OcempGUI Manual

The following applications and libraries are needed before OcempGUI can be installed:

Please refer to the documentation of the respective package about how to install it.

OcempGUI is not provided as binary package by the author. However, it might be that someone else set up such a package for your wanted operating system or distribution. Those packages are usually not supported by the author, what means that installation problems or similar issues, which do not target the library directly, should be escalated to the respective supplier of that package.

The unpacked package contains two possibilities of building and installing it. Both ways are mostly identical but the one or other user might tend to prefer a specific way. The first is the python way of installing software, the second follows the tradition of the unix environment and uses a Makefile (which actually simply starts the python way). While being in the top source directory, it is possible to type either

make install

for the traditional unix way or

python setup.py install

for the python way.

The package might have some special options, which are described in the README file shipped with it.

Using the latest development sources is possible via CVS. More information about how to use the SourceForge repository can be found on SourceForge.

The following OcempGUI modules are currently available:

The ocempgui.draw.Draw submodule contains several functions for geometric objects. Although most of them are only wrappers around the respective pygame functions, some of them can be used to create more complex geometric objects. The following list gives an overview about the functions defined within this submodule.

The ocempgui.draw.Image submodule contains image related functions, such as loading or saving image data.

The ocempgui.draw.String submodule contains functions, which allow the creation and manipulation of fonts and string surfaces. It includes a simple font caching system, which provides a fast availability of fonts, which were created earlier. Besides this feature, the string surface related functions are mostly wrappers around the respective pygame functions.

The event management system of OcempGUI uses a simple approach using signal slots. This means, that objects will register themselves only for specific event types, of which they want to be notified. Any other event will not be sent to them.This reduces the overhead of events the objects have to deal with (either by dropping or processing them) and improves the performance and scalability of the event management system (especially with many objects).

Figure 1. Event management

An object, which shall be event aware, has to implement a notify() method, which will receive events distributed by the EventManager. The signature of the method looks like the following:

class OwnObject:
    ...
    def notify (self, event):
        ...

The event argument of the method will be an Event object, which can be used to perform certain actions within the method body then:

class OwnObject:
    ...
    def move (self, coords):
       # Moves the object to the desired coordinates (x, y).
       self.x = coords[0]
       self.y = coords[1]
       print "Moved to %d,%d" % (self.x, self.y)

    def notify (self, event):
        # Check the event signal and run a certain action with its data.
        if event.signal == "clicked":
           print "Something was clicked!"
        elif event.signal == "move":
           # Assuming that the event.data contains a coordinate tuple.
           self.move (event.data)
        

Setting up the main event management is nearly as easy as enhancing the objects. To add objects to the EventManager, the add_object() method has to be invoked. It receives the object to add and a list of signal ids as arguments. The signal ids will cause the object to be registered in specific queues, to which events with matching signal ids then will be sent.

Objects can be removed from the EventManager using the remove_object() method. The method allows you to either remove the object from specific slots or from all slots, it is registered for, at once.

# Create an EventManager and OwnObject instance.
manager = EventManager ()
myobj = OwnObject ()

# Add the object to the EventManager.
manager.add_object (myobj, "move", "clicked")

# Remove the object from the 'clicked' slot.
manager.remove_object (myobj, "clicked")

# Remove the object from _all_ slots it still listens on.
manager.remove_object (myobj)
        

Now let us proceed to the most important: sending events. To send events to the objects of the EventManager, you can use the emit() method. It receives two arguments, which will become the signal and data of a Event object. The Event will be created by the EventManager and then sent to the matching objects. So all you have to do is to pass the emit() method the correct information for the event.

Both arguments the emit() receives, have no limitations of type, length or whatsoever. It is up to you to to send correct information through the event management system and to check for correct information on the object side.

# Send events to the registered objects via the emit() method.
manager.emit ("clicked", None)
manager.emit ("move", (10, 10))
        

The following example is a complete example based on the excerpts from above. You can find it as python script under examples/eventmanager.py

# EventManager usage example.
from ocempgui.events import EventManager

# Create a new event capable object. This can be acquired by adding a
# 'notify ()' method to the object, which receives a single argument.
class OwnObject:
    def __init__ (self):
        self.x = 0
        self.y = 0
    
    def move (self, coords):
       # Moves the object to the desired coordinates (x, y).
       self.x = coords[0]
       self.y = coords[1]
       print "Moved to %d,%d" % (self.x, self.y)

    def notify (self, event):
        # Check the event signal and run a certain action with its data.
        if event.signal == "clicked":
           print "Something was clicked!"
        elif event.signal == "move":
           # Assuming that the event.data contains a coordinate tuple.
           self.move (event.data)

# Create an EventManager and OwnObject instance.
manager = EventManager ()
myobj = OwnObject ()

# Add the object to the EventManager.
manager.add_object (myobj, "move", "clicked")

# Send events to the registered objects via the emit() method.
manager.emit ("clicked", None)
manager.emit ("move", (10, 10))

# Remove the object from the 'clicked' slot.
manager.remove_object (myobj, "clicked")

# Send the 'clicked' event once more.
manager.emit ("clicked", None)

# Remove the object from _all_ slots it still listens on.
manager.remove_object (myobj)

# Send the 'move' event again.
manager.emit ("move", (40, 40))

The BaseObject is event driven, which means, that it acts and reacts upon events it receives and that it can raise events. Hereby you have to distinguish between Signals and Events. Signals are certain identifiers, a BaseObject can listen to, while Events are a combination of a Signal and additional data.

