Re: [pygtk] ANN: PyGUI 2.1.1

[John Finlay]

I'm disappointed that I appear to be the only one on this list that 
finds these announcements objectionable. And contrary to speculation I 
find them all objectionable (yes that means yours as well Roberto). So 
far it's been lucky that not all of the hundreds of projects based on 
pygtk have felt it necessary to send announcements to the list and I 
thank them for their restraint. I would suggest that an announcement 
only maillist be setup for those who wish to send and receive these 
announcements but I fear that many would continue to spam this list anyway.

John

Greg Ewing wrote:
> PyGUI 2.1.1 is available:
>
>http://www.cosc.canterbury.ac.nz/greg.ewing/python_gui/
>
> This is an emergency bugfix release to repair some major
> breakage in the gtk version. Also corrects some other
> problems.
>
>
> What is PyGUI?
> --
>
> PyGUI is a cross-platform GUI toolkit designed to be lightweight
> and have a highly Pythonic API.
>
>   

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] ANN: PyGUI 2.1

[John Finlay]

Pietro Battiston wrote:
> Il giorno mar, 17/11/2009 alle 11.19 +1300, Greg Ewing ha scritto:
>   
>> Pietro Battiston wrote:
>>
>> 
>>> And I promise I won't come there just saying that PyGUI API sucks
>>>   
>> Okay, that particular remark was a bit rude, and I
>> apologise for it.
>>
>> Let me rephrase: One of the reasons I created PyGUI
>> is that none of the existing cross-platform GUI APIs,
>> including pygtk, are entirely to my taste. So making
>> PyGUI emulate pygtk wouldn't make sense from my
>> point of view.
>>
>> I certainly don't mean to dissuade anyone from
>> using pygtk. If you like it, by all means continue
>> to use it.
>>
>> 
>
>
> OK, that's more diplomatic.
>
> My opinion is that if you create a mailing list for PyGUI (and I suggest
> Mailman [0], the software running this mailing list, for it), a message
> "if anybody is interested in a different approach to pygtk, a PyGUI
> mailing list has been created at this address" on this mailing list will
> cause no complaints (notice your first email didn't mention pygtk at
> all).
>
>   
I have to disagree. I consider attempts to build a competing community 
by spamming maillists with announcements to be annoying and rude. The 
criterion should be does this post add value to the community. I believe 
that Greg's posts do not add value to the pygtk community and he should 
refrain from posting his announcement to the pygtk list.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] ANN: PyGUI 2.1

[John Finlay]

Greg Ewing wrote:
> John Finlay wrote:
>   
>> Greg,
>>
>> Why do you post to mailing lists that are unrelated to your project? I 
>> would appreciate it if in future you didn't post a message about your 
>> project ot the PyGTK mailing list.
>> 
>
> I posted the announcement to the pyobjc, pygtk and pywin32
> lists because PyGUI uses all of those libraries, and because
> I don't know of any single mailing list where people interested
> in Python GUIs in general can be found.
>
> However, if the consensus is that PyGUI announcements are
> not welcome on those lists, I will be happy to cease posting
> them there.
>
> What is the general feeling out there? Should I stop posting
> PyGUI messages to these lists? Is there another GUI-related
> list that would be more appropriate?
>
>   
Start your own list for the community that is interested in your project.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] ANN: PyGUI 2.1

[John Finlay]

Greg Ewing wrote:
> Luis A. Bastiao Silva wrote:
>
>   
>> But it just works for application wroted from stratch, otherwise 
>> application wroted in pygtk need to be ported to work cross-platform.
>> 
>
> It's already possible to run pygtk programs on all three
> platforms, if you're willing to install the required
> libraries. API compatibility with pygtk (or any other
> gui library) is not one of PyGUI's goals.
>
>   
>> What are you wrapping in Win32? Windows API directly?
>> 
>
> Currently it uses the MFC layer of pywin32. I hope
> eventually to replace that with raw win32 via ctypes.
>
>   
>> Anyway seems a good project, although it will be better have a pygtk 
>> cross platform :)
>> 
>
> That's a matter of opinion. One of the major goals of
> PyGUI is to provide an API that doesn't suck. The
> pygtk API doesn't meet that requirement, IMO.
>   
Greg,

Why do you post to mailing lists that are unrelated to your project? I 
would appreciate it if in future you didn't post a message about your 
project ot the PyGTK mailing list. I think you are rude to post your 
off-topic messages to the PyGTK mailing list mo to mention the other 
lists. Please don't post your project adverts here in future.

Thanks

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Saving and reading images from database

[John Finlay]

Timo List wrote:
> My program uses the SQLite database. The database holds some 
> information about persons and I want to add a picture for each person. 
> Not the path to the file, but the actual image so that the image also 
> works if the file is deleted on the harddisk.
>
> I can save the image in a BLOB column as:
> imgfile = open('/path/to/image.png')
> db.save_image(imgfile.read())
>
> Works fine as far as I see. (Maybe any comments on this?)
>
> Retrieving also works, but I see no possibility to show this image in 
> a gtk.Image widget. I can't seem to find anything related to this on 
> the net, maybe someone reading this has done this before?
One possibility is to save the image as a compressed in-line image in 
the db and create a pixbuf from the in-line data after it's retrieved 
and decompressed. Here's some code to create an in-line image and 
compress it:

import struct, bz2
...
# pb is a pixbuf containing the image to be in-lined

pix = pb.get_pixels()
inline_img = 
'GdkP'+struct.pack('!iBBHiii',len(pix)+24,1,1,pb.get_has_alpha()+1,
  
pb.get_rowstride(),pb.get_width(),pb.get_height())+pix
bz2_inline_img = bz2.compress(inline_img)

# save bz2_inline_ob in the db

To use the compressed in-line image use something like:

# the compressed in-line image is in bz2_inline_img
inline_img = bz2.decompress(bz2_inline_img)
pb = gtk.gdk.pixbuf_new_from_inline(-1, inline_img, False)

Copy the pixels if you are going to delete the inline_img string.

Then you can use the pixbuf to create your image.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] How to add 'clicked' event to treeview cell?

[John Finlay]

Taras wrote:
> Hello, all!
>
>
> There is a table with some rows. For example, it has 3 columns. 
> The first one is icon column 'Add/Remove to favorites' with star icon.
> I read about pixbuf renderer but I can't find how can I connect
> 'clicked' event to this column :( I also research how it made in Exaile
> (music player written in Python/GTK). 
>
> There is workaround with:
> ...
>   self.list.connect('button-release-event', self.update_rating)
> ...
> and
> ...
>  (x, y) = e.get_coords()
>#check if the click is within rating column and on a list entry
>  if self.list.get_path_at_pos(int(x), int(y)) \
>  and left_edge < x < left_edge + rating_col_width:
> ...
> But it looks like unbeautiful workaround. Is there more beautiful
> solution to connect 'clicked' event to icon cell?
>
>   
I don't think there is a built-in way to connect to a clicked event on a 
TreeView cell.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] pyGtk segfault comparing gtk.Image with 0

[John Finlay]

Pietro Battiston wrote:
> Il giorno sab, 17/10/2009 alle 20.49 +0300, Paul Pogonyshev ha scritto:
>   
>> Pietro Battiston wrote:
>> 
>>> Il giorno sab, 17/10/2009 alle 17.30 +0100, Tomeu Vizoso ha scritto:
>>>   
 On Sat, Oct 17, 2009 at 17:12, René 'Necoro' Neumann  
 wrote:
 
> Well -- you are not initializing gtk.Image. So it's your very own
> mistake. I don't see a pygtk issue here.
>   
 This is Python, any crash is a bug.
 
>>> Wait, wait.
>>>
>>> Read http://bugzilla.gnome.org/show_bug.cgi?id=561130 , then come back
>>> here and let's talk :-)
>>>
>>> Pietro
>>>
>>> (P.S: as you can desume from the page, I _would like_ to agree with you)
>>>   
>> You can still rightfully agree with him that it _is_ a bug.  Whether
>> it gets fixed is another point.
>> 
>
> What I mean is: where is it written that every crash is a bug, if Pygtk
> developers themselves are not so clear on that (and I'm possibly too
> ignorant)?
>
>   
this question has come around a couple of times before and I recall that 
each time it's deemed too impractical to catch every possible error to 
prevent a segfault.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Styling a text view across the visible line

[John Finlay]

Scott Ferguson wrote:
> Hi,
>
> I'm trying to build an integrated diff tool for an app that I'm 
> writing, and I am styling the text to make certain parts of the diff 
> more visible (adds are green, removes are red, etc).  Typical look and 
> feel of a diff viewer.  
>
> The styling includes a background and foreground color.  Currently I'm 
> iterating over each line of the gtk.TextBuffer, creating an ending 
> iterator at the next line, and styling everything between the line 
> start, and the start of the following line.  My issue is that I would 
> like the background color to carry straight across the gtk.TextView, 
> not end where the line ends (on the '\n' character).  A single line 
> should have a highlight from left to right.
>
> Is this possible with a stock gtk.TextView?  My current formatting 
> function can be found here: http://gist.github.com/206387
>
Try setting the 'paragraph-background' property of the tag.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] gtk.entry copy and paste via ctrl-c & ctrl-v

[John Finlay]

Kim Adil wrote:
> I have an entry and want to use  the keyboard shortcuts for 
> cut/copy/paste. Without any coding by me, popup menu works fine, but 
> keyboard shortcuts don't work. the tutorial suggests it should be 
> working by default:
>
>
>   15.1.2. Using Clipboards with Entry, Spinbutton and TextView
>
> Entry, SpinButton and TextView widgets have popup menus that provide the 
> ability to cut and copy the selected text to and paste from the 
> "CLIPBOARD" clipboard. In addition key bindings are set to allow 
> keyboard accelerators to cut, copy and paste. Cut is activated by 
> *Control*+*X*; copy, by *Control*+*C*; and, paste, by *Control*+*V*.
>
>
> Am I missing something obvious?
>
> BTW using ubuntu jaunty standard packages.
>   
works for me on:

Python 2.6.2, PyGTK 2.14.1 (Gtk+ 2.16.1)

on jaunty

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] image on a Button

[John Finlay]

Gorse Emmanuel wrote:
> Thanks for the answer.
>
> I'm checking the imgFullFile with os.path.exists() before loading it,
> so the absolute path of the image is correct.
>
> As img.set_from_file() doesn't raise an exception when it can't load
> the file, i've tried with this:
>
> img = gtk.Image()
> pixBuf = gtk.gdk.pixbuf_new_from_file(imgFullFile)
> img.set_from_pixbuf(pixBuf)
> button.set_image(img)
>
> and gtk.gdk.pixbuf_new_from_file() didn't raise an exception, but the
> image is still not displayed.
>
> Furthermore, i'm also using an IconView to display the images, and
> they are displayed just fine. The code is:
> store = gtk.ListStore(str, gtk.gdk.Pixbuf)
> for name in xxx:
>   pixbuf = gtk.gdk.pixbuf_new_from_file(imgFullFile)
>   store.append([name, pixbuf])
>
> So the image loading seems fine but images don't get displayed on the 
> buttons...
>
> 2009/9/30, sandbox_mail :
>   
>> I did not try it out but my first guess is: the image is not found in
>> Windows due to the different path syntax.
>>
>> In linux you will have a path like /home/user/picture.jpg but in windows
>> 1.) there is no /, instead \ is used
>> 2.) absolute paths start at the drive letter like C:\Users\picure.jpg
>>
>> I believe python's "os" module will help you build the correct paths
>> depending on which system the code is running.
>>
>> Gorse Emmanuel wrote:
>> 
>>> Hye,
>>>
>>> I'm using this code to display an image on a button:
>>>
>>> img = gtk.Image()
>>> img.set_from_file(imgFullFile)
>>> button.set_image(img)
>>>
>>> with imgFullFile beeing the absolute path of the image.
>>>
>>> The code works fine under Linux, but the image is not displayed under
>>> Windows.
>>> Do you have any idea why ?
>>>
>>>   
Try adding:

img.show()

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Image Resize

[John Finlay]

Pete Stapley wrote:
> I have a menu bar, an image and a status bar in a vbox. Is there a way 
> to find out the new size of the image widget when the window is 
> resized? I can get the size of the complete window with 
> configure-event signal, but I want just the size of the image widget. 
> Thanks!
http://www.daa.com.au/pipermail/pygtk/2009-September/017635.html
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Resizing an hbox

[John Finlay]

Fabrice DELENTE wrote:
> Hello.
>
> I have an interface which is a simple toplevel window containing an hbox
> which contains a gtk.DrawingArea, designed in glade.
>
> How can I catch the fact that the hbox or the DrawingArea are resized when
> the toplevel is resized in the window manager?
>
> I thought that connecting the check-resize event of the hbox would be
> enough, but it doesn't work. Is there a flag to set somewhere?
>
> Thanks!
>
>   
Try the "size-allocate" signal.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Drawing a PNG on a gtk.DrawingArea

[John Finlay]

Fabrice DELENTE wrote:
> Hello.
>
> I'm still struggling with displaying a PNG image on a gtk.DrawingArea. I
> designed a glade interface that is basically a toplevel window with an hbox
> in it. I then add() a gtk.DrawingArea in the hbox.
>
> I'm trying to display the image this way:
>
> fichier = 
> "/home/fab/tex/courante/TES/partie1/equations_et_inequations_du_second_degre.png"
> self.imageOriginelle = gtk.gdk.pixbuf_new_from_file(fichier)
> self.imageReduite = gtk.gdk.Pixbuf(gtk.gdk.COLORSPACE_RGB, False, 8, 500, 
> 700)
> self.imageOriginelle.scale(self.imageReduite, 0, 0, 500, 700, 0, 0,
>  self.imageOriginelle.get_width()/500.0, 
>  self.imageOriginelle.get_height()/700.0,
>  gtk.gdk.INTERP_HYPER)
>
> I do this because I need to rescale the image so that it fits in my
> interface. The 500 and 700 values are arbitrary and for tests purposes only.
>
> I then try to display it:
>
> self.GC = gtk.gdk.GC(self.zoneDeDessin.window)
> self.zoneDeDessin.window.draw_pixbuf(self.GC, self.imageReduite, 0, 0, 0, 
> 0)
>
> where self.zoneDeDessin is the gtk.DrawingArea object I add()ed to the hbox.
>
> I can see a white rectangle, but no image in it. What did I miss?
>
>   
What happens if you try drawing the original pixbuf?

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problems with button.

[John Finlay]

Hidura wrote:
> I'm trying to make a button just with the icon, I want to give the 
> relief of the button to the icon for have more stetic in my project
>
> On Tue, Sep 8, 2009 at 1:24 PM, John Finlay  <mailto:fin...@moeraki.com>> wrote:
>
> Hidura wrote:
>
> Hello, List i have this sub-class of a Button and make me the
> icon what i pass in the button but lose the releif when the
> icon replace the button how could i change that and give the
> relief back again.
>
>
> Thanks.
>
>
> ##CODE
> class Button(gtk.Button):
>"""
>This is the constructor of the Vertical Scale.
>"""
>
>def __init__(self):
>gtk.Button.__init__(self)
>self.gc = None  # initialized in realize-event handler
>self.width  = 0 # updated in size-allocate handler
>self.height = 0 # idem
>self.connect('size-allocate', self.on_size_allocate)
>self.icon =
> '/usr/local/lib/python2.5/HidalgoP/.Logos/Buttons/add.png'
>  
>def do_expose_event(self, event):
>
>self.window.draw_pixbuf(self.gc,
> gtk.gdk.pixbuf_new_from_file_at_size(self.icon, 60, 60),
>0, 0, -1, -1, -1, -1, gtk.gdk.RGB_DITHER_NORMAL, 0, 0)
>self.set_relief(gtk.RELIEF_NORMAL)
>  
>def on_size_allocate(self, widget, allocation):
>self.width = allocation.width
>self.height = allocation.height
>
>
> Are you trying to put the Pixbuf inside the Button or undeneath
> the button? I'm having trouble understanding what you are trying
> to do.
>
> John
>
>
>
Try putting the pixbuf inside an Image and then put the Image inside a 
Button.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Multiple Gtk.DrawingArea

[John Finlay]

Antoine Cailliau wrote:
> Dear,
>
>   
>> don't use this since it returns the allocation within the Window not 
>> within the DA. Use self.window.get_size() to get the width and height 
>> and use 0, 0 for the x and y in the following.
>> 
> Thanks you a lot for your answer. 
>
> I've on other small problem (but I do not think it is useful to start a
> new thread): 
>
> I would like to set the size of the bottom DA. Using da2.size(X, Y) it
> seems that the application become greedy when I resize the window.
> Therefore, I guess it is not the right way to set the size.
>
>   
Spend some time reading the PyGTK Tutorial and reviewing the appropriate 
Reference Manual pages and the FAQ at pygtk.org

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Multiple Gtk.DrawingArea

[John Finlay]

Antoine Cailliau wrote:
> Dear all,
>
> First of all, a short presentation since I'm new on the ML. I'm student
> in Belgium (UCLouvain) in Computer Science (especially computer langages
> and software engineering). I just started a small project (a game for
> gnome) to get familiar with Cairo and Pango libs.
>
> I need two distinct drawing area in a gtk window. I tried a naive way
> but I only get one working area (the first is accordingly filled with a
> black rectangle, the other seems to be not displayed). I presume this is
> not the right way to do it.
>
> Here is the code:
>
> #!/usr/bin/env python
> #
>
> import gtk, cairo
>
> class DA(gtk.DrawingArea):
>
>   def __init__(self):
>   gtk.DrawingArea.__init__(self)
>   self.connect("expose_event", self.expose)
>
>   def expose(self, widget, event):
>   self.context = widget.window.cairo_create()
>   
>   # set a clip region for the expose event
>   self.context.rectangle(event.area.x, event.area.y,
>   event.area.width, 
> event.area.height)
>   self.context.clip()
>   
>   self.draw(self.context)
>   
>   return False
>   
>   def draw(self, context):
>   rect = self.get_allocation()
>   
don't use this since it returns the allocation within the Window not 
within the DA. Use self.window.get_size() to get the width and height 
and use 0, 0 for the x and y in the following.
>   context.rectangle(rect.x, rect.y, rect.width, rect.height)
>   context.fill()
>   return False
>
> def main():
>   window = gtk.Window()
>   da1 = DA()
>   da2 = DA()
>   
>   vbox = gtk.VBox()
>   vbox.add(da1)
>   vbox.add(da2)
>   window.add(vbox)
>
>   window.connect("destroy", gtk.main_quit)
>   window.show_all()
>   gtk.main()
>   
> if __name__ == "__main__":
>   main()
>
>   

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Disable signals and events to use a button as a container, not as a button?

