introduced into the code by my recent changes.

I'm planning to install a later version of wxPython on another machine 
and repeat my test there to try to rule out any bug in wx that my code 
might be tickling.

My understanding of the event.Skip() call is that it simply leaves the 
current event unchanged and allows it to pass back 'up the chain' to be 
dealt with elsewhere. It seems as though my code is somehow swallowing 
the on_close event.

Any ideas?


-- 

Regards

Phil Edwards
Brighton, UK

[Pythoncard-users] Error when hitting key while menu open....just need to upgrade?

[XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX]

Before I start digging into my code, I'd be grateful if somebody could 
just confirm that they're not seeing this happen as well...if you have a 
PythonCard application with multiple menus, and one of those menus has 
multiple menu items, could you try:

1. Opening the menu so that the items appear;
2. Hitting any key other than one that might be a shortcut to one of the 
items.

With XP + wxPython 2.8.x + Python 2.5 nothing happens....as I'd expect.
With NT + wxPython 2.5.x + Python 2.3 the application falls over due to 
illegal memory overwrite.

This is happening with all of my applications with multiple menu items.

I still need my old NT setup with Python 2.3 because I'm building 
applications that use ctypes to link to DLLs built with Visual Studio 6. 
But if the problem is restricted to old versions of wxPython then I can 
upgrade that part.

So if anybody is still using Python 2.3 in particular I would be 
interested to know if you see this crash. In the meantime, I'll start 
debugging my code to make sure I haven't done anything stupid that is 
sneaking through on XP without generating a warning.

Thanks
-- 
XXXXXXXXXXX

Re: [Pythoncard-users] Calendar compopnent

[Alex T. <al...@tw...>]

hwg wrote:
> I'm using the Calendar component to enter dates into a text field.  
> (Thanks to the example Calendar Dialog provided by Alex Tweedly on his 
> home page). 
>
> I just want to change the functionality a bit, and I'm not sure how to 
> do it.
>
> When the Calendar pops up, it is always sitting at today's date.  I 
> would like it to be sitting at whatever date was already in the text 
> box.  How do I change the "current date" on the Calendar programatically?
>
Certainly not the only way,
possibly not the best way,
but you can do something very like this .....

I made modifications to caldialog.py  I think to the same version as is 
on the web site) -  here's notes and the relevant section that I changed

1. add the import of datetime
2. added year, month and day parameters to the __init__
3. set the date as soon as the custom dialog is created  (i.e. before it 
is displayed)

and then you'll need to make changes to the calling code to extract the 
date from the text box and pass it in ...

import wx
import datetime

class CalDialog(model.CustomDialog):
    def __init__(self, parent, txt='', y=2006, m=1, d=1):
        model.CustomDialog.__init__(self, parent)
        self.components.Calendar1.PySetDate(datetime.date(y,m,d))


    def on_Calendar1_mouseDoubleClick(self, event):
        self.EndModal(wx.ID_OK)
       
    def on_btnOK_mouseClick(self, event):
        print "in ok click"
        event.Skip()
       
def calDialog(parent, y=2006, m=1, d=2):
    dlg = CalDialog(parent, y=y, m=m, d=d)
    result = dlg.showModal()



-- 

Alex Tweedly      mailto:al...@tw...      www.tweedly.net

[Pythoncard-users] Calendar compopnent

[hwg <hw...@ya...>]

I'm using the Calendar component to enter dates into a text field.  (Thanks=
 to the example Calendar Dialog provided by Alex Tweedly on his home page).=
  =0A=0AI just want to change the functionality a bit, and I'm not sure how=
 to do it.=0A=0AWhen the Calendar pops up, it is always sitting at today's =
date.  I would like it to be sitting at whatever date was already in the te=
xt box.  How do I change the "current date" on the Calendar programatically=
?=0A=0AThanks,=0Ahwg=0A=0A =0A=0A=0A=0A      ______________________________=
______________________________________________________=0ATonight's top pick=
s. What will you watch tonight? Preview the hottest shows on Yahoo! TV.=0Ah=
ttp://tv.yahoo.com/ =0A

[Pythoncard-users] possible to pass parameter to Notebook page?

[Alec B. <wry...@gm...>]

I've been experimenting with the Notebook component and love it. One problem
though is I can't figure out how to pass a parameter to one of my Notebook
pages. For example, in the below example I'm loading teste.py into two
notebook pages. Is there some way to pass a paramter to it, as if I were
typing on the commandline "teste.py whatever"? In other words, to pass the
word "whatever" to teste.py?

Here's a zip containing all the files necessary to run this test:

http://remindmyass.com/simplenotebook.zip

And here's the simple code in simplenotebook.py contained in that zip file:


#!/usr/bin/python

import os, sys
import wx
from PythonCard import model

class simpleNotebook(model.Background):

    def on_initialize(self, event):

        self.components.notebook.SetMinSize(self.components.notebook.size)

        import teste
        win1 = model.childWindow(self.components.notebook, teste.MyApp)   #
how to pass a parameter to this?
        self.components.notebook.AddPage(win1, 'teste', True)

        win2 = model.childWindow(self.components.notebook, teste.MyApp)   #
how to pass a parameter to this?
        self.components.notebook.AddPage(win2, 'teste2', True)


if __name__ == '__main__':
    app = model.Application(simpleNotebook)
    app.MainLoop()

[Pythoncard-users] Looking for a PythonCard/Twisted SSH connection.