What does that mean in practice? As you already know from the section called “Building event driven systems”, objects will register themselves at an event manager with a specific signal, they listen to. Events in turn carry a specific signal id and additional data. A BaseObject will register itself at an event manager with its signals. If events are passed to the event manager, it will distribute them to the object, if needed. The object in turn will react according to its programming instructions.

To allow your object to react according to the application needs on certain events easily, the BaseObject supports the connection of callbacks to signals you can define for it. A callback is a method or function, which should be invoked, when the object receives a certain event.

Figure 2.  Signal handling of the BaseObject class

The first thing you should do is to let your existing object or the newly created one inherit from the BaseObject class. Afterwards you can unleash its full power by adding just a minimal set of code.

from ocempgui.object import BaseObject

class OwnObject (BaseObject):
    def __init__ (self):
        BaseObject.__init__ (self)
    
    ...
        

That is not all of course. You also have to set up the signals the object has to listen to and create callbacks. Let us create a small ping-pong example, where two objects react upon a 'ping' and 'pong' signal.

The BaseObject has a _signals attribute, which basically is a dictionary with the signal ids it listens to as keys and list for the callbacks as values. To allow your object to listen to the 'ping' or 'pong' signal, you have to add those to this dictionary.

from ocempgui.object import BaseObject

class OwnObject (BaseObject):
    def __init__ (self):
        BaseObject.__init__ (self)
        self._signals["ping"] = []
        self._signals["pong"] = []
    ...
        

Now we just need to make the notify() aware of those signal types and let it invoke the appropriate callbacks, which will be connected to those signals.

class OwnObject (BaseObject):
    ...
    def notify (self, event):
        if event.signal == "ping":
            self.run_signal_handlers ("ping")
        elif event.signal == "pong":
            self.run_signal_handlers ("pong")
        

As you see, the run_signal_handlers() takes care of invoking the connected callbacks. Now you have anything set up to allow the object to listen to specific events, to connect callbacks to it and let it invoke them, when it receives the specific signal.

To connect methods or functions as callbacks to a specific signal, the connect_signal() method of the BaseObject can be used. It allows additional data to be passed to the callback by specifiying the data right after the signal and callback. If the callbacks is not needed anymore, it can be disconnected using the disconnect_signal() method.

class OwnObject (BaseObject):
    ...

my_obj = OwnObject ()
ev_callback1 = my_obj.connect_signal ("ping", ping_callback, data)
ev_callback2 = my_obj.connect_signal ("pong", pong_callback, data1, data2)
...

my_obj.disconnect_signal (ev_callback1)
my_obj.disconnect_signal (ev_callback2)
        

Now it just needs to be connected to an event manager. In contrast to the earlier section, you do not need to register any signal of your BaseObject inheritor manually. When you connect it to an event manager, it will automatically do that for you.

class OwnObject (BaseObject):
    ...


manager = EventManager
my_obj = OwnObject ()

# Any signal of the object will be registered automatically.
my_obj.manager = manager
        

The last important thing to know about the BaseObject is its ability to emit events. If the object is connected to an event manager, you can let it send events through the manager with the object its emit() method. The syntax is the same as if you would emit events on the EventManager directly.

Now let us look at the example we just went through (the following oone is slightly modified only). You can find the example as python script under examples/baseobject.py

# BaseObject usage example.
from ocempgui.object import BaseObject
from ocempgui.events import EventManager

# Callbacks, which should be invoked for the object.
def ping_callback (obj, additional_data):
    print "The object is: %s" % obj.name
    print "Passed data is: %s" % additional_data

def pong_callback ():
    print "Another callback with no arguments."

