The PEAK Developers' Center   Trellis UserPreferences
HelpContents Search Diffs Info Edit Subscribe XML Print View

Event-Driven Programming The Easy Way, with

(NOTE: As of 0.7a1, many new features have been added to the Trellis API, and some old ones have been deprecated. If you are upgrading from an older version, please see the porting guide for details.)

Whether it's an application server or a desktop application, any sufficiently complex system is event-driven -- and that usually means callbacks.

Unfortunately, explicit callback management is to event-driven programming what explicit memory management is to most other kinds of programming: a tedious hassle and a significant source of unnecessary bugs.

For example, even in a single-threaded program, callbacks can create race conditions, if the callbacks are fired in an unexpected order. If a piece of code can cause callbacks to be fired "in the middle of something", both that code and the callbacks can get confused.

Of course, that's why most GUI libraries and other large event-driven systems usually have some way for you to temporarily block callbacks from happening. This lets you fix or workaround your callback order dependency bugs... at the cost of adding even more tedious callback management. And it still doesn't fix the problem of forgetting to cancel callbacks... or register needed ones in the first place!

The Trellis solves all of these problems by introducing automatic callback management, in much the same way that Python does automatic memory management. Instead of worrying about subscribing or "listening" to events and managing the order of callbacks, you just write rules to compute values. The Trellis "sees" what values your rules access, and thus knows what rules may need to be rerun when something changes -- not unlike the operation of a spreadsheet.

But even more important, it also ensures that callbacks can't happen while code is "in the middle of something". Any action a rule takes that would cause a new event to fire is automatically deferred until all of the applicable rules have had a chance to respond to the event(s) in progress. And, if you try to access the value of a rule that hasn't been updated yet, it's automatically updated on-the-fly so that it reflects the current event in progress.

No stale data. No race conditions. No callback management. That's what the Trellis gives you.

Here's a super-trivial example:

>>> from import trellis

>>> class TempConverter(trellis.Component):
...     F = trellis.maintain(
...         lambda self: self.C * 1.8 + 32,
...         initially = 32
...     )
...     C = trellis.maintain(
...         lambda self: (self.F - 32)/1.8,
...         initially = 0
...     )
...     @trellis.perform
...     def show_values(self):
...         print "Celsius......", self.C
...         print "Fahrenheit...", self.F

>>> tc = TempConverter(C=100)
Celsius...... 100
Fahrenheit... 212.0

>>> tc.F = 32
Celsius...... 0.0
Fahrenheit... 32

>>> tc.C = -40
Celsius...... -40
Fahrenheit... -40.0

As you can see, each attribute is updated if the other one changes, and the show_values action is invoked any time the dependent values change... but not if they don't:

>>> tc.C = -40

Since the value didn't change, none of the rules based on it were recalculated.

Now, imagine all this, but scaled up to include rules that can depend on things like how long it's been since something happened... whether a mouse button was clicked... whether a socket is readable... or whether a Twisted "deferred" object has fired. With automatic dependency tracking that spans function calls, so you don't even need to know what values your rule depends on, let alone having to explicitly code any dependencies in!

Imagine painless MVC, where you simply write rules like the above to update GUI widgets with application values... and vice versa.

And then, you'll have the tiny beginning of a mere glimpse... of what the Trellis can do for you.

Other Python libraries exist which attempt to do similar things, of course; PyCells and Cellulose are two. However, only the Trellis supports fully circular rules (like the temperature conversion example above), and intra-pulse write conflict detection. The Trellis also uses less memory for each cell (rule/value object), and offers many other features that either PyCells or Cellulose lack.

The Trellis package can can be downloaded from the Python Package Index or installed using Easy Install, and it has a fair amount of documentation, including the following manuals:

Release highlights for 0.7a2:

Questions, discussion, and bug reports for the Trellis should be directed to the PEAK mailing list.

Table of Contents

Developer's Guide and Tutorial

Creating Components, Cells, and Rules

A trellis.Component is an object that can have its attributes automatically maintained by rules, the way a spreadsheet is maintained by its formulas.

These managed attributes are called "cell attributes", because the attribute values are stored in "cell" (trellis.AbstractCell) objects. Cell objects can be variable or constant, and either computed by a rule or explicitly set to a value -- possibly both, as in the temperature converter example!

There are five basic types of cell attributes:

Passive, Settable Values (attr() and attrs())
These are simple read-write attributes, with a specified default value. Rules that read these values will be automatically recalculated after the attribute is changed.
Computed Constants Or Initialized Values (make() and make.attrs())
These attributes are usually used to hold a mutable object, such as a list or dictionary (e.g. cache = trellis.make(dict)). The callable (passed in when you define the attribute) will be called at most once for each instance, in order to initialize the attribute's value. After that, the same object will be returned each time. (Unless you make the attribute writable, and set the attribute to a new value.)
Computed, Observable Values (@compute and compute.attrs())
These attributes are used to compute simple formulas, much like those in a spreadsheet. That is, ones that calculate a current state based on the current state of other values. Formulas used in @compute attributes must be non-circular, side-effect free, and cannot depend on the attribute's previous value. They are automatically recalculated when their dependencies change, but only if a maintenance or action-performing rule depends upon the result, either directly or indirectly. (This avoids unnecessary recalculation of values that nobody cares about.)
Maintenance Rules/Maintained Values (@maintain and maintain.attrs())
These rules or attribute values are used to reflect changes in state. A maintenance rule can modify other values or use its own previous value in a calculation. It is re-invoked any time a value it has previously used changes, even if no other rule depends upon it. Maintenance rules can be circular, as in the temperature converter example, as their values can be explicitly set -- both as an initial value, and at runtime. They are also used to implement "push" or "pull" rules that update one data structure in response to changes made in another data structure. All side-effects in maintenance rules must be undo-able using the Trellis's undo API. (Which is automatic if the side-effects happen only on trellis attributes or data structures.) But if you must change non-trellis data structures inside a maintenance rule, you will need to log undo actions. We'll discuss the undo log mechanism in more detail later, in the section on Creating Your Own Data Structures.
Action-Performing Rules (@perform)

These rules are used to perform non-undoable actions on non-trellis data or systems, such as output I/O and calls to other libraries. Like maintenance rules, they are automatically re-invoked whenever a value they've previously read has changed. Unlike maintenance rules, however, they cannot return a value or modify any trellis data.

Note, by the way, that this means performing rules should never raise errors. If they do, the changes that caused the rule to run will be rolled back, but if any other performing rules were run first, their actions will not be rolled back, leaving your application in an inconsistent state.

For each of the attribute types, you can use the plural attrs() form (if there is one) to define multiple attributes at once in the body of a class. The singular forms (except for attr()) can be used either inline or as function decorators wrapping a method to be used as the attribute's rule.

Let's take a look at a sample class that uses some of these ways to define different attributes, being deliberately inconsistent just to highlight some of the possible options:

>>> class Rectangle(trellis.Component):
...     trellis.attrs(
...         top = 0,
...         width = 20,
...     )
...     left = trellis.attr(0)
...     height = trellis.attr(30)
...     trellis.compute.attrs(
...         bottom = lambda self: + self.height,
...     )
...     @trellis.compute
...     def right(self):
...         return self.left + self.width
...     @trellis.perform
...     def show(self):
...         print self
...     def __repr__(self):
...         return "Rectangle"+repr(
...             ((self.left,, (self.width,self.height),
...              (self.right,self.bottom))
...         )

>>> r = Rectangle(width=40, height=10)
Rectangle((0, 0), (40, 10), (40, 10))

>>> r.width = 17
Rectangle((0, 0), (17, 10), (17, 10))

>>> r.left = 25
Rectangle((25, 0), (17, 10), (42, 10))

By the way, note that computed attributes (as well as make attributes by default) will be read-only:

>>> r.bottom = 99
Traceback (most recent call last):
AttributeError: can't set attribute

However, "maintained" attributes will be writable if you supply an initial value, as we did in the TemperatureConverter example. (Plain attr attributes are always writable, and make attributes can be made writable by passing in writable=True when creating them.)