[John Finlay]

Ryan Martin wrote:
> When you skin a button (and certain other widgets), you can control 
> the amount of pixels on top, left right and bottom to remain intact. 
> Those pixels do not scale when the widget is resized. The rest of the 
> pixels inside the widget scale as normal. This is what keeps button 
> borders from scaling horizontally or vertically and distorting, 
> creating ugly edges.
>
> This example should show you what I mean:
> http://www.ensomniac.com/pygtk/button_example8.jpg
That's nice. What's the API to control that?

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Disable signals and events to use a button as a container, not as a button?

[John Finlay]

Ryan Martin wrote:
> Hey Guys
>
> For all intensive purposes, an eventBox can have widgets packed into 
> it as I need and it can also have the background set as a style 
> property ala "bg_pixmap[NORMAL] = 'something.png'". My problem is that 
> the background needs to stretch, not tile. Like this:
>
> http://www.ensomniac.com/pygtk/button_example3.jpg
>
> When looking at the image above, please consider that I need to pack 
> widgets inside the blue graphic area. That is why I need to use some 
> form of container. I can't use an image widget, I can't use anything 
> that won't let me pack widgets inside the graphic area.
>
> The reason I initially decided to go with a button was because I knew 
> I could skin it (it has advanced skinning functionality allowing you 
> to stretch the pixmap while leaving the borders of the graphic intact) 
> and it could act as a container for other widgets. The problem with 
> using a button is that each time you hover over the main container 
> button, it steals all of the events that I want to go to the child 
> widgets. My initial post was looking for ideas on how I could disable 
> the main button from events and just use it as a graphic container for 
> my child widgets which would still need to work as expected. This has 
> proven to be impossible?


What is the advanced skinning functionality of a Button? This is 
something I'm not aware of.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problems with button.

[John Finlay]

Hidura wrote:
> Hello, List i have this sub-class of a Button and make me the icon 
> what i pass in the button but lose the releif when the icon replace 
> the button how could i change that and give the relief back again.
>
>
> Thanks.
>
>
> ##CODE
> class Button(gtk.Button):
> """
> This is the constructor of the Vertical Scale.
> """
>
> def __init__(self):
> gtk.Button.__init__(self)
> self.gc = None  # initialized in realize-event handler
> self.width  = 0 # updated in size-allocate handler
> self.height = 0 # idem
> self.connect('size-allocate', self.on_size_allocate)
> self.icon = 
> '/usr/local/lib/python2.5/HidalgoP/.Logos/Buttons/add.png'
>
>
>
>
> def do_expose_event(self, event):
>
> self.window.draw_pixbuf(self.gc, 
> gtk.gdk.pixbuf_new_from_file_at_size(self.icon, 60, 60),
> 0, 0, -1, -1, -1, -1, gtk.gdk.RGB_DITHER_NORMAL, 0, 0)
> self.set_relief(gtk.RELIEF_NORMAL)
>
>
> def on_size_allocate(self, widget, allocation):
> self.width = allocation.width
> self.height = allocation.height
>
>
Are you trying to put the Pixbuf inside the Button or undeneath the 
button? I'm having trouble understanding what you are trying to do.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Disable signals and events to use a button as a container, not as a button?

[John Finlay]

Ryan Martin wrote:
> Hey John,
> Thanks for the response. Yes, technically that would work, except 
> that you can't stretch the background of an event box. It only tiles. 
> As the style is applied to other sized event boxes, the background 
> texture would not stretch and resize.
>
> If anybody knows a way to make the background pixmap for an event box 
> stretch, I'd be all set. Any ideas?
>
> Thanks
> Ryan
>
>
>
What do you mean by stretch and resize? An example would help to 
visualize how this would work.
>
>
>
>
>
> John Finlay wrote:
>> Have you tried using an eventbox inside the frame to hold the vbox 
>> that holds the other stuff? I believe that the purpose of the 
>> eventbox is to do just the kind of thing you want i.e. set the 
>> bg_pixmap for the eventbox to some custom pixmap that represents your 
>> style.
>>
>> John
>>
>> Ryan Martin wrote:
>>> That's a good question. I didn't explain why I needed to do this 
>>> because I thought I'd confuse more. In the software I'm writing, I'm 
>>> combining theme elements with some widget tricks to accomplish a 
>>> visual task that I don't know is possible by any other means.
>>>
>>> I need to have a container than I can 'skin' that can contain other 
>>> widgets. My initial thought was to theme gtk.Frame to have borders 
>>> and a background (like a button), and I can pack all of my widgets 
>>> into that. Unfortunately, to my knowledge, you can't have a 
>>> background on a frame widget. So I started cycling through other gtk 
>>> widgets to see if there was some sort of container that I could 
>>> theme, but was unable to find something that worked.
>>>
>>>
>>> (http://www.ensomniac.com/pygtk/button_example2.jpg)
>>>
>>> That led me to my latest theory, that if I were able to use a button 
>>> as a container, I could theme gtk.Button to get the visual style I 
>>> wanted, but also the functionality of having the widgets inside. I 
>>> would need to disable the core button (container) from receiving and 
>>> processing signals, but somehow pass them on to any of the widgets 
>>> inside the button:
>>>
>>>
>>> (http://www.ensomniac.com/pygtk/button_example.jpg)
>>>
>>> And that leaves me where I'm at now. The following code will disable 
>>> my base button from working:
>>> *print* self.widgets.main_container_button.get_property("above-child")
>>> >> *False**
>>> print* 
>>> self.widgets.main_container_button.set_property("above-child", True)
>>> *print* self.widgets.main_container_button.get_property("above-child")
>>> >> *True*
>>> But I have no idea how to pass the events to the widgets that are 
>>> children of the button (the 'button_2' widget and the "entry" widget 
>>> in the image above).
>>>
>>> If there is some other way to accomplish my task, I'm all ears. 
>>> Currently, I'm struggling trying to get my events passed to the 
>>> child widgets.
>>>
>>>
>>> Thanks for the help!
>>>
>>> Ryan
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> John Finlay wrote:
>>>> Ryan Martin wrote:
>>>>> Hey Guys,
>>>>> I have a strange situation that I'm having trouble finding a 
>>>>> solution for. In the application I'm building, I would like to use 
>>>>> a button as a container and not necessarily a button. Currently, I 
>>>>> have a text field and another button inside my main button in 
>>>>> question. What I would like is to disable all of the innate 
>>>>> callbacks (prelight/hover, clicked, etc.) on that main button and 
>>>>> just have it sit there while still being able to click the child 
>>>>> button inside the main button or enter text into the text field.
>>>>>
>>>>> One thing to consider is that I need to disable callbacks on the 
>>>>> main button but not on any of the widgets that are added to the 
>>>>> main button as a child.
>>>>>
>>>>> Is this possible? Did that make sense?
>>>>>
>>>>> Any help would be greatly appreciated. I have scoured through old 
>>>>> threads looking for clues.
>>>> Why do you need to use a button as a container when you don't want 
>>>> to use any of the button's features? Are there no GTK containers 
>>>> that work?
>>>>
>>>> John
>>>>
>>>
>>> -- 
>>> Ryan Martin
>>> *Industrial Light + Magic
>>> *Assistant Technical Director
>>> cell: 973-632-1417 / desk: 415-746-2117
>>>
>>> *
>>> *
>>
>>
>
> -- 
> Ryan Martin
> *Industrial Light + Magic
> *Assistant Technical Director
> cell: 973-632-1417 / desk: 415-746-2117
>
> *
> *
> 
>
> ___
> pygtk mailing list   pygtk@daa.com.au
> http://www.daa.com.au/mailman/listinfo/pygtk
> Read the PyGTK FAQ: http://faq.pygtk.org/

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Color a ToggleButton

[John Finlay]

E.R. Uber wrote:
> I agree, it works fine on my Fedora 11 box as well, but not under 
> Windows XP SP3.
>
> So do you think this is impossible under XP or that more effort is 
> required??
>
> Thanks
> E.R. Uber
>
> On Sun, Sep 6, 2009 at 8:26 AM, John Finlay  <mailto:fin...@moeraki.com>> wrote:
>
> E.R. Uber wrote:
>
> As in?
>
>widget.modify_bg(gtk.STATE_NORMAL, color_off)
> andwidget.modify_bg(gtk.STATE_ACTIVE, color_on)
>
> That doesn't work either.
>
> On Sat, Sep 5, 2009 at 7:41 PM, John Finlay
> mailto:fin...@moeraki.com>
> <mailto:fin...@moeraki.com <mailto:fin...@moeraki.com>>> wrote:
>
>E.R. Uber wrote:
>
>After reading PyGTK FAQ 4.6 at
>  
>  http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp>
>  
>  <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp>>
>  
>  <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp>
>  
>  <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp>>> I
>thought perhaps I could change the color of a
> ToggleButton by
>copying its style, updating the style, and setting the
> style.
>This did not work.
>
>Then reading PyGTK FAQ 4.16 at
>  
>  http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp>
>  
>  <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp>>
>  
>  <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp>
>  
>  <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp>>>
>it mentions that gtk.Button (and I assume gtk.ToggleButton)
>are windowless widgets that do not allow changing their
>background and base color unless you insert them into an
>EventBox.
>
>
>Using a parent EventBox succeeds in changing the color
> of the
>event box, but not the child ToggleButton, the button just
>appears to be surrounded by the color change of the
> event box.
>I tried the same with a Button and it works the same way.
>
>I am using pygtk 2.12.1, gtk 2.16.4,  glade 3.6.7, and
> python
>2.6 on both windows and fedora core 9. I have only
> tried this
>to date on windows though.
>
>You don't need the EventBox just use toggle.modify_bg()
> with the
>appropriate state and color. Note in your example you need
> to set
>STATE_NORMAL and STATE_ACTIVE to have different colors for the
>active and inactive states.
>
>John
>
>
> Works fine on ubuntu 9.04 (Python 2.6.2, PyGTK 2.14.1 (Gtk+
> 2.16.1)) and Fedora 10.21 (Python 2.5.2, PyGTK 2.13.0 (Gtk+
> 2.14.7)). Maybe Windows overrides the change. Note the color
> change isn't visible when the cursor is over the button because it
> is in the prelight state.
>
> John
>
>
Try asking your question on one of the gtk mail lists (see: 
http://www.gtk.org/mailing-lists.html)

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] wrapping custom functions

[John Finlay]

gabriele lanaro wrote:
> Hi list,
>
> I'm wrapping a function like that:
>
> int gtk_scintilla_base_find_text( GtkScintillaBase *sci,int flags, 
> long start, long end,
> char *text, long *ostart, long *oend);
>
> but codegen can't wrap the function because of the long * type.
>
> In addition I have also to modify this function overriding it in the 
> override file ( ostart and oend are output parameters) with the %% 
> override directive.
>
> There's a way to define a brand new function in the override file 
> instead of defining-in-header/overriding it?
You can add a new function or method using the "define" directive in an 
override file.

define funcname [kwargs|noargs|onearg] 
[classmethod|staticmethod]
define Class.method [kwargs|noargs|onearg] 
[classmethod|staticmethod]

There are some examples in the pygtk override files.

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Color a ToggleButton

[John Finlay]

E.R. Uber wrote:
> As in?
>
> widget.modify_bg(gtk.STATE_NORMAL, color_off)
> and 
> widget.modify_bg(gtk.STATE_ACTIVE, color_on)
>
> That doesn't work either.
>
> On Sat, Sep 5, 2009 at 7:41 PM, John Finlay  <mailto:fin...@moeraki.com>> wrote:
>
> E.R. Uber wrote:
>
> After reading PyGTK FAQ 4.6 at
> http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp>
> <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp>> I
> thought perhaps I could change the color of a ToggleButton by
> copying its style, updating the style, and setting the style.
> This did not work.
>
> Then reading PyGTK FAQ 4.16 at
> http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp>
> <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp
> <http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp>>
> it mentions that gtk.Button (and I assume gtk.ToggleButton)
> are windowless widgets that do not allow changing their
> background and base color unless you insert them into an
> EventBox.
>
>
> Using a parent EventBox succeeds in changing the color of the
> event box, but not the child ToggleButton, the button just
> appears to be surrounded by the color change of the event box.
> I tried the same with a Button and it works the same way.
>
> I am using pygtk 2.12.1, gtk 2.16.4,  glade 3.6.7, and python
> 2.6 on both windows and fedora core 9. I have only tried this
> to date on windows though.
>
> You don't need the EventBox just use toggle.modify_bg() with the
> appropriate state and color. Note in your example you need to set
> STATE_NORMAL and STATE_ACTIVE to have different colors for the
> active and inactive states.
>
> John
>
>
Works fine on ubuntu 9.04 (Python 2.6.2, PyGTK 2.14.1 (Gtk+ 2.16.1)) and 
Fedora 10.21 (Python 2.5.2, PyGTK 2.13.0 (Gtk+ 2.14.7)). Maybe Windows 
overrides the change. Note the color change isn't visible when the 
cursor is over the button because it is in the prelight state.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Color a ToggleButton

[John Finlay]

E.R. Uber wrote:
> After reading PyGTK FAQ 4.6 
> at http://faq.pygtk.org/index.py?req=show&file=faq04.006.htp 
>  I thought 
> perhaps I could change the color of a ToggleButton by copying its 
> style, updating the style, and setting the style. This did not work.
>
> Then reading PyGTK FAQ 4.16 
> at http://faq.pygtk.org/index.py?req=show&file=faq04.016.htp 
>  it 
> mentions that gtk.Button (and I assume gtk.ToggleButton) are 
> windowless widgets that do not allow changing their background and 
> base color unless you insert them into an EventBox.
>
> Using a parent EventBox succeeds in changing the color of the event 
> box, but not the child ToggleButton, the button just appears to be 
> surrounded by the color change of the event box. I tried the same with 
> a Button and it works the same way.
>
> I am using pygtk 2.12.1, gtk 2.16.4,  glade 3.6.7, and python 2.6 on 
> both windows and fedora core 9. I have only tried this to date on 
> windows though.
You don't need the EventBox just use toggle.modify_bg() with the 
appropriate state and color. Note in your example you need to set 
STATE_NORMAL and STATE_ACTIVE to have different colors for the active 
and inactive states.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Disable signals and events to use a button as a container, not as a button?

[John Finlay]

Have you tried using an eventbox inside the frame to hold the vbox that 
holds the other stuff? I believe that the purpose of the eventbox is to 
do just the kind of thing you want i.e. set the bg_pixmap for the 
eventbox to some custom pixmap that represents your style.

John

Ryan Martin wrote:
> That's a good question. I didn't explain why I needed to do this 
> because I thought I'd confuse more. In the software I'm writing, I'm 
> combining theme elements with some widget tricks to accomplish a 
> visual task that I don't know is possible by any other means.
>
> I need to have a container than I can 'skin' that can contain other 
> widgets. My initial thought was to theme gtk.Frame to have borders and 
> a background (like a button), and I can pack all of my widgets into 
> that. Unfortunately, to my knowledge, you can't have a background on a 
> frame widget. So I started cycling through other gtk widgets to see if 
> there was some sort of container that I could theme, but was unable to 
> find something that worked.
>
>
> (http://www.ensomniac.com/pygtk/button_example2.jpg)
>
> That led me to my latest theory, that if I were able to use a button 
> as a container, I could theme gtk.Button to get the visual style I 
> wanted, but also the functionality of having the widgets inside. I 
> would need to disable the core button (container) from receiving and 
> processing signals, but somehow pass them on to any of the widgets 
> inside the button:
>
>
> (http://www.ensomniac.com/pygtk/button_example.jpg)
>
> And that leaves me where I'm at now. The following code will disable 
> my base button from working:
> *print* self.widgets.main_container_button.get_property("above-child")
> >> *False**
> print* self.widgets.main_container_button.set_property("above-child", True)
> *print* self.widgets.main_container_button.get_property("above-child")
> >> *True*
> But I have no idea how to pass the events to the widgets that are 
> children of the button (the 'button_2' widget and the "entry" widget 
> in the image above).
>
> If there is some other way to accomplish my task, I'm all ears. 
> Currently, I'm struggling trying to get my events passed to the child 
> widgets.
>
>
> Thanks for the help!
>
> Ryan
>
>
>
>
>
>
>
>
> John Finlay wrote:
>> Ryan Martin wrote:
>>> Hey Guys,
>>> I have a strange situation that I'm having trouble finding a 
>>> solution for. In the application I'm building, I would like to use a 
>>> button as a container and not necessarily a button. Currently, I 
>>> have a text field and another button inside my main button in 
>>> question. What I would like is to disable all of the innate 
>>> callbacks (prelight/hover, clicked, etc.) on that main button and 
>>> just have it sit there while still being able to click the child 
>>> button inside the main button or enter text into the text field.
>>>
>>> One thing to consider is that I need to disable callbacks on the 
>>> main button but not on any of the widgets that are added to the main 
>>> button as a child.
>>>
>>> Is this possible? Did that make sense?
>>>
>>> Any help would be greatly appreciated. I have scoured through old 
>>> threads looking for clues.
>> Why do you need to use a button as a container when you don't want to 
>> use any of the button's features? Are there no GTK containers that work?
>>
>> John
>>
>
> -- 
> Ryan Martin
> *Industrial Light + Magic
> *Assistant Technical Director
> cell: 973-632-1417 / desk: 415-746-2117
>
> *
> *

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Disable signals and events to use a button as a container, not as a button?