# Object implementation, which can listen to specific events.
class OwnObject (BaseObject):
    def __init__ (self, name):
        BaseObject.__init__ (self)
        self.name = name
        # The object should be able to listen to 'ping' and 'pong'
        # events.
        self._signals["ping"] = []
        self._signals["pong"] = []

    def notify (self, event):
        # This simple notify method will not be used in this
        # example. Instead, the signals are invoked directly.
        if event.signal == "ping":
            self.run_signal_handlers ("ping")
        elif event.signal == "pong":
            self.run_signal_handlers ("pong")

manager = EventManager ()

# Create an object and connect callbacks to its both events.
my_obj = OwnObject ("First object")
ev1 = my_obj.connect_signal ("ping", ping_callback, my_obj, "data")
ev2 = my_obj.connect_signal ("pong", pong_callback)

# Connect it to the event manager
my_obj.manager = manager

# Invoke the connected signals handlers for a specific event.
manager.emit ("ping", None)
manager.emit ("pong", None)

# After disconnecting a callback, it will not be invoked anymore.
my_obj.disconnect_signal (ev1)
my_obj.disconnect_signal (ev2)
manager.emit ("ping", None)
manager.emit ("pong", None)

The ocempgui.object module includes another event capable object class, the ActionListener, which inherits from the BaseObject class, but allows you to create and delete signals and listening queues as you need them without the necessity to subclass.

The ActionListener creates the signal id and a callback queue, when you connect a callback to it and registers itself for this signal automatically at its event manager. This can be extremeley useful, if a more flexible event capable object type is needed, which does not need to do any sanity checks on the event data. Instead it will send the event data to the callback as well, which then can work with it.

Once more let us create a ping-ping example using the ActionListener class instead of a BaseObject now.

You can find the example as python script under examples/actionlistener.py

# ActionListener usage example.
import sys
from ocempgui.events import EventManager
from ocempgui.object import ActionListener

count = 0

def emit_pong (event, manager):
    print "emit_pong received: [%s] - emitting pong..." % event
    manager.emit ("pong", "pong_event")

def emit_ping (event, manager):
    global count
    if count > 10:
        sys.exit ()
    count += 1
    print "emit_ping received: [%s] - emitting ping..." % event
    manager.emit ("ping", "ping_event")

# Create an event manager and two ping-pong listeners.
manager = EventManager ()

listener1 = ActionListener ()
listener1.connect_signal ("ping", emit_pong, manager)
listener1.manager = manager

listener2 = ActionListener ()
listener2.connect_signal ("pong", emit_ping, manager)
listener2.manager = manager

# start ping-pong actions
print "Starting Ping-Pong"
manager.emit ("ping", "ping_event")

Looking at the first interesting line, line eight,

def emit_pong (event, manager):
    print "emit_pong received: [%s] - emitting pong..." % event
    manager.emit ("pong", "pong_event")
        

Line twelve and following does the same, but breaks, if it was invoked more than ten times.

The next lines of interest are line twenty-three to twenty-nine,

listener1 = ActionListener ()
listener1.connect_signal ("ping", emit_pong, manager)
listener1.manager = manager
        

The first thing an application using OcempGUI should do is to initialize the renderering system. The Renderer class from the ocempgui.widgets package contains all necessary parts to take care of this. It includes

The Renderer can be set up with only three lines of code.

from ocempgui.widgets import *
re = Renderer ()
re.create_screen (200, 200) # Creates the pygame window
        

Now that the Renderer is set up, you can start to place widgets on the pygame window by adding them using the Renderer.add_widget() method. Let us do this by building a simple (and well-known) application with a Button widget on it, that displays 'Hello World'.

You can find the example as python script under examples/hello_world.py

# Hello World example.
from ocempgui.widgets import *

# Initialize the drawing window.
re = Renderer ()
re.create_screen (100, 50)
re.title = "Hello World"
re.color = (250, 250, 250)

button = Button ("Hello World")
button.position = (10, 10)
re.add_widget (button)

# Start the main rendering loop.
re.start ()

The second line

from ocempgui.widgets import*
        

The fifth and sixth lines

re = Renderer ()
re.create_screen (100, 50)
        

The seventh and eight line

re.title = "Hello World"
re.color = (250, 250, 250)
        

In line ten

button = Button ("Hello World")

The next line will place the Button at a specific position.

button.position = (10, 10)

Line twelve

re.add_widget (button)

The last line will start the main processing loop of the Renderer.

re.start ()

Every widget of OcempGUI inherits from the ocempgui.object.BaseObject class and its event handling makes heavy usage of the BaseObject features.

To cause a Button to print a message upon a mouse click, you would connect a message printing function to the click signal of the Button. In turn, if this callback is not needed anymore in the later program flow, you would disconnect the function from the Button's signal.

This theoretical model is used in many different toolkits and OcempGUI stays with it. We will enhance our 'Hello world' example application from the previous chapter with a callback now, wich prints a message each time the Button is clicked.

You can find the example as python script under examples/hello_world_signals.py

# Hello World example.
from ocempgui.widgets import *
from ocempgui.widgets.Constants import *

def print_message  ():
    print "The button was clicked!"