Note, by the way, that you aren't required to make everything in your program a trellis.Component in order to use the Trellis. The Component class does only four things, and you are free to accomplish these things some other way if you need or want to:

  1. It sets self.__cells__ = trellis.Cells(self). This creates a special dictionary that will hold all the Cell objects used to implement cell attributes.
  2. The __init__ method takes any keyword arguments it receives, and uses them to initialize any named attributes. (Note that this is the only thing the __init__ method does, so you don't have to call it unless you want this behavior.)
  3. It creates a cell for each of the object's non-optional cell attributes, in order to initialize their rules and set up their dependencies. We'll cover this in more detail in the next section, Automatic Activation and Dependencies.
  4. It wraps the entire object creation process in a @modifier, so that all of the above operations occur in a single logical transaction. We'll cover this more in a later section on Managing State Changes.

In addition to doing these things another way, you can also use Cell objects directly, without any Component classes. This is discussed more in the section below on Working With Cell Objects.

Automatic Activation and Dependencies

You'll notice that each time we change an attribute value, our Rectangle instance above prints itself -- including when the instance is first created. That's because of two important Trellis principles:

  1. When a Component instance is created, all its "non-optional" cell attributes are calculated after initialization is finished. That is, if the attribute is a maintenance or performing rule, and has not been marked optional, then the rule is invoked, and the result is used to determine the cell's initial value.
  2. While a cell's rule is running, any trellis cell whose value is looked at becomes a dependency of that rule. If the looked-at cell changes later, it triggers recalculation of the rule that "looked". In Trellis terms, we say that the first cell has become a "listener" of the second cell.

The first of these principles explains why the rectangle printed itself immediately: the show performer cell was activated. We can see this if we look at the rectangle's show attribute:

>>> print

(The show rule is a performer, so the resulting attribute value is None. Also notice that rules are not methods -- they are more like properties.)

The second principle explains why the rectangle re-prints itself any time one of the attributes changes value: all six attributes are referenced by the __repr__ method, which is called when the show performer prints the rectangle. Since the cells that store those attributes are being looked at during the execution of another cell's rule, they become dependencies, and the show rule is thus re-run whenever the listened-to cells change.

Each time a rule runs, its dependencies are automatically re-calculated -- which means that if you have more complex rules, they can actually depend on different cells every time they're calculated. That way, the rule is only re-run when it's absolutely necessary.

By the way, a listened-to cell has to actually change its value (as determined by the != operator), in order to trigger recalculation. Merely setting a cell doesn't cause its observers to recalculate:

>>> r.width = 17    # doesn't trigger ``show``

But changing it to a non-equal value does:

>>> r.width = 18
Rectangle((25, 0), (18, 10), (43, 10))

"Optional" Rules and Subclassing

The show rule we've been playing with on our Rectangle class is kind of handy for debugging, but it's kind of annoying when you don't need it. Let's turn it into an "optional" performer, so that it won't run unless we ask it to:

>>> class QuietRectangle(Rectangle):
...     @trellis.perform(optional=True)
...     def show(self):
...         print self

By subclassing Rectangle, we inherit all of its cell attribute definitions. We call our new optional rule show so that its definition overrides the noisy version of the rule. And, because it's marked optional, it isn't automatically activated when the instance is created. So we don't get any announcements when we create an instance or change its values:

>>> q = QuietRectangle(width=18, left=25)
>>> q.width = 17

Unless, of course, we activate the show rule ourselves:

Rectangle((25, 0), (17, 30), (42, 30))

And from now on, it'll be just as chatty as the previous rectangle object:

>>> q.left = 0
Rectangle((0, 0), (17, 30), (17, 30))

While any other QuietRectangle objects we create will of course remain silent, since we haven't activated their show cells:

>>> q2 = QuietRectangle()
>>> = 99

@compute rules are always "optional". make() attributes are optional by default, but can be made non-optional by passing in optional=False. @maintain and @perform are non-optional by default, but can be made optional using optional=True.

Notice, by the way, that rule attributes are more like properties than methods, which means you can't use super() to call the inherited version of a rule. (Later, we'll look at other ways to access rule definitions.)

Read-Only and Read-Write Attributes

Attributes can vary as to whether they're settable:

  • Passive values (attr(), attrs()) and @maintain rules are always settable
  • make() attributes are settable only if created with writable=True
  • @compute and @perform attributes are never settable

For example, here's a class with a non-settable aDict attribute:

>>> class Demo(trellis.Component):
...     aDict = trellis.make(dict)

>>> d = Demo()
>>> d.aDict
>>> d.aDict[1] = 2
>>> d.aDict
{1: 2}

>>> d.aDict = {}
Traceback (most recent call last):
AttributeError: Constants can't be changed

Note, however, that even if an attribute isn't settable, you can still initialize the attribute value, before the attribute's cell is created:

>>> d = Demo(aDict={3:4})
>>> d.aDict
{3: 4}

>>> d = Demo()
>>> d.aDict = {1:2}
>>> d.aDict
{1: 2}

Since the aDict attribute is "optional" (make attributes are optional by default), it wasn't initialized when the Demo instance was created. So we were able to set an alternate initialization value. But, if we make it non-optional, we can't do this, because the attribute will be initialized during instance construction:

>>> class Demo(trellis.Component):
...     aDict = trellis.make(dict, optional=False)

>>> d = Demo()
>>> d.aDict = {1:2}
Traceback (most recent call last):
AttributeError: Constants can't be changed

And so, non-optional read-only attributes can only be set while an instance is being created:

>>> d = Demo(aDict={3:4})
>>> d.aDict
{3: 4}

But if an attribute is settable, it can be set at any time, whether the attribute is optional or not:

>>> class Demo(trellis.Component):
...     aDict = trellis.make(dict, writable=True)

>>> d = Demo()
>>> d.aDict = {1:2}
>>> d.aDict = {3:4}

Model-View-Controller and the "Observer" Pattern

As you can imagine, the ability to create rules like this can come in handy for debugging. Heck, there's no reason you have to print the values, either. If you're making a GUI application, you can define rules that update displayed fields to match application object values.

For that matter, you don't even need to define the rule in the same class! For example:

>>> class Viewer(trellis.Component):
...     model = trellis.attr(None)
...     @trellis.perform
...     def view_it(self):
...         if self.model is not None:
...             print self.model

>>> view = Viewer(model=q2)
Rectangle((0, 99), (20, 30), (20, 129))

Now, any time we change q2, it will be printed by the Viewer's view_it rule, even though we haven't activated q2's show rule:

>>> q2.left = 66
Rectangle((66, 99), (20, 30), (86, 129))

This means that we can automatically update a GUI (or whatever else might need updating), without adding any code to the thing we want to "observe". Just use cell attributes, and everything can use the "observer pattern" or be a "Model-View-Controller" architecture. Just define rules that can read from the "model", and they'll automatically be invoked when there are any changes to "view".

Notice, by the way, that our Viewer object can be repointed to any object we want. For example:

>>> q3 = QuietRectangle()
>>> view.model = q3
Rectangle((0, 0), (20, 30), (20, 30))

>>> q2.width = 59       # it's not watching us any more, so no output

>>> view.model = q2     # watching q2 again
Rectangle((66, 99), (59, 30), (125, 129))

>>> = 77         # but we're not watching q3 any more

See how each time we change the model attribute, the view_it rule is recalculated? The rule references self.model, which is a value cell attribute. So if you change view.model, this triggers a recalculation, too.