[John Finlay]

Ryan Martin wrote:
> Hey Guys,
> I have a strange situation that I'm having trouble finding a 
> solution for. In the application I'm building, I would like to use a 
> button as a container and not necessarily a button. Currently, I have 
> a text field and another button inside my main button in question. 
> What I would like is to disable all of the innate callbacks 
> (prelight/hover, clicked, etc.) on that main button and just have it 
> sit there while still being able to click the child button inside the 
> main button or enter text into the text field.
>
> One thing to consider is that I need to disable callbacks on the main 
> button but not on any of the widgets that are added to the main button 
> as a child.
>
> Is this possible? Did that make sense?
>
> Any help would be greatly appreciated. I have scoured through old 
> threads looking for clues.
Why do you need to use a button as a container when you don't want to 
use any of the button's features? Are there no GTK containers that work?

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] PyGTK problem

[John Finlay]

nipun batra wrote:
> I need from this code that var1,var2,var3,var4 should be continuosly 
> read from file(or serial port) and there value be updated 
> continuosly.Also i wish to use these 4 variables for plotting in 
> pygame/matplotlib/VPython .How can i do that.Moreover for every read 
> cycle there has to be a write cycle on to a file(or serial port) and 
> the user must be able to set this value via an entry.have not been 
> able to figure out all this.Moreover with my code i am able to 
> presently display updated information on a button click(where as i 
> wanna use button just to start things of)
> have attached program and source file 1.txt
IO to a disk file has quite different characteristics than IO to a 
serial port or network connection. You probably have to code these 
separately because of the different needs.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Updating widgets within threads

[John Finlay]

Bottiger, Maxwell - AES wrote:
> Hi,
>
> I have an implementation methodology question.  I'm putting together an app 
> which runs a bunch of external devices.  All my devices are on different 
> serial ports and run at different frequencies.  I'd like to use one thread to 
> monitor each device, then have the thread update it's relevant section of the 
> GUI.  I've been playing with updating a progress bar with a thread this 
> morning and I've found once I start the thread the main window freezes until 
> the thread completely exits.
>
> Obviously this isn't working.  What is the best way to go about this?  I 
> don't want to freeze my main window by putting sleep calls in the main loop, 
> but I do need to update the display when new data becomes available.
>
>   
Have you considered using using gobject.io_add_watch() instead of using 
threads? Unless there is some really long calculation this would seem 
like an ideal event driven program.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Existing C/GTK application

[John Finlay]

Rick Hightower wrote:
> I have an existing C/GTK application. I would like to write all new 
> features in Python using PyGTK.
> Is there any way to write widgets in PyGTK and use them easily in a 
> C/GTK based application?
>
> Are there examples of this somewhere?
>
> Essentially I want to add a Widget to a parent where the Widget is 
> managed by PyGTK and the parent is in the C/GTK application.
It might be easier to rewrite the C/GTK app in  Python if possible.

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Trying to change the "pen color" just before I draw a line

[John Finlay]

Rick Hightower wrote:


> Here is the full program. It changes the background color and redraws.
> I need to be able to draw different color lines.
>
> import pygtk
> pygtk.require("2.0")
> import gtk
> from gtk import Window, Button
>
> class Base:
> def handle_color_change(self, widget, data) :
> self.color = data
> if self.color == "Red":
> gdk_color = gtk.gdk.color_parse("Red")
> elif self.color == "Green":
> gdk_color = gtk.gdk.color_parse("Green")
> else:
> gdk_color = gtk.gdk.color_parse("Blue")
>  style = self.area.get_style().copy()
>  style.bg [gtk.STATE_NORMAL] = gdk_color
>  self.area.set_style(style)
> self.area.queue_draw()
>
> def __init__(self):
> self.color = None
> self.window = Window(gtk.WINDOW_TOPLEVEL)
>  self.window.connect("destroy", lambda w: gtk.main_quit())
>
>  self.mainPane = gtk.VBox(False, 0)
> self.window.add(self.mainPane)
> # create the toolbar
>  self.toolbar = gtk.HBox(False, 0)
> self.green = Button('Green')
> self.red = Button ('Red')
> self.toolbar.pack_start(self.green, True, True, 0)
> self.toolbar.pack_start(self.red, True, True, 0)
> self.mainPane.pack_start(self.toolbar, True, True, 0)
> self.red.connect("clicked", self.handle_color_change, "Red")
> self.green.connect("clicked", self.handle_color_change, "Green")
>
>
> #Create the drawing area
>  self.area = gtk.DrawingArea()
>  self.area.set_size_request(400, 300)
> self.area.connect("expose-event", self.area_expose)
> self.mainPane.pack_start(self.area, True, True, 0)
>
> self.window.show_all()
>
> def area_expose(self, area, event):
> print ("area exposed")
v delete vv
>  gc = self.area.get_style().fg_gc[gtk.STATE_NORMAL]
> gdk_color = gtk.gdk.color_parse("Blue")
> gc.get_colormap().alloc_color(gdk_color)
^^ delete ^^^

v add v

   gc=self.area.window.new_gc()
   gdk_color = gc.get_colormap().alloc_color("Blue")

^^ add ^^
> gc.set_foreground(gdk_color)
>self.area.window.draw_line(gc, 0, 0, 80, 70)
> return False
>
> def main(self):
> gtk.main()
>
> if __name__ == "__main__":
> base = Base()
> base.main()
>

You could create the GC  and color up front and reuse it in the expose 
event handler to simplify.

john
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Animation performance - limitation or my code?

[John Finlay]

Kim Adil wrote:
> I once wrote a gtk program in C that displayed about 30 animations on a
> drawing area and they moved around the screen, and performance was
> excellent. The drawingarea size was 1600x1200 and the animations were
> very smooth.
>
> I have started to write a similar app using pygtk in a more object
> oriented way, and the drawing area size really seems to have a large
> impact on performance. When the drawing area is 100x100, anims are
> smooth. If the drawing area is dragged to 1000x1000 the frame rate drops
> very quickly. Is this to be expected with the python being interpreted,
> or is it just that my code is inefficient. See sample below (requires 2
> animated gif files a.gif and a2.gif, just substitute any from the web 
> for demonstartion, and a glade file attached):
>
>
> #!/usr/bin/env python
> import sys
> import pygtk
> import glib
> import gtk
> import gtk.glade
> # main window to display objects in motion
>
> i=1
> class MainWin:
> def __init__(self):
> self.gladefile = "hv.glade"
> self.wTree = gtk.glade.XML(self.gladefile, "main_win")
> dic = {"on_drawingarea1_expose_event" : self.expose,
>"on_drawingarea1_configure_event": self.configure,
>}
> self.wTree.signal_autoconnect(dic)
> self.window = self.wTree.get_widget("main_win")
> self.da = self.wTree.get_widget("drawingarea1")
> self.window.show_all()
> x, y, width, height = self.da.get_allocation()
> self.pixmap = gtk.gdk.Pixmap(self.da.window, width, height)
> self.machlist = []
> self.setup_objects()
> glib.timeout_add(30, self.draw_objects)
>
> def setup_objects(self):
> self.machlist.append(Object('t1','v','stopped',10,10,'a.gif'))
> 
> self.machlist.append(Object('t2','v','travelling',300,10,'a2.gif'))
> self.machlist.append(Object('t3','v','stopped',100,300,'a.gif'))
> self.machlist.append(Object('t4','v','stopped',600,600,'a2.gif'))
> self.machlist.append(Object('t5','v','stopped',100,60,'a.gif'))
> self.machlist.append(Object('t6','f','stopped',60,100,'a.gif'))
> 
> self.machlist.append(Object('t7','f','travelling',100,100,'a.gif'))
>
> def draw_objects(self):
> x, y, width, height = self.da.get_allocation()
> self.pixmap = gtk.gdk.Pixmap(self.da.window, width, height)
> self.pixmap.draw_rectangle(self.da.get_style().white_gc,True, 0,
> 0, width, height)
> for object in self.machlist:
>
> self.pixmap.draw_pixbuf(self.da.get_style().fg_gc[gtk.STATE_NORMAL],object.pb,
>  
>
> 0, 0, object.x,object.y)
Add this line:
   self.da.queue_draw_area(x, y, width, height)
> return True
> def expose(self,a,b):
> x, y, width, height = self.da.get_allocation()
>
> self.da.window.draw_drawable(self.da.get_style().fg_gc[gtk.STATE_NORMAL],self.pixmap,
>  
>
> x, y, x, y, width, height)
> self.da.queue_draw_area(x, y, width, height)
Remove the above line.
> return False
>
> def configure(self,a,b):
> x, y, width, height = self.da.get_allocation()
>
> self.da.window.draw_drawable(self.da.get_style().fg_gc[gtk.STATE_NORMAL],self.pixmap,
>  
>
> x, y, x, y, width, height)
> return False
>
> def cleanup(self):
> for object in self.machlist:
> glib.source_remove(object.timeout)
>
> def list_objects(self):
> self.draw_objects()
> for object in self.machlist:
>
> self.pixmap.draw_pixbuf(self.da.get_style().fg_gc[gtk.STATE_NORMAL],object.pb,
>  
>
> 0, 0, 0, 0)
> #self.da.queue_draw()
> class Object:
> def __init__(self,name,model,state,x,y,anim):
> self.name=name
> self.model=model
> self.state=state
> self.x=x
> self.y=y
> self.anim=anim
> self.image='l'
> self.frame=0
> self.imagef = gtk.Image()
> self.pixbufanim = gtk.gdk.PixbufAnimation(anim)
> print 'width %d - height
> %d'%(self.pixbufanim.get_width(),self.pixbufanim.get_height())
> self.pbiter = self.pixbufanim.get_iter()
> print 'delay time %d'%(self.pbiter.get_delay_time())
> self.pb=self.pbiter.get_pixbuf()
> self.t=glib.timeout_add(self.pbiter.get_delay_time(),
> self.update_pb)
>
> def update_pb(self):
> self.pbiter.advance()
> self.pb=self.pbiter.get_pixbuf()
> self.t=glib.timeout_add(self.pbiter.get_delay_time(),
> self.update_pb)
> return False
>
>
> print 'create mainwin'
> m=MainWin()
> gtk.main()
>
I think the program is generating too many expose events and too many 
draw_pixbuf calls are slowing down the animations when a large pixbuf 
draw is needed.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Non-blocking subprocess.Popen stop GUI

[John Finlay]

Bou Baris wrote:
>> Instead of using select() use gobject.io_add_watch() on proc.stdout to
>> register a callback function that is used to read the % and set the progress
>> bar.
>> 
>
> I've modified my code to test gobject.io_add_watch()
> I expect that test_io_watch() print in console line readed from proc.stdout
> But i see nothing: look like a blocking behaviour (as when use Popen
> without use select)
> If I stop program with CTRL-C, now I see the output to console/terminal
>
>
>   def test_io_watch(self, file, condition):
> print str(file.readline())
> return True
>
>   def mkv_dts_extraction(self,dtstrack,pgbar=None):
>
> command=['mkvextract', \
> 'tracks', \
> self.INPUTFILE, \
> dtstrack + ':' + self.TEMP_DTSFILENAME]
>
> proc=subprocess.Popen(command,stdout=subprocess.PIPE)
>
> outfile=proc.stdout
> outfd=outfile.fileno()
> file_flags = fcntl.fcntl(outfd, fcntl.F_GETFL)
> fcntl.fcntl(outfd, fcntl.F_SETFL, file_flags | os.O_NDELAY)
>
> gobject.io_add_watch(proc.stdout,gobject.IO_IN |
> gobject.IO_HUP,self.test_io_watch)
>
> err = proc.wait()
>
> return True
> ___
>   
Also I forgot you must use file.read() instead of file.readline() in 
your callback since you need to read all the available input in the 
callback.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Non-blocking subprocess.Popen stop GUI

[John Finlay]

Bou Baris wrote:
>> Instead of using select() use gobject.io_add_watch() on proc.stdout to
>> register a callback function that is used to read the % and set the progress
>> bar.
>> 
>
> I've modified my code to test gobject.io_add_watch()
> I expect that test_io_watch() print in console line readed from proc.stdout
> But i see nothing: look like a blocking behaviour (as when use Popen
> without use select)
> If I stop program with CTRL-C, now I see the output to console/terminal
>
>
>   def test_io_watch(self, file, condition):
> print str(file.readline())
> return True
>
>   def mkv_dts_extraction(self,dtstrack,pgbar=None):
>
> command=['mkvextract', \
> 'tracks', \
> self.INPUTFILE, \
> dtstrack + ':' + self.TEMP_DTSFILENAME]
>
> proc=subprocess.Popen(command,stdout=subprocess.PIPE)
>
> outfile=proc.stdout
> outfd=outfile.fileno()
> file_flags = fcntl.fcntl(outfd, fcntl.F_GETFL)
> fcntl.fcntl(outfd, fcntl.F_SETFL, file_flags | os.O_NDELAY)
>
> gobject.io_add_watch(proc.stdout,gobject.IO_IN |
> gobject.IO_HUP,self.test_io_watch)
>
> err = proc.wait()
>
> return True
>   
I was surprised at your result so I looked back at this message and 
realized that the reason it didn't work is because you are waiting for 
the process to complete before returning from the mkv_dts_extraction() 
method. The above code should work if you remove the line:

err = proc.wait()

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Non-blocking subprocess.Popen stop GUI

[John Finlay]

Bou Baris wrote:
>> I think this is due to mkvextract not flushing its output because it's not
>> connected to a tty. Try using a pty in between to fool mkvextract. Something
>> like:
>>
>> from_mkve, to me = pty.openpty()
>> to_me = os.fdopen(to_me, 'r')
>> file_flags = fcntl.fcntl(to_me, fcntl.F_GETFL)
>> fcntl.fcntl(to_me, fcntl.F_SETFL, file_flags|os.O_NDELAY)
>> proc = subprocess.Popen(command, stdout=from_mkve)
>> gobject.io_add_watch(to_me, gobject.IO_IN|gobject.IO_HUP,
>> self.test_io_watch)
>>
>> 
> Thanks for your suggestion. This evening when I come back @home I'll try it.
>
> BTW if 'mkvextract' doesn't flush its output, I should have the same
> problem running my original code that use 'select'.
> Using 'select' is fine, there is no block to the flow due to lack of
> stdout flush, BUT the problem is GUI that is stopped.
>
> One hour ago I've also tried with thread (using module thread and
> thread.start_new_thread() function).
> The same behaviour.. GUI is stopped until subprocess.Popen() end.
>
>   
I'm guessing that you are printing a buffer full of output at a time but 
the progressbar isn't getting updated because the program is stuck in 
the select() loop and can't run the gtk event processing loop until it 
exits the loop i.e. when the mkvextract program ends.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Glade interface and Dialog: buttons don't work

[John Finlay]

Fabrice DELENTE wrote:
>> Look at:
>>
>> http://library.gnome.org/devel/pygtk/stable/class-gladexml.html#method-gladexml--signal-autoconnect
>> 
>
> I think I don't need this method if I define the callbacks directly in the
> glade file.
>
>   
You need to call this method to connect the callbacks named in the glade 
file to the callbacks in the python program.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Glade interface and Dialog: buttons don't work

[John Finlay]

Fabrice DELENTE wrote:
> Hello.
>
> I designed an interface with Glade, and I want to use it with PyGtk.
>
> In this interface, there is a Window and a Dialog. The Dialog is popped up
> when I click a button in the Window.
>
> I have a reference to the Dialog:
>
> dialog = gtk.glade.XML(gladeFile, "dialog1").get_widget("dialog1")
>
> and references to the two button that are in the action area:
>
> okButton = gtk.glade.XML(gladeFile, "dialog1").get_widget("okButton")
> cancelButton = gtk.glade.XML(gladeFile, "dialog1").get_widget("cancelButton")
>
> These buttons are connected to handlers in Glade: okButton's activate and
> clicked signals are connected to handleOkButton, and cancelButton's activate
> and clicked signals to handleCancelButton.
>
> I have defined these handlesr:
>
> def handleOkButton(widget):
>   print "ok button pressed"
>
> def handleCancelButton(widget):
>   print "cancel button pressed"
>
> but nothing happens when I pop up my Dialog with dialog.show_all() and I
> click either Ok or Cancel... I tried with dialog.run(), and it didn't worked
> either.
>
> Any hint? Thanks!
>
>   
Did you connect the handlers to the signals using autoconnect?