# Initialize the drawing window.
re = Renderer ()
re.create_screen (100, 50)
re.title = "Hello World"
re.color = (250, 250, 250)

button = Button ("Hello World")
button.position = (10, 10)
button.connect_signal (SIG_CLICKED, print_message)
re.add_widget (button)

# Start the main rendering loop.
re.start ()

The first change you note is the new import directive in line three.

from ocempgui.widgets.Constants import *
        

Line five and six contain the function, which will be used as the callback for the button. It is indifferent, if the callback is a class or object method or a function. Both cases will work in the same way.

The next notably change was done in line 16

button.connect_signal (SIG_CLICKED, print_message)
        

The ocempgui.widgets module contains some globally accessed settings, that influence its complete behaviour. Those can be found in the ocempgui.widgets.base part and contain the Style, which is currently in use, the timer rate for double-clicks and a debugging flag. You can adjust those settings easily by simply binding them to new values.

Before you go ahead and change base.GlobalStyle, you should read the section called “Changing the widget appearance”.

The base.debug setting is for additional debugging output and should not be set to True usually (unless you are using a development version).

The last one, base.DoubleClickRate should be set using the base.set_doubleclick_rate() method only and denotes the maximum time to elaps between two click operations to identify them as a double-click.

A widget can be set to a specific position on the main screen using the position attribute.

widget.position = 10, 15
        

If you read the current position value of a widget, keep in mind, that it will return a tuple containing the both, x and y, coordinates.

Every widget supports a minimum size, that will be respected by the default drawing methods.

widget.size = 80, 20
        

The currently used width and height, which can differ from its size can be retrieved using the width and height attributes.

if (widget.width == widget.size[0]) and (widget.height == widget.size[1]):
    print "Widget does not exceed its minimum size."
        

To allow an easy and logical keyboard navigation, widgets have an index attribute, which influences the navigation order using the keyboard.

widget.index = 3
        

The input focus mentioned above denotes a state of the widget, in which the user can interact with it using the keyboard only. A Button widget will react upon pressing the space bar with a click while an Entry widget will let the user type text. You can set the input focus of a widget manually with the focus attribute.

widget.focus = True
        

frame = VFrame ()
focusok = frame.set_focus (True)
if not focusok:
    print "Focus could not be set."
        

Widgets can be disabled from user interaction and receiving events, if you set their sensitive attribute to False.

widget.sensitive = False
        

The renderering system of the ocempgui.widgets module can use different layers, on which widgets are drawn. This is especially useful and often necessary to place widgets above others (e.g. to place a Window widget above another one).

widget.depth = 3
        

You already learned about the Button widget in an earlier section, so let us look at some interesting details of it now.

The Button widget is a interactive widget, which reacts upon mouse input such as clicks. Basically it is a container (which will be explained detailled later on), which holds a Label widget to display its text.

To create a Button, you usually will type

button = Button (text)
        

The usage of mnemonics for the Button is easy to achieve by simply supplying a mnemonic text as described in the section called “Labels”. You can set the text directly through the text attribute or set_text() method.

button.text = "#Mnemonic"
button.set_text ("#Mnemonic")
        

As one of only few widgets the Button supports different border styles to adjust its look and feel without the need to override its drawing methods.

button.border = BORDER_NONE
button.set_border (BORDER_NONE)
        

The Button widget has some default signals, it listens to. Those are

Below you will find an example to illustrate most of the abilities of the Button widget class. You do not need to care about other widgets like the Frame class for now as those are explained later on.

You can find the following example as a python script under examples/button.py.

# Button examples.
import os
from ocempgui.widgets import *
from ocempgui.widgets.Constants import *

def _create_vframe (text):
    frame = VFrame (Label (text))
    frame.spacing = 5
    frame.align = ALIGN_LEFT
    return frame

def create_button_view ():
    states = ("STATE_NORMAL", "STATE_ENTERED", "STATE_ACTIVE",
              "STATE_INSENSITIVE")

    table = Table (2, 3)
    table.spacing = 5

    # Frame with the states.
    frm_states = _create_vframe ("States")
    for i in xrange (len (states)):
        btn = Button (states[i])
        if STATE_TYPES[i] == STATE_INSENSITIVE:
            btn.sensitive = False
        else:
            btn.state = STATE_TYPES[i]
        frm_states.add_child (btn)
    table.add_child (0, 0, frm_states)
    table.set_align (0, 0, ALIGN_TOP)

    # Frame with different padding.
    frm_padding = _create_vframe ("Padding")
    for i in xrange (5):
        btn = Button ("Padding: %dpx" % (i * 2))
        btn.padding = i * 2
        frm_padding.add_child (btn)
    table.add_child (0, 1, frm_padding)
    table.set_align (0, 1, ALIGN_TOP)

    # Mnemonics.
    frm_mnemonic = _create_vframe ("Mnemonics")
    btn = Button ("#Simple Mnemonic")
    btn2 = Button ("#Activate using <ALT><Underlined Key>")
    frm_mnemonic.add_child (btn, btn2)
    table.add_child (0, 2, frm_mnemonic)
    table.set_align (0, 2, ALIGN_TOP)