Remember: once a rule reads another cell, it will be recalculated whenever the previously-read value changes. Each time view_it is invoked, it renews its dependency on self.model, but also acquires new dependencies on whatever the repr() of self.model looks at. Meanwhile, any dependencies on the attributes of the previous self.model are dropped, so changing them doesn't cause the perform rule to be re-invoked any more. This means we can even do things like set model to a non-component object, like this:

>>> view.model = {}

But since dictionaries don't use any cells, changing the dictionary won't do anything:

>>> view.model[1] = 2

To be able to observe mutable data structures, you need to use data types like trellis.Dict and trellis.List instead of the built-in Python types. We'll cover how that works in the section below on Mutable Data Structures.

By the way, the links from a cell to its listeners are defined using weak references. This means that views (and cells or components in general) can be garbage collected even if they have dependencies. For more information about how Trellis objects are garbage collected, see the later section on Garbage Collection.

Accessing a Rule's Previous Value

Sometimes it's useful to create a maintained value that's based in part on its previous value. For example, a rule that produces an average over time, or that ignores "noise" in an input value, by only returning a new value when the input changes more than a certain threshhold since the last value. It's fairly easy to do this, using a @maintain rule that refers to its previous value:

>>> class NoiseFilter(trellis.Component):
...     trellis.attrs(
...         value = 0,
...         threshhold = 5,
...     )
...     @trellis.maintain(initially=0)
...     def filtered(self):
...         if abs(self.value - self.filtered) > self.threshhold:
...             return self.value
...         return self.filtered

>>> nf = NoiseFilter()
>>> nf.filtered
>>> nf.value = 1
>>> nf.filtered
>>> nf.value = 6
>>> nf.filtered
>>> nf.value = 2
>>> nf.filtered
>>> nf.value = 10
>>> nf.filtered
>>> nf.threshhold = 3   # changing the threshhold re-runs the filter...
>>> nf.filtered
>>> nf.value = -3
>>> nf.filtered

As you can see, referring to the value of a cell from inside the rule that computes the value of that cell, will return the previous value of the cell. (Note: this is only possible in @maintain rules.)

Beyond The Spreadsheet: "Resetting" Cells

So far, all the stuff we've been doing isn't really any different than what you can do with a spreadsheet, except maybe in degree. Spreadsheets usually don't allow the sort of circular calculations we've been doing, but that's not really too big of a leap.

But practical programs often need to do more than just reflect the values of things. They need to do things, too.

So far, we've seen only attributes that reflect a current "state" of things. But attributes can also represent things that are "happening", by automatically resetting to some sort of null or default value. In this way, you can use an attribute's value as a trigger to cause some action, following which it resets to an "empty" or "inactive" value. And this can then help us handle the "Controller" part of "Model-View-Controller".

For example, suppose we want to have a controller that lets you change the size of a rectangle. We can use "resetting" attributes to do this, in a way similar to an "event", "message", or "command" in a GUI or other event-driven system:

>>> class ChangeableRectangle(QuietRectangle):
...     trellis.attrs.resetting_to(
...         wider    = 0,
...         narrower = 0,
...         taller   = 0,
...         shorter  = 0
...     )
...     width = trellis.maintain(
...         lambda self: self.width  + self.wider - self.narrower,
...         initially = 20
...     )
...     height = trellis.maintain(
...         lambda self: self.height + self.taller - self.shorter,
...         initially = 30
...     )

>>> c = ChangeableRectangle()
>>> view.model = c
Rectangle((0, 0), (20, 30), (20, 30))

A resetting attribute (created with attr(resetting_to=value) or attrs.resetting_to()) works by receiving an input value, and then automatically resetting to its default value after its dependencies are updated. For example:

>>> c.wider

>>> c.wider = 1
Rectangle((0, 0), (21, 30), (21, 30))

>>> c.wider

>>> c.wider = 1
Rectangle((0, 0), (22, 30), (22, 30))

Notice that setting c.wider = 1 updated the rectangle as expected, but as soon as all updates were finished, the attribute reset to its default value of zero. In this way, every time you put a value into a resetting attribute, it gets processed and discarded. And each time you set it to a non-default value, it's treated as a change. Which means that any maintenance or performing rules that depends on the attribute will be recalculated (along with any @compute rules in between). If we'd used a normal trellis.attr here, and then set c.wider = 1 twice in a row, nothing would have happen the second time, because the value would not have changed.

Now, we could write methods for changing value cells that would do this sort of resetting for us, but it wouldn't be a good idea. We'd need to have both the attribute and the method, and we'd need to remember to never set the attribute directly. (What's more, it wouldn't even work correctly, for reasons we'll see later.) It's much easier to just use a resetting attribute as an "event sink" -- that is, to receive, consume, and dispose of any messages or commands you want to send to an object.

But why do we need such a thing at all? Why not just write code that directly manipulates the model's width and height? Well, sometimes you can, but it limits your ability to create generic views and controllers, makes it impossible to "subscribe" to an event from multiple places, and increases the likelihood that your program will have bugs -- especially order-dependency bugs.

If you use rules to compute values instead of writing code to manipulate values, then all the code that affects a value is in exactly one place. This makes it very easy to verify whether that code is correct, because the way the value is arrived at doesn't depend on what order a bunch of manipulation methods are being called in, and whether those methods are correctly updating everything they should.

Thus, as long as a cell's rule doesn't modify anything except local variables, there is no way for it to become "corrupt" or "out of sync" with the rest of the program. This is a form of something called "referential transparency", which roughly means "order independent". We'll cover this topic in more detail in the later section on Managing State Changes. But in the meantime, let's look at how using attributes instead of methods also helps us implement generic controllers.

Creating Generic Controllers by Sharing Cells

Let's create a couple of generic "Spinner" controllers, that take a pair of "increase" and "decrease" command attributes, and hook them up to our changeable rectangle:

>>> class Spinner(trellis.Component):
...     """Increase or decrease a value"""
...     increase = trellis.attr(resetting_to=0)
...     decrease = trellis.attr(resetting_to=0)
...     by = trellis.attr(1)
...     def up(self):
...         self.increase =
...     def down(self):
...         self.decrease =

>>> cells = trellis.Cells(c)
>>> width = Spinner(increase=cells['wider'], decrease=cells['narrower'])
>>> height =  Spinner(increase=cells['taller'], decrease=cells['shorter'])

The trellis.Cells() API returns a dictionary containing all active cells for the object. (We'll cover more about this in the section below on Working With Cell Objects_.) You can then access them directly, assigning them to other components' attributes.

Assigning a Cell object to a cell attribute allows two components to share the same cell. In this case, that means setting the .increase and .decrease attributes of our Spinner objects will set the corresponding attributes on the rectangle object, too:

>>> width.up()
Rectangle((0, 0), (23, 30), (23, 30))

>>> width.down()
Rectangle((0, 0), (22, 30), (22, 30))

>>> = 5

>>> height.down()
Rectangle((0, 0), (22, 25), (22, 25))

>>> height.up()
Rectangle((0, 0), (22, 30), (22, 30))

Could you do the same thing with methods? Maybe. But can methods be linked the other way?:

>>> width2 = Spinner()
>>> height2 = Spinner()
>>> controlled_rectangle = ChangeableRectangle(
...     wider = trellis.Cells(width2)['increase'],
...     narrower = trellis.Cells(width2)['decrease'],
...     taller = trellis.Cells(height2)['increase'],
...     shorter = trellis.Cells(height2)['decrease'],
... )

>>> view.model = controlled_rectangle
Rectangle((0, 0), (20, 30), (20, 30))

>>> = 10
>>> height2.up()
Rectangle((0, 0), (20, 40), (20, 40))

A shared cell is a shared cell: it doesn't matter which "direction" you share it in! It's a simple way to create an automatic link between two parts of your program, usually between a view or controller and a model. For example, if you create a text editing widget for a GUI application, you can define a value cell for the text in its class:

>>> class TextEditor(trellis.Component):
...     text = trellis.attr('')
...     @trellis.perform
...     def display(self):
...         print "updating GUI to show", repr(self.text)

>>> te = TextEditor()
updating GUI to show ''

>>> te.text = 'blah'
updating GUI to show 'blah'

And then you'd write some additional code to automatically set self.text when there's accepted input from the GUI. An instance of this editor can then either maintain its own text cell, or be given a cell from an object whose attributes are being edited.

