pyobjc-dev — CVS commit notices go here

Re: [Pyobjc-dev] SIGBUS with Twisted

[Bob I. <bo...@re...>]

Ok, I've confirmed that it does indeed crash here.  Unfortunately I 
can't fix it until next week at the earliest.

-bob

On Jul 9, 2004, at 7:03 PM, burris ewell wrote:

> FWIW, this is the same behavior I'm seeing when using listenUDP...
>
> burris
>
> On Jul 7, 2004, at 12:03 AM, Bob Ippolito wrote:
>
>>
>> On Jul 7, 2004, at 2:14 AM, Jordan Krushen wrote:
>>
>>> ...this is what I get:
>>>
>>> """
>>> 2004-07-06 23:05:05.848 t[2681] Application did finish launching
>>> 2004-07-06 23:05:05.858 t[2681] Starting reactor
>>> 2004-07-06 23:05:05.891 t[2681] Reactor started
>>>
>>> t has exited due to signal 10 (SIGBUS).
>>> """
>>>
>>> It crashes about 1 second after it logs the 'Reactor started' 
>>> message.  Commenting out reactor.run() calms it down.  I figured 
>>> this would be about as simple as it gets.. am I missing something to 
>>> do with the Xcode templates?
>>>
>>> OS X 10.3.4
>>> Apple standard Python, with MacPython for Panther addons
>>> PyObjc 1.1
>>> Twisted 1.3.0 (from source tarball -- python setup.py install)
>>> Xcode 1.2
>>
>> Can I get the traceback for that?  Should be in 
>> ~/Library/Logs/CrashReporter/
>>
>> Also, try the svn trunk of Twisted.  I made some changes after the 
>> last release.
>>
>> If it's still broken, try actually doing something with the reactor.  
>> A reactor with no timers or sockets might be an edge case.
>>
>> -bob
>
>
>
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital 
> self defense, top technical experts, no vendor pitches, unmatched 
> networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Pyobjc-dev mailing list
> Pyo...@li...
> https://lists.sourceforge.net/lists/listinfo/pyobjc-dev

Re: [Pyobjc-dev] SIGBUS with Twisted

[burris e. <bu...@gi...>]

FWIW, this is the same behavior I'm seeing when using listenUDP...

burris

On Jul 7, 2004, at 12:03 AM, Bob Ippolito wrote:

>
> On Jul 7, 2004, at 2:14 AM, Jordan Krushen wrote:
>
>> ...this is what I get:
>>
>> """
>> 2004-07-06 23:05:05.848 t[2681] Application did finish launching
>> 2004-07-06 23:05:05.858 t[2681] Starting reactor
>> 2004-07-06 23:05:05.891 t[2681] Reactor started
>>
>> t has exited due to signal 10 (SIGBUS).
>> """
>>
>> It crashes about 1 second after it logs the 'Reactor started' 
>> message.  Commenting out reactor.run() calms it down.  I figured this 
>> would be about as simple as it gets.. am I missing something to do 
>> with the Xcode templates?
>>
>> OS X 10.3.4
>> Apple standard Python, with MacPython for Panther addons
>> PyObjc 1.1
>> Twisted 1.3.0 (from source tarball -- python setup.py install)
>> Xcode 1.2
>
> Can I get the traceback for that?  Should be in 
> ~/Library/Logs/CrashReporter/
>
> Also, try the svn trunk of Twisted.  I made some changes after the 
> last release.
>
> If it's still broken, try actually doing something with the reactor.  
> A reactor with no timers or sockets might be an edge case.
>
> -bob

Re: [Pyobjc-dev] Inconsistant problems with OpenGL

[Matt D. <mat...@ya...>]

Okay - will do.

My PyObjC 1.0 was installed by MacPython's
PackageManager. 

There seems to be no uninstall option, and the note on
the PyObjC 1.1. download page says:

NOTE: upgrading earlier versions of PyObjC using the
binary package doesn't seem to work correctly. Please
use the source installer when you upgrade an earlier
version of PyObjC

I am wondering how best in this case to deal with the
upgrade in order to avoid introducing any more
problems into debugging this problem.

Thanks
Matt Duignan


--- Ronald Oussoren <ron...@ma...> wrote:
> Could you try again using PyObjC 1.1. I'm pretty
> sure that several 
> issues with OpenGL support were fixed in that
> release.
> 
> Ronald



		
__________________________________
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail 

Re: [Pyobjc-dev] Update progressbar from another class

[Ray S. <ki...@sd...>]

>>

Thanks Ronald for your quick reply, below is my changes/findings:

>> One of the functions defined as:
>>
>>      def updateDownloadProgress(percentDone):
> There should be a 'self' argument in this definition.
>

Line now reads:

def updateDownloadProgress(self, percentDone):

> Add 'import objc; objc.setVerbose(1)' to the top of your script. This 
> might increase the amount of noise when something goes wrong.
>
> Ronald
>

Added the setVerbose(1) line, but got no further output in the console 
log.

To add more to history, updateDownloadProgress is being called from

h = HeadlessDisplayer()

in HeadlessDisplayer is the call to updateDownloadProgress.  As 
mentioned by Bob, I am probably going about it the wrong way.