    # Borders.
    frm_borders = _create_vframe ("Borders")
    btn_raised = Button ("Raised border")
    btn_sunken = Button ("Sunken border")
    btn_sunken.border = BORDER_SUNKEN
    btn_flat = Button ("Flat border")
    btn_flat.border = BORDER_FLAT
    btn_none = Button ("No border")
    btn_none.border = BORDER_NONE
    btn_etchedin = Button ("Etched in")
    btn_etchedin.border = BORDER_ETCHED_IN
    btn_etchedout = Button ("Etched out")
    btn_etchedout.border = BORDER_ETCHED_OUT
    frm_borders.add_child (btn_raised, btn_sunken, btn_flat, btn_none,
                           btn_etchedin, btn_etchedout)
    table.add_child (1, 0, frm_borders)
    table.set_align (1, 0, ALIGN_TOP)

    # Multiline labeled buttons
    frm_multiline = _create_vframe ("Multiline labels")
    strings = ("Single lined Button", "Two lines on" + os.linesep + "a Button",
               "Two lines with a" + os.linesep + "#mnemonic")
    for i in xrange (len (strings)):
        button = Button (strings[i])
        button.child.multiline = True
        frm_multiline.add_child (button)
    table.add_child (1, 1, frm_multiline)
    table.set_align (1, 1, ALIGN_TOP)
    
    return table
    
if __name__ == "__main__":
    # Initialize the drawing window.
    re = Renderer ()
    re.create_screen (550, 470)
    re.title = "Button examples"
    re.color = (234, 228, 223)
    re.add_widget (create_button_view ())
    # Start the main rendering loop.
    re.start ()

The ImageButton is basically a subclass of the Button, but enhances it by the ability to load and display image data. It supports any image data format, that can be handled by the underlying pygame library. Due to its inheritance, everything said about the Button widget applies to the ImageButton as well.

The creation of an ImageButton is slightly different to the Button. Instead of passing the text to display, you can pass either the name of a file to load (including the full path to it) or a pygame.Surface object to display.

button = ImageButton ("path/to/an/image.png")
button = ImageButton (pygame_surface)
        

button.text = "Additional text"
        

Below you will find an example to illustrate most of the abilities of the ImageButton widget class. You do not need to care about other widgets like the Frame class for now as those are explained later on.

You can find the following example as a python script under examples/imagebutton.py.

# ImageButton examples.
import pygame, os
from ocempgui.draw import Image
from ocempgui.widgets import *
from ocempgui.widgets.Constants import *

def _create_vframe (text):
    frame = VFrame (Label (text))
    frame.spacing = 5
    frame.align = ALIGN_LEFT
    return frame

def create_button_view ():
    states = ("STATE_NORMAL", "STATE_ENTERED", "STATE_ACTIVE",
              "STATE_INSENSITIVE")

    image = Image.load_image ("./image.png")
    table = Table (2, 2)
    table.spacing = 5

    # Frame with the states.
    frm_states = _create_vframe ("States")
    for i in xrange (len (states)):
        btn = ImageButton (image)
        if STATE_TYPES[i] == STATE_INSENSITIVE:
            btn.sensitive = False
        else:
            btn.state = STATE_TYPES[i]
        btn.text = states[i]
        frm_states.add_child (btn)
    table.add_child (0, 0, frm_states)
    table.set_align (0, 0, ALIGN_TOP)

    # Frame with different padding.
    frm_padding = _create_vframe ("Padding")
    for i in xrange (4):
        btn = ImageButton (image)
        btn.padding = i * 2
        frm_padding.add_child (btn)
    table.add_child (0, 1, frm_padding)
    table.set_align (0, 0, ALIGN_TOP)

    # Mnemonics.
    frm_mnemonic = _create_vframe ("Mnemonics")
    btn = ImageButton (image)
    btn.text = "#Simple Mnemonic"
    btn2 = ImageButton (image)
    btn2.text = "#Activate using <ALT><Underlined Key>"
    frm_mnemonic.add_child (btn, btn2)
    table.add_child (1, 0, frm_mnemonic)
    table.set_align (1, 0, ALIGN_TOP)

    # Multiline labeled ImageButton
    frm_multiline = _create_vframe ("Multiline label")
    button = ImageButton (image)
    button.text = "Multiple lines" + os.linesep + "with a #mnemonic"
    button.child.multiline = True
    frm_multiline.add_child (button)
    table.add_child (1, 1, frm_multiline)
    table.set_align (1, 1, ALIGN_TOP)

    return table

if __name__ == "__main__":
    # Initialize the drawing window.
    re = Renderer ()
    re.create_screen (570, 400)
    re.title = "ImageButton examples"
    re.color = (234, 228, 223)
    re.add_widget (create_button_view ())
    # Start the main rendering loop.
    re.start ()

Inherited from the Button the ToggleButton widget does not differ from it except that it is always in one state, either active or inactive, which are alternated by a click. By default it is displayed in a depressed state on a click and pops up after clicking it again.