This allows you to independently test your models, views, and controllers, then simply link them together at runtime in any way that's useful.

Resetting Rules

Resetting attributes are designed to "accept" what might be called events, messages, or commands. But what if you want to generate or transform such events instead?

Let's look at an example. Suppose you'd like to trigger an action whenever a new high temperature is seen:

>>> class HighDetector(trellis.Component):
...     value = trellis.attr(0)
...     last_max = trellis.attr(None)
...     @trellis.maintain
...     def new_high(self):
...         last_max = self.last_max
...         if last_max is None:
...             self.last_max = self.value
...             return False    # first seen isn't a new high
...         elif self.value > last_max:
...             self.last_max = self.value
...             return True
...         return False
...     @trellis.perform
...     def monitor(self):
...         if self.new_high:
...             print "New high"

The new_high rule runs whenever value changes, and checks to see if it's greater than the current highest value. If so, it returns true and updates the maximum value. Let's try it out:

>>> hd = HighDetector()

>>> hd.value = 7
New high

>>> hd.value = 9

Oops! We set a new high value, but the monitor rule didn't detect a new high, because new_high was already True from the previous high.

Just as with a regular attribute, rules normally return what might be called "continuous" or "steady state" values. That is, their value remains the same until something causes them to be recalculated. In this case, the second recalculation of new_high returns True, just like the first one... meaning that there's no change, and thus the performing rule isn't triggered.

But, just as with regular attributes, @compute and @maintain rules can be made "resetting", using the resetting_to= keyword, allowing the value to reset to a default as soon as all of the value's listeners have "seen" the original value. Let's try a new version of our high detector:

>>> class HighDetector2(HighDetector):
...     @trellis.maintain(resetting_to=False)
...     def new_high(self):
...         # this is a bit like a super() call, but for a rule:
...         return HighDetector.new_high.rule(self)

>>> hd = HighDetector2()

>>> hd.value = 7
New high

>>> hd.value = 9
New high

>>> hd.value = 3

>>> hd.value = 16
New high

As you can see, each new high is detected correctly now, because the value of new_high is silently reset to False after it's calculated as (or set to) any other value:

>>> hd.new_high

>>> hd.new_high = True
New high

>>> hd.new_high

(By the way, that HighDetector.new_high.rule in the new new_high rule retrieves the base class version of the rule. We could also have done the same thing this way:

>>> class HighDetector2(HighDetector):
...     new_high = trellis.maintain(
...         HighDetector.new_high.rule, resetting_to = False
...     )

and the result would have been the same, except it would run faster since the lookup of the inherited rule only happens once.)

Wiring Up Multiple Components

Over the course of this tutorial, we've created a whole bunch of different objects, like the temperature converter, high detector, changeable rectangle, and a simple viewer. Let's link them up together to make a rectangle that gets wider and taller whenever the Celsius temperature reaches a new high:

>>> tc = TempConverter()
Celsius...... 0.0
Fahrenheit... 32.0

>>> hd = HighDetector2(value = trellis.Cells(tc)['C'])
>>> cr = ChangeableRectangle(
...     wider  = trellis.Cells(hd)['new_high'],
...     taller = trellis.Cells(hd)['new_high'],
... )

>>> viewer = Viewer(model = cr)
Rectangle((0, 0), (20, 30), (20, 30))

>>> tc.F = -40
Celsius...... -40.0
Fahrenheit... -40

>>> tc.F = 50
Celsius...... 10.0
Fahrenheit... 50
New high
Rectangle((0, 0), (21, 31), (21, 31))

Crazy, huh? None of these components were designed with any of the others in mind, but because they all "speak Trellis", you can link them up like building blocks to do new and imaginative things.

By the way, although in this demonstration we saw the three outputs in one particular order, in general the Trellis does not guarantee what order rules will be recalculated in, so it's unwise to assume that your program will always produce results in a certain order, unless you've taken steps to ensure that it will.

That's why managing the order of Trellis output (and dealing with state changes in general) is the subject of our next major section.

Managing State Changes

Time is the enemy of event-driven programs. They say that time is "nature's way of keeping everything from happening at once", but in event-driven programs we usually want certain things to happen "all at once"!

For example, suppose we want to change a rectangle's top and left co-ordinates:

>>> = 66
Rectangle((25, 66), (18, 10), (43, 76))

>>> r.left = 53
Rectangle((53, 66), (18, 10), (71, 76))

Oops! If we were updating a GUI like this, we would see the rectangle move first down and then sideways, instead of just going to where it belongs in one movement.

Therefore, in most practical event-driven systems, certain kinds of changes are automatically deferred, usually by adding them to some kind of event queue so that they can happen later, after all the desired changes have happened. That way, they don't take effect until the current event is completely finished.


The Trellis actually does something similar, but its internal "event queue" is automatically flushed whenever you set a value from outside a rule. If you want to set multiple values, you need to use a @modifier function or method like this one, which we could've made a method of Rectangle, but didn't:

>>> @trellis.modifier
... def set_position(rectangle, left, top):
...     rectangle.left = left
... = top

>>> set_position(r, 55, 22)
Rectangle((55, 22), (18, 10), (73, 32))

Notifications of changes made by a modifier do not take effect until the outermost active modifier function returns. (In other words, if one modifier directly or indirectly calls another modifier, the inner modifier's changes don't cause notifications to occur until the same time as the outer modifier's changes do.)

Now, notice that this means that within a modifier, you can't rely on any values controlled by rules to be updated when you make changes. This means it's generally a bad idea for a rule to look at what it's changing. For example:

>>> @trellis.modifier
... def set_position(rectangle, left, top):
...     rectangle.left = left
... = top
...     print rectangle

>>> set_position(r, 22, 55)
Rectangle((22, 55), (18, 10), (73, 32))
Rectangle((22, 55), (18, 10), (40, 65))

The first print is from inside the rule, showing that from the rule's perspective, the bottom/right co-ordinates are not updated to reflect the changed top/left co-ordinates. The second print is from a perform rule, showing that the values do get updated after the modifier has exited.

The Evil of Order Dependency

The reason that time is the enemy of event driven programs is because time implies order, and order implies order dependency -- a major source of bugs in event-driven and GUI programs.

Writing a polished GUI program that has no visual glitches or behavioral quirks is difficult precisely because such things are the result of changes in the order that events occur in.

Worse still, the most seemingly-minor change to a previously working version of such a program can introduce a whole slew of new bugs, making it hard to predict how long it will take to implement new features. And as a program gets more complex, even fixing bugs can introduce new bugs!

Indeed, Adobe Systems Inc. estimates that nearly half of all their reported desktop application bugs (across all their applications!) are caused by such event-management problems.

So a major goal of the Trellis' is to not only wipe out these kinds of bugs, but to prevent most of them from happening in the first place.