Ray Slakinski

Re: [Pyobjc-dev] Update progressbar from another class

[Ronald O. <ron...@ma...>]

On 7-jul-04, at 21:40, Ray Slakinski wrote:

> I have a class:
>
> class NucleusInterface(NibClassBuilder.AutoBaseClass):
>     # the actual base class is NSObject
>     # The following outlets are added to the class:
>     # currentlyDownloading
>     # downloadProgress
>     # nextUpdate
>     # updatingRSS
>
> One of the functions defined as:
>
>      def updateDownloadProgress(percentDone):
There should be a 'self' argument in this definition.

> 		self.downloadProgress.setDoubleValue(percentDone)
>
>
> In another class I want to call updateDownloadProgress, or just update 
> the downloadProgress directly.  Currently with the method that I am 
> doing, my app just stops with no errors outputted to the console.  
> I've placed debug messages in updateDownloadProgress() but they are 
> not printed prior to the program exiting.

Add 'import objc; objc.setVerbose(1)' to the top of your script. This 
might increase the amount of noise when something goes wrong.

Ronald

>
> Any help would be greatly appreciated,
> Ray
>
> [----------]
> Ray Slakinski
>
> GnuPG Fingerprints:
> ki...@sd... - C8AD 4847 2DA8 3469 079D  13F9 135D F0CF 1CFC FD03
> ra...@sd... - 286D BBE9 9110 64E4 3E04  BE02 B771 A58E 7488 65D5
>
> Blog: http://ddll.sdf1.net
>
> How do you explain school to a higher intelligence?
>  		-- Elliot, "E.T."
>
>
>
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital 
> self defense, top technical experts, no vendor pitches, unmatched 
> networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Pyobjc-dev mailing list
> Pyo...@li...
> https://lists.sourceforge.net/lists/listinfo/pyobjc-dev
>
>
--
X|support bv            http://www.xsupport.nl/
T:  +31 610271479       F:  +31 204416173

[Pyobjc-dev] Update progressbar from another class

[Ray S. <ki...@sd...>]

I have a class:

class NucleusInterface(NibClassBuilder.AutoBaseClass):
     # the actual base class is NSObject
     # The following outlets are added to the class:
     # currentlyDownloading
     # downloadProgress
     # nextUpdate
     # updatingRSS

One of the functions defined as:

      def updateDownloadProgress(percentDone):
		self.downloadProgress.setDoubleValue(percentDone)


In another class I want to call updateDownloadProgress, or just update 
the downloadProgress directly.  Currently with the method that I am 
doing, my app just stops with no errors outputted to the console.  I've 
placed debug messages in updateDownloadProgress() but they are not 
printed prior to the program exiting.

Any help would be greatly appreciated,
Ray

[----------]
Ray Slakinski

GnuPG Fingerprints:
ki...@sd... - C8AD 4847 2DA8 3469 079D  13F9 135D F0CF 1CFC FD03
ra...@sd... - 286D BBE9 9110 64E4 3E04  BE02 B771 A58E 7488 65D5

Blog: http://ddll.sdf1.net

How do you explain school to a higher intelligence?
  		-- Elliot, "E.T."

Re: [Pyobjc-dev] calling outlet from different classes

[Ray S. <ki...@sd...>]

Thats what I've tried and my program quits with the following error:

AttributeError: No attribute downloadProgress

This is how I'm calling it:

myInterface.downloadProgress.doubleValue_(int(round(float(self.percentDo 
ne),0)))

Thoughts?

[----------]
Ray Slakinski

GnuPG Fingerprints:
ki...@sd... - C8AD 4847 2DA8 3469 079D  13F9 135D F0CF 1CFC FD03
ra...@sd... - 286D BBE9 9110 64E4 3E04  BE02 B771 A58E 7488 65D5

Blog: http://ddll.sdf1.net

There is a green, multi-legged creature crawling on your shoulder.


On 6-Jul-04, at 5:50 PM, Bob Ippolito wrote:

> I'm not sure if I understand you quite correctly, but outlets are just  
> attributes from Python.  Instead of  
> self.outletName.setObjectValue_(...) you would do  
> referenceToSomeInstanceOfA.outletName.setObjectValue_(...).  Of  
> course, that's not really very good encapsulation, but it works  
> anyway.
>
> If you were doing this from Objective C, it's probably best to use Key  
> Value Coding to access outlets from another instance.



[----------]
Ray Slakinski

GnuPG Fingerprints:
ki...@sd... - C8AD 4847 2DA8 3469 079D  13F9 135D F0CF 1CFC FD03
ra...@sd... - 286D BBE9 9110 64E4 3E04  BE02 B771 A58E 7488 65D5

Blog: http://ddll.sdf1.net

Good leaders being scarce, following yourself is allowed.