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Non-blocking subprocess.Popen stop GUI

[John Finlay]

Bou Baris wrote:
>> Instead of using select() use gobject.io_add_watch() on proc.stdout to
>> register a callback function that is used to read the % and set the progress
>> bar.
>> 
>
> I've modified my code to test gobject.io_add_watch()
> I expect that test_io_watch() print in console line readed from proc.stdout
> But i see nothing: look like a blocking behaviour (as when use Popen
> without use select)
> If I stop program with CTRL-C, now I see the output to console/terminal
>
>
>   def test_io_watch(self, file, condition):
> print str(file.readline())
> return True
>
>   def mkv_dts_extraction(self,dtstrack,pgbar=None):
>
> command=['mkvextract', \
> 'tracks', \
> self.INPUTFILE, \
> dtstrack + ':' + self.TEMP_DTSFILENAME]
>
> proc=subprocess.Popen(command,stdout=subprocess.PIPE)
>
> outfile=proc.stdout
> outfd=outfile.fileno()
> file_flags = fcntl.fcntl(outfd, fcntl.F_GETFL)
> fcntl.fcntl(outfd, fcntl.F_SETFL, file_flags | os.O_NDELAY)
>
> gobject.io_add_watch(proc.stdout,gobject.IO_IN |
> gobject.IO_HUP,self.test_io_watch)
>
> err = proc.wait()
>
> return True
>   
I think this is due to mkvextract not flushing its output because it's 
not connected to a tty. Try using a pty in between to fool mkvextract. 
Something like:

from_mkve, to me = pty.openpty()
to_me = os.fdopen(to_me, 'r')
file_flags = fcntl.fcntl(to_me, fcntl.F_GETFL)
fcntl.fcntl(to_me, fcntl.F_SETFL, file_flags|os.O_NDELAY)
proc = subprocess.Popen(command, stdout=from_mkve)
gobject.io_add_watch(to_me, gobject.IO_IN|gobject.IO_HUP, 
self.test_io_watch)

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Non-blocking subprocess.Popen stop GUI

[John Finlay]

Bou Baris wrote:
> I'm a new user of PyGTK.
> I'm writing with Python and PyGTK a little software for decoding MKV
> video files using external tool MKV TOOL UNIX. It's a simple graphical
> front-end to ffmpeg and MKVTOOLNIX.
> I use Ubuntu 9.04 shipped with Python 2.6.2 and PyGTK 2.x
>
> I've a problem (obviously).
>
> I've made a non blocking function (using subprocess.Popen and select)
> for calling the external program MKVEXTRACT. When works, 'mkvextract'
> reports its progress state as a percentage:
> Progress: 0%
> Progress: 1%
> Progress: 2%
> 
>
> I want read 'mkvextract' stdout in a real-time (non-blocking) fashion,
> decode progress number from string 'Progress: xx%' and control a PyGTK
> ProgressBar.
>
>
> This is the code and non-blocking is working (printing the progress to
> console it's ok).
> The problem is that when code run, I see in terminal the percentage
> number decoded correctly (printed from my code, not by 'mkvextract',
> see code below)  but progress bar isn't moving.
> Really ALL GUI is blocked.
> It's look that Popen block GUI job until 'mkvextraxt' end.
> When Popen/mkvextract ends its job, GUI restart to normal behaviour
> and I see the progress bar to 100%
>
>
> def mkv_dts_extraction(self,dtstrack,pgbar=None):
>
>  command=[ 'mkvextract', \
> 'tracks', \
> self.INPUTFILE, \
> dtstrack + ':' + self.DTSFILENAME ]
>
> # start subprocess
> proc = subprocess.Popen(command,stdout=subprocess.PIPE)
>
> # set non-blocking mode
> outfile = proc.stdout
> outfd = outfile.fileno()
> file_flags = fcntl.fcntl(outfd, fcntl.F_GETFL)
> fcntl.fcntl(outfd, fcntl.F_SETFL, file_flags | os.O_NDELAY)
>
> # use 'select' for reading
> while True:
> ready = select.select([outfd],[],[]) # wait for input
> if outfd in ready[0]:
> outchunk = outfile.read()
> if outchunk == '' :
> break
> select.select([],[],[],.1) # give a little time for buffers to 
> fill
>
> perc =
> re.compile("Progress:\s+(\d+)",re.IGNORECASE).findall(outchunk)
> if len(perc) > 0:
> completed = int(perc[len(perc)-1]) / 100.00  # take
> last 'Progress' number
>
> print completed  # this print WORKS: in terminal I can
> see the percentage growing
>
> # progress bar DON'T WORK : GUI is blocked
> pgbar.set_fraction(completed)
> pgbar.show()
>
> err = proc.wait()
>
> return True
>   
Instead of using select() use gobject.io_add_watch() on proc.stdout to 
register a callback function that is used to read the % and set the 
progress bar.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Sorting every row of a tree regardless of its hierarchy

[John Finlay]

Jiri Bajer wrote:
> On Wed, 2009-08-26 at 01:47 +0200, John Finlay wrote:
>   
>>> If I store the data in a TreeModel, wrap a TreeModelSort over it and
>>> display the result via TreeView, the sorting mechanism is not able to
>>> put "Task 2" before "Task 1.2" even if "Task 2" has higher priority. It
>>> is caused by respecting the hierarchy - sorting won't break a child from
>>> its parent (but I need to do so).
>>>
>>> Is there any way how to achieve such sorting without maintaining two
>>> copies of all data - one in TreeStore and one in ListStore and
>>> sorting/displaying the ListStore instead?
>>>   
>>>   
>> Look at the TreeModelFilter. With the right set_modify_func() function 
>> you may be able to massage the look for your purposes.
>> 
>
> Thanks for a hint! I've managed to create a 1:1 mapping between filtered
> and original model - but I don't see how to change the row hierarchy in
> the filtered model. It seems to me it is not possible to do anything
> else than setting a value for filtered model's cell and thus using
> PyGTK's native tree sorting mechanism over the filtered model won't
> work.
>
> Did you mean I can reorder the filtered tree rows as if they were sorted
> by a given column (with original model's hierarchy that doesn't respect
> my re-ordering) and then hide the expanders? That would also mean I have
> to maintain a path-to-path/row-to-row mapping for every tree row and
> every column I want to sort-by.
>
> Mapping example for sort by Priority:
>
> tree -> sorted "flat" list
> (0,) -> (0,)
> (0,0) -> (1,)
> (0,1) -> (3,)
> (1,) -> (2,)
>
> Do you have any tips for the mapping? If I use something like the
> example before I have to go through most of the tree and re-number the
> paths with every row insert/delete (but finding the right path is easy).
>
> Or I could store "pointers" like iterators or row references for every
> row and every sortable column - they wouldn't need any additional care
> in case of row insert/delete. But the iterators are not guaranteed to be
> persistent and large number of row references would slow down the tree
> to crawl...
>
> That looks like having two copies of all data (a tree and a list)
> doesn't help either because I would have to maintain the path-to-path
> mapping anyway (or compute it on a fly by walking through all tree rows,
> counting the rows traversed and searching for a particular hierarchy
> number - which is too slow, O(N) for N=number of tree rows).
>
> I can also imagine a hybrid solution where every parent tree node keeps
> a number of nodes in his subtree which would speed up the tree
> traversing to something like O(log N) but would also add O(log N)
> operation of re-calculating the counts in a part of subtree on every row
> add/delete.
>
> >From what I've thought about so far it seems the easiest way would be to
> hack into the TreeModelSort and add a property that would control if the
> rows compared during the sort operation will respect the hierarchy or if
> every row in the model gets compared. Can it be done directly from
> Python or will I have to roll up the sleeves and dive into the C code?
> Are there any tutorials available?
>
> Thanks for any further ideas!
>
>   
I think that you have to use a ListStore instead of a TreeStore and then 
work out some way to simulate a hierarchy with hiding "children" which I 
imagine is the major benefit of using a TreeStore. I was thinking a 
TreeModelFilter could do this.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Sorting every row of a tree regardless of its hierarchy

[John Finlay]

Jiri Bajer wrote:
> Hi,
>
> I am writing an editor for multi-columned hierarchical data (for example
> project tasks or software requirements) and would like to allow the user
> to sort the rows as if the data had a flat structure.
>
> Example TreeView:
>
> Hierarchy | Task name | Priority   | Man-days
> --+---++-
>   
>> 1   | Task 1| High   | 5
>> 
> 1.1   | Subtask 1 | High   | 3
> 1.2   | Subtask 2 | Low| 2
> 2 | Task 2| High   | 1
>
> Now I would like to sort all the project tasks and subtasks by priority.
>
> If I store the data in a TreeModel, wrap a TreeModelSort over it and
> display the result via TreeView, the sorting mechanism is not able to
> put "Task 2" before "Task 1.2" even if "Task 2" has higher priority. It
> is caused by respecting the hierarchy - sorting won't break a child from
> its parent (but I need to do so).
>
> Required result:
>
> Hierarchy | Task name | Priority v | Man-days
> --+---++-
> 1 | Task 1| High   | 5
> 1.1   | Subtask 1 | High   | 3
> 2 | Task 2| High   | 1
> 1.2   | Subtask 2 | Low| 2   <--- breaks the hierarchy
>
> Is there any way how to achieve such sorting without maintaining two
> copies of all data - one in TreeStore and one in ListStore and
> sorting/displaying the ListStore instead?
>
> I was able to hide the expanders via setting the expander column to a
> hidden one so the view part seems to be OK.
>
> I don't think TreeSortable.set_sort_func() would help here as the
> comparison works but the rows to be compared are chosen only from one
> parent and within the same tree depth (i.e. 1.1 is compared only with
> 1.2 but not with 1 or 2).
>
> Maybe I could write a wrapper providing a ListStore-like behavior while
> having no own data but taking them from a TreeStore + put this ListStore
> in its own TreeView? Or inheriting from a plain ListStore and implement
> "child" collapse/expand on my own?
>
> Any ideas are welcome, thanks!
>   
Look at the TreeModelFilter. With the right set_modify_func() function 
you may be able to massage the look for your purposes.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] set_use_markup and gtk.Button

[John Finlay]

Art Hunkins wrote:
> I have been happily using the set_use_markup property of gtk.Label and 
> gtk.Frame with XML markup, to create bolded, italic and various sized text.
>
> I need to do this with gtk.Button as well - to do the same with (callback) 
> button labels. However, I've not been able. I've employed the two general 
> approaches below (with results immediately below them):
>
> #1:
>
> def cbbutton(self,box,callback,title=""):
>  """Creates a callbackbutton
>  box: parent box
>  callback: click callback
>  title: if given, the button name
>  returns the widget instance"""
>  self.cbbutts = self.cbbutts + 1
>  label = gtk.Label(title)
>  label.set_use_markup(True)
>  butt = gtk.Button(" %s " % label)
>  box.pack_start(butt, False, False, 5)
>  self.cbbuttons.append([butt,label,0])
>  butt.connect("clicked", callback)
>  butt.show()
>  return butt
>
> The above version runs fine, but gives unwanted button labels, of this 
> variety:
> 
>
>
> #2:
>
> def cbbutton(self,box,callback,title=""):
>  """Creates a callbackbutton
>  box: parent box
>  callback: click callback
>  title: if given, the button name
>  returns the widget instance"""
>  self.cbbutts = self.cbbutts + 1
>  label = gtk.Label()
>  label.set_markup(title)
>  butt = gtk.Button()
>  butt.child = label
>  label.show()
>  box.pack_start(butt, False, False, 5)
>  self.cbbuttons.append([butt,label,0])
>  butt.connect("clicked", callback)
>  butt.show()
>  return butt
>
> The above version crashes, with the following error log:
>
> /usr/lib/python2.6/site-packages/sugar/activity/main.py:35: GtkWarning:
> gtk_widget_realize: assertion `GTK_WIDGET_ANCHORED (widget) ||
> GTK_IS_INVISIBLE (widget)' failed
> activity.show()
> Gtk:ERROR:gtkwidget.c:7967:gtk_widget_real_map: assertion failed:
> (GTK_WIDGET_REALIZED (widget))
>
> Can someone please suggest what changes might give me a proper label?
>
> FWIW, including XML tags in "title" makes no difference whatever in the 
> above results.
>
>   
How about:

butt = gtk.Button(title)
butt.child.set_use_markup(True)

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] size request in scrolled windows + viewport

[John Finlay]

On 08/03/2009 01:27 PM, Alessandro Dentella wrote:
> Hi
>
>I have a table with many widgets inside, so that total dimenstions are
>bigger that the screen so that I put it in a ScrolledWindow + ViewPort.
>
>So far so good. ow I have a Window with all widgets in a pane that scrolls
>fine... but starts very little indeed.
>
>I'd like to know how to propagate the dimentions that the table would
>have requested to set dimentions of the Viewport.
>
>I'm a little lost between size_request/get_child_requisition and similar
>methds.
>
>
If the table is larger than the screen then I would assume you want the 
window to be smaller than the table. I usually set the default size of 
the Window so it starts up bigger than the minimum size.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Help with set_cursor() ---- Invisible cursor option

[John Finlay]

On 08/03/2009 03:05 PM, DINESHBABU DINAKARABABU wrote:
> Hi,
>
> Now, it throws:
>
> AttributeError: 'NoneType' object has no attribute 'set_cursor'
>
> I agree to the fact that it is a gtk.gdk.Window object. Does 
> win.window invoke a gtk.gdk.Window object? Correct me if am wrong as I 
> am still learning the details and tricks associated with python and gtk.
>
This means that there is no gtk.gdk.Window associated with the 
gtk.Window (win in this case). This is usually because the gtk.Window 
has not been realized (have a look at the tutorial which has an 
description of the widget display methods). You can call the widget 
realize() or show() methods before setting the cursor, or connect to the 
realize signal of win and set the cursor in the callback.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Help with set_cursor() ---- Invisible cursor option

[John Finlay]

On 08/03/2009 09:23 PM, John Finlay wrote:

DINESHBABU DINAKARABABU wrote:
   

Hi everyone,

I am pretty new to Python and Pygtk. I would be grateful if someone
could help me out with this issue.

I am trying to get a full screen display of an image from some raw
data using the pixbuf_new_from_data() method. I want to do away with
the default cursor option and implement an invisible cursor so that we
don't see the cursor during the full screen display. To implement the
invisible cursor, I am doing the following:

pixmap = gtk.gdk.Pixmap(None, 1, 1, 1)
color = gtk.gdk.Color()
cursor = gtk.gdk.Cursor(pixmap, pixmap, color, color, 0, 0)
gtk.gdk.Window.set_cursor(cursor)

However, I am getting this error:

TypeError: descriptor 'set_cursor' requires a 'gtk.gdk.Window' object
but receives a 'gtk.gdk.Cursor'.

 

Try using:

win.set_cursor(cursor)

The error message is indicating that you have passed the wrong object to
the function gtk.gdk.Window.set_cursor() which requires a Window object
and a Cursor object as args. You should invoke the method of the Window
object win instead.

   

My mistake it should be:

win.window.set_cursor(cursor)

since it's the gtk.gdk.Window object not the gtk.Window object.

john
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Help with set_cursor() ---- Invisible cursor option

[John Finlay]

DINESHBABU DINAKARABABU wrote:
> Hi everyone,
>
> I am pretty new to Python and Pygtk. I would be grateful if someone 
> could help me out with this issue.
>
> I am trying to get a full screen display of an image from some raw 
> data using the pixbuf_new_from_data() method. I want to do away with 
> the default cursor option and implement an invisible cursor so that we 
> don't see the cursor during the full screen display. To implement the 
> invisible cursor, I am doing the following:
>
> pixmap = gtk.gdk.Pixmap(None, 1, 1, 1)
> color = gtk.gdk.Color()
> cursor = gtk.gdk.Cursor(pixmap, pixmap, color, color, 0, 0)
> gtk.gdk.Window.set_cursor(cursor) 
>
> However, I am getting this error:
>
> TypeError: descriptor 'set_cursor' requires a 'gtk.gdk.Window' object 
> but receives a 'gtk.gdk.Cursor'.
>
Try using:

win.set_cursor(cursor)

The error message is indicating that you have passed the wrong object to 
the function gtk.gdk.Window.set_cursor() which requires a Window object 
and a Cursor object as args. You should invoke the method of the Window 
object win instead.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problem in fetching Unicode from URL and displaying it in PyGTK widget

[John Finlay]

Bertrand Kintanar wrote:
> On 7/17/09 6:39 PM, John Finlay wrote:
>   
>> Bertrand Kintanar wrote:
>> 
>>> On 7/17/09 1:12 PM, John Finlay wrote:
>>>   
>>>> Maybe you could gzip it and base64 encode it before stuffing it into 
>>>> the DB?
>>>>
>>>> John
>>>> 
>>> is gzipping it necessary for this to work?
>>>
>>>   
>> No but it'll make it a lot smaller than just base64 alone.
>>
>> John
>> 
> I'm really not concerned with database size here. What I'm really 
> concerned, which I'm having trouble implementing is how to convert the 
> unicode to display the right character text rather than the coded form.
> ___
>   
I misunderstood what you wanted. I thought you just wanted to save the 
html file contents into the DB and were having a problem with the text 
encoding between the html and the DB but it sounds like you want to do 
something different.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] gtk.ToggleButton not working.

[John Finlay]

Neil Dugan wrote:
> With pygtk version 2.12.9 this works
> "button = gtk.ToggleButton('test',False)"
> but on pygtk version 2.16.1 it no longer works :)
>
> I get the error message
> RuntimeError: more argument specifiers than keyword list entries 
> (remaining format:'):GtkToggleButton.__init__')
>
> what do I need to do for it to work on both versions?
>   
It's a bug. The workaround is:

button = gtk.ToggleButton('test')
button.props.use_underline = False

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problem in fetching Unicode from URL and displaying it in PyGTK widget

[John Finlay]

Bertrand Kintanar wrote:
> On 7/17/09 1:12 PM, John Finlay wrote:
>   
>> Maybe you could gzip it and base64 encode it before stuffing it into 
>> the DB?
>>
>> John
>> 
> is gzipping it necessary for this to work?
>
>   
No but it'll make it a lot smaller than just base64 alone.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problem in fetching Unicode from URL and displaying it in PyGTK widget

[John Finlay]

Bertrand Kintanar wrote:
> Hi list
>
> I've been working on a personal project and for some reason I need to 
> read a html file and then store what I've read into database. And 
> since there is huge possibility that the html file contains some 
> unicode in it (e.g. `ñ' ) I'm having a hard time 
> converting/displaying the right ascii equivalent to it (e.g. letter 
> `ñ'). I don't care what format to store it to database, may it be in 
> coded form or the ascii form. I can just make a function to convert it 
> later. My problem is how to convert it.
>
Maybe you could gzip it and base64 encode it before stuffing it into the DB?