And all you have to do to get the benefits, is to divide your code three ways:

  • Input code, that sets trellis cells or calls modifier methods, but does not run inside trellis rules. This kind of code is usually invoked by GUI or other I/O callbacks, or by top-level non-trellis code.
  • Processing rules that compute values, and/or make undo-able changes to cells or other data structures. (i.e. @compute and @maintain rules.)
  • Output rules that send data on to other systems (like the screen, a socket, a database, etc.). This code may appear in @perform rules, or it can be "application" code that reads results from a finished trellis calculation.

The first and third kinds of code are inherently order-dependent, since information comes in (and must go out) in a meaningful order. However, by putting related outputs in the same performer (or non-trellis code), you can ensure that the required order is enforced by a single piece of code. This approach is highly bug-resistant.

Second, you can reduce the order dependency of input code by making it do as little as possible, simply dumping data into input cells, where it can be handled by processing rules. And, since input controllers can be very generic and highly-reusable, there's a natural limit to how much input code you will need.

By using these approaches, you can maximize the portion of your application that appears in side effect-free (or at least undo-able) processing rules, which the Trellis makes 100% immune to order dependencies. Anything that happens in Trellis rules, happens instantaneously, in a logical sense. Ther is no "order", and thus no order dependency.

In truth, of course, rules do execute in some order. However, as long as the rules don't do anything but compute their own values, then it cannot matter what order they do it in. (The trellis guarantees this by automatically recalculating rules whenever their dependencies change, and undoing any calculations that "saw" out-of-date or inconsistent values.)

The Side-Effect Rules

To sum up the recommended approach to handling side-effects in Trellis-based programs, here are a few brief guidelines that will keep your code easy to write, understand, and debug.

Rule 1 - If Order Matters, Use Only One Rule

If you care what order some modifications to a trellis data structure occur in, then code them both in the same maintenance rule. If you care what order two "outside world" side-effects happen in, code them both in the same perform rule.

For example, in the TempConverter demo, we had a performer that printed the Celsius and Fahrenheit temperatures. If we'd put those two print statements in separate rules, we'd have had no control over the output order; either Celsius or Fahrenheit might have come first on any given change to the temperatures. So, if you care about the relative order of certain output or actions, you must put them all in one rule. If that makes the code too big or complex, you can always refactor to extract computing or maintenance rules to calculate the intermediate values. (Just don't put any of the external actions in the other rules, only the calculations. Then have a perform rule that only does the external actions.)

Rule 2 - When Setting Or Changing, Use One Rule or One Value

If you set a value from more than one place, you are introducing an order dependency. In fact, if you set a cell value from more than one rule, the Trellis will stop you, unless the values are equal. For example:

>>> class Conflict(trellis.Component):
...     value = trellis.attr(99)
...     @trellis.maintain
...     def ruleA(self):
...         self.value = 22
...     @trellis.maintain
...     def ruleB(self):
...         self.value = 33

>>> Conflict()
Traceback (most recent call last):
InputConflict: (33, 22)

This example fails because the two rules set different values for the value attribute, causing a conflict error. Since the rules don't agree, the result would depend on the order in which the rules happened to run -- which again is precisely what we don't want in an event-driven program!

So this rule is for your protection, because it makes it impossible for you to accidentally set the same thing in two different places in response to an event, and then miss the bug or be unable to reproduce it because the second change masks the first!

Instead, what happens is that assigning two different values to the same cell in response to the same event always produces an error message, making it easier to find the problem. Of course, if you arrange your input code so that only one piece of input code is setting trellis values for a given event, or only one piece of code ever modifies a given cell or data structure, then you'll never have this problem.

Of course, if all of your code is setting a cell to the same value, you won't get a conflict error either. This is mostly useful for e.g. receiver cells that represent a command the program should do. If you have GUI input code that triggers a command by setting some receiver to True whenever that command is selected from a menu, invoked by a keyboard shorcut, or accessed with a toolbar button click, then it doesn't matter which event happens or even if all three could somehow happen at the same time, because the end result is exactly the same: the receiver processes the True message once and then discards it.

Rule 3 - Rule Side-Effects MUST Be Logged For Undo

If your rules only set cell values or modify trellis-managed data structures, you don't need to worry about undo logging, as it's taken care of for you.

However, if you implement any other kind of side-effects in a maintenance rule (such as updating a mutable data structure that's not trellis-managed), you must record undo actions to allow the trellis to roll back your rule's action(s), in the event that it must be recalculated due to an order inconsistency, or if an error occurs during recalculation. If you don't do this, you risk corrupting your program's state. This is especially important if you are creating a new trellis-managed data structure type.

In general, it's best to keep side-effects in rules to a minimum, and use only cells and other trellis-managed data structures. And of course, any side effects that can't easily be undone should be placed in a @perform rule, which is guaranteed to run no more than once per overall recalculation of the trellis.

However, if you are creating your own trellis-managed data structure type, you may need to use the trellis.on_undo() API to register undo callbacks, to protect your data structure's integrity. See the section below on Creating Your Own Data Structures for more details on how this works.

Rule 4 - If You Write, Don't Read!

Be aware that rules with side-effects cannot see the ultimate effect of their changes, and so should avoid reading anything but their minimum required inputs. For example:

>>> import sys

>>> class ChangeTakesTime(trellis.Component):
...     v1 = trellis.attr(2)
...     v2 = trellis.compute(lambda self: self.v1*2)
...     @trellis.maintain
...     def update(self):
...         if self.v1!=3:
...             print "before", self.v1, self.v2
...             self.v1 = 3
...             print "after", self.v1, self.v2

>>> x = ChangeTakesTime()
before 2 4
after 3 4

>>> x.v2

Here's what's happening: first, v2 is calculated as 2*2 == 4. Then, the update rule sets v1 to 3. However, v2 is NOT immediately updated. Instead, it's put on a schedule of rules to be re-run. So the update rule still sees the OLD value of v2.

So, if you are making any kind of changes from inside a rule, beware of trying to read anything that might be affected by those changes, as you will likely see something that's out of date. This is particularly important when changing trellis-managed data structures, since many data structures rely on rules for their internal consistency. So if you first write and then read the same data structure from a single rule, you will almost certainly see inconsistent results.

Mutable Data Structures

So far, all of our Trellis examples have worked with atomic cell values, like integers, strings, and so forth. We've avoided working with lists, sets, dictionaries, and similar structures, because the standard Python implementations of these types can't be "observed" by rules, which means that they won't be automatically updated.

But this doesn't mean you can't use sets, lists, and dictionaries. You just need to use Trellis-managed ones. (Of course, all the warnings above about changing values still apply; just because you're modifying something other than attributes, doesn't mean you're not still modifying things!)

The Trellis package provides three primary mutable types for you to use in your components: Set, List, and Dict. You can also subclass them or create your own mutable types, as we'll discuss in a later section. (And, the module also provides some fancier data structures; see the Collections manual for details.)


The trellis.Dict type looks pretty much like any dictionary, but it can be observed by rules. Any change to the dictionary's contents will result in its observers being recalculated. For example, if we use our view object (defined way back in the section on Model-View-Controller and the "Observer" Pattern), we can print it whenever it changes, no matter how it changes:

>>> d = trellis.Dict(a=1)
>>> view.model = d
{'a': 1}

>>> del d['a']

>>> d['a'] = 2
{'a': 2}

Unlike normal values, however, even changing a dictionary entry to the same value will trigger a recalculation:

>>> d['a'] = 2
{'a': 2}

This is because the Dict type doesn't try to compare the values you put into it. If you need to prevent such recalculations from happening, you can always check the dictionary contents first, or create a subclass and override __setitem__ (but be sure to read the section on Creating Your Own Data Structures for some important information first).