To create a ToggleButton, you can do the same as with the Button class.

button = ToggleButton (text)
        

You can retrieve the current state as boolean value of the ToggleButton through its active attribute and set its state programmatically via this attribute or the set_active() method.

if button.active:
    print "The ToggleButton is currently active!"
button.set_active (False)
        

To track changes of this state, the ToggleButton supplies a SIG_TOGGLED signal, which will be raised, if the state is changed via a mouse input or the accelerator action of a Label.

def state_changed (togglebutton):
    state = "active"
    if not togglebutton.active:
       state = "inactive"
    out = "The state of the ToggleButton has been set to %s" % state

button = ToggleButton ("A ToggleButton")
button.connect_signal (SIG_TOGGLED, state_changed, button)
        

Below you will find an example to illustrate most of the abilities of the ToggleButton widget class. You do not need to care about other widgets like the Frame class for now as those are explained later on.

You can find the following example as a python script under examples/togglebutton.py.

# ToggleButton examples.
import os
from ocempgui.widgets import *
from ocempgui.widgets.Constants import *

def _create_vframe (text):
    frame = VFrame (Label (text))
    frame.spacing = 5
    frame.align = ALIGN_LEFT
    return frame

def create_button_view ():
    states = ("STATE_NORMAL", "STATE_ENTERED", "STATE_ACTIVE",
              "STATE_INSENSITIVE")

    table = Table (2, 3)
    table.spacing = 5

    # Frame with the states.
    frm_states = _create_vframe ("States")
    for i in xrange (len (states)):
        btn = ToggleButton (states[i])
        if STATE_TYPES[i] == STATE_INSENSITIVE:
            btn.sensitive = False
        else:
            btn.state = STATE_TYPES[i]
        frm_states.add_child (btn)
    table.add_child (0, 0, frm_states)
    table.set_align (0, 0, ALIGN_TOP)

    # Frame with different padding.
    frm_padding = _create_vframe ("Padding")
    for i in xrange (5):
        btn = ToggleButton ("Padding: %dpx" % (i * 2))
        btn.padding = i * 2
        frm_padding.add_child (btn)
    table.add_child (0, 1, frm_padding)
    table.set_align (0, 0, ALIGN_TOP)

    # Mnemonics.
    frm_mnemonic = _create_vframe ("Mnemonics")
    btn = ToggleButton ("#Simple Mnemonic")
    btn2 = ToggleButton ("#Activate using <ALT><Underlined Key>")
    frm_mnemonic.add_child (btn, btn2)
    table.add_child (0, 2, frm_mnemonic)
    table.set_align (0, 2, ALIGN_TOP)


    # Multiline labeled buttons
    frm_multiline = _create_vframe ("Multiline labels")
    strings = ("Single lined ToggleButton",
               "Two lines on" + os.linesep + "a ToggleButton",
               "Two lines with a" + os.linesep + "#mnemonic")
    for i in xrange (len (strings)):
        button = ToggleButton (strings[i])
        button.child.multiline = True
        frm_multiline.add_child (button)
    table.add_child (1, 0, frm_multiline)
    table.set_align (1, 0, ALIGN_TOP)

    return table

if __name__ == "__main__":
    # Initialize the drawing window.
    re = Renderer ()
    re.create_screen (530, 400)
    re.title = "ToggleButton examples"
    re.color = (234, 228, 223)
    re.add_widget (create_button_view ())
    # Start the main rendering loop.
    re.start ()

The CheckButton, which inherits from the ToggleButton, does not bring in any new features. Instead it just uses a different look to display its set state. The state of the CheckButton is indicated by a small (usually 10x10 px) square, which is either checked or unchecked.

To create a CheckButton widget, you usually will do the same as with the ToggleButton.

button = CheckButton (text)
        

Below you will find an example to illustrate most of the abilities of the CheckButton widget class. You do not need to care about other widgets like the Frame class for now as those are explained later on.

You can find the following example as a python script under examples/checkbutton.py.