[Pyobjc-dev] [ pyobjc-Patches-986395 ] PyDocBrowser buildapp omissions

[SourceForge.net <no...@so...>]

Patches item #986395, was opened at 2004-07-07 16:14
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=314534&aid=986395&group_id=14534

Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Jim Tittsler (jtittsler)
Assigned to: Nobody/Anonymous (nobody)
Summary: PyDocBrowser buildapp omissions

Initial Comment:
The PyDocBrowser example appears to omit two modules from its 
list of resources.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=314534&aid=986395&group_id=14534

Re: [Pyobjc-dev] SIGBUS with Twisted

[Bob I. <bo...@re...>]

On Jul 7, 2004, at 2:14 AM, Jordan Krushen wrote:

> ...this is what I get:
>
> """
> 2004-07-06 23:05:05.848 t[2681] Application did finish launching
> 2004-07-06 23:05:05.858 t[2681] Starting reactor
> 2004-07-06 23:05:05.891 t[2681] Reactor started
>
> t has exited due to signal 10 (SIGBUS).
> """
>
> It crashes about 1 second after it logs the 'Reactor started' message. 
>  Commenting out reactor.run() calms it down.  I figured this would be 
> about as simple as it gets.. am I missing something to do with the 
> Xcode templates?
>
> OS X 10.3.4
> Apple standard Python, with MacPython for Panther addons
> PyObjc 1.1
> Twisted 1.3.0 (from source tarball -- python setup.py install)
> Xcode 1.2

Can I get the traceback for that?  Should be in 
~/Library/Logs/CrashReporter/

Also, try the svn trunk of Twisted.  I made some changes after the last 
release.

If it's still broken, try actually doing something with the reactor.  A 
reactor with no timers or sockets might be an edge case.

-bob

[Pyobjc-dev] SIGBUS with Twisted

[Jordan K. <jo...@kr...>]

Finally got around to playing more with PyObjc/Twisted, and figured I'd 
try the Cocoa-Python template in Xcode as a starting point, calling it 
't'.

If I replace the contents of tAppDelegate.py with the following..

"""
from Foundation import NSLog
from PyObjCTools import NibClassBuilder

import twisted.internet.cfreactor
reactor = twisted.internet.cfreactor.install()

NibClassBuilder.extractClasses("MainMenu")

class tAppDelegate(NibClassBuilder.AutoBaseClass):
	def applicationDidFinishLaunching_(self, aNotification):
		NSLog("Application did finish launching")
		NSLog("Starting reactor")
		reactor.run()
		NSLog("Reactor started")
"""


...this is what I get:

"""
2004-07-06 23:05:05.848 t[2681] Application did finish launching
2004-07-06 23:05:05.858 t[2681] Starting reactor
2004-07-06 23:05:05.891 t[2681] Reactor started

t has exited due to signal 10 (SIGBUS).
"""

It crashes about 1 second after it logs the 'Reactor started' message.  
Commenting out reactor.run() calms it down.  I figured this would be 
about as simple as it gets.. am I missing something to do with the 
Xcode templates?

OS X 10.3.4
Apple standard Python, with MacPython for Panther addons
PyObjc 1.1
Twisted 1.3.0 (from source tarball -- python setup.py install)
Xcode 1.2

Thanks,

J.

Re: [Pyobjc-dev] calling outlet from different classes

[Bob I. <bo...@re...>]

On Jul 6, 2004, at 5:18 PM, Ray Slakinski wrote:

> I have what I hope is a simple question, but I couldn't find an answer 
> searching the list nor the docs.
>
> I have class A and class B
>
> class A has all the specifications for the actions and outlets, where 
> class B has some processing functions among other things.  I need to 
> update an outlet (a progress bar) in class A from class B, but I can't 
> seem to wrap my head around it.
>
> Any ideas to help me out?

I'm not sure if I understand you quite correctly, but outlets are just 
attributes from Python.  Instead of 
self.outletName.setObjectValue_(...) you would do 
referenceToSomeInstanceOfA.outletName.setObjectValue_(...).  Of course, 
that's not really very good encapsulation, but it works anyway.

If you were doing this from Objective C, it's probably best to use Key 
Value Coding to access outlets from another instance.

-bob

[Pyobjc-dev] calling outlet from different classes

[Ray S. <ki...@sd...>]

I have what I hope is a simple question, but I couldn't find an answer 
searching the list nor the docs.

I have class A and class B

class A has all the specifications for the actions and outlets, where 
class B has some processing functions among other things.  I need to 
update an outlet (a progress bar) in class A from class B, but I can't 
seem to wrap my head around it.

Any ideas to help me out?

Re: [Pyobjc-dev] Inconsistant problems with OpenGL

[Ronald O. <ron...@ma...>]

On 6-jul-04, at 11:31, Matt Duignan wrote:

> Hello all,
>
> I am having some trouble with running PyObjC OpenGL on
> my G3 iBook.
>
> I have the same problem with two apps - the first is
> the OpenGLDemo Python/ObjectiveC demo application, the
> other is one I have written based on this.
>
> Both of these apps work find on my G4 PowerMac
>
> When I run the apps on the iBook I get the warning on
> the command line:

Could you try again using PyObjC 1.1. I'm pretty sure that several=20
issues with OpenGL support were fixed in that release.