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] warnings from dialog.run in pygtk2-2.14.1

[John Finlay]

On 07/15/2009 01:09 AM, Chris Rouch wrote:
> Hi,
>
> I recently upgraded to Fedora 11. Since then every pygtk program I
> have has started to produce a warning when dialog.run() is called:
>
> ./tmp.py:12: GtkWarning: gdk_x11_atom_to_xatom_for_display: assertion
> `atom != GDK_NONE' failed
>   response=dialog.run()
>
> A very simple example is:
>
>
> #!/usr/bin/python
> import gtk,gobject
>
> window = gtk.Window(gtk.WINDOW_TOPLEVEL)
> dialog = gtk.Dialog("title",window)
> renametable=gtk.Table(2, 2, False)
> dialog.add_button(gtk.STOCK_OK, gtk.RESPONSE_OK)
> dialog.add_button(gtk.STOCK_CANCEL, gtk.RESPONSE_CANCEL)
> dialog.set_position(gtk.WIN_POS_MOUSE)
> dialog.set_default_response(gtk.RESPONSE_OK)
> dialog.show_all()
> response=dialog.run()
> dialog.destroy()
>
>
> Is this a problem with my code, or a bug in pygtk?
>
>
Neither. It's the result of a change in gtk 16.2.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Disappearing pango attributes

[John Finlay]

Krzysztof Kotlenga wrote:
> Hi,
>
> I have a program which, I believe, should do the same thing in both
> Python and C - set pango attributes on a label. The C version works
> as expected. In PyGTK version attributes get applied only on the first
> character. Any ideas why does this happen? I attach both test cases.
>
>   
The pygtk pango docs say that the end_index is 1 by default. You need to 
specify the end_index=-1 to get the same behavior as the C code. Pango C 
code started to set the attribute  end_index to a default value in 
version 1.20 and the default value used was end of string.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] gtk.TreeModel.iter_children() method, how often should it be called?

[John Finlay]

Try asking this question on the gtk list gtk-l...@gnome.org. It's more 
likely someone there would have the answer.

John

Gerald Britton wrote:
> I don't think that makes sense.  For example, it does stop calling
> iter_children at some point and display the tree.  In my case, the
> first time the tree is build, it makes two calls to iter_children and
> then displays the tree and waits for something to happen (I still
> think that two is one too many, but I can live with it).  The next
> time, it calls iter_children over 2000 times before it displays the
> tree.
>
> That can't be right!
>
> So, the question remains, "When does the TreeModel stop calling
> iter_children and display the tree?"
>
> It obviously does stop calling iter_children at some point,  I can
> show that. But, what are the circumstances?  What is it looking for?
>
> On Mon, Jun 22, 2009 at 6:28 AM, Chris Camacho wrote:
>   
>> As far as I could tell that's just the way it works *if* you could
>> suspend the calls to iter_children it would stop working...
>>
>> --- On Sat, 20/6/09, Gerald Britton  wrote:
>>
>> 
>>> From: Gerald Britton 
>>> Subject: Re: [pygtk] gtk.TreeModel.iter_children() method, how often should 
>>> it  be called?
>>> To: "Chris Camacho" 
>>> Cc: pygtk@daa.com.au
>>> Date: Saturday, 20 June, 2009, 11:11 PM
>>> I hear you, but I still need a direct
>>> answer to the question, "What
>>> makes the TreeModel stop calling iter_children?  I
>>> can't find a clear
>>> explanation and I can't figure it out by looking at the
>>> calls that
>>> come into my method either.
>>>
>>> On Thu, Jun 18, 2009 at 11:18 AM, Chris Camacho
>>> wrote:
>>>   
 jeeze not being able to just reply is a pain
 
>>> anyhow
>>>   
 I looked into using treemodels connecting to
 
>>> databases
>>>   
 from what I could tell it seems to initially iterate
 
>>> all items
>>>   
 and either copies them or creates meta data about the
 
>>> items
>>>   
 (I didnt look into the implementation)
 Depending on how things were added it would also seem
 
>>> to iterate the whole
>>>   
 set again!
 I think the standard treemodel is ok for small numbers
 
>>> (sub 200)
>>>   
 but I had to end up paging 1,000's of items 100 at a
 
>>> time
>>>   
 (interestingly this is what the gnome db widgets
 
>>> do...)
>>>   
 I've been tempted to weld together my own data grid
 
>>> with a viewport
>>>   
 and moving entry widgets but this would at a guess
 
>>> need C and I haven't
>>>   
 had the time to look into it


 --- On Thu, 18/6/09, Gerald Britton 
 
>>> wrote:
>>>   
> From: Gerald Britton 
> Subject: [pygtk] gtk.TreeModel.iter_children()
>   
>>> method, how often should it be called?
>>>   
> To: pygtk@daa.com.au
> Date: Thursday, 18 June, 2009, 2:50 PM
> Hi -- I think I have a problem but
> I'm not sure since I'm relatively
> new to the TreeModel and how it works.  Here's
>   
>>> the
>>>   
> scenario:
>
> I use TreeModel to display rows of data from a
> database.  Upon
> startup, the TreeModel is built and I notice two
>   
>>> calls to
>>>   
> the
> gtk.TreeModel.iter_children() method.  Later, I
>   
>>> change
>>>   
> some data in
> one of the rows.  The next time the TreeModel is
> built,
> gtk.TreeModel.iter_children() is called thousands
>   
>>> of times
>>>   
> before the
> display is refreshed. Oh, the database has less
>   
>>> than 50
>>>   
> rows of data
> in it!
>
> So, two questions:
>
> 1. Is this normal (I hope not!) behavior?
>
> 2. What tells the TreeModel to stop calling
> gtk.TreeModel.iter_children()?  How does the TM
>   
>>> know
>>>   
> that it has done
> everything it needs to do?
>
> --
> Gerald Britton
> ___
> pygtk mailing list   pygtk@daa.com.au
> http://www.daa.com.au/mailman/listinfo/pygtk
> Read the PyGTK FAQ: http://faq.pygtk.org/
>
>   


 
>>>
>>> --
>>> Gerald Britton
>>>
>>>   
>>
>>
>> 
>
>
>
>   

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] button-press-event on a TreeViewColumn Header?

[John Finlay]

Alessandro Dentella wrote:
> On Wed, Jun 17, 2009 at 08:16:37PM +0430, saeed wrote:
>   
>> I even tried to use a custom class:
>>   class MyTreeviewColumn(gtk.TreeViewColumn, gtk.EventBox):
>> def __init__(self, *args):
>>   gtk.TreeViewColumn.__init__(self, *args)
>>   gtk.EventBox.__init__(self)
>>   self.connect('button-press-event', self.buttonPress)
>>
>> But it doesn't detect 'button-press-event'.
>>
>> Would be the header of treeview was accessible as a separate widget,
>> or the TreeViewColums was a gtk.Widget not only a gtk.Object! :-D
>>
>> The only way that I found is that: if you don't need to resize columns,
>> 
>
> no, I definitely *need* to resize columns...
>
>
> sandro
> *:-)
>
>   
Could you use a MenuToolButton instead of a Button for the header using 
TreeViewColumn.set_widget()?

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Difficulty with rich text in gtk.TextView

[John Finlay]

Dennis Honeyman wrote:
> Hi everyone.  I've been working on a simple rich text component for a
> project I've been working on using a gtk.TextView, and while I have
> something that's usable (in a very loose sense of the word =), there
> are a couple of problems I'm having that I can't figure out how to
> solve.  This is what I have so far...
>
> http://pastebin.com/m6d3eb0b9
> (The toggle buttons and text view are part of a glade file)
>
> For one, if there isn't text selected and the bold, italic, or
> underline button is pressed, I tried to apply the tag
> (buffer.apply_tag_by_name) by using the cursor's position as the start
> and end gtk.TextIter, although it seems to do nothing, as any text I
> type after I apply the tag is still in the same font as before.
>
> Also, when I have the cursor at the last character of a section of
> text with an applied style and I type more text, it doesn't keep the
> formatting like word processors do.
>
> (If text in underscores is bold)
> _This is bold te_
> (But if I keep typing, the formatting isn't kept)
> _This is bold te_xt
> (It should be like this)
> _This is bold text_
>
> I found a rich text demo application in Pygtk while I was searching
> for a solution (http://www.mail-archive.com/pygtk@daa.com.au/msg10283.html),
> but that one has the same limitation as mine. (Tags are not applied to
> text added directly after a section with the tag applied)
>
> And lastly, when the cursor is at the beginning of a block of text
> with an applied style, the gtk.TextIter at the cursor position says
> that the tag for the style is applied, but when I type it shows up in
> the default font
>
> ( '|' is the cursor)
> |_ere is bold text_
> (iter.get_tags() has the bold tag, but when I type something...)
> h_ere is bold text_
> (The tag is not applied)
>
> Why does it say that the bold tag is applied at the cursor if it
> doesn't apply the tag to text typed at that location?
>
> I think I might be missing something fundamental about how tags work... >_<
>
>
> I'm sorry if any of this was unclear.  If anyone could point me in the
> right direction I would be grateful. =)
>
>   
Your application will have to handle the bookkeeping required to manage 
the tags in the TextBuffer. When text is inserted at the cursor your app 
has to catch a signal indicating the text is being inserted and then 
apply the required tags to the inserted text. I think this can be done 
by catching theTextBuffer  "insert-text" signal (using connect_after() 
to make sure the text has already been inserted) and then applying the 
tags to the text in the callback.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Getting string with in pixels

[John Finlay]

vlad...@atlas.cz wrote:
> Walter Leibbrandt  napsal(a) :
>
>   
>> Hi,
>>
>> Vladimír Jícha wrote:
>> 
>>> Hello,
>>>
>>> I'm trying to get width of a text string. I found a FAQ 4.15 
>>> (http://faq.pygtk.org/index.py?req=show&file=faq04.015.htp) which has a 
>>> solution for this. But unfortunately it doesn't work. If I use "font = 
>>> widget.get_style().font", I get following error message: AttributeError: 
>>> 'gtk.Style' has no attribute 'font'.
>>>
>>> Is the FAQ outdated? What syntax should I use?
>>>
>>> Thank yo
>>>   
>> Try widget.get_style().get_font().
>>
>> 
>
> Thanks for your answer. Unfortunately it doens't work correctly for me. I 
> have following 2 problems:
> 1) I get a DeprecationWarning for the line.
> 2) No matter what fond and size I use, I get the same width.
>
> Sample code:
>
> label = gtk.Label()
> name_font = pango.FontDescription("Times 10")
> label.modify_font(name_font)
> font = label.get_style().get_font()
> label2 = gtk.Label()
> name_font2 = pango.FontDescription("Sans Bold 12")
> label2.modify_font(name_font2)
> font2 = label2.get_style().get_font()
> print font.width("string"), font2.width("string")
>
> I always get the same numbers. Is something wrong in my code?
>   
You should use the PangoLayout to retrieve the pixel size of the text. Use:

label.get_layout().get_pixel_size()

to get the width and height of the label string in pixels.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Fwd: Question

[John Finlay]

Johan Dahlin wrote:
> -- Forwarded message --
> From: Rodrigo Almeida 
> Date: Tue, Feb 10, 2009 at 6:37 PM
> Subject: Question
> To: pygtk-...@async.com.br
>
>
> Hello Friends,
>
> I'm a programmer from Brasil, and i am doing an application using pygtk.
> I would like to know if you can help me. My question is: is there a
> way to use an image as a cursor?
>
> something like this:
>
> self.image.set_from_file("myimage.png")
> self.Tela.window.set_cursor(self.image)
>
> I hope you can help me.
>
> Greetings.
>
> Rodrigo Almeida.
>   
pb = gtk.gdk.pixbuf_new_from_file("myimage.png")
c = gtk.gdk.Cursor(gtk.gdk.display_get_default(), pb, 0, 0)
widget.window.set_cursor(c)

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Simple question on reusing stock item

[John Finlay]

Alessandro Dentella wrote:
> Hi,
>
>   i'd like to use stock item in menu entries with modified labels to better
>   suit context.
>
>   *  MenuItem doesn't allow to use images (reasonable)
>   *  ImageMenuItem doesn't appearently allow to change label (really???)
>
>   I think I could go with IconFactory but it seems to me it's more complex
>   than what is should be and I wandered if there is a simple way I was not
>   able to work out.
>
>   Thanks
>   
>   sandro
>   *:-)
>
>   
A MenuItem is a Container so you can add anything to it. In your case 
create an empty MenuItem and add an HBox containing an Image and a Label.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] synchronizing HPaned widgets

[John Finlay]

Stephen Langer wrote:
> On Feb 5, 2009, at 3:15 PM, John Finlay wrote:
>
>   
>> Stephen Langer wrote:
>> 
>>> Hi --
>>>
>>> I'm hoping that someone can tell me if this is a bug or if I'm  
>>> doing  something wrong.
>>>
>>> I have two HPaned widgets whose separators I'd like to keep   
>>> synchronized.  Both Paneds are in the same VBox, so they have the  
>>> same  width.
>>>
>>> As described at http://faq.pygtk.org/index.py?req=show&file=faq19.013.htp 
>>>  , I connect to the 'notify' signal for each pane to find out when  
>>> the  separator has been moved.  The callback function blocks the  
>>> signal for  the other  pane, changes its position to match the  
>>> first pane, and  then unblocks the signal.
>>>
>>> The problem is that the signal doesn't seem to be blocked  
>>> sometimes,  so the program goes into an infinite loop.  There are  
>>> easy workarounds  for this, but they're ugly and unsatisfying.  In  
>>> particular, the  program loops when the panes are first shown if  
>>> the contents of two  diagonally opposite subpanes are larger than  
>>> the contents of the other  two.
>>>
>>> I've appended short program that illustrates the problem.  Just   
>>> starting the program will put it into an infinite loop.  If you   
>>> replace paneMoved by paneMoved_workaround, the problem goes away.
>>>
>>> Why doesn't blocking the signal work?  Am I doing something wrong?
>>>
>>> I've run this test on OS X with python 2.5, gtk 2.14.7, and pygtk   
>>> 2.12.1, and on Linux with python 2.4, gtk 2.8.20, and pygtk 2.8.6.
>>>
>>>
>>>   
>
> [ code deleted ]
>
>   
>> You have to allow the widgets to shrink - remove the "shrink=0" args  
>> from the pack1() and pack2() calls.
>>
>> John
>> 
>
> Thanks for the response, but what if I don't want the widgets to be  
> able to shrink?  Is that possible?  Why isn't calling handler_block  
> before calling set_position sufficient?
>
>   -- Steve
>   
Yes but you have to set the default size of the window as large or 
larger than the required size of the largest pane. Your code doesn't set 
the window size so it starts up with the minimum required size to fit 
the paned widgets. Since the two separators can't line up because the 
panes are different sizes and because the panes can't shrink the 
set_position() calls fail and apparently trigger a resize calculation 
which trys to set the position and so on. Your choices are allow 
something to shrink or set a window size that will allow the separators 
to line up initially.

Also you can add a detail to the "notify" signal connection that only 
signals when the position is changed - use "notify::position" and you 
can remove the gparamspec.name check in paneMoved().

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] synchronizing HPaned widgets

[John Finlay]

Stephen Langer wrote:
> Hi --
>
> I'm hoping that someone can tell me if this is a bug or if I'm doing  
> something wrong.
>
> I have two HPaned widgets whose separators I'd like to keep  
> synchronized.  Both Paneds are in the same VBox, so they have the same  
> width.
>
> As described at http://faq.pygtk.org/index.py?req=show&file=faq19.013.htp 
> , I connect to the 'notify' signal for each pane to find out when the  
> separator has been moved.  The callback function blocks the signal for  
> the other  pane, changes its position to match the first pane, and  
> then unblocks the signal.
>
> The problem is that the signal doesn't seem to be blocked sometimes,  
> so the program goes into an infinite loop.  There are easy workarounds  
> for this, but they're ugly and unsatisfying.  In particular, the  
> program loops when the panes are first shown if the contents of two  
> diagonally opposite subpanes are larger than the contents of the other  
> two.
>
> I've appended short program that illustrates the problem.  Just  
> starting the program will put it into an infinite loop.  If you  
> replace paneMoved by paneMoved_workaround, the problem goes away.
>
> Why doesn't blocking the signal work?  Am I doing something wrong?
>
> I've run this test on OS X with python 2.5, gtk 2.14.7, and pygtk  
> 2.12.1, and on Linux with python 2.4, gtk 2.8.20, and pygtk 2.8.6.
>
> Thanks,
>Steve
>
>
>
> import gtk
>
> def setPos(pane, signal, pos):
>  "Set a Paned position without propagating signals"
>  pane.handler_block(signal)
>  pane.set_position(pos)
>  pane.handler_unblock(signal)
>
> def paneMoved(pane, gparamspec):
>  "Callback for synchronizing Paned position changes"
>  if gparamspec.name == 'position':
>  pos = pane.get_position()
>  if pane is topPane:
>  print "top"
>  setPos(botPane, botSignal, pos)
>  else:
>  print "bottom"
>  setPos(topPane, topSignal, pos)
>
> count = 0
> def paneMoved_workaround(pane, gparamspec):
>  "Callback for synchronizing Paned position changes"
>  global count
>  if gparamspec.name == 'position':
>  count += 1
>  pos = pane.get_position()
>  if count == 1:
>  if pane is topPane:
>  print "top"
>  setPos(botPane, botSignal, pos)
>  else:
>  print "bottom"
>  setPos(topPane, topSignal, pos)
>  elif count > 1:
>  count = 0
>
> window = gtk.Window()
> box = gtk.VBox()
> window.add(box)
> window.connect("delete-event", gtk.main_quit)
>
> # Create two panes, one on top of the other
> topPane = gtk.HPaned()
> box.pack_start(topPane, expand=1, fill=1)
> botPane = gtk.HPaned()
> box.pack_start(botPane, expand=1, fill=1)
>
> # Synchronize the pane positions
> topSignal = topPane.connect('notify', paneMoved)
> botSignal = botPane.connect('notify', paneMoved)
>
> # If the displayed sizes of shortstring and longstring are different,
> # the program will enter an infinite loop after show_all().  If the
> # strings are identical, the program behaves correctly (no infinite
> # loop, the panes can be resized and stay synchronized).
> shortstring = "string"
> longstring = "long string"  # Try "String" too.
>
> # Put each string into two diagonally opposite labels in the panes.
> topPane.pack1(gtk.Label(longstring), resize=1, shrink=0)
> topPane.pack2(gtk.Label(shortstring), resize=1, shrink=0)
>
> botPane.pack1(gtk.Label(shortstring), resize=1, shrink=0)
> botPane.pack2(gtk.Label(longstring), resize=1, shrink=0)
>
> window.show_all()
>
> gtk.main()
>
>   
You have to allow the widgets to shrink - remove the "shrink=0" args 
from the pack1() and pack2() calls.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Treeview: How to set Font Size in Column Titles?