In addition to these basic features, the Dict type provides three receiver attributes (added, changed, and deleted) that reflect changes currently in progress. Ordinarily, they are empty dictionaries, but while a change is taking place they temporarily become non-empty. For example:

>>> view.model = None

>>> class Dumper(trellis.Component):
...     @trellis.perform
...     def dump(self):
...         for name in 'added', 'changed', 'deleted':
...             if getattr(d, name):
...                 print name, '=', getattr(d, name)
>>> dumper=Dumper()

>>> del d['a']
deleted = {'a': 2}

>>> d[3] = 4
added = {3: 4}

>>> d[3] = 5
changed = {3: 5}

>>> @trellis.modifier
... def two_at_once():
...     del d[3]
...     d[4] = 5

>>> two_at_once()
added = {4: 5}
deleted = {3: 5}

These dictionaries immediately reset to empty as soon as a change has been fully processed, so you'll never see anything in them if you look from non-rule code:

>>> d.added

Also note that you cannot use the .pop(), .popitem(), or .setdefault() methods of Dict objects:

>>> d.setdefault(1, 2)
Traceback (most recent call last):
InputConflict: Can't read and write in the same operation

Remember: the trellis wants all changes to be deferred until the next recalculation. That means you can't see the effect of a change in the same moment during which you make the change, so operations like pop() are disallowed, because they would have to return the same value no matter how many times you called it during the same recalculation! (Otherwise, the change hasn't really been deferred.)

This limitation also applies to the pop() method of List and Set objects, as we'll see in the next two sections.


Trellis Set objects offer nearly all the comforts of the Python standard library's sets.Set objects (minus .pop(), and support for sets of mutable sets), but with observability:

>>> s = trellis.Set("abc")
>>> view.model = s
Set(['a', 'c', 'b'])

>>> s.add('d')
Set(['a', 'c', 'b', 'd'])

>>> s.remove('c')
Set(['a', 'b', 'd'])

>>> s -= trellis.Set(['a', 'b'])

Similar to the Dict type, the Set type offers receiver set attributes, added and removed, that reflect changes-in-progress to the set:

>>> view.model = None

>>> class Dumper(trellis.Component):
...     @trellis.perform
...     def dump(self):
...         for name in 'added', 'removed':
...             if getattr(s, name):
...                 print name, '=', list(getattr(s, name))
>>> dumper=Dumper()

>>> s.add('a')
added = ['a']

>>> s.add('a')      # duplicates are ignored

>>> s.remove('d')
removed = ['d']

Note, however, that you cannot use the .pop() method of Set objects:

>>> s.pop()
Traceback (most recent call last):
InputConflict: Can't read and write in the same operation

Remember: as with trellis.Dict, operations like pop() are disallowed here because they would require reading the effect of a change, before the logical future moment in which the change actually takes effect.


A trellis.List looks and works pretty much the same as a normal Python list, except that it can be observed by rules:

>>> myList = trellis.List([1,2,3])
>>> myList
[1, 2, 3]

>>> myList.reverse()    # no output while not being observed

>>> view.model = myList
[3, 2, 1]

>>> myList.reverse()    # but now we're being watched
[1, 2, 3]

>>> myList.insert(0, 4)
[4, 1, 2, 3]

>>> myList.sort()
[1, 2, 3, 4]

trellis.List objects also have a receiver attribute called changed. It's normally false, but is temporarily True during the recalculation triggered by a change to the list. But as with all receiver attributes, you'll never see a value in it from non-rule code:

>>> myList.changed

Only in rule code will you ever see it true, a moment before it becomes false:

>>> view.model = None   # quiet, please

>>> class Watcher(trellis.Component):
...     @trellis.perform
...     def dump(self):
...         print myList.changed

>>> watcher=Watcher()

>>> del myList[0]

>>> myList
[2, 3, 4]

Note, however, that you cannot use the .pop() method of List objects:

>>> myList.pop()
Traceback (most recent call last):
InputConflict: Can't read and write in the same operation

Remember: as with trellis.Dict and trellis.Set, operations like pop() are disallowed here because they would require reading the effect of a change, before the logical future moment in which the change actually takes effect.

trellis.List objects also have some inherent inefficiencies due to the wide variety of operations supported by Python lists. While trellis.Set and trellis.Dict objects update themselves in place by applying change logs, trellis.List has to use a copy-on-write strategy to manage updates, because there isn't any simple way to reduce operations like sort(), reverse(), remove(), etc. to a meaningful change log. (That's why it only provides a simple changed flag.)

So if you need to use large lists in an application, you may be better off creating a custom data structure of your own design. That way, if you only need a subset of the list interface, you can implement a changelog-based structure. For example, the Trellis package includes a SortedSet type that maintains an index of items sorted by keys, with a cell that lists changed regions. (See the Collections manual for more details.)


A trellis.Pipe is a little bit like a Python list, except it only has supports for 5 methods: append, extend, __iter__, __len__, and __contains__. Its purpose is to allow you to easily interconnect components that communicate streams of objects or data, not unlike an operating system pipe. You can use append() and extend() to put data in the pipe, and use the other methods to get it back out. And it resets itself to being empty after all of its observers have had a chance to see the contents:

>>> p = trellis.Pipe()

>>> view.model = p

>>> p.append(42)

>>> p.extend([27, 59])
[27, 59]

One common use for pipes is to allow you to create objects that communicate via sockets or other IPC. If you write a component so that it expects to receive its inputs via one pipe, and sends output to another, then those pipes can be connected at runtime to a socket. And at test time, you can just append data to the input pipe, and have a performer spit out what gets written to the output pipe.

The Pipe type is the trellis's simplest data structure type -- so you may want to have a peek at its source code after you read the next section. (Better still, try to write your own Pipe clone first, and then compare it to the real one!)

Creating Your Own Data Structures

If you want to create your own data structures along the lines of Dict, List, and Set, you have a few options. First, you can just build components that use those existing data types, and use @modifier methods to perform operations on them. (If you just directly perform operations, then listeners of your data structure may be recalculated in the middle of your changes, and see an inconsistent state.)

Depending on the nature of the data structure you need, however, this may not be sufficient. For example, when you perform multiple operations on a trellis.Dict, the later operations need to know about changes made by the earlier ones. If you add some items and then delete one, for example, the dict needs to know whether the item you're deleting is one of the ones that you added.

But, if you use normal read operations on the dictionary (like .has_key()), these will only reflect the "before" state -- what the dictionary had in it during the current recalculation, before any new changes were made.

So, the Trellis-supplied data types use a couple of special tools to allow them to "see the future" (and change it).

Let's suppose that we're creating a simple "queue" type, that keeps track of items added to it. Its output is a list of the most-recently added items, and the list becomes empty in the next recalculation if nobody adds anything to it:

>>> class Queue(trellis.Component):
...     items = trellis.todo(list)
...     to_add = items.future
...     @trellis.modifier
...     def add(self, item):
...         self.to_add.append(item)
...     def __repr__(self):
...         return str(self.items)

>>> q = Queue()
>>> view.model = q

>>> q.add(1)

Let's break down the pieces here. First, we create a "todo" cell. A todo cell is basically a resetting_to attribute, except that it resets to a calculated value instead of a constant. It takes a function or type, just like make. That is, if you use a function (or other object with a __get__ method), it's called with the object the attribute belongs to, and if you use a type (or other object lacking a __get__ method), it's called with no arguments.

When the "todo" cell is created, the rule is called to create the resetting value, just as with a make attribute. Unlike a make attribute, however, its rule will be called again each time a "future" (i.e. modified) value is required.

(By the way, you can define todo cells with either a direct call as shown above, a @trellis.todo decorator on a function, or by using trellis.todos(attr=func, ...)` in your class body.)

The second thing that we did in this class above is create a "future" property. Todo cell descriptors have a .future attribute that returns a new property. (This property accesses the "future" version of the todo cell's value -- causing the rule to be called to generate a new value, and various undo-log operations to be performed.)