Ronald

>
> ----
> ./OpenGLDemo.app/Contents/Resources/OpenGLDemo.py:43:
> RuntimeWarning: Calling method (description) on
> unitialized object 0x578b80 of class
> NSOpenGLPixelFormat
>
>   fmt =3D
> NSOpenGLPixelFormat.alloc().initWithAttributes_(attribs)
> ----
>
> And the error in a cocoa message box:
>
>
> ----
> An unexpected error has occurred
>
> (TypeError: Cannot call initWithAttributes: on
> <NSOpenGLPixelFormat: 0x578b80> from Python
>
>         Quit    Continue
>
>
> ----
>
> This is exactly the same application(s) running on the
> iBook and Powermac. The Powermac runs them without
> error.
>
> As far as I can tell I have the same versions of
> python etc. installed - from memory pyobjc (1.0) and
> pyopengl were both installed via the macpython package
> manager.
>
> The only obvious difference is that my PowerMac has
> the /Library/Documentation/PyObjC Documentation &
> Examples/ files whereas the iBook does not. I cannot
> remember how I installed these files.
>
> Any help would be very appreciated.
> Matt Duignan
>
>
> (Some system info for the two macs follows)
>
>
> -------
> G4 Powermac
>
> Hardware:
>
>     Hardware Overview:
>
>       Machine Model: Power Mac G4
>       CPU Type: PowerPC G4  (2.1)
>       Number Of CPUs: 1
>       CPU Speed: 867 MHz
>       L2 Cache (per CPU): 256 KB
>       L3 Cache (per CPU): 2 MB
>       Memory: 512 MB
>       Bus Speed: 133 MHz
>       Boot ROM Version: 4.3.3f2
>       Serial Number:
>
>
>
>
>
>
> Software:
>
>     System Software Overview:
>
>       System Version: Mac OS X 10.3.4 (7H63)
>       Kernel Version: Darwin 7.4.0
>       Boot Volume: OSX
>       Computer Name: Haumia
>       User Name: mduignan (mduignan)
>
>
>
>
> Memory:
>
>     DIMM0/J21:
>
>       Size: 256 MB
>       Type: SDRAM
>       Speed: PC100-322S
>
>     DIMM1/J22:
>
>       Size: 128 MB
>       Type: SDRAM
>       Speed: PC100-222S
>
>     DIMM2/J23:
>
>       Size: 128 MB
>       Type: SDRAM
>       Speed: PC100-322S
>
>
>
>
>
>
> PCI/AGP Cards:
>
>     GeForce2 MX:
>
>       Type: display
>       Bus: AGP
>       Display Type: CRT
>       Slot: SLOT-1
>       VRAM (Total): 32 MB
>       Vendor: nVIDIA (0x10de)
>       Device ID: 0x0110
>       Revision ID: 0x00b2
>       ROM Revision: 1055
>
>
> ----
> G3 iBook
>
> Hardware:
>
>     Hardware Overview:
>
>       Machine Model: iBook
>       CPU Type: PowerPC 750  (22.15)
>       Number Of CPUs: 1
>       CPU Speed: 500 MHz
>       L2 Cache (per CPU): 256 KB
>       Memory: 320 MB
>       Bus Speed: 67 MHz
>       Boot ROM Version: 4.2.0f4
>       Serial Number: UV12820QLAL
>       Sales Order Number: M7698X/A
>
>
>
>
>
>
> Software:
>
>     System Software Overview:
>
>       System Version: Mac OS X 10.3.4 (7H63)
>       Kernel Version: Darwin 7.4.0
>       Boot Volume: OSX
>       Computer Name: jagadmin=82=C4=F4s Computer
>       User Name: mduignan (mduignan)
>
>
>
>
> Memory:
>
>     DIMM0/BUILT-IN:
>
>       Size: 64 MB
>       Type: Built-in
>       Speed: Built-in
>
>     DIMM1/J12:
>
>       Size: 256 MB
>       Type: SDRAM
>       Speed: PC100-222S
>
>
>
>
>
>
> PCI/AGP Cards:
>
>     ATY,RageM3:
>
>       Type: display
>       Bus: AGP
>       Display Type: LCD
>       Slot: ATI
>       VRAM (Total): 8 MB
>       Vendor: ATI (0x1002)
>       Device ID: 0x4c46
>       Revision ID: 0x0002
>       ROM Revision: 113-XXXXX-125
>
>
>
>
>
> 	=09
> __________________________________
> Do you Yahoo!?
> Yahoo! Mail is new and improved - Check it out!
> http://promotions.yahoo.com/new_mail
>
>
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 -
> digital self defense, top technical experts, no vendor pitches,
> unmatched networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Pyobjc-dev mailing list
> Pyo...@li...
> https://lists.sourceforge.net/lists/listinfo/pyobjc-dev
>
>
--
X|support bv            http://www.xsupport.nl/
T:  +31 610271479       F:  +31 204416173

[Pyobjc-dev] Inconsistant problems with OpenGL

[Matt D. <mat...@ya...>]

Hello all,

I am having some trouble with running PyObjC OpenGL on
my G3 iBook.

I have the same problem with two apps - the first is
the OpenGLDemo Python/ObjectiveC demo application, the
other is one I have written based on this.