[<Pau...@mn...>]

Anyone have such code?

I see in the example there is a twistedEchoClient example that has a 
Twisted-specific Perspective Broker included. Twisted is a bit daunting so 
tweaking it for my purposes is out of reach of my skill set at the moment.

Paul.

[Pythoncard-users] Python Card compile issue

[Andrew G. <AG...@ev...>]

Hello all

 

I am wondering if any one would be able to help me with a compile issue I am
having.

 

I am using Python Card 1.9 with Python 2.4.4.  I am using Stand Alone
Builder 0.3.0-10

 

When I build my scripts and do the install, none of the drop down boxes drop
down.  You can click them and scroll through the selections, you just don't
see the list drop down to see all the selections.

 

When I run the program in the Stand Alone Builder compiler every thing runs
fine.  I have no debugging enable during compile and have tried all the
options in there to see if that helps (Encoding, compress..)

 

I am running this on Windows XP.

 

Any help would be appreciated (remember, I am a beginner)

 

Cheers

 

Andrew Grist

Production Engineering

Master Control and Branding

Evertz Microsystems LTD

Reading UK

ph: +44 118 935 0200

cell: +44 791 937 7497

Quartz, an Evertz Brand

 

Re: [Pythoncard-users] Color tuple wxPython issue?

[XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX]

On 05/09/2007 07:15, Kevin Altis wrote:
> When the long 4 tuple is passed as an arg is where the actual  
> exception occurs under wxPython earlier than 2.7.1.1. You can try  
> this out in the shell yourself. Just run a sample like minimal.py  
> with the shell using an older wxPython...
> 
> python minimal.py -s
> 
> then type something like:
> 
>  >>> comp.field1.SetForegroundColour((0, 0, 0, 255))
> 
> to see if it throws an exception.

Yes...

File "C:\Python23\lib\site-packages\wx\_core.py", line 6982, in 
SetForegroundColour
     return _core_.Window_SetForegroundColour(*args, **kwargs)
TypeError: Expected a wxColour object or a string containing a colour 
name or '#RRGGBB'.

-- 
XXXXXXXXXXX

Re: [Pythoncard-users] Color tuple wxPython issue?

[XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX]

On 05/09/2007 07:15, Kevin Altis wrote:
> Checking wx.ALPHA_OPAQUE in the shell shows...
> 
>  >>> wx.ALPHA_OPAQUE
> 255

Yes, I'm getting 255 alpha all the time. In fact, the Windows XP color 
picker does not allow you to set transparency and when you set a color 
it comes back as an (R,G,B) tuple in the property editor. But when you 
save the file, or click on a different property and return to the color 
one it has the transparency added.

> I haven't dug into the other code in the resourceEditor or other places 
> that might make assumptions about the format of the color in the 
> framework, but I'm pretty sure there will be a number of problem spots 
> for trying to run on an older wxPython.

I'm guessing that I'm an unusual case and most developers are using a 
single machine or update all their machines to the same version. Is it 
worth trying to handle this in code, or should we just class it as a 
very-low-priority issue that awkward s*!ds like me have to work around 
manually?

> Alternatively, you could identify the 4 tuple strings in your .rsrc.py 
> files and do a global replace on them before copying/using on an older 
> version of wxPython.

Exactly what I did when I hit the issue - I only had 2 color properties 
in the resource file so I chopped the transparency element off the end.

Thanks for checking...

-- 
XXXXXXXXXXX

Re: [Pythoncard-users] Color tuple wxPython issue?

[Kevin A. <al...@se...>]

On Sep 4, 2007, at 6:14 AM, XXXXXXXXXXX wrote:

> On 10/08/2007 15:01, Kevin Altis wrote:
>> ... I have moved to Python 2.5.x and wxPython 2.8.x for my own work,
>> so that is what I'm testing against.
>
> Kevin...could you check whether wxPython 2.8.x on your machine is
> returning 4-element wxColor tuples, i.e. RGB + alpha, when you try
> changing the foreground/background colors in the resource/layout  
> editor?

Yep, this looks like a 2.7/2.8 API change.

http://www.wxpython.org/recentchanges.php

2.7.1.1

18-Oct-2006
...
wx.Colour now includes an alpha component, which defaults to  
wx.ALPHA_OPAQUE. This is in preparation for allowing various new  
alpha blening functionality using wx.Colour objects, such as drawing  
with pens and brushes on a wx.DC.

Checking wx.ALPHA_OPAQUE in the shell shows...