[John Finlay]

darethehair wrote:
> Hello all!
>
>  From the PyGTK docs, I know the various ways to change the font size in 
> 'treeview' columns/cells, but they appear to only work on the column 
> *data*, and not the column *titles*.  The effect of this is that I can 
> make my fonts (for example) smaller and get more rows in the screen, but 
> the title font sizes do not change, and so it looks silly.
>
> Any tips?
>   
Try using a custom widget for the header using TreeViewColumn.set_widget()

JOhn
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] TreeView selection issue

[John Finlay]

A minor nit. The cursor is not the same as the selection. To see the 
difference press Control-n. The cursor is identified by the outlined 
box. Of course it's usually the same as the selection.

John

Alan F wrote:
>
>
> > Is there some reason why you are not using the TreeSelection "changed"
> > signal to track the selction changes and update your display? In your
> > case I assume that only one selection at a time is allowed so the
> > "changed" signal would seem to be ideal.
> >
> > John
>
> Thanks John. Finally got it working.
>
> If anybodys interested, I've just created a program that displays the 
> tags of mp3 files in a selected folder. It's quite basic as you have 
> to enter path into script file.
>
>
> # # Start of File
> #!/usr/bin/env python
>
> # tag-tree.py
>
> import pygtk
> pygtk.require('2.0')
> import gtk
> import os
> import glob
> from mutagen.id3 import ID3, TPE2, TIT2, TPE1, TALB, TCOM, TDRC, TCON, 
> TRCK, TPOS
>
> MUSIC_DIR = "ENTER MUSIC FOLDER PATH HERE"
>
> class tag_tree:
>
> # close the window and quit
> def delete_event(self, widget, event, data=None):
> gtk.main_quit()
> return False
>
> def add_data(self):
>
> os.chdir(MUSIC_DIR)
> file_list = glob.glob('*.mp3')
> first_time = True
>
> if len(file_list) == 0:
> print "No mp3 files in folder"
> else:
> file_list.sort()
> for x in file_list:
> curr_iter = self.liststore.append([x,x] )
> if first_time == True:
> first_iter = curr_iter
> first_time = False
> return first_iter
>
> def change_details(self, widget, txt_area):
>  #get data from highlighted selection  
> treeselection = self.treeview.get_selection()
>
> (model, iter) = treeselection.get_selected()
> if iter != None :
> # Add audio data
> txt_buffer = gtk.TextBuffer()
> file_name = self.liststore.get_value(iter, 0)
> audio = ID3(file_name)   
> if audio.has_key('TIT2'):
> txt_buffer.insert_at_cursor("Title: " + 
> unicode(audio["TIT2"]) + '\n')
> if audio.has_key('TPE1'):
> txt_buffer.insert_at_cursor( "Artist: " + 
> unicode(audio["TPE1"]) + '\n')
> if audio.has_key('TCOM'):
> txt_buffer.insert_at_cursor("Composer: " + 
> unicode(audio["TCOM"]) + '\n')
> if audio.has_key('TPE2'):
> txt_buffer.insert_at_cursor("AlbumArtist: " + 
> unicode(audio["TPE2"]) + '\n')
> if audio.has_key('TALB'):
> txt_buffer.insert_at_cursor("Album: " + 
> unicode(audio["TALB"]) + '\n')
> if audio.has_key('TCON'):
> txt_buffer.insert_at_cursor("Genre: " + 
> unicode(audio["TCON"]) + '\n')
> if audio.has_key('TDRC'):
> txt_buffer.insert_at_cursor("Year: " + 
> unicode(audio["TDRC"]) + '\n')
> if audio.has_key('TRCK'):
> txt_buffer.insert_at_cursor("Track num: " + 
> unicode(audio["TRCK"]) + '\n')
> if audio.has_key('TPOS'):
> txt_buffer.insert_at_cursor("Disc num: " + 
> unicode(audio["TPOS"]) + '\n')   
> txt_area.set_buffer(txt_buffer)
>
>  def __init__(self):
> # Create a new window
> self.window = gtk.Window(gtk.WINDOW_TOPLEVEL)
>
> self.window.set_title("Tag Display")
> self.window.set_size_request(500, 200)
> self.window.connect("delete_event", self.delete_event)
>
> table1 = gtk.Table(1, 2, False)
> table1.set_row_spacings( 5)
> table1.set_col_spacings(5)
> self.window.add(table1)
>   
> # create a ListStore with one string column to use as the model
> self.liststore = gtk.ListStore(str, str)
> first_iter = self.add_data()
>
> # create the TreeView using liststore
> self.treeview = gtk.TreeView(self.liststore)
> self.treeview.set_headers_visible(False)
>
> # create the TreeViewColumn to display the data
> self.tvcolumn = gtk.TreeViewColumn()
>
> # add tvcolumn to treeview
> self.treeview.append_column(self.tvcolumn)
>
> # create a CellRendererText to render the data
> self.cell = gtk.CellRendererText()
>
> # add the cell to the tvcolumn and allow it to expand
> self.tvcolumn.pack_start(self.cell, True)
>
> # set the cell "text" attribute to column 0 - retrieve text
> # from that column in treestore
> self.tvcolumn.add_attribute(self.cell, 'text', 0)
>
> # make it searchable
> self.treeview.set_search_column(0)
>
>
>  # Allow drag and drop reordering of rows
> self.treeview.set_reorderable(False)
>
> # values colx, coly, roxx, r

Re: [pygtk] Treeview/model: Problems with Expanding and Selecting Rows

[John Finlay]

darethehair wrote:
> :
>
> I do not understand why my mouse clicks on the 'expansion' arrow is
> interpreted as an entire 'row selection' instead :(
>   
Bug in your program? On my system clicking the expander does not change 
the seslection.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Treeview/model: Problems with Expanding and Selecting Rows

[John Finlay]

darethehair wrote:
> Hello Folks!
>
> The PyGTK tutorial example named 'basictreeview.py' works fine for me.
> When I click on the 'expander' arrow for a row, it instantly expands to
> show the child rows under it.
>
> In my case, I am using a 'lazy treeview' to only populate the child rows
> as I need to do so -- mostly because of time, but also because I want to
> allow unlimited 'nesting'.  My problem is two-fold:
>
> 1) When I use a 'row-expanded' signal detector, I have to click two or
> three times on the 'expander' arrow in order for the child rows to
> actually show up.  In fact, out of desperation I have even added the
> following to my routine to 'force' the issue:
>
> self.treeview.expand_to_path(path)
> self.treeview.expand_row(path, True)
>
> 2) When I add a selection 'changed' signal, it works, but it 'gets in
> the way' of my 'row-expanded' detector i.e. even if I am very careful to
> click just the 'expand arrow' on a row, it triggers my 'changed'
> selection routine the first time I touch a row.
>
> I would appreciate any tips to produce the desired behavior! :)
>   
Connect to the "test-expand-row" signal and add the rows in the 
callback. You have to put a dummy child row to get the expander to show 
and replace it with the real children in the callback.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] TreeView selection issue

[John Finlay]

Alan F wrote:
> Hi
>
> I am new to GTK programming. I want to create a program that displays 
> the sub-folders of a main folder in a tree view format. Once I find 
> the particular sub-folder , I want to be able to click on it to 
> display the contents of that folder. When the contents of that folder 
> is displayed, I want to click on any file to display the mp3 tags of 
> that file.
>
> I have managed to get a program working that displays all the 
> sub-folders of the main folder as a treeview. The problem is I am 
> having trouble trying to get it to do the next step i.e. when I click 
> on a sub-folder I want to display the contents of that folder.
>
> After googling around, I found a post that used a solution of adding a 
> button event to the treeview. I put it in an example code below and it 
> works but with a problem. 
>
> Firstly, to describe the actions of the example code, I've created a 
> treeeview that has 3 parents and 3 childs, when I click on a selection 
> (either parent or child), the value of that selection is printed in 
> the console.
>
> The problem is that when I click on any selection, nothing happens. 
> When I click on it again, the value of that selection is displayed. 
> When I click on another selection, the value of the previous selection 
> is displayed. When I click on that selection again, the values are 
> displayed. On clicking on any selection, the value of the previous one 
> is displayed.
>
> I've put console print statements throughout the code and the contents 
> are definately being displayed when I single click on selection. I can 
> only think it is something to do with the lines  treeselection = 
> self.treeview.get_selection(), (model, iter) = 
> treeselection.get_selected().
>
> I've checked the API and in TreeSelection there is text that may have 
> something to do my issue:
>
> "One of the important things to remember when monitoring the selection 
> of a view is that the "changed" signal is mostly a hint. That is, it 
> may only emit one signal when a range of rows is selected. 
> Additionally, it may on occasion emit a "changed" signal when nothing 
> has happened (mostly as a result of programmers calling the 
> |select_path|() 
> 
>  
> or |select_iter|() 
> 
>  
> methods on an already selected row)."
>
> I don't really know enough about GTK programming to understand where 
> to start to look to fix this. Can someone please point me in the right 
> direction?
>
>
> # ###Start of file
> #!/usr/bin/env python
>
>
> import pygtk
> pygtk.require('2.0')
> import gtk
> import os
>
> class BasicTreeViewExample:
>
> # close the window and quit
> def delete_event(self, widget, event, data=None):
> gtk.main_quit()
> return False
>
> def add_data(self):
>
> # we'll add some data now - 4 rows with 3 child rows each
> for parent in range(4):
> piter = self.treestore.append(None, ['parent %i' % 
> parent,  'value %i' %parent])
> #print "this is path of parent " + str(parent)
> #print self.treestore.get_path(piter)
>
> for child in range(3):
> citr = self.treestore.append(piter, ['parent  
> %i child %i' % (parent, child, ),  'value %i' %parent])
> #print "this is path of parent " + str(parent) + "child " 
> + str(child)
> #print self.treestore.get_path(citr)
>
>
> def selectTest(self, widget, event):
> print "This is event button"
> print event.button
> # Mouse press confirmed ok
> if event.button == 1:
>  
> #get data from highlighted selection  
> treeselection = self.treeview.get_selection()
>
> (model, iter) = treeselection.get_selected()
> if iter != None :
> print "this is path of selected "
>
> print self.treestore.get_path(iter)
> print "value or select store"
> print self.treestore.get_value(iter, 0)
> print model.get_value(iter, 0)
>
> else:
> print "no selection made"
>
> def __init__(self):
> # Create a new window
> self.window = gtk.Window(gtk.WINDOW_TOPLEVEL)
>
> self.window.set_title("Music Collection")
>
> self.window.set_size_request(300, 200)
>
> self.window.connect("delete_event", self.delete_event)
>   
> # create a new scrolled window.
> scrolled_window = gtk.ScrolledWindow()
> scrolled_window.set_border_width(10)
>
> # the policy is one of POLICY AUTOMATIC, or POLICY_ALWAYS.
> # POLICY_AUTOMATIC will automatically decide whether you need
> # scrollbars, whereas POLICY_ALWAYS will always leave the 
> scrollbars
> # there. The first one is the horizontal scrollbar, the 
> second, the
> # vertical.

Re: [pygtk] TreeView/cell/signals

[John Finlay]

Alessandro Dentella wrote:
> ... the same problem, a different point of view...
>
> If I enter a CellRendererText in editing mode I can press return and the
> mode switched and signal 'edited'is emitted.
>
> Who is responsable for that?  I guess is the editable as I can't imagine
> anything else... but how can I be sure? I mean is there a way to "listen all
> signals" and who is emitting them?
>
> Which is the signal emitted? 
>
> What can I do from within the 'edited' callback to stop switching from
> editing mode? 
>
> sandro
>
>
>   
The signals are described in the Reference Manual but sometimes can be 
hard to find. The devhelp application can help find them because it 
provides a search mode.

If I understand your question I don't think that you can prevent the 
Entry that is used as the CellEditable for the CellRendererText from 
closing when activated. You can only prevent the edits from being 
applied which is the default.

John

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Updating a TextBuffer line by line

[John Finlay]

Daniel Roesler wrote:
> Ok, I tried that, but it's giving an error about the number of
> arguments sent to read_output. "TypeError: display_details() takes
> exactly 2 arguments (3 given)"
>
> Here's my code:
> --
>   gobject.io_add_watch(command.stdout, gobject.IO_IN | gobject.IO_HUP,
> self.read_output)
> 
>
> def read_output(source, condition):
>   if condition == gobject.IO_IN:
> line = source.readline()
> self.txtbuffer.insert_at_cursor(line)
>   if condition == gobject.IO_HUP:
> self.txtbuffer.insert_at_cursor("Command finished.")
> return False
>   return True
> --
>
> Any ideas on why this is occurring?
>   
You passed io_add_watch a class method as the callback - the first arg 
of the classmethod must be self so prepend self to the method definition 
params:

def read_output(self, source, condition):

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Updating a TextBuffer line by line

[John Finlay]

Daniel Roesler wrote:
> Howdy again,
>
> I am running into a problem displaying stdout from a subprocess
> command. I have a loop that checks to see if the subprocess is still
> active, then reads a line in from stdout and sends it to the text
> buffer. However, I can't seem to get my text buffer to print except
> when the command ends. Obviously, I'm missing some logic behind how
> loops work with pygtk.
>
> Here's my code:
> ---
> command = subprocess.Popen(cmd, stdout=subprocess.PIPE)
>
> while command.poll() is None:
>   line = command.stdout.readline()
>   self.txtbuffer.insert_at_cursor(line)
> ---
>
> Any ideas on how to read stdout line by line with subprocesses?
>
>   
Don't use command.poll() rather use gobject.io_add_watch to register a 
callback when data is available on the stdout pipe or the pipe is 
closed. Something like:

gobject.io_add_watch(command.stdout, gobject.IO-IN | gobject.IO_HUP, 
read_output)

def read_output(source, condition):
   if condition == gobject.IO_IN:
   line = source.readline()
   
   if condition == gobject.IO_HUP:
  
  return False

   return True


John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] TreeViewColumn sizing

[John Finlay]

The "expand" property of a TreeViewColumn is False by default. You have 
to set_expand(True)  the first two columns in your TreeView.