Next, we define a modifier method, add(). This method accesses the to_add attribute, thereby getting the future value of the items attribute. This future value is initially created by calling the "todo" cell's rule. In this case, the rule returns an empty list, so that's what add() sees, and adds a value to it.

(Note, by the way, that you cannot access future values except from inside a @modifier function.)

Next, let's create another @modifier that adds more than one item to the to_add attribute. This will works because only a single "future value" is created during a given recalculation sweep, and @modifier methods guarantee that no new sweeps can occur while they are running. Thus, the changes made in the modifier won't take effect until it returns:

>>> @trellis.modifier
... def add_many(*args):
...     for arg in args: q.add(arg)

>>> add_many(1,2,3)
[1, 2, 3]

Finally, notice that after each change, the queue resets itself to empty, because the default value of the items cell is the empty list that was created when the cell was initialized.

Of course, since "todo" attributes are automatically resetting, what we've seen so far isn't enough to create a data structure that actually keeps any data around. To do that, we need to combine "todo" attributes with a rule to maintain an existing data structure:

>>> class Queue2(Queue):
...     added = trellis.todo(list)
...     to_add = added.future
...     @trellis.maintain(make=list)
...     def items(self):
...         if self.added:
...             return self.items + self.added
...         return self.items

>>> q = Queue2()
>>> view.model = q

>>> q.add(1)

>>> add_many(2, 3, 4)
[1, 2, 3, 4]

This version is very similar to the first version, but it separates added from items, and the items rule is set up to compute a new value that includes the added items. (Notice also the use of the make keyword to initialize items to an empty list before the items rule is run for the first time.)

Notice, by the way, that the items rule returns a new list every time there is a change. If it didn't, the updates wouldn't be tracked:

>>> class Queue3(Queue2):
...     @trellis.maintain(make=list)
...     def items(self):
...         if self.added:
...             self.items.extend(self.added)
...         return self.items

>>> q = Queue3()
>>> view.model = q

>>> q.add(1)

>>> add_many(2, 3, 4)

Why are no updates displayed here? Because items is being modified in-place, and when the trellis compares the "before" and "after" versions of its value, it concludes they are the same. This didn't happen when we returned a new list, because the old list still had its old contents, and the new list was different.

If you are modifying a return value in place like this, you should use the the trellis.mark_dirty() API to flag that your return value has changed, even though it's the same object. In addition, you should log an undo action so that if the trellis needs to roll back some calculations involving your data structure, it can do so:

>>> class Queue4(Queue2):
...     @trellis.maintain(make=list)
...     def items(self):
...         items = self.items
...         if self.added:
...             trellis.on_undo(items.__delitem__, slice(len(items),None))
...             items.extend(self.added)
...             trellis.mark_dirty()
...         return items

>>> q = Queue4()
>>> view.model = q

>>> q.add(1)

>>> add_many(2, 3, 4)
[1, 2, 3, 4]

As you can see, calling mark_dirty() caused the trellis to notice the change to the list, even though the newly-returned list is (by definition) still equal to the previous value of the rule (i.e., the same list).

The on_undo() function lets you register a callback function (with optional positional arguments) that will be invoked if the trellis needs to roll back changes due to an error, or due to an out-of-order calculation. (If a rule makes a change to a data structure that has already been read by another rule, the trellis has to undo any changes made by the earlier rule and re-run it to ensure consistent results.)

Registered functions are called in reverse order, so that callbacks registered by later on_undo() calls will run before earlier ones. The Trellis keeps track of what callbacks were registered during each rule's execution, so that it can roll back the minimum number of changes needed to resolve a calculation order conflict. In the event of an error, however, all changes are rolled back:

>>> @trellis.modifier
... def error_demo():
...     @trellis.Performer   # make a standalone performing rule
...     def bad_observer():
...         print q
...         raise RuntimeError("ha!")
...     q.add(5)

>>> try: error_demo()
... except RuntimeError: print "caught error"
[1, 2, 3, 4, 5]
caught error

>>> print q
[1, 2, 3, 4]

This example is a bit odd, because it's somewhat difficult to force the trellis to get an error in such a way as to test your undo logging. If we had simply raised an error in the modifier, the change would appear to have been rolled back, when in fact it hadn't happened yet! (It's easy to see this if you add a "print" to the items rule -- if you raise an error in the modifier, it will never be called, because the rules don't run until the modifier is over.)

So to actually test the undo-ing, we have to raise the error in a new performer cell, which then runs after q.items is updated. (Performers don't run until/unless there are no other kinds of rules pending.)

In later sections on Working with Cell Objects, we'll see more about how to create and use one-off cells like this Performer, without needing to make them part of a component.

In the meantime, please note that creating good trellis data structures can be tricky: be sure to write automated tests for your code, and verify that they actually test what you think they test. This is one situation where it's REALLY a good idea to write your tests first, and try to make them fail before you add any mark_dirty() or on_undo() calls to your code. Otherwise, you won't be sure that your tests are really testing anything!

Of course, you don't need to deal with mark_dirty() and undo() at all, if you stick to using immutable values as a basis for your data structure, or use a copy-on-write approach like that shown in our Queue2 example above. Such data structures are less efficient than updating in-place, if they contain large amounts of data, but not every data structure needs to contain large quantities of data!

Therefore, we suggest that you start with simpler data structures first, and only add in-place updates if and when you can prove that the data copying is unacceptable overhead, since such updates are harder to write in a provably-correct way. (Note, too, that Python's built-in data types can often copy data a lot faster than you'd expect...)

A Practical Example -- and trellis.Pipe

System Message: ERROR/3 (data/text/Trellis, line 1761)

Section empty; must have contents.

Other Things You Can Do With A Trellis

XXX This section isn't written yet and should include examples

  • MVC/Live UI Updates
  • Testable UI Models
  • Live Object Validation
  • Persistence/ORM
  • Async I/O
  • Process Monitoring
  • Live Business Statistics

Advanced Features and API Details

Working With Cell Objects

Throughout the main tutorial, we worked only with component attributes. But it's also possible to work directly with Cell objects. For example, here's a temperature converter implemented directly with cells:

>>> F = trellis.Cell(lambda: C.value * 1.8 + 32, 32)
>>> C = trellis.Cell(lambda: (F.value - 32)/1.8, 0)

>>> F.value
>>> C.value
>>> F.value = 212
>>> C.value

The trellis.Cell() constructor takes three arguments: a zero-argument callable (or None), an optional value, and an optional "discrete" flag. In our example above, we created a pair of cells with both rules and values, that are not discrete.

Notice, by the way, that when you are directly creating cells, you must use zero-argument callables. That is, Cell objects don't pass in a "self" argument to their rules. (The reason rules in a component use a "self" is that those rules are turned into methods before the cell is created. The Cell doesn't pass in a "self", but it's already bound to the method, so it shows up anyway.)

The value attribute of a Cell can be read or set, to get or change the value of the cell, and it works just like getting or setting a component cell attribute (except that setting a cell's value to another cell doesn't cause the cell to be replaced!). In addition to the .value attribute, there are also get_value() and set_value() methods:

>>> C.set_value(-40)

>>> F.get_value()

These can be useful if you need to register callbacks with other systems. For example, you could use a cell's set_value() method as a callback for a Twisted "deferred" object, so that the cell would receive the deferred's value when it became available.