# CheckButton examples.
import os
from ocempgui.widgets import *
from ocempgui.widgets.Constants import *

def _create_vframe (text):
    frame = VFrame (Label (text))
    frame.spacing = 5
    frame.align = ALIGN_LEFT
    return frame

def create_button_view ():
    states = ("STATE_NORMAL", "STATE_ENTERED", "STATE_ACTIVE",
              "STATE_INSENSITIVE")

    table = Table (2, 3)
    table.spacing = 5

    # Frame with the states.
    frm_states = _create_vframe ("States")
    for i in xrange (len (states)):
        btn = CheckButton (states[i])
        if STATE_TYPES[i] == STATE_INSENSITIVE:
            btn.sensitive = False
        else:
            btn.state = STATE_TYPES[i]
        frm_states.add_child (btn)
    table.add_child (0, 0, frm_states)
    table.set_align (0, 0, ALIGN_TOP)

    # Frame with different padding.
    frm_padding = _create_vframe ("Padding")
    frm_padding.spacing = 5
    frm_padding.align = ALIGN_LEFT
    for i in xrange (5):
        btn = CheckButton ("Padding: %dpx" % (i * 2))
        btn.padding = i * 2
        frm_padding.add_child (btn)
    table.add_child (0, 1, frm_padding)
    table.set_align (0, 0, ALIGN_TOP)

    # Mnemonics.
    frm_mnemonic = _create_vframe ("Mnemonics")
    btn = CheckButton ("#Simple Mnemonic")
    btn2 = CheckButton ("#Activate using <ALT><Underlined Key>")
    frm_mnemonic.add_child (btn, btn2)
    table.add_child (0, 2, frm_mnemonic)
    table.set_align (0, 2, ALIGN_TOP)

    # Multiline labeled buttons
    frm_multiline = _create_vframe ("Multiline labels")
    strings = ("Single lined CheckButton",
               "Two lines on" + os.linesep + "a CheckButton",
               "Two lines with a" + os.linesep + "#mnemonic")
    for i in xrange (len (strings)):
        button = CheckButton (strings[i])
        button.child.multiline = True
        frm_multiline.add_child (button)
    table.add_child (1, 0, frm_multiline)
    table.set_align (1, 0, ALIGN_TOP)

    return table

if __name__ == "__main__":
    # Initialize the drawing window.
    re = Renderer ()
    re.create_screen (550, 350)
    re.title = "CheckButton examples"
    re.color = (234, 228, 223)
    re.add_widget (create_button_view ())
    # Start the main rendering loop.
    re.start ()

RadioButton widgets are similar to the CheckButton widgets, except that they can grouped, so that only one button of a group can be activated at a time. This is especially helpful, if you need to have the user to choose between a small amount of options.

The creation of a RadioButton is done using

button = RadioButton (text, group)
        

Alternatively to the group argument of the constructor, a RadioButton can be assigned to a group after its creation with the group attribute or set_group() method.

button.group = other_radio_button
button.set_goup (other_radio_button)
        

Given those possibilities a group of four choices can be created like the following example.

group = RadioButton ("Choice 1")
button1 = RadioButton ("Choice 2", group)

button2 = RadioButton ("Choice 3")
button2.group = group

button3 = RadioButton ("Choice 4")
group.add_button (button3)
        

In contrast to its parent classes, activating a RadioButton causes the other buttons in its group to loose their active state.

Below you will find an example to illustrate most of the abilities of the RadioButton widget class. You do not need to care about other widgets like the Frame class for now as those are explained later on.

You can find the following example as a python script under examples/radiobutton.py.

# RadioButton examples.
import os
from ocempgui.widgets import *
from ocempgui.widgets.Constants import *

def _create_vframe (text):
    frame = VFrame (Label (text))
    frame.spacing = 5
    frame.align = ALIGN_LEFT
    return frame

def create_button_view ():
    states = ("STATE_NORMAL", "STATE_ENTERED", "STATE_ACTIVE",
              "STATE_INSENSITIVE")

    table = Table (2, 3)
    table.spacing = 5

    # Frame with the states.
    frm_states = _create_vframe ("States")
    group = None
    for i in xrange (len (states)):
        btn = RadioButton (states[i], group)
        if i == 0:
            group = btn
        if STATE_TYPES[i] == STATE_INSENSITIVE:
            btn.sensitive = False
        else:
            btn.state = STATE_TYPES[i]
        frm_states.add_child (btn)
    table.add_child (0, 0, frm_states)
    table.set_align (0, 0, ALIGN_TOP)

    # Frame with different padding.
    frm_padding = _create_vframe ("Padding")
    group = None
    for i in xrange (5):
        btn = RadioButton ("Padding: %dpx" % (i * 2), group)
        if i == 0:
            group = btn
        btn.padding = i * 2
        frm_padding.add_child (btn)
    table.add_child (0, 1, frm_padding)
    table.set_align (0, 0, ALIGN_TOP)

    # Mnemonics.
    frm_mnemonic = _create_vframe ("Mnemonics")
    btn = RadioButton ("#Simple Mnemonic")
    btn2 = RadioButton ("#Activate using <ALT><Underlined Key>", btn)
    frm_mnemonic.add_child (btn, btn2)
    table.add_child (0, 2, frm_mnemonic)
    table.set_align (0, 2, ALIGN_TOP)