 >>> wx.ALPHA_OPAQUE
255

I'm guessing that the exceptions you are probably seeing can be  
traced back to the _getDefaultColor method in PythonCard/widget.py.  
Note that the check the code is doing is historical for a very old  
version of wxPython and this code could probably be safely changed.

     def _getDefaultColor( self, aColor ) :
         if aColor is None :
             return wx.NullColour
         else :
             # KEA 2001-07-27
             # is the right place for this check?
             if isinstance(aColor, tuple) and len(aColor) == 3:
                 return wx.Colour(aColor[0], aColor[1], aColor[2])
             else:
                 return aColor

If you have a resource that contains 4 element tuples (0, 0, 0, 255)  
instead of (0, 0, 0) then the code above is just going to return the  
original input and then it will end up calling the wxPython method  
SetForegroundColour with the 4 element tuple.

     def _setForegroundColor( self, aColor ) :
         aColor = self._getDefaultColor( aColor )
         self.SetForegroundColour( aColor )
         self.Refresh()   # KEA wxPython bug?

When the long 4 tuple is passed as an arg is where the actual  
exception occurs under wxPython earlier than 2.7.1.1. You can try  
this out in the shell yourself. Just run a sample like minimal.py  
with the shell using an older wxPython...

python minimal.py -s

then type something like:

 >>> comp.field1.SetForegroundColour((0, 0, 0, 255))

to see if it throws an exception.

I haven't dug into the other code in the resourceEditor or other  
places that might make assumptions about the format of the color in  
the framework, but I'm pretty sure there will be a number of problem  
spots for trying to run on an older wxPython. I'm guessing the code  
would have to specifically check for a color that wouldn't work on an  
older version of wxPython and remove the last element or do some  
generic try/except block to try using a color and then chop to a 3- 
tuple and try again.

Alternatively, you could identify the 4 tuple strings in  
your .rsrc.py files and do a global replace on them before copying/ 
using on an older version of wxPython.

ka

> Anybody else using 2.5.x or 2.6.x - could you check whether the color
> selector is just returning RGB?
>
> My situation is unusual, in that I'm mainly developing on Windows XP +
> wxPython 2.8.x alongside Windows NT + wxPython 2.5.x and yesterday I
> copied a project from the XP box onto the NT box only to find that the
> resource editor failed after importing half the components. My initial
> examination of the stack trace and code makes me think that the 2.5.x
> system cannot cope when it reads in a colour tuple that looks like
> (255,0,0,255) instead of (255,0,0).
>
> Unfortunately when I looked at the on-line wxPython docs they make  
> very
> little reference to the alpha channel, so I want to make sure there
> isn't something weird going on that is isolated to my XP machine.
>
> -- 
> XXXXXXXXXXX

[Pythoncard-users] Color tuple wxPython issue?

[XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX]

On 10/08/2007 15:01, Kevin Altis wrote:
> ... I have moved to Python 2.5.x and wxPython 2.8.x for my own work, 
 > so that is what I'm testing against.

Kevin...could you check whether wxPython 2.8.x on your machine is 
returning 4-element wxColor tuples, i.e. RGB + alpha, when you try 
changing the foreground/background colors in the resource/layout editor?

Anybody else using 2.5.x or 2.6.x - could you check whether the color 
selector is just returning RGB?

My situation is unusual, in that I'm mainly developing on Windows XP + 
wxPython 2.8.x alongside Windows NT + wxPython 2.5.x and yesterday I 
copied a project from the XP box onto the NT box only to find that the 
resource editor failed after importing half the components. My initial 
examination of the stack trace and code makes me think that the 2.5.x 
system cannot cope when it reads in a colour tuple that looks like 
(255,0,0,255) instead of (255,0,0).

Unfortunately when I looked at the on-line wxPython docs they make very 
little reference to the alpha channel, so I want to make sure there 
isn't something weird going on that is isolated to my XP machine.

-- 
XXXXXXXXXXX

Re: [Pythoncard-users] Problem with Resource Editor on Debian / Kubuntu

[Christoph S. <cs...@bo...>]

Kevin,

Thanks a lot for your detailed reply. Actually its not that big problem,
since the GUI design is not the biggest deal and i am still able to code
and test on linux, just when adjustements to the GUI are necessary I'll
connect to a nearby Terminal Server and will do the adjustements there.
Also I thought I exhausted google on this matter but I overlooked the
most obvious source of information: local docs ;-)