Both of these apps work find on my G4 PowerMac 

When I run the apps on the iBook I get the warning on
the command line:

----
./OpenGLDemo.app/Contents/Resources/OpenGLDemo.py:43:
RuntimeWarning: Calling method (description) on
unitialized object 0x578b80 of class
NSOpenGLPixelFormat

  fmt =
NSOpenGLPixelFormat.alloc().initWithAttributes_(attribs)
----

And the error in a cocoa message box:


----
An unexpected error has occurred

(TypeError: Cannot call initWithAttributes: on
<NSOpenGLPixelFormat: 0x578b80> from Python

        Quit    Continue


----

This is exactly the same application(s) running on the
iBook and Powermac. The Powermac runs them without
error.

As far as I can tell I have the same versions of
python etc. installed - from memory pyobjc (1.0) and
pyopengl were both installed via the macpython package
manager. 

The only obvious difference is that my PowerMac has
the /Library/Documentation/PyObjC Documentation &
Examples/ files whereas the iBook does not. I cannot
remember how I installed these files.

Any help would be very appreciated.
Matt Duignan


(Some system info for the two macs follows)


-------
G4 Powermac

Hardware:

    Hardware Overview:
    
      Machine Model: Power Mac G4
      CPU Type: PowerPC G4  (2.1)
      Number Of CPUs: 1
      CPU Speed: 867 MHz
      L2 Cache (per CPU): 256 KB
      L3 Cache (per CPU): 2 MB
      Memory: 512 MB
      Bus Speed: 133 MHz
      Boot ROM Version: 4.3.3f2
      Serial Number: 
    
    




Software:

    System Software Overview:
    
      System Version: Mac OS X 10.3.4 (7H63)
      Kernel Version: Darwin 7.4.0
      Boot Volume: OSX
      Computer Name: Haumia
      User Name: mduignan (mduignan)
    
    


Memory:

    DIMM0/J21:
    
      Size: 256 MB
      Type: SDRAM
      Speed: PC100-322S
    
    DIMM1/J22:
    
      Size: 128 MB
      Type: SDRAM
      Speed: PC100-222S
    
    DIMM2/J23:
    
      Size: 128 MB
      Type: SDRAM
      Speed: PC100-322S
    
    




PCI/AGP Cards:

    GeForce2 MX:
    
      Type: display
      Bus: AGP
      Display Type: CRT
      Slot: SLOT-1
      VRAM (Total): 32 MB
      Vendor: nVIDIA (0x10de)
      Device ID: 0x0110
      Revision ID: 0x00b2
      ROM Revision: 1055
    
    
----
G3 iBook

Hardware:

    Hardware Overview:
    
      Machine Model: iBook
      CPU Type: PowerPC 750  (22.15)
      Number Of CPUs: 1
      CPU Speed: 500 MHz
      L2 Cache (per CPU): 256 KB
      Memory: 320 MB
      Bus Speed: 67 MHz
      Boot ROM Version: 4.2.0f4
      Serial Number: UV12820QLAL
      Sales Order Number: M7698X/A
    
    




Software:

    System Software Overview:
    
      System Version: Mac OS X 10.3.4 (7H63)
      Kernel Version: Darwin 7.4.0
      Boot Volume: OSX
      Computer Name: jagadmin’s Computer
      User Name: mduignan (mduignan)
    
    


Memory:

    DIMM0/BUILT-IN:
    
      Size: 64 MB
      Type: Built-in
      Speed: Built-in
    
    DIMM1/J12:
    
      Size: 256 MB
      Type: SDRAM
      Speed: PC100-222S
    
    




PCI/AGP Cards:

    ATY,RageM3:
    
      Type: display
      Bus: AGP
      Display Type: LCD
      Slot: ATI
      VRAM (Total): 8 MB
      Vendor: ATI (0x1002)
      Device ID: 0x4c46
      Revision ID: 0x0002
      ROM Revision: 113-XXXXX-125
    
    



		
__________________________________
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail

[Pyobjc-dev] [ pyobjc-Patches-985194 ] Install.txt typos

[SourceForge.net <no...@so...>]

Patches item #985194, was opened at 2004-07-05 15:04
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=314534&aid=985194&group_id=14534

Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Jim Tittsler (jtittsler)
Assigned to: Nobody/Anonymous (nobody)
Summary: Install.txt typos

Initial Comment:
'Development Tools' should be 'Developer Tools'


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=314534&aid=985194&group_id=14534

[Pyobjc-dev] [ pyobjc-Bugs-984350 ] HelloWorld bug

[SourceForge.net <no...@so...>]

Bugs item #984350, was opened at 2004-07-02 23:34
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=114534&aid=984350&group_id=14534

Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: McErnie (mcernie)
Assigned to: Nobody/Anonymous (nobody)
Summary: HelloWorld bug

Initial Comment:
The example HelloWorld.py as supplied does not work.

Easy to fix:
on line 41, change NSApp.setDelegate_ into app.setDelegate_

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=114534&aid=984350&group_id=14534

Re: [Pyobjc-dev] HIServices framework

[Jack J. <Jac...@cw...>]

>>> bgen is the standard wrapping tool for MacPython. The framework 
>>> seems to be based on CoreFoundation.
>>
>> bgen sucks, though :)
>
> I know that :-)

And so do I:-)

Seriously, though: we've had discussion many times, and so far no-one 
has come up with a reasonable replacement for bgen. swig sucks even 
more, pyrex doesn't cut it because it's hand-written, ctypes/ffi plus 
some generated Python code might do the trick, but it's likely to 
become as ugly as bgen (and less efficient too).