    # Multiline labeled buttons
    frm_multiline = _create_vframe ("Multiline labels")
    strings = ("Single lined RadioButton",
               "Two lines on" + os.linesep + "a RadioButton",
               "Two lines with a" + os.linesep + "#mnemonic")
    group = None
    for i in xrange (len (strings)):
        btn = RadioButton (strings[i], group)
        if i == 0:
            group = btn
        btn.child.multiline = True
        frm_multiline.add_child (btn)
    table.add_child (1, 0, frm_multiline)
    table.set_align (1, 0, ALIGN_TOP)

    return table

if __name__ == "__main__":
    # Initialize the drawing window.
    re = Renderer ()
    re.create_screen (550, 330)
    re.title = "RadioButton examples"
    re.color = (234, 228, 223)
    re.add_widget (create_button_view ())
    # Start the main rendering loop.
    re.start ()

The Editable class is an abstract base class, which takes care of dealing with keyboard events and text caret positioning. It is used as backend for the Entry widget class, which will be explained in the next section.

You usually will not create an Editable object directly, but rather use it as parent for your own classes.

It can store and operate on (unicode) text through its text attribute. The text usually will be set via this attribute or the set_text() method.

editable.text = "Text"
editable.set_text ("Text")
        

It also supports a virtual text caret position, which will be used by its internals to determine the position for editing operations. The caret attribute and set_caret() allow you to adjust the caret position.

editable.caret = 7
editable.set_caret (7)
        

Given both the text and caret attribute, the internals of the Editable will modify the text of it upon receiving keyboard events, if it has the input focus. Given an Editable, which contains the text "This is a Test." and its caret is set to 4, the processing of a pressed key ("D") causes:

The previous list mentioned a check, whether the text of an Editable can be modified. To allow or disallow editing the hold text, you can adjust the value of the editable attribute.

editable.editable = True
editable.set_editable (True)
        

The Editable listens by default to the following signals:

The Entry widget is a single line text input box, that inherits from the Editable class. It fully supports the Editable features and enhances it by mouse event sensitivity.

To create a Entry widget, you usually will type

entry = Entry (text)
        

Entry widgets support a password-like mode, in which any typed character will be displayed as an asterisk ('*').

entry.password = True
        

To allow a better look for some special fonts, the Entry can place an additional amount of pixels between its border and the text input area via the padding attribute or set_padding() method.

entry.padding = 5
entry.set_padding (10)
        

The Entry supports the SIG_MOUSEDOWN signal to get the input focus upon a left mouse button press.

Below you will find an example to illustrate most of the abilities of the Entry widget class. You do not need to care about other widgets like the Frame class for now as those are explained later on.

You can find the following example as a python script under examples/entry.py.

# Entry examples.
from ocempgui.widgets import *
from ocempgui.widgets.Constants import *

def create_entry_view ():
    states = ("STATE_NORMAL", "STATE_ENTERED", "STATE_ACTIVE",
              "STATE_INSENSITIVE")

    table = Table (10, 10)
    table.spacing = 5

    # Frame with the states.
    frm_states = VFrame (Label ("States"))
    frm_states.spacing = 5
    frm_states.align = ALIGN_LEFT
    for i in xrange (len (states)):
        entry = Entry (states[i])
        if STATE_TYPES[i] == STATE_INSENSITIVE:
            entry.sensitive = False
        else:
            entry.state = STATE_TYPES[i]
        frm_states.add_child (entry)
    table.add_child (0, 0, frm_states)
    table.set_align (0, 0, ALIGN_TOP)

    # Frame with different padding.
    frm_padding = VFrame (Label ("Padding"))
    frm_padding.spacing = 5
    frm_padding.align = ALIGN_LEFT
    for i in xrange (5):
        entry = Entry ("Padding: %dpx" % (i * 2))
        entry.padding = i * 2
        frm_padding.add_child (entry)
    table.add_child (0, 1, frm_padding)
    table.set_align (0, 0, ALIGN_TOP)

    return table

if __name__ == "__main__":
    # Initialize the drawing window.
    re = Renderer ()
    re.create_screen (400, 400)
    re.title = "Entry examples"
    re.color = (234, 228, 223)
    re.add_widget (create_entry_view ())
    # Start the main rendering loop.
    re.start ()

The Range widget class is an abstract class, which enables inheriting classes to set and use value ranges. It can make use of float values, thus providing a high and for most cases exact resolution of values and it supports setting minimum and maximum values as well as stepwise increments.

You usually will not create a Range object directly, but inherit from it in your own widget classes. The constructor of a Range

range = Range (minimum, maximum, step)
        

The minimum attribute defines the lower limit of the value range it serves, while the maximum attribute defines the upper limit of it. Both can be set either via the attribute or the set_minimum() or set_maximum() methods.

if range.minimum < 0:
    range.set_minimum (0)
if range.maximum > 100:
    range.set_maximum (100)
        

The step attribute of the Range is useful to inc