Cheers,
Christoph

Kevin Altis wrote:
> Sadly I don't know if there is any solution on GTK other than using a
> GTK1 build instead of GTK2. I don't currently have any Linux builds
> installed so I can't check myself, but maybe another user on the list
> can respond.
>
> The following text has been sitting at the top of the changelog.txt
> file for the last 3 years along with the link to my SF bug report and
> test program for wxPython that shows the problem.
>
> """
> SPECIAL NOTE: at least as of wxPython 2.5.2.8 you must still
> use a GTK1 build of wxWidgets/wxPython on Linux in order for
> component dragging to work in the resourceEditor. You can track
> the following bug report to be notified when this issue is fixed.
> http://sourceforge.net/tracker/?func=detail&aid=1024777&group_id=9863&atid=109863
>
> """
>
> Since nobody on the wxWidgets team seems to have dealt with the issue
> I'm assuming the low-level mouse handling for native and compound
> controls under GTK2 remains a problem, which impacts wxPython and thus
> PythonCard which sits on top of wxPython. In order for the
> resourceEditor dragging to work, the low-level mouse drag events need
> to be available.
>
> You can run the widgets sample, testevents sample or any other sample
> for that matter with the Message Watcher window (-m on the
> command-line) and it will show you every event PythonCard is seeing.
> You'll see different events depending on whether you are dragging
> (mouse is down) or moving (mouse is up) in and out of the various
> components on the different platforms (GTK1 and GTK2 included) with
> different builds of wxPython. This is the primary reason the current
> strategy used for building interfaces with the resourceEditor is
> problematic; the low-level mouse events are just not guaranteed by
> wxWidgets to work the same on all platforms. Almost every other
> interface builder for wx is sizer-based and doesn't allow the user to
> move and resize widgets except via the sizers they are contained in,
> so they avoid this problem.
>
> I've never found an adequate work-around, so I'm always up for
> suggestions. It would probably be possible to make a rock-solid
> resourceEditor just for Windows, which is the most-consistent of the
> supported wx platforms with the idea that you build on Windows and
> then are able to run on Linux and Mac, but given that we've tried to
> remain platform-neutral since the beginning of the project, that isn't
> very appealing to me.
>
> ka
>
>
> On Aug 29, 2007, at 5:03 AM, Christoph Schneeberger wrote:
>
>> Hi,
>>
>> I am new to PythonCard but already heavily fallen in love (2 weeks ago
>> with Python and 1 week ago with PythonCard - until then, pls forgive me,
>> i thought Perl::Tk was really cool *lol*). I just love it and things
>> work out of the box, including all my code i wrote yet -  a complete new
>> experience for somebody scripting since roughly 15yrs in all kind of
>> languages. So first, thanks a lot for this wonderful tool.
>>
>> But I have a small problem when running the resource editor in my
>> Kubuntu feisty Notebook: I can move certain elements with drag and drop
>> while others are not moveable at all (on my windows box, i had no
>> problems at all). I tested this also on our terminal server running
>> Debian Etch it behaves the same.
>> The elements I tested that i can move are i.e.:
>> - textarea
>> - statictext
>> - textfield
>> - tree
>>
>> Elements that I cant move around at all are i.e.:
>> - button
>> - list (can't even be selected after being deselected for the first
>> time)
>> - checkbox
>> - imagebutton
>>
>> Now I am almost sure I lack some library or doing something inredibly
>> stupid but I didn't find any information searching the archives for the
>> list and the website.
>>
>> Any tip or hints, reading instructions etc. will be highly appreciated.
>> As said, I don't think thats a bug, I think I miss something completely,
>> but I can't find it myself.
>>
>> Thanks in advance and have a nice day!
>>
>> Christoph Schneeberger
>>
>>
>>
>> -- 
>>           "Quis custodiet ipsos custodes?"

Re: [Pythoncard-users] Some questions regarding cross platform development

[John H. <ec...@ya...>]

> The second thing was that once I had that fixed that
> I noticed that
> the position of many items was off from the way it
> had been set up in
> windows.  I was actually expecting this since the
> size of GTK controls
> are different enough from windows controls to cause
> a problem.  Is
> there a commonly accepted way to handle this?
> 

I don't run Pythoncard under Linux but I am guessing
that you need to use a sizer.  For simple layouts,
it's not that hard to do.  Search for the web and you
should turn up on-line sample of how to use sizer
under PythonCard.




--
John Henry

[Pythoncard-users] Some questions regarding cross platform development

[Chris N. <chn...@gm...>]

Hello everyone, I am new to Python, PythonCard and this mailing list
so bear with me as I a few questions that will probably seem somewhat
trivial.

I am currently developing an interface that I intend to use in both
Linux and Windows and I am noticing a few 'quirks'.  I did the
original prototype for the interface in Windows and then loaded it up
in the resourceEditor in Linux to take a look at it.  A number of the
components did not show up at all, several errors had occurred with
regards to the default background color settings that the windows
resourceEditor had added to components like multi-column lists.

The resourceEditor in Linux handled this by not loading these
components them but it also removed them entirely (i.e., if I saved
the work without doing anything other than opening the file it would
remove the offending components).  I ended up fixing this by removing
all of the lines which set the background color and, just in case the
font lines as well (as Linux systems generally don't have Tahoma by
default anyways).  I am uncertain as to why, in Linux, PythonCard does
not seem to handle the background color instruction though so perhaps
someone can explain why this might be happening.

The second thing was that once I had that fixed that I noticed that
the position of many items was off from the way it had been set up in
windows.  I was actually expecting this since the size of GTK controls
are different enough from windows controls to cause a problem.  Is
there a commonly accepted way to handle this?

That's it for now, I am sure these are beginner questions but any help
would be greatly appreciated.

Thanks,
Chris

Re: [Pythoncard-users] Problem with Resource Editor on Debian / Kubuntu

[Kevin A. <al...@se...>]

Sadly I don't know if there is any solution on GTK other than using a  
GTK1 build instead of GTK2. I don't currently have any Linux builds  
installed so I can't check myself, but maybe another user on the list  
can respond.

The following text has been sitting at the top of the changelog.txt  
file for the last 3 years along with the link to my SF bug report and  
test program for wxPython that shows the problem.