>>> What would be the right way for wrapping CF types? A naive wrapping 
>>> using the template scripts would create a number of opaque types + 
>>> some global functions, just like in C. If you'd ask me I'd go for 
>>> this + some sanely named methods for the opaque types. But that 
>>> doesn't integrate nicely with the MacPython CF wrappers, that uses 
>>> CFArray.CFArrayGetElement (name made up) but that's ugly.

If someone comes up with a decent replacement for CF I'm more than 
willing to ditch the current implementation.
--
Jack Jansen, <Jac...@cw...>, http://www.cwi.nl/~jack
If I can't dance I don't want to be part of your revolution -- Emma 
Goldman

Re: [Pyobjc-dev] HIServices framework

[Ronald O. <ron...@ma...>]

On 2-jul-04, at 0:14, Bob Ippolito wrote:

> On Jul 1, 2004, at 2:15 PM, Ronald Oussoren wrote:
>
>> On 1-jul-04, at 22:01, Bob Ippolito wrote:
>>
>>> On Jul 1, 2004, at 12:06 PM, Ronald Oussoren wrote:
>>>
>>>> On 1-jul-04, at 18:48, Bob Ippolito wrote:
>>>>
>>>>> On Jul 1, 2004, at 8:19 AM, Kag Gyllstrom wrote:
>>>>>
>>>>>> I am trying to use the functionality of the
>>>>>> Assessibility functions revolving around AXUIElement.
>>>>>> I have not been able to find this in the pyobjc
>>>>>> libraries ... does it exist?  Is there some other way
>>>>>> to extract information about widgets using pyobjc?
>>>>>
>>>>> PyObjC can only automatically bridge Objective C classes.   
>>>>> AXUIElement is not an Objective C class.
>>>>>
>>>>> I'm not familiar enough with the accessibility support in OS X to  
>>>>> say whether or not there is an equivalent Objective C way to do  
>>>>> things.
>>>>
>>>> I found a Cocoa-based example at developer.apple.com, and that's  
>>>> using the C-level API. I guess that means there is no Obj-C level  
>>>> API for that framework, and hence no chance for PyObjC support :-)
>>>>
>>>> If you really want support you could file a feature-request in the  
>>>> python bugreporter. If you include an implementation that uses bgen  
>>>> this is likely to be accepted :-)
>>>
>>> Using bgen sounds like a horrible idea.  If it talks to Cocoa, it's  
>>> probably all CFTypeRef, can't we use our template stuff to wrap the  
>>> functions and constants?
>>
>> bgen is the standard wrapping tool for MacPython. The framework seems  
>> to be based on CoreFoundation.
>
> bgen sucks, though :)

I know that :-)

>  It would take 10x less time to just write an ObjC framework to wrap  
> this functionality by hand, and it would work great through PyObjC  
> (and may be useful for people using ObjC or other languages with  
> bridges like Java or Ruby).

I don't really care for users of other languages when thinking about  
making features available to python users.

>
>> I'd like to enhance our template stuff to work with CF (including  
>> creating the right inheritance tree), but that requires some time.
>>
>> BTW. CF is not really well-suited for runtime inspection. Even though  
>> NSMutableArray and NSArray are mapped to two different CF types you  
>> cannot tell those apart using the CF calls. That's a bit annoying.
>
> Are you sure CFGetTypeID won't be different for NSMutableArray and  
> NSArray?

I'm pretty sure they have the same typeid.

>
>> What would be the right way for wrapping CF types? A naive wrapping  
>> using the template scripts would create a number of opaque types +  
>> some global functions, just like in C. If you'd ask me I'd go for  
>> this + some sanely named methods for the opaque types. But that  
>> doesn't integrate nicely with the MacPython CF wrappers, that uses  
>> CFArray.CFArrayGetElement (name made up) but that's ugly.
>
> You can generally just treat CF types as though they subclass  
> NSObject.. In this case, all of the CFTypeRefs you're getting are  
> bridged to more useful ObjC classes than that.  What's the issue?

Huh? The example that I saw used CF types that were not bridged to  
public ObjC classes.

But there *does* seem to be a Cocoa equivalent of this api:  
http://developer.apple.com/documentation/Cocoa/Reference/ 
ApplicationKit/ObjC_classic/Protocols/NSAccessibility.html#//apple_ref/ 
doc/uid/20000945

Ronald
--
X|support bv            http://www.xsupport.nl/
T:  +31 610271479       F:  +31 204416173

Re: [Pyobjc-dev] HIServices framework

[Bob I. <bo...@re...>]

On Jul 1, 2004, at 2:15 PM, Ronald Oussoren wrote:

> On 1-jul-04, at 22:01, Bob Ippolito wrote:
>
>> On Jul 1, 2004, at 12:06 PM, Ronald Oussoren wrote:
>>
>>> On 1-jul-04, at 18:48, Bob Ippolito wrote:
>>>
>>>> On Jul 1, 2004, at 8:19 AM, Kag Gyllstrom wrote:
>>>>
>>>>> I am trying to use the functionality of the
>>>>> Assessibility functions revolving around AXUIElement.
>>>>> I have not been able to find this in the pyobjc
>>>>> libraries ... does it exist?  Is there some other way
>>>>> to extract information about widgets using pyobjc?
>>>>
>>>> PyObjC can only automatically bridge Objective C classes.  
>>>> AXUIElement is not an Objective C class.
>>>>
>>>> I'm not familiar enough with the accessibility support in OS X to 
>>>> say whether or not there is an equivalent Objective C way to do 
>>>> things.
>>>
>>> I found a Cocoa-based example at developer.apple.com, and that's 
>>> using the C-level API. I guess that means there is no Obj-C level 
>>> API for that framework, and hence no chance for PyObjC support :-)
>>>
>>> If you really want support you could file a feature-request in the 
>>> python bugreporter. If you include an implementation that uses bgen 
>>> this is likely to be accepted :-)
>>
>> Using bgen sounds like a horrible idea.  If it talks to Cocoa, it's 
>> probably all CFTypeRef, can't we use our template stuff to wrap the 
>> functions and constants?
>
> bgen is the standard wrapping tool for MacPython. The framework seems 
> to be based on CoreFoundation.

bgen sucks, though :)  It would take 10x less time to just write an 
ObjC framework to wrap this functionality by hand, and it would work 
great through PyObjC (and may be useful for people using ObjC or other 
languages with bridges like Java or Ruby).

> I'd like to enhance our template stuff to work with CF (including 
> creating the right inheritance tree), but that requires some time.
>
> BTW. CF is not really well-suited for runtime inspection. Even though 
> NSMutableArray and NSArray are mapped to two different CF types you 
> cannot tell those apart using the CF calls. That's a bit annoying.

Are you sure CFGetTypeID won't be different for NSMutableArray and 
NSArray?

> What would be the right way for wrapping CF types? A naive wrapping 
> using the template scripts would create a number of opaque types + 
> some global functions, just like in C. If you'd ask me I'd go for this 
> + some sanely named methods for the opaque types. But that doesn't 
> integrate nicely with the MacPython CF wrappers, that uses 
> CFArray.CFArrayGetElement (name made up) but that's ugly.

You can generally just treat CF types as though they subclass 
NSObject.. In this case, all of the CFTypeRefs you're getting are 
bridged to more useful ObjC classes than that.  What's the issue?

-bob

Re: [Pyobjc-dev] HIServices framework

[Ronald O. <ron...@ma...>]

On 1-jul-04, at 22:01, Bob Ippolito wrote:

> On Jul 1, 2004, at 12:06 PM, Ronald Oussoren wrote:
>
>> On 1-jul-04, at 18:48, Bob Ippolito wrote:
>>
>>> On Jul 1, 2004, at 8:19 AM, Kag Gyllstrom wrote:
>>>
>>>> I am trying to use the functionality of the
>>>> Assessibility functions revolving around AXUIElement.
>>>> I have not been able to find this in the pyobjc
>>>> libraries ... does it exist?  Is there some other way
>>>> to extract information about widgets using pyobjc?
>>>
>>> PyObjC can only automatically bridge Objective C classes.  
>>> AXUIElement is not an Objective C class.
>>>
>>> I'm not familiar enough with the accessibility support in OS X to 
>>> say whether or not there is an equivalent Objective C way to do 
>>> things.
>>
>> I found a Cocoa-based example at developer.apple.com, and that's 
>> using the C-level API. I guess that means there is no Obj-C level API 
>> for that framework, and hence no chance for PyObjC support :-)
>>
>> If you really want support you could file a feature-request in the 
>> python bugreporter. If you include an implementation that uses bgen 
>> this is likely to be accepted :-)
>
> Using bgen sounds like a horrible idea.  If it talks to Cocoa, it's 
> probably all CFTypeRef, can't we use our template stuff to wrap the 
> functions and constants?

bgen is the standard wrapping tool for MacPython. The framework seems 
to be based on CoreFoundation.

I'd like to enhance our template stuff to work with CF (including 
creating the right inheritance tree), but that requires some time.

BTW. CF is not really well-suited for runtime inspection. Even though 
NSMutableArray and NSArray are mapped to two different CF types you 
cannot tell those apart using the CF calls. That's a bit annoying.

What would be the right way for wrapping CF types? A naive wrapping 
using the template scripts would create a number of opaque types + some 
global functions, just like in C. If you'd ask me I'd go for this + 
some sanely named methods for the opaque types. But that doesn't 
integrate nicely with the MacPython CF wrappers, that uses 
CFArray.CFArrayGetElement (name made up) but that's ugly.