John

Daniel Hernández Bahr wrote:
> this is the thing:
>
> def init_files():
> for column in self.updates_view.get_columns():
> self.updates_view.remove_column(column)
> 
> headers = [_("New File"),_("Old File"), _("Revision")]
> index = 0
> 
> for header in headers[:3]:
> index += 1
> column = gtk.TreeViewColumn(header)
> self.updates_view.append_column(column)
> cell = gtk.CellRendererText()
> column.pack_start(cell, True)
> column.add_attribute(cell, 'text', index)
> column.set_sort_column_id(index)
>
> column = gtk.TreeViewColumn(_("Edit"))
> self.updates_view.append_column(column)
> cell = gtk.CellRendererPixbuf()
> column.pack_start(cell)
> column.add_attribute(cell, 'stock_id', 4)
> column.set_name('edit_column')
> column.set_expand(False)
> 
> selection = self.updates_view.get_selection()
> selection.set_mode(gtk.SELECTION_MULTIPLE)
> selection.connect('changed', self.on_files_selection_changed)
>
> - Original Message -
> From: "John Stowers" 
> To: "Daniel Hernández Bahr" 
> Cc: "pygtk" 
> Sent: Wednesday, January 14, 2009 4:29:07 PM (GMT-0500) Auto-Detected
> Subject: Re: [pygtk] TreeViewColumn sizing
>
> On Wed, 2009-01-14 at 16:34 -0500, Daniel Hernández Bahr wrote:
>   
>> Should, but won't .. 
>> 
>
> Please provide your code so we may take a look
>
> Regards,
>
> John
>
>   
>> best regards .. 
>>
>> - Original Message -
>> From: "Pietro Battiston" 
>> To: "pygtk" 
>> Sent: Wednesday, January 14, 2009 2:10:43 PM (GMT-0500) Auto-Detected
>> Subject: Re: [pygtk] TreeViewColumn sizing
>>
>> Il giorno mer, 14/01/2009 alle 15.50 -0500, Daniel Hernández Bahr ha
>> scritto:
>> 
>>> Hullo, everyone!
>>>
>>> my issue is, I have a TreeView with 4 Columns, and it is my interest
>>> that the last 2 should have the minimal size required (as the last one
>>> is a small pixbuf and the one before that a 4 digit number), and let
>>> the first two to get all the size they need (as they represent file
>>> paths).
>>>
>>> I have tried two set fixed sizes already for the small ones, but it
>>> doesn't work.
>>>   
>> column.set_expand(False) for the two small should do the job
>>
>> Pietro
>>
>> ___
>> pygtk mailing list   pygtk@daa.com.au
>> http://www.daa.com.au/mailman/listinfo/pygtk
>> Read the PyGTK FAQ: http://faq.pygtk.org/
>>
>> -- 
>>
>> Daniel Hernández Bahr 
>> Universidad de las Ciencias Informáticas. 
>>
>> ___
>> pygtk mailing list   pygtk@daa.com.au
>> http://www.daa.com.au/mailman/listinfo/pygtk
>> Read the PyGTK FAQ: http://faq.pygtk.org/
>> 
>
>
>
>
>   

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Entry.connect_after('changed', ...) doesn't see updated cursor position

[John Finlay]

Yang Zhang wrote:
> Hi, I have a signal handler connected to an Entry's 'changed' signal, 
> but it never sees the updated cursor position, only the previous cursor 
> position.  E.g., if I start with an empty Entry and then type 'a', 
> Entry.get_text() returns 'a', but Entry.get_position() returns 0, not 1. 
>   I tried both .connect() and .connect_after(), but there doesn't seem 
> to be any difference.  How can I get the updated cursor position?
>   
Connect to the "notify::cursor-position" signal. This should work for 
your app.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problem with Notebook - stop-emission doubts

[John Finlay]

Alessandro Dentella wrote:
> On Wed, Dec 24, 2008 at 01:22:49PM +0100, Gian Mario Tagliaretti wrote:
>   
>> On Wed, Dec 24, 2008 at 9:57 AM, Frédéric
>>  wrote:
>>
>> 
>>> PS: BTW, where did you find the stop_emission() method? I can't retreive it
>>> in the doc...
>>>   
>> it a gobject method:
>> http://library.gnome.org/devel/pygobject/stable/class-gobject.html#method-gobject--stop-emission
>> 
>
> following the docs I have several questions?
>
>   1. the explanation is exactly the same as for "emit_stop_by_name": which
>  is the difference? 
>
>   
They are the same.
> """
>   The stop_emission() method stops the current emission of the signal
>   specified by detailed_signal
> """
>
>   2. "current emission" means the emission of the signal that is now
>  handled? if this is the case, what's the need for specifying the name?
>   
A signal could be emitted within a signal emission.
>   3. How is it different from returning True? 
>  I tested it with::
>
> b = gtk.Button("Press me")
> e = gtk.Entry()
> b.connect('clicked', self.on_clicked1)
> b.connect('clicked', self.on_clicked2)
>
> e.connect('key-press-event', self.on_key_press_event1)
> e.connect('key-press-event', self.on_key_press_event2)
>
>
>  on_clicked2 cannot be inhibited returning True from on_clicked1 while 
>  on_key_press_event1 *can* be inhibited simply returning True.
>
>  Both can be inhibited using stop_emission.
>  
Some signals provide an explicit means of stopping the signal emission 
using a return value from a handler. The "clicked" handler is not 
supposed to provide a return value so returning True does nothing.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problem with Notebook

[John Finlay]

Frédéric wrote:
> On mercredi 24 décembre 2008, John Finlay wrote:
>
>   
>> I don't see how that would happen. I guess a small self-contained
>> example would illustrate the problem.
>> 
>
> Here is a little example...
>
> I block switching to the Page 2, with your trick. The first time you select 
> that page, the callback is called, but the page is not displayed. Good.
>
> But then, if you click again on that page, the callback is not called 
> anymore, until you switch to another page first (or just click on the 
> current selected one). It acts like if this page was selected, but not 
> displayed. Also note that if you click the currnt page again, the callback 
> is not called neither. But is change internal stuff, as it calls the 
> callback when clicking on Page 2!
>   
The current page number hasn't changed so it seems that it remembers the 
last clicked tab.
> Sounds like a bug to me...
>   
Maybe. Might be a good idea to ask about this on the gtk mailing list.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problem with Notebook

[John Finlay]

Frédéric wrote:
> On mercredi 24 décembre 2008, John Finlay wrote:
>
>   
>> Since you just want to stay on the same page (assuming you mean the
>> currently displayed page) then you should be able to stop the emission
>> of the 'switch-page' signal using:
>>
>> self.notebook.stop_emission('switch-page')
>>
>> which will prevent the page change.
>> 
>
> Ok, it works, thanks :)
>
> There is just a strange behaviour: once I have clicked on the blocked page 
> and showed the warning, then if I click again on the same page, 
> the 'switch-page' is not emitted anymore. I have to click on the current 
> page, and then click again on the other one.
>   
> It seems that the internal page_num var. is set to the new page. Blocking 
> the event just prevents to visualy switch to that page, but internally, 
> PyGTK thinks that he is on that page... I tried to use set_current_page(), 
> but it does not change anything.
>   
I don't see how that would happen. I guess a small self-contained 
example would illustrate the problem.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Problem with Notebook

[John Finlay]

Frédéric wrote:
> Hello,
>
> I use a Notebook widget. I added the following callback to 
> the 'switch-page' signal:
>
> def __onNoteBookSwitchedPage(self, widget, page, page_num):
> if page_num == 0 and self._model.camera.lens.type_ == 'fisheye':
> controller = WarningMessageController(_("Warning"),
>   _("Can't set shooting mode"))
> controller.run()
> self.notebook.set_current_page(1) # Does not work!!!
> else:
> if page_num == 0:
> self._model.mode = 'mosaic'
> else:
> self._model.mode = 'preset'
>
> As you can see, depending of a config. parameter 
> (self._model.camera.lens.type_), I show a warning saying that the selected 
> TAB can't be used. I would like to stay on the current tab, but as soon as 
> the warning is closed, the selected tab is shown; the 
> self.notebook.set_current_page(1) call does not seem to work. Why ? If I 
> change some widgets, here, it works; why not the Notebook?
>
> Thanks,
>
>   
Since you just want to stay on the same page (assuming you mean the 
currently displayed page) then you should be able to stop the emission 
of the 'switch-page' signal using:

self.notebook.stop_emission('switch-page')

which will prevent the page change.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] hide on delete_event()

[John Finlay]

Frédéric wrote:
> On vendredi 19 décembre 2008, Frédéric wrote:
>
>   
>>> Then don't use run(), connect a callback to the "response" signal and
>>> put all the dialog processing in that. You still need to handle the
>>> "delete-event" signal to prevent the dialog closing and put the
>>> destroy() in the "response" signal handler assuming the close button
>>> is part of the dialog; if not then put it in the button's "clicked"
>>> handler.
>>>   
>> Ok, I got it!
>> 
>
> Things work fine without the run() call...
>
> A last question: is it possible to simulate the run() call, to get a 
> response id, as I need to make different things depending which button 
> closed the dialog?
>
>   

That's what connecting a callback to the "response" signal allows you to do.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] hide on delete_event()

[John Finlay]

Frédéric wrote:
> On vendredi 19 décembre 2008, John Finlay wrote:
>
>   
>>> dialog.connect("destroy", destroy)
>>>   
>> You probably mean to connect to the "destroy-event" signal in the above
>> though it's not likely that you need to handle it.
>> 
>
> The event is called 'delete', and not 'delete-event'...
>   
Check the docs for the gtk.Widget to see the signals:
http://pygtk.org/docs/pygtk/class-gtkwidget.html#signal-prototypes-gtkwidget

Both "delete-event" and "destroy-event" are Widget signals. "destroy" is 
a gtk.Object signal.
>>> dialog.destroy()
>>>   
>> The above line is what is destroying your dialog - remove it.
>> 
>
> Yes, but I need to destroy it once I leave it (but only using a special 
> close button, not shown here)...
>
>   
>> You are mixing the two ways of using a dialog - show() and run(). You
>> should only use one of these ways. If you use the show() method then add
>> a call in delete() to hide() the dialog when you are finished the delete
>> processing. If you use the run() method then you don't need to handle
>> the delete-event and you should retrieve the return response from the
>> run() method to figure out what to do (you also don't need to make the
>> dialog modal since run() blocks the UI until it returns).
>> 
>
> I forgot to remove the show() call from my previous tests. But it does not 
> work neither. As soon as I click on the X dialog, the run() returns. This 
> is not what I want. I really want to avoid the dialog to be ended; I want 
> to stay in its loop...
>   
Then don't use run(), connect a callback to the "response" signal and 
put all the dialog processing in that. You still need to handle the 
"delete-event" signal to prevent the dialog closing and put the 
destroy() in the "response" signal handler assuming the close button is 
part of the dialog; if not then put it in the button's "clicked" handler.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] hide on delete_event()

[John Finlay]

Frédéric wrote:
> Le 19/12/2008, "John Finlay"  a écrit:
>
>   
>> Frédéric wrote:
>> 
>>> It seems that all documentation related to that point refer to gtk.Window
>>> widgets, and it works fine. But if the main window opens a gtk.Dialog,
>>> then, it is impossible to avoid this dialog to be destroyed when hitting
>>> its X box: even if event-delete callback returns True, the signal is
>>> propagated, and the destroy signal is emited.
>>>   
>> Could you create a small, self-contained program to illustrate the
>> problem you are having?
>> 
>
> Sure:
>
> import gtk
>
> def delete(widget, event):
> print "delete()"
> return True # Should not propagate delete-event
>
> def destroy(widget):
> print "destroy()"
> return True
>
> def clicked(widget):
> print "clicked()"
> dialog = gtk.Dialog("My dialog", win,
>  gtk.DIALOG_MODAL,
>  (gtk.STOCK_OK, gtk.RESPONSE_ACCEPT))
> dialog.set_size_request(200, 50)
> dialog.show()
> dialog.connect("delete-event", delete)
> dialog.connect("destroy", destroy)
>   
You probably mean to connect to the "destroy-event" signal in the above 
though it's not likely that you need to handle it.
> dialog.run()
> dialog.destroy()
>   
The above line is what is destroying your dialog - remove it.
> win = gtk.Window()
> win.set_size_request(200, 50)
> button = gtk.Button("Open dialog...")
> button.connect("clicked", clicked)
> win.add(button)
> win.connect("delete-event", gtk.main_quit)
> win.show_all()
> gtk.main()
>
>   
You are mixing the two ways of using a dialog - show() and run(). You 
should only use one of these ways. If you use the show() method then add 
a call in delete() to hide() the dialog when you are finished the delete 
processing. If you use the run() method then you don't need to handle 
the delete-event and you should retrieve the return response from the 
run() method to figure out what to do (you also don't need to make the 
dialog modal since run() blocks the UI until it returns).

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] hide on delete_event()

[John Finlay]

Frédéric wrote:
> Le 5/12/2008, "Christian Becke"  a écrit:
>
>   
>> Peyman schrieb:
>> 
>>> I have tried every variation, but I will try it once again...Nope,
>>> still doesn't work. I tried returning: True, False, 0, 1, gtk.TRUE,
>>> and gtk.FALSE
>>>   
>> This works for me (python 2.5.2, pygtk 2.13.0, gtk 2.14.4):
>> 
>
> I'm trying to block the delete event when hitting the X box of a dialog.
>
> It seems that all documentation related to that point refer to gtk.Window
> widgets, and it works fine. But if the main window opens a gtk.Dialog,
> then, it is impossible to avoid this dialog to be destroyed when hitting
> its X box: even if event-delete callback returns True, the signal is
> propagated, and the destroy signal is emited.
>
> Why? Is there a way to avoid that?
>
>   
Frederic,

Could you create a small, self-contained program to illustrate the 
problem you are having?

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Unable to run your tutorial examples

[John Finlay]

A.K.Karthikeyan wrote:
> Hello PyGTK people,
>
> I have tried to run *pygtkconsole.py* 
>  and 
> *gpython.py*  
> found in http://www.pygtk.org/pygtk2tutorial/ch-Introduction.html on my 
> Ubuntu 8.04, but I got the following error message
>
> 
> Traceback (most recent call last):
>   File "pygtkconsole.py", line 14, in 
> import gtk
>   File "/media/sda7/python/py/gtk.py", line 2
> SyntaxError: Non-ASCII character '\xe2' in file 
> /media/sda7/python/py/gtk.py on line 2, but no encoding declared; see 
> http://www.python.org/peps/pep-0263.html for details
>
> and
>
> Traceback (most recent call last):
>   File "gpython.py", line 16, in 
> import gtk, gobject
>   File "/media/sda7/python/py/gtk.py", line 2
> SyntaxError: Non-ASCII character '\xe2' in file 
> /media/sda7/python/py/gtk.py on line 2, but no encoding declared; see 
> http://www.python.org/peps/pep-0263.html for details
>
> I am clueless what to even after visiting 
> http://www.python.org/peps/pep-0263.html as there are no non-ASCII 
> characters in program. Can any one help me?
>   
The problem is the file "/media/sda7/python/py/gtk.py" that interferes 
with the importing of the gtk module. If possible you should rename this 
file or make sure it isn't in the import path of python.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] TreeView - Making whole row colored, Again

[John Finlay]

Kevin wrote:
> Greetings,
>
> I came across an old message in the archives at 
> http://www.mail-archive.com/pygtk@daa.com.au/msg15480.html , which sounds a 
> lot like I want to do (though I want to work with background colors rather 
> than foreground colors).  But I was thinking: couldn't I just add a column to 
> my model containing the color values I want for each row, and use values of 
> #ff for all the rows I wanted to leave white?
>
> The problem is that without a column of booleans, I can't seem to set the 
> background-set property of the columns.  Specifically, I'm using:
>
> leftjust = gtk.CellRendererText()
> column = gtk.TreeViewColumn('Filename', leftjust, text=COLUMN_FILENAME)
> column.set_sort_column_id(COLUMN_FILENAME)
> treeview.append_column(column)
>
> column.add_attribute(leftjust, 'background-set', True)
>   
remove the above line.
> column.add_attribute(leftjust, 'background', COLUMN_FLAG)
>
> but when I run the program, I keep getting 
> Warning: unable to set property `background-set' of type `gboolean' from 
> value of type `gchararray'
>
> Is there really no way of doing this without creating another column of 
> booleans?
>
>   
You don't have to set "background-set" - it's set automatically when 
"background" is set.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] gtk.Image write text inside Image

[John Finlay]

Luis Gonzalez wrote:
> Hi to all ,
>
> I have a gtk.Image load from file.
>
> I want to write text inside this Image and then write to a file (jpg or png).
>
> I try to do use gtk.gdk.Pixbuf but to write text , draw lines , draw 
> rectangle I need a drawable object like gtk.gdk.Pixmap.
>
> How can i do this?
>   
If you want to read an image from a file draw on it and write it back 
you don't need to use a gtk.Image - that's only used for displaying an 
image. In fact you don't need PyGTK you can use cairo or PIL to do the 
job. If however you are using PyGTK anyway or are also going to use the 
image within PyGTK, the basic procedure is:

- load a pixbuf with the image from a file using 
gtk.gdk.pixbuf_new_from_file()
- create a pixmap of the same size using gtk.gdk.Pixmap() or 
gtk.gdk.Pixbuf.render_pixmap_and_mask()
- draw the pixbuf image on the pixmap using 
gtk.gdk.Drawable.draw_pixbuf() (not needed if render_pixmap_and_mask was 
used)
- draw your text, lines, etc. in the pixmap using either the cairo or 
gtk methods
- transfer the pixmap to the pixbuf using gtk.gdk.Pixbuf.get_from_drawable()
- write out the image using gtk.gdk.Pixbuf.save()

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] spinbutton set value and signal

[John Finlay]

Frédéric wrote:
> On mercredi 05 novembre 2008, Frédéric wrote:
>
>   
>> Is there a way to set the value of a spinbutton without having
>> its 'value-changed' signal emitted?
>> 
>
> I mean set the value from the code, not from the GUI...
>
>   
You could save the handler id and use handler_block() and 
handler_unblock() around the code that sets the value to prevent the 
handler from running. Other handlers connected to the signal will still 
run though. Otherwise stop_emission should prevent all handlers from 
being called.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] inhibate menu of window manager

[John Finlay]

awalter1 wrote:
> Thanks for your help
>
> I take a look to the set_functions() method and I've found the constants as
> gtk.gdk.FUNC_CLOSE that allows the window to be closeable. In fact I don't
> know how to allow the window to be NOT closeable and it seems that all
> functions (minimize, close, move ...) are set by default.
>
> I have found also a function set_deletable that should be adapted to my
> needs, but I'm not using the right pygtk version.
>
>   
If you set the functions and don't include the gtk.gdk.FUNC_CLOSE that 
tells the WM that you don't want to display the close button in the 
window frame.
> John Finlay wrote:
>   
>> awalter1 wrote:
>> 
>>> Hello,
>>>
>>> In my application python/gtk, I want to inhibate the item "close" from
>>> the
>>> standard menu available in all windows (bottom up of the window) and that
>>> allows actions as : minimize, maximise, ... close.
>>> If it is not possible to do that, which signal may I connect to be wake
>>> up ?
>>>
>>> I've found the signal "delete-event" to catch the "close" action, but it
>>> is
>>> required to connect it for each windows that is a little bit heavy for my
>>> Application.
>>> Thanks for your help
>>>   
>>>   
>> Try the gtk.gdk.Window.set_functions() method to see if that works with 
>> your window manager. Not all window managers will honor the request.
>>
>> John.
>>
>> John
>> ___
>> pygtk mailing list   pygtk@daa.com.au
>> http://www.daa.com.au/mailman/listinfo/pygtk
>> Read the PyGTK FAQ: http://faq.pygtk.org/
>>
>>
>> 
>
>   

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] inhibate menu of window manager

[John Finlay]

awalter1 wrote:
> Hello,
>
> In my application python/gtk, I want to inhibate the item "close" from the
> standard menu available in all windows (bottom up of the window) and that
> allows actions as : minimize, maximise, ... close.
> If it is not possible to do that, which signal may I connect to be wake up ?
>
> I've found the signal "delete-event" to catch the "close" action, but it is
> required to connect it for each windows that is a little bit heavy for my
> Application.
> Thanks for your help
>   
Try the gtk.gdk.Window.set_functions() method to see if that works with 
your window manager. Not all window managers will honor the request.