"""
SPECIAL NOTE: at least as of wxPython 2.5.2.8 you must still
use a GTK1 build of wxWidgets/wxPython on Linux in order for
component dragging to work in the resourceEditor. You can track
the following bug report to be notified when this issue is fixed.
http://sourceforge.net/tracker/? 
func=detail&aid=1024777&group_id=9863&atid=109863
"""

Since nobody on the wxWidgets team seems to have dealt with the issue  
I'm assuming the low-level mouse handling for native and compound  
controls under GTK2 remains a problem, which impacts wxPython and  
thus PythonCard which sits on top of wxPython. In order for the  
resourceEditor dragging to work, the low-level mouse drag events need  
to be available.

You can run the widgets sample, testevents sample or any other sample  
for that matter with the Message Watcher window (-m on the command- 
line) and it will show you every event PythonCard is seeing. You'll  
see different events depending on whether you are dragging (mouse is  
down) or moving (mouse is up) in and out of the various components on  
the different platforms (GTK1 and GTK2 included) with different  
builds of wxPython. This is the primary reason the current strategy  
used for building interfaces with the resourceEditor is problematic;  
the low-level mouse events are just not guaranteed by wxWidgets to  
work the same on all platforms. Almost every other interface builder  
for wx is sizer-based and doesn't allow the user to move and resize  
widgets except via the sizers they are contained in, so they avoid  
this problem.

I've never found an adequate work-around, so I'm always up for  
suggestions. It would probably be possible to make a rock-solid  
resourceEditor just for Windows, which is the most-consistent of the  
supported wx platforms with the idea that you build on Windows and  
then are able to run on Linux and Mac, but given that we've tried to  
remain platform-neutral since the beginning of the project, that  
isn't very appealing to me.

ka


On Aug 29, 2007, at 5:03 AM, Christoph Schneeberger wrote:

> Hi,
>
> I am new to PythonCard but already heavily fallen in love (2 weeks ago
> with Python and 1 week ago with PythonCard - until then, pls  
> forgive me,
> i thought Perl::Tk was really cool *lol*). I just love it and things
> work out of the box, including all my code i wrote yet -  a  
> complete new
> experience for somebody scripting since roughly 15yrs in all kind of
> languages. So first, thanks a lot for this wonderful tool.
>
> But I have a small problem when running the resource editor in my
> Kubuntu feisty Notebook: I can move certain elements with drag and  
> drop
> while others are not moveable at all (on my windows box, i had no
> problems at all). I tested this also on our terminal server running
> Debian Etch it behaves the same.
> The elements I tested that i can move are i.e.:
> - textarea
> - statictext
> - textfield
> - tree
>
> Elements that I cant move around at all are i.e.:
> - button
> - list (can't even be selected after being deselected for the first  
> time)
> - checkbox
> - imagebutton
>
> Now I am almost sure I lack some library or doing something inredibly
> stupid but I didn't find any information searching the archives for  
> the
> list and the website.
>
> Any tip or hints, reading instructions etc. will be highly  
> appreciated.
> As said, I don't think thats a bug, I think I miss something  
> completely,
> but I can't find it myself.
>
> Thanks in advance and have a nice day!
>
> Christoph Schneeberger
>
>
>
> -- 
>
>           "Quis custodiet ipsos custodes?"

[Pythoncard-users] Problem with Resource Editor on Debian / Kubuntu

[Christoph S. <cs...@bo...>]

Hi,

I am new to PythonCard but already heavily fallen in love (2 weeks ago
with Python and 1 week ago with PythonCard - until then, pls forgive me,
i thought Perl::Tk was really cool *lol*). I just love it and things
work out of the box, including all my code i wrote yet -  a complete new
experience for somebody scripting since roughly 15yrs in all kind of
languages. So first, thanks a lot for this wonderful tool.

But I have a small problem when running the resource editor in my
Kubuntu feisty Notebook: I can move certain elements with drag and drop
while others are not moveable at all (on my windows box, i had no
problems at all). I tested this also on our terminal server running
Debian Etch it behaves the same.
The elements I tested that i can move are i.e.:
- textarea
- statictext
- textfield
- tree

Elements that I cant move around at all are i.e.:
- button
- list (can't even be selected after being deselected for the first time)
- checkbox
- imagebutton

Now I am almost sure I lack some library or doing something inredibly
stupid but I didn't find any information searching the archives for the
list and the website.

Any tip or hints, reading instructions etc. will be highly appreciated.
As said, I don't think thats a bug, I think I miss something completely,
but I can't find it myself.

Thanks in advance and have a nice day!

Christoph Schneeberger



-- 

          "Quis custodiet ipsos custodes?"