Ronald
>
> -bob
>
--
X|support bv            http://www.xsupport.nl/
T:  +31 610271479       F:  +31 204416173

Re: [Pyobjc-dev] HIServices framework

[Bob I. <bo...@re...>]

On Jul 1, 2004, at 12:06 PM, Ronald Oussoren wrote:

> On 1-jul-04, at 18:48, Bob Ippolito wrote:
>
>> On Jul 1, 2004, at 8:19 AM, Kag Gyllstrom wrote:
>>
>>> I am trying to use the functionality of the
>>> Assessibility functions revolving around AXUIElement.
>>> I have not been able to find this in the pyobjc
>>> libraries ... does it exist?  Is there some other way
>>> to extract information about widgets using pyobjc?
>>
>> PyObjC can only automatically bridge Objective C classes.  
>> AXUIElement is not an Objective C class.
>>
>> I'm not familiar enough with the accessibility support in OS X to say 
>> whether or not there is an equivalent Objective C way to do things.
>
> I found a Cocoa-based example at developer.apple.com, and that's using 
> the C-level API. I guess that means there is no Obj-C level API for 
> that framework, and hence no chance for PyObjC support :-)
>
> If you really want support you could file a feature-request in the 
> python bugreporter. If you include an implementation that uses bgen 
> this is likely to be accepted :-)

Using bgen sounds like a horrible idea.  If it talks to Cocoa, it's 
probably all CFTypeRef, can't we use our template stuff to wrap the 
functions and constants?

-bob

Re: [Pyobjc-dev] How to debug?

[Zachery B. <zb...@ur...>]

On Jul 1, 2004, at 2:25 PM, Pierre Thibault wrote:

> Le 30 juin 2004, =E0 22:57, Zachery Bir a =E9crit :
>
>> If you're used to using pdb, you can set a trace point, and then=20
>> start the application from Terminal like so:
>
> What it is pdb?

A Python debugger included with Python.

~
zbir@gorilla $ python
Python 2.3 (#1, Sep 13 2003, 00:49:11)
[GCC 3.3 20030304 (Apple Computer, Inc. build 1495)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
 >>> import pdb
 >>> help(pdb)

Help on module pdb:

NAME
     pdb - A Python debugger.
...

> Could you be more specific please?

I can try, but it might be more helpful to understand how JEdit or=20
PyOxyd handle debugging. If you're looking for a fully integrated=20
graphical debugger for PyObjC, we're not there yet, but Donovan, Bob,=20
and Bill (and others?) have both been working on parts that might make=20=

that a reality.

When I need to debug a process, I'll find a good entry point and add=20
this to the code:

import pdb;pdb.set_trace()

Then, when running the application from the Terminal, via it's actual=20
executable in AppName.app/Contents/MacOS/AppName, it will stay attached=20=

to the terminal, and when the break point happens, I'll get a pdb=20
prompt and can introspect the app. It's not terribly pretty without=20
some help (like the aforementioned XEmacs + python-mode.el), but it's=20
very effective for what I need (made more so now that Python supports=20
nesting debugging).

Zac

Re: [Pyobjc-dev] How to debug?

[Pierre T. <p.t...@bi...>]

Le 30 juin 2004, =E0 22:57, Zachery Bir a =E9crit :

> If you're used to using pdb, you can set a trace point, and then start=20=

> the application from Terminal like so:
>
>

What it is pdb?

> ~/Applications
> zbir@gorilla $ ./ZopeEditManager.app/Contents/MacOS/ZopeEditManager
>
> That will keep the app running in the current tty and let you attach=20=

> to the running instance when it gets to your trace.
>
> That's how I do it, anyway, but I'm a sucker for `import pdb;=20
> pdb.set_trace()` - I do it all day, every day, with Zope[1].
>
> Zac
>
> [1] Especially in Xemacs shell-mode with a decent python-mode.el -=20
> then you get to step through with code tracking in a separate=20
> buffer...

I don't understand. I don't use Zope or Xemacs.

Could you be more specific please?

Regards.

---------------------
Pierre=

Re: [Pyobjc-dev] HIServices framework

[Ronald O. <ron...@ma...>]

On 1-jul-04, at 18:48, Bob Ippolito wrote:

> On Jul 1, 2004, at 8:19 AM, Kag Gyllstrom wrote:
>
>> I am trying to use the functionality of the
>> Assessibility functions revolving around AXUIElement.
>> I have not been able to find this in the pyobjc
>> libraries ... does it exist?  Is there some other way
>> to extract information about widgets using pyobjc?
>
> PyObjC can only automatically bridge Objective C classes.  AXUIElement 
> is not an Objective C class.
>
> I'm not familiar enough with the accessibility support in OS X to say 
> whether or not there is an equivalent Objective C way to do things.

I found a Cocoa-based example at developer.apple.com, and that's using 
the C-level API. I guess that means there is no Obj-C level API for 
that framework, and hence no chance for PyObjC support :-)

If you really want support you could file a feature-request in the 
python bugreporter. If you include an implementation that uses bgen 
this is likely to be accepted :-)

Ronald
--
X|support bv            http://www.xsupport.nl/
T:  +31 610271479       F:  +31 204416173

Re: [Pyobjc-dev] HIServices framework

[Bob I. <bo...@re...>]

On Jul 1, 2004, at 8:19 AM, Kag Gyllstrom wrote:

> I am trying to use the functionality of the
> Assessibility functions revolving around AXUIElement.
> I have not been able to find this in the pyobjc
> libraries ... does it exist?  Is there some other way
> to extract information about widgets using pyobjc?

PyObjC can only automatically bridge Objective C classes.  AXUIElement 
is not an Objective C class.

I'm not familiar enough with the accessibility support in OS X to say 
whether or not there is an equivalent Objective C way to do things.

-bob