John.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] pygtk, drawing area

[John Finlay]

Tomáš Dohnálek wrote:
> Hello,
> i have some trouble with drawing area. I am actually not able to
> configure colours of lines and other object.
>
> self.gc.set_foreground(gtk.gdk.Color(1,1,6))
> widget.window.draw_rectangle(self.gc, True, x*70, y*70, 70, 70)
>
> I thought that this will work fine, but the line is black.
>   
You need to allocate the color. see:

http://faq.pygtk.org/index.py?req=show&file=faq04.008.htp 


John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Adding progressbar into a liststore

[John Finlay]

Timo wrote:
> Hello, I want to make some sort of downloadmanager like Firefox has, but 
> not that advanced. Just a window that adds a progressbar (and maybe a 
> label) for each download. But I already get stuck by adding a 
> progressbar into the liststore.
> This is what I have:
>
>   
You could use set_cell_data_func() to register a function that will 
provide the text data for display like:
> liststore  = gtk.ListStore(object)
> renderer = gtk.CellRendererText()
> column   = gtk.TreeViewColumn("Downloads",renderer, text=0)
>   
def func(column, cell, model, iter, func_data):
   # create some text string to display
   cell.set_property('text', string)
   return

column   = gtk.TreeViewColumn("Downloads",renderer)
column.set_cell_data_func(renderer, func, func_data)

> treeview.set_model(self.liststore)
> treeview.append_column(column)
>
> liststore.append([gtk.ProgressBar])
>
> Then I get this error:
>
> Warning: unable to set property `text' of type `gchararray' from 
> value of type `PyObject'
>
> There is a row in the treeview, cause I can select one line, but nothing 
> is in it.
>   

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Two models: can I have only one?

[John Finlay]

José Luis wrote:
> Hello to all!
>
> First, I explain my situation:
>
> I have one treeview and one combobox on my GUI. The treeview model 
> contains objects (it is a ListStore(object)). The ComboBox model has 
> only strings (ListStore(str)).
>
> My problem is that the information represented by the second model is 
> a subset of the informationon the treeview model.
>
> Is one method that I need to implement on my treeview model objects to 
> make it representable by the combobox?
>
You could use a TreeModelFilter to get what you wan. Seet:

http://pygtk.org/pygtk2tutorial/sec-TreeModelSortAndTreeModelFilter.html#sec-TreeModelFilter

http://pygtk.org/docs/pygtk/class-gtktreemodelfilter.html

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] returning window.handle from drawingarea

[John Finlay]

Patty Ackermann wrote:
> hi,
>
> I have initialized a gtk.DrawingArea() object and am trying to return 
> window.handle, but it's always returning None.
> Anyone have any ideas?
If you mean that the drawingarea.window attribute is None it's usually 
because the drawingarea hasn't been realized so call show() or realize() 
before accessing the window attribure.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] How can I suppress redraw of a gtk.ScrolledWindow and children until I'm done changing stuff?

[John Finlay]

Joel Hedlund wrote:
> Hi all!
>
> I've got a large widget (W) in an alignment in a viewport in a small 
> scrolledwindow (S). S overrides the do_scroll_event method, so that 
> ctrl-scroll zooms by doing W.set_size_request(). Since what's under the 
> mouse pointer should stay put and not be scrolled off the screen as W 
> resizes, the values on the adjustments on S need to be changed to 
> compensate.
>
> Now, my problem. GTK redraws after W.set_size_request() and after 
> S.props.vadjustment.set_value(), making zooming a very flickery and 
> migraine inducing experience that I don't want to subject my users to.
>
> So, how can I suppress redraws on a gtk.ScrolledWindow and its children 
> until I'm done changing stuff?
>   
Try:

http://pygtk.org/docs/pygtk/class-gdkwindow.html#method-gdkwindow--freeze-updates

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Filechooserbutton callback

[John Finlay]

Frédéric wrote:
> On jeudi 23 octobre 2008, John Finlay wrote:
>
>   
>>> Using the selection-changed signal does not work; it is never emited.
>>>   
>> Works for me on:
>>
>> Python 2.5.1, PyGTK 2.12.0 (Gtk+ 2.12.5)
>> 
>
> Wich plateform?
>   
Fedora 8 and Ubuntu Hardy

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Filechooserbutton callback

[John Finlay]

Frédéric wrote:
> On mercredi 22 octobre 2008, John Finlay wrote:
>
>   
>> AFAICT the FileChooserButton is a replacement for using an entry/label
>> and associated file browser button. It only allows selecting one
>> existing file or folder and displays the name of the selected file in
>> the button label. The most likely application is to use the FCB in a
>> dialog where a file or folder is to be selected (along with other info)
>> and finally another action (e.g. clicking an OK button) is used to
>> retrieve the currently selected file or folder and perform some
>> operations. I don't think it works very well if you want to perform some
>> actions immediately after selecting a file or folder unless you pass in
>> a FileChooserDialog and add the buttons and handle its responses and
>> signals. You can use the "selection-changed" signal with the FCB to be
>> notified when a different file has been selected and the dialog is
>> closed.
>> 
>
> Using the selection-changed signal does not work; it is never emited. 
>   
Works for me on:

Python 2.5.1, PyGTK 2.12.0 (Gtk+ 2.12.5)

> Anyway, as you suggested, I just get the value when the entire config 
> dialog is closed, and it works fine... except under Windows! 
> get_current_folder() always returns the first dir I set in the chooser, 
> with set_current_folder() method. Note that it works fine under linux and 
> Nokia/maemo.
>
>   

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] How to do_size_allocate properly in a gtk.Viewport subclass?

[John Finlay]

Joel Hedlund wrote:
> Hi!
>
> I was directed here from comp.lang.python, so sorry for the cross-post. 
>   I've also raised this issue on #pygtk and #gtk+ but with no luck. I 
> haven't been able to solve this even with aid of google, the pygtk 
> reference and the gtk C source, so pretty please help?
>
> I'm making an application that you can think of as an image viewer. I
> want to display a widget in a gtk.Viewport. The widget can have any size
> from tiny to humungous. I don't want the viewport to ever give the
> widget a larger size allocation than requested, and I don't want the
> viewport to ever resize to accomodate a large widget. It should rather
> leave grey areas around the widget/show only a portion of the widget.
>
> To do this I have subclassed gtk.Viewport (MyViewport) and overrided the
> do_size_allocate method:
>
>   
>> def do_size_allocate(self, allocation):
>> self.allocation = allocation
>> child_req = self.child.get_child_requisition()
>> child_alloc = gtk.gdk.Rectangle(0, 0, *child_req)
>> self.child.size_allocate(child_alloc)
>> self.props.hadjustment.update(allocation.width, child_alloc.width)
>> self.props.vadjustment.update(allocation.height, child_alloc.height)
>> if self.flags() & gtk.REALIZED:
>> self.window.move_resize(*self.allocation)
>> self.child.window()
>> 
>
> When I add a very large widget (a gtk.DrawingArea) to MyViewport only
> the originally visible portion of the widget is redrawn when I resize
> the window using the mouse, and the grey area around widget gets
> littered with grey lines that are not redrawn if you minimize and
> restore the window. I assume this comes from that the proper gdk windows
> haven't been updated, and that the grey lines are remnants of old
> Viewport borders.
>
> In gtk_viewport_size_allocate in gtkviewport.c, gdk_window_move_resize
> is called on three gdk windows: viewport->window, viewport->view_window
> and viewport->bin_window, but in pygtk I only have gtk.Viewport.window.
> I assume that this is the problem? If so, how can I fix this? Or is
> there something else that I have overlooked?
>
> And another relevant question: am I overcomplicating this? Is there some
> kind of flag that I could set on a vanilla viewport to accomplish this?
>   
I'm not sure what behavior you are looking for so I'll assume that you 
have one widget (e.g. Image) that you want to put inside a Viewport in a 
ScrolledWindow in a Window. I'll also assume that you want the widget to 
float in the center of the Window when the widget is smaller than the 
Window. Finally I assume that you want the widget to be partially 
displayed but scrollable within the window when the widget is larger 
than the Window. If these assumptions are wrong that ignore the 
following suggestion.

If so then I suggest that you put the widget inside a Table of size 1,1 
and attach it with EXPAND but not FILL. Put the table in a Viewport 
inside a ScrolledWindow inside a Window. This should keep the widget to 
a fixed size within the Window.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Filechooserbutton callback

[John Finlay]

Frédéric wrote:
> Hello,
>
> I added a Filechooserbutton in my GUI (using glade-3). When I clik on the
> button, it opens the file dialog, and I can choose a file.
>
> But how can I be notified that the user has used that button? What signal
> to bind, so I can retreive what was the user action (Accept/Cancel), and
> the file he chosed?
>   
AFAICT the FileChooserButton is a replacement for using an entry/label 
and associated file browser button. It only allows selecting one 
existing file or folder and displays the name of the selected file in 
the button label. The most likely application is to use the FCB in a 
dialog where a file or folder is to be selected (along with other info) 
and finally another action (e.g. clicking an OK button) is used to 
retrieve the currently selected file or folder and perform some 
operations. I don't think it works very well if you want to perform some 
actions immediately after selecting a file or folder unless you pass in 
a FileChooserDialog and add the buttons and handle its responses and 
signals. You can use the "selection-changed" signal with the FCB to be 
notified when a different file has been selected and the dialog is closed.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Possible error in the reference manual

[John Finlay]

Eric S. Raymond wrote:
> I think I've spotted an error in the Reference:
>
> --- pygtk-gobject.xml (revision 3046)
> +++ pygtk-gobject.xml (working copy)
> @@ -513,7 +513,7 @@
>  connect_object() method. For example, a call with a
>  function handler:
>
> -  handler_id = object("signal_name", handler, gobject)
> +  handler_id = connect_object("signal_name", handler, gobject)
>  
>will cause the handler to be invoked
>  as:
>
> If this is not an error, it need more explanation than it presently has.
>   
I think you're looking at the defunct version of the pygobject docs 
which used to be part of the pygtk docs. Pygobject has it's own source 
tree and its docs live there.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Is the pygtk tutorial still being maintained?

[John Finlay]

Eric S. Raymond wrote:
> In the course of a recent project, I found myself consultung the pygtk
> tutorial and developed some updates abnd additions for it.  I mailed
> one to John Findlay but received no response.
>
>   
Oops. Your email got buried until you reminded me of it.
> I have written documentation and an example program for the
> MessageDialog widget.
>   
Thanks.
> I have the following list of issues about the tutorial:
>
> 1. What is "private tooltip data" and how is it used.
> 2. The Toolbar request to add space is deprected.  What should replace it?
> 3. Entire toobar section should be rewritten for new API.
>   
Yes there are a lot of sections of the tutorial that have become 
out-of-date and should be rewritten. I lost interest in maintaining the 
tutorial (and time) a couple of years ago due to other commitments, etc. 
I hope someone will volunteer to take up the challenge.
> 4. How can I set a single dimension on a widget to be fixed in the
>presence of resizes?
>   
I don't know of any easy way to do this.
> 5. What, exactly triggers the appearance of scrollbars with automatic policy?
>   
If I understand your question correctly I've assumed that when the 
window became smaller than the widget required allocation in a direction 
that the scrollbar would appear.
> I note that the last update was in 2006.  Is this document still being
> maintained?  Are the sources in a version-control system that I might
> get write access to?
>   
I think Gian answered this. You'd have to get checkin access (ssh) for 
svn on the gnone servers.

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

Re: [pygtk] Geting a window/dialog to appear on a different screen.

[finlay]

Neil Dugan wrote:

Hi,

I have a LTSP set-up here.  I would like to put a window or dialog on 
the client screens, but I can't seem to get this to happen I am 
getting weird errors from gtk.main().


See attachment for the code.

The errors I am getting are
/media/tmp/testgui:43: GtkWarning: gdk_screen_get_display: assertion 
`GDK_IS_SCREEN (screen)' failed

  gtk.main()
/media/tmp/testgui:43: GtkWarning: gdk_keymap_get_for_display: 
assertion `GDK_IS_DISPLAY (display)' failed

  gtk.main()
/media/tmp/testgui:43: Warning: invalid (NULL) pointer instance
  gtk.main()
/media/tmp/testgui:43: Warning: g_signal_connect_data: assertion 
`G_TYPE_CHECK_INSTANCE (instance)' failed

  gtk.main()
Traceback (most recent call last):
  File "/media/tmp/testgui", line 43, in 
gtk.main()


If I comment out line 20 
"window.set_screen(display.get_screen(screen))" all works as expected.


Any help appreciated.
The Display object (and probably the Screen object) created by Window() 
are dereferenced after the function returns and being destroyed. A 
reference to display needs to be maintained somewhere e.g. return it 
from Window() or save it as an attribute of window or something similar.


John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://www.async.com.br/faq/pygtk/

Re: [pygtk] Table resize

[John Finlay]

[EMAIL PROTECTED] wrote:

Hi!

I have a table with 4 cells. A scrolledwindow attached to the top-left and 
custom
scrollbars top-right and bottom-left. On resizing, the left (vertical) 
scrollbar should
retain its width. The bottom (horizontal) scrollbar should retain its height.

I tried various things with FILL and EXPAND to no avail.

The code:

# Layout
window.add(table)
table.attach(view, 0, 1, 0, 1, gtk.EXPAND | gtk.FILL, gtk.EXPAND | gtk.FILL, 0 
, 0)
table.attach(scroller_v, 1, 2, 0, 1)
table.attach(scroller_h, 0, 1, 1, 2)

# Set scrollbar width and minimum height, then resize window to something 
practical
scroller_v.set_size_request(16, 60)
scroller_h.set_size_request(500, 16)
view.set_size_request(500, 400)
window.resize(516, 416)

---

What can I do?

  


table.attach(scroller_v, 1, 2, 0, 1, xoptions=0)
table.attach(scroller_h, 0, 1, 1, 2, yoptions=0)

John

___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://www.async.com.br/faq/pygtk/

Re: [pygtk] setting a filter for gtk.FontSelection

[John Finlay]

Le Roux Bodenstein wrote:

 This feature was removed from gtk after 2.2. There should be a note similar
to the note in the description of the FontSelectionDialog documentation.




Aah thanks. So it is a bug in the documentation? I don't mind just
using a normal widget and populating it with the right fonts. Any idea
how to list and filter fonts so that you get all the monospace fonts
on the system? I can't find it in the documentation.
  
Retrieve the pango context from the main widget and then get the list of 
the font families in the pango context and filter them for monospace 
fonts e.g.:


context = gtk.Window().get_pango_context()
monofonts = [fam for fam in context.list_families() if fam.is_monospace()]

John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://www.async.com.br/faq/pygtk/

Re: [pygtk] setting a filter for gtk.FontSelection

[finlay]

Le Roux Bodenstein wrote:

I'm trying to display a font selection widget that only lists monospace
fonts. I found the documentation for gtk.FontSelection here:
http://www.pygtk.org/docs/pygtk/class-gtkfontselection.html

And it mentions base and user filters:

"Filters can be used to limit the font selections. There are 2 filters
in the gtk.FontSelection - a base filter and a user filter. The base
filter cannot be changed by the user, so this can be used when the
user must choose from the restricted set of fonts (e.g. for a
terminal-type application you may want to force the user to select a
fixed-width font). The user filter can be changed or reset by the
user, by using the Reset Filter button or changing the options on the
Filter page of the widget."

..but I cannot find any documentation or other reference to these
filters. How do I set them? How do they work? Is this something that's
not implemented in pygtk, was removed from gtk at some stage, is just
not documented or am I losing my mind? :)
  
This feature was removed from gtk after 2.2. There should be a note 
similar to the note in the description of the FontSelectionDialog 
documentation.


John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://www.async.com.br/faq/pygtk/

Re: [pygtk] Pygtk does not play nicely with speech-dispatcher

[John Finlay]

James Simmons wrote:

Michael,

Thanks for the tip!  It was a big help.  I put gtk.gdk.threads_init()  
in the __init__ method of my program and I did get callbacks.  It 
looks like I didn't get callbacks for every word, but Pygtk seems to 
not be the problem anymore.  Thanks again,
The problem is that gtk and by extension pygtk has an event loop, and 
the python speechd api has implemented its own event loop within a 
thread. The speechd protocol is synchronous and sounds similar to RPCs. 
This event loop conflict was in general a problem with RPC programming 
with GUIs. The best solution is to integrate the two event loops using 
io_add_watch but that would entail throwing away the python speechd api 
and you would not have to use threads. This is the course I would take 
if I was doing your project.


If you want to use the speechd python api you have to make your pygtk 
program threaded and that is more than just calling threads_init. Others 
on this list can point you to a list of best practices for threading in 
pygtk.


John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://www.async.com.br/faq/pygtk/

Re: [pygtk] CellRendererCombo Question

[John Finlay]

Andrea Caminiti wrote:

hi john:

is there a way to do that wrapping but for rows in the combobox? becuase i have 
a combobox with 8 elements (image + label) and so the pop down list is very 
large. it would be great to do that.
  
I must have missed a message. Is the problem that you have a combobox 
but the list is too long? Amd each row has 8 cellrenderers in it?


John
___
pygtk mailing list   pygtk@daa.com.au
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://www.async.com.br/faq/pygtk/