Re: [Pythoncard-users] codeEditor installation problem

[Kevin A. <al...@se...>]

On Aug 27, 2007, at 12:26 PM, Root, Richard wrote:

> Kevin,
>
> Seems like we fix one problem and immediately another arises!  In =20
> the exercises in the book, I have to switch between Python Shell =20
> and codeEditor, cutting, pasting etc. When I dbl click on =20
> codeEditor.py, and run the script, the codeEditor code is =20
> apparently all that the Python shell can handle, because I can no =20
> longer use it for the exercises. In the book, the only info =20
> provided on running codeEditor is on page 3:
>
>
>
> =93Once it is installed on your system with PythonCard, on linux or =20=

> unix based systems, you can just type codeEditor in a terminal =20
> shell window and it will start.=94
>
>
>
> I have tried that, and its not so. So how can I have codeEditor =20
> running, and a =93live=94 Python Shell window also? Am I making any =20=

> sense here?
>
> Thanks,
>
> Rich

The codeEditor has a built-in shell. There are a number of different =20
ways that you could startup the codeEditor, but if they aren't giving =20=

you specific instructions about how they want you to run it with =20
specific examples here are some things to try. Since you're on a Mac =20
you can run it from the Terminal by typing:

python codeEditor.py

This is assuming that the default python is 2.5.x. You can double-=20
check this by just typing:

python

at the Terminal prompt. You should see something like this:

[machine:PythonCard/tools/codeEditor] bob% which python
/Library/Frameworks/Python.framework/Versions/Current/bin/python

[machine:PythonCard/tools/codeEditor] bob% python
Python 2.5.1 (r251:54869, Apr 18 2007, 22:08:04)
[GCC 4.0.1 (Apple Computer, Inc. build 5367)] on darwin
Type "help", "copyright", "credits" or "license" for more information.


Type Ctrl-D to quit the Python shell. So, assuming you have 2.5.x you =20=

can just use "python" instead of pythonw to run GUI PythonCard =20
programs. Note the machine name, user name and directories will be =20
different on your setup. You can also provide a filename directly for =20=

codeEditor to open up like this:

python ~/python/PythonCard/tools/codeEditor/codeEditor.py somefile.py

What I'm doing above is changing to a directory for a project I'm =20
working on and then providing a path to the codeEditor application so =20=

Python can find it and a command-line filename argument for the file =20
I want to open, which in the example above is called somefile.py.

If you have configured your Python Launcher correctly on the Mac, =20
which is located in your Applications->MacPython 2.5 directory then =20
you can also just double-click a Python file to run it. See the =20
instructions at:

http://pythoncard.sourceforge.net/macosx_tiger_installation.html

Basically, you want your interpreter set to /usr/local/bin/python

Back in the Terminal you can see that this is a link to the 2.5.x Python

ls -la /usr/local/bin/python
lrwxr-xr-x   1 root  wheel  68 Jul 23 13:35 /usr/local/bin/python -=20
 > ../../../Library/Frameworks/Python.framework/Versions/2.5/bin/python

I know it seems confusing, but all these links are the *nix way :)

Anyway, what you'll want to do is make an alias of the PythonCard/=20
tools/codeEditor/codeEditor.py file a keep that on your desktop. Then =20=

you can just double-click it to open it whenever you want. It keeps a =20=

history of the last files you've been editing, so it should be easy =20
then to get back to what you were working on.

Now finally, as far as the shell goes there is a Shell menu in the =20
codeEditor that you can toggle via the F5 key to hide/show it. This =20
is a full Python shell, so you can type in code there to try stuff =20
out. Note that the shell can be somewhat confusing for copying and =20
pasting on the Mac because it uses control instead of command (Ctrl-=20
C, Ctrl-V) for copy and paste and the codeEditor menu doesn't =20
actually deal correctly with shell selections so you need to rely on =20
the control keys. That is a bit of a Mac-specific bug and window =20
context, that hopefully one of these days we can workaround. Check =20
out the following page for more on the codeEditor shell:

http://pythoncard.sourceforge.net/shell.html

Note that you generally wouldn't run complete programs within the =20
shell, instead you would be editing them with the codeEditor and then =20=

typing Command-R (run) or Command-Shift-R (run with interpreter). =20
Assuming you run the codeEditor from the Finder using Python Launcher =20=

you'll probably want to keep the Console, which is in Applications-=20
 >Utilities->Console open to see additional error messages that =20
otherwise would show up in the Terminal.

I hope that answers your questions.

ka=

Re: [Pythoncard-users] codeEditor installation problem

[Kevin A. <al...@se...>]

On Aug 27, 2007, at 9:31 AM, Root, Richard wrote:

> I am having some problem with getting codeEditor to run. I have =20
> followed Kevin=92s great installation guide at =20
> PythonCard.sourceforge, and the =93acid test=94 worked fine. I can =
load =20
> and run the other samples besides minimal, just fine. However, in =20
> the components folder, nothing seems to work, not ever codeEditor. =20
> I do get a restart, and then the following shows in the shell window:
>
>
> Last login: Sun Aug 26 17:47:22 on ttyp1
>
> Welcome to Darwin!
>
> Richard-Roots-Computer:~ richardroot$ cd '/Library/Frameworks/=20
> Python.framework/Versions/2.5/lib/python2.5/site-packages/=20
> PythonCard/components/' && '/usr/local/bin/pythonw'  '/Library/=20
> Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-=20
> packages/PythonCard/components/codeeditor.py'  && echo Exit status: =20=

> $? && exit 1
>
> Exit status: 0
>
> logout
>
> [Process completed]

The codeEditor tool is located at PythonCard/tools/codeEditor/=20
codeEditor.py not the components directory. The modules in the =20
components directory are just that, components used by the PythonCard =20=

framework, samples, and tools, they are not applications. If the =20
instructions you were following refer to running from the components =20
directory, please send me a URL so I can get that fixed.

If you are unable to run PythonCard/tools/codeEditor/codeEditor.py =20
please send a follow-up and we'll investigate further.

Thanks,

ka
> If it matters, I am running a G4 laptop, with OS 10.4.8. I have =20
> installed the packages on another Mac and had a little trouble at =20
> first with a file that could not be found, moved it to the =20
> PythonCard folder, and was able to get codeEditor to run. But now  =20
> I find that nothing I do will launch it. What do I need to know to =20
> get codeEditor to launch?
>
> Thanks,
>
>
> Richard T. Root, M.Sc.
>
> Principal Scientist/ Manager
>
> Advisory Editor, IVD Technology
>
> Antibody Research and Development Lab
>
> Roche Centralized & Molecular Diagnostics,
>
> Roche Diagnostics Corp.
>
> 9115 Hague Road,
>
> Indianapolis, IN 46250
>
> office: 317.521.7483
>
> cell: 317.727.1510
>
> Ric...@Ro...

[Pythoncard-users] codeEditor installation problem

[Root, R. <ric...@ro...>]

I am having some problem with getting codeEditor to run. I have followed
Kevin's great installation guide at PythonCard.sourceforge, and the
"acid test" worked fine. I can load and run the other samples besides
minimal, just fine. However, in the components folder, nothing seems to
work, not ever codeEditor. I do get a restart, and then the following
shows in the shell window:

Last login: Sun Aug 26 17:47:22 on ttyp1
Welcome to Darwin!
Richard-Roots-Computer:~ richardroot$ cd
'/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-pa
ckages/PythonCard/components/' && '/usr/local/bin/pythonw'
'/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-pa
ckages/PythonCard/components/codeeditor.py'  && echo Exit status: $? &&
exit 1
Exit status: 0
logout
[Process completed]

If it matters, I am running a G4 laptop, with OS 10.4.8. I have
installed the packages on another Mac and had a little trouble at first
with a file that could not be found, moved it to the PythonCard folder,
and was able to get codeEditor to run. But now  I find that nothing I do
will launch it. What do I need to know to get codeEditor to launch?
Thanks,=20

Richard T. Root, M.Sc.=20
Principal Scientist/ Manager=20
Advisory Editor, IVD Technology
Antibody Research and Development Lab
Roche Centralized & Molecular Diagnostics,
Roche Diagnostics Corp.=20
9115 Hague Road,
Indianapolis, IN 46250
office: 317.521.7483
cell: 317.727.1510=20
Ric...@Ro...

Re: [Pythoncard-users] The Future of Python Card

[John H. <ec...@ya...>]

--- "Smith, Frank" <F....@te...> wrote:

> I'm also a "serious beginner" and have been lurking
> on the list for
> awhile now. PythonCard is a great tool as it stands.
> Despite its
> apparent "limitations" it is suitable for most
> applications. 

Agree totally.  I can't say enough thanks to the
developers of Pythoncard.  I don't know how "complex"
and "big" a project has to be before one have to
resort to wxpython but recently I coded some 10,000
lines of purely GUI code (some 50 to 60 screens) in
about 2 week's time (see my previous postings about
creating Pythoncard controls on the fly).  May be it's
not the sexiest looking code in the universe - but it
surely made my customer (and my accountant) very
happy.  THANKS - Pythoncard.


>I would be
> interested having some sort of plotting capability
> supported - perhaps I
> might even be able to help, although I'm not sure
> how much work or skill
> it would require! Despite Kevin's note, I'm not sure
> how to go about it.
> 

For the kind of 2-D plotting I do, nothing beats
matplotlib.  There was some rumblings about
integrating matplotlib with pythoncard (there is
direct support of wxpython already) but nothing came
out of it.



--
John Henry

Re: [Pythoncard-users] The Future of Python Card

[Smith, F. <F....@te...>]

I'm also a "serious beginner" and have been lurking on the list for
awhile now. PythonCard is a great tool as it stands. Despite its
apparent "limitations" it is suitable for most applications. I would be
interested having some sort of plotting capability supported - perhaps I
might even be able to help, although I'm not sure how much work or skill
it would require! Despite Kevin's note, I'm not sure how to go about it.

Frank

-----Original Message-----
From: pyt...@li...
[mailto:pyt...@li...] On Behalf Of
Lance Haig
Sent: Friday, August 10, 2007 9:41 AM
To: pyt...@li...
Subject: Re: [Pythoncard-users] The Future of Python Card


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I have been on this list for a while now lurking in the shadows.

This question has been asked a few time on the list in the last two
years, the answer as I remember it was  (so if it is wrong don't shoot
me)

That Python card has loads of features and that anyone who is interested
in either fixing bugs or enhancing the tool is free to do so.

Perhaps we should set-up a list of things we think it will need and then
perhaps one of the list members who is proficient enough will grab one
of these items and add it to the tool.

What do you guys think?

Lance

I am a serious beginner on this so I would not be of any help at all.


phil jones wrote:
> Yeah, I'm interested in this question too.
>=20
> Although I'd like to ask a slight variation. What's the "problem" (if=20
> there is one) with PythonCard development? Not enough people involved?

> Have the original people lost interest and moved on to something else?
>=20
> Is it really to do with a fall off in interest in building desktop GUI

> based applications.
>=20
> How many people here are also looking at using the browser as their=20
> GUI these days?
>=20
> phil
>=20
>=20
>=20
> On 8/9/07, John Henry <ec...@ya...> wrote:
>> --- Tony Cappellini <cap...@gm...> wrote:
>>
>>> What is the future of Python Card?
>>>
>>> wx has many widgets that are not part of Python
>>> Card.
>>>
>>> Will Python Card be modified to include more
>>> widgets?
>>>
>> That would be nice.  I wish I know enough to help.  I
>> have become very productive with Pythoncard.
>>
>> --
>> John Henry
>>
>> ---------------------------------------------------------------------
>> ----
>> This SF.net email is sponsored by: Splunk Inc.
>> Still grepping through log files to find problems?  Stop.
>> Now Search log events and configuration files using AJAX and a
browser.
>> Download your FREE copy of Splunk now >>  http://get.splunk.com/
>> _______________________________________________
>> Pythoncard-users mailing list
>> Pyt...@li...
>> https://lists.sourceforge.net/lists/listinfo/pythoncard-users
>>
>=20
> ----------------------------------------------------------------------
> ---
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a
browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> Pythoncard-users mailing list
> Pyt...@li...
> https://lists.sourceforge.net/lists/listinfo/pythoncard-users
>=20
> --
> This message was scanned by HaigMail and is believed to be clean.=20
> Click here to report this message as spam.=20
> =
http://mail03.redarmour.co.uk/cgi-bin/learn-msg.cgi?id=3DE061D29B56.E72D
> 8
>=20
>=20
>=20

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org

iD8DBQFGvGrhOw09RVRgt9wRAtSJAKCBoGogUD9yPHv9++Lvyp5qzhpfRACgtHAs
hmS0DUYUDEpuFOW1iifAnWI=3D
=3DY832
-----END PGP SIGNATURE-----

------------------------------------------------------------------------
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Pythoncard-users mailing list Pyt...@li...
https://lists.sourceforge.net/lists/listinfo/pythoncard-users

Re: [Pythoncard-users] The Future of Python Card

[Lance H. <lh...@ha...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I have been on this list for a while now lurking in the shadows.

This question has been asked a few time on the list in the last two
years, the answer as I remember it was  (so if it is wrong don't shoot me)

That Python card has loads of features and that anyone who is interested
in either fixing bugs or enhancing the tool is free to do so.

Perhaps we should set-up a list of things we think it will need and then
perhaps one of the list members who is proficient enough will grab one
of these items and add it to the tool.

What do you guys think?

Lance

I am a serious beginner on this so I would not be of any help at all.


phil jones wrote:
> Yeah, I'm interested in this question too.
> 
> Although I'd like to ask a slight variation. What's the "problem" (if
> there is one) with PythonCard development? Not enough people involved?
> Have the original people lost interest and moved on to something else?
> 
> Is it really to do with a fall off in interest in building desktop GUI
> based applications.
> 
> How many people here are also looking at using the browser as their
> GUI these days?
> 
> phil
> 
> 
> 
> On 8/9/07, John Henry <ec...@ya...> wrote:
>> --- Tony Cappellini <cap...@gm...> wrote:
>>
>>> What is the future of Python Card?
>>>
>>> wx has many widgets that are not part of Python
>>> Card.
>>>
>>> Will Python Card be modified to include more
>>> widgets?
>>>
>> That would be nice.  I wish I know enough to help.  I
>> have become very productive with Pythoncard.
>>
>> --
>> John Henry
>>
>> -------------------------------------------------------------------------
>> This SF.net email is sponsored by: Splunk Inc.
>> Still grepping through log files to find problems?  Stop.
>> Now Search log events and configuration files using AJAX and a browser.
>> Download your FREE copy of Splunk now >>  http://get.splunk.com/
>> _______________________________________________
>> Pythoncard-users mailing list
>> Pyt...@li...
>> https://lists.sourceforge.net/lists/listinfo/pythoncard-users
>>
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> Pythoncard-users mailing list
> Pyt...@li...
> https://lists.sourceforge.net/lists/listinfo/pythoncard-users
> 
> --
> This message was scanned by HaigMail and is believed to be clean.
> Click here to report this message as spam. 
> http://mail03.redarmour.co.uk/cgi-bin/learn-msg.cgi?id=E061D29B56.E72D8
> 
> 
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org

iD8DBQFGvGrhOw09RVRgt9wRAtSJAKCBoGogUD9yPHv9++Lvyp5qzhpfRACgtHAs
hmS0DUYUDEpuFOW1iifAnWI=
=Y832
-----END PGP SIGNATURE-----