Here's our earlier "noise filter" example, reconstituted as a set of cells:

>>> value = trellis.Cell(value=0)
>>> threshhold = trellis.Cell(value=5)
>>> def filtered():
...     if abs(value.value - filtered.value) > threshhold.value:
...         return value.value
...     return filtered.value

>>> filtered = trellis.Cell(filtered, 0)

>>> filtered.value
>>> value.value = 1
>>> filtered.value
>>> value.value = 6
>>> filtered.value

Read-Only and Constant Cells

As you can see, you can provide either a value only, or a rule and a value when you create a cell. However, if you provide just a rule and no value, you end up with a read-only cell whose value can't be set:

>>> roc = trellis.Cell(lambda: 123)

>>> roc.value = 456
Traceback (most recent call last):
AttributeError: can't set attribute

In fact, it's not even a Cell instance, but of a different type altogether:

>>> roc
ReadOnlyCell(<function <lambda> at ...>, None [uninitialized])

What the above means is that you have a read-only cell whose current value is None, but has not yet been initialized. This means that if you actually try to read the value of this cell, it may or may not match what the repr() showed. (This is because simply looking at the cell shouldn't cause the cell's value to be calculated; that could be very painful when debugging).

If we actually read the value of this cell, the rule will be run:

>>> roc.value

But since the rule doesn't depend on any other cells, the cell changes type again, to a Constant:

>>> roc

Since the rule didn't depend on any other cells, there is never any way that it could be meaningfully recalculated. Thus, it becomes constant, and cannot be listened-to by any other rules. If we create another rule that reads this cell, it will not end up depending on it:

>>> cell2 = trellis.Cell(lambda: roc.value)
>>> cell2.value

>>> cell2

Thus, constant values propagate automatically through the cell network, eliminating dependencies on things that can't possibly change. Of course, if a read-only cell depends on a cell that can change, it remains a read-only cell, and will be recalculated whenever its dependencies change:

>>> c1 = trellis.Cell(value=0)
>>> c2 = trellis.Cell(lambda: c1.value * 2)

>>> c2.value
>>> c1.value = 27
>>> c2
ReadOnlyCell(<function <lambda>...>, 54)

Note that you can take advantage of constant propagation by explicitly setting a component attribute to a trellis.Constant at creation time. For example, if for some reason you wanted a temperature converter that could only be used once:

>>> tc = TempConverter(C=trellis.Constant(100))
Celsius...... 100
Fahrenheit... 212.0

>>> tc.C = -40
Traceback (most recent call last):
AttributeError: Constants can't be changed

(This would probably be more useful with something like the NoiseFilter example, in that you could set its threshhold to a Constant(), eliminating the need for the filtered rule to check for changes to the threshhold in order to know if it should be recalculated.)

Working With A Component's Cells

As we saw in the main tutorial, the trellis.Cells() API returns a dictionary of active cells for a component:

>>> trellis.Cells(view)
{'model': Value([1, 2, 3, 4]),
 'view_it': Performer(<bound method Viewer.view_it of
                         <Viewer object at 0x...>>, None)}

In the case of a Component, this data is also stored in the component's __cells__ attribute:

>>> trellis.Cells(view) is view.__cells__

This makes it possible for you to set up direct links between components using shared cells. It also lets you access cell objects directly, in order to e.g. register their set_value() methods as callbacks for other systems.

Discrete and Performer Cells

To make a cell "discrete" (i.e. automatically resetting to its initial value), you set its third constructor argument (i.e., discrete) to true:

>>> aReceiver = trellis.Cell(value=0, discrete=True)
>>> aReceiver.value

>>> v = Viewer(model = aReceiver)
>>> aReceiver.value = 42

As you can see, the value a discrete cell is created with, is the default value it resets to between set (or calculated) values. If you want to make a resetting rule, just include a rule in addition to the default value and the discrete flag.

@perform rules are implemented with the trellis.Performer class:

>>> trellis.Cells(view)['view_it']
Performer(<bound method Viewer.view_it of
           <Viewer object at 0x...>>, None)

The Performer constructor takes only one parameter: a zero-argument callable, such as a bound method or a function with no parameters. You can't set a value for a Performer (because it's not writable), nor can you make it discrete (since that would imply a readable value, and performer cells exist only for their side-effects). Creating a Performer cell schedules it for execution as soon as the current modifier is complete and any normal rules are finished. It will then be re-executed in the future, after any cells or other trellis- managed data structures it depended on are changed. (As long as the Performer isn't garbage collected, of course.)

Garbage Collection

Cells keep strong references to all of the cells whose values they accessed during rule calculation, and weak references to all of the cells that accessed them. This ensures that as long as a listener exists, its most-recently read subject(s) will also continue to exist.

Cells whose rules are effectively methods (i.e., cells that represent component attributes) also keep a strong reference to the object that owns them, by way of the method's im_self attribute. This means that as long as some attribute of a component is being observed, the whole component will continue to exist.

In addition, a component's __cells__ dictionary keeps a reference to all its cells, creating a reference cycle between the cells and the component. Thus, Component instances can only be reclaimed by Python's cycle collector, and are not destroyed as soon as they go out of scope. You should therefore avoid giving Component objects a __del__ method, and should explicitly dispose of any resources that you want to reclaim early.

You should NOT, however, attempt to break the cycle between a component and its cells. If the cells have any observers, this will just cause the rules to break upon recalculation, or else recreate some of the cells, depending on how you tried to break the cycle. It's better to simply let Python detect the cycle and get rid of it itself.

However, if you absolutely MUST mess with this, the best thing to do is delete the component's __cells__ attribute with del ob.__cells__, as this will ensure that any dangling observers will at least get attribute errors when recalculation occurs. Thus, if the component is really still in use, at least you'll get an error message, instead of weird results. But it still won't be a fun problem to debug, so it's highly recommended that you leave the garbage collection to Python. Python always knows more about what's happening in your program than you do!

Additional Documentation

There's a lot more to the Trellis package than what's in this brief guide and tutorial. Here are some links to other documentation provided with the package:

Time, Event Loops, and Tasks (Activity.txt in the source distribution)
This manual explains how to create generator-based pseudo-threads, schedule activities for idle moments, integrate the Trellis with other event loops (e.g. Twisted and wxPython), and implement things like timeouts, delays, alarms, etc. in Trellis rules.
Event-Driven Collections with the Trellis (Collections.txt in the source)
This document provides a brief overview of some additional Trellis data structures provided by the package, such as the SortedSet type.
Software Transactional Memory (STM) And Observers (aka STM-Observer.txt)
This document shows how some of the underlying Trellis pieces work, and in future revisions, it'll include some hints on how to create your own custom cell types, etc.


The "Trellis" Name

The "Trellis" name comes from Dr. David Gelernter's 1991 book, "Mirror Worlds", where he describes a parallel programming architecture he called "The Trellis". In the excerpted passages below, he describes the portions of his architecture that are roughly the same as in this Python implementation:

"Consider an upward-stretching network of infomachines tethered together, rung-upon-rung (billowing slightly in the breeze?) No two rungs need have exactly the same number of machines.... There might be ten rungs in all or hundreds or thousands, and the average rung might have anywhere from a handful to hundreds of members. This architecture spans a huge range of shapes and sizes....

So, these things are "tethered together" -- meaning? Those lines are lines of communication. Each member of the Trellis is tethered to some lower-down machines and to some higher-ups.... A machine deals only with the machines to which it is tethered. So far as it's concerned, the rest don't exist. It deals with inferiors in a certain way and superiors in a certain other way, and that's it....

Information rushes upward through the network, and the machines on each rung respond to it on their own terms.... Each machine focuses on one piece of the problem -- on answering a single question about the thing out there...that is being monitored. Each machine's entire and continuous effort is thrown into answering its one question. You can query a machine at any time -- what's the current best answer to your particular question? -- and it will produce an up-to-the-second response....

So data flows upward through the ensemble; there's also a reverse, downward flow of what you might call "anti-data" -- inquiries about what's going on. A high ranking element might attempt to generate a new value, only to discover it's missing some key datum from an inferior. It sends a query downward.... The inferior tries to come up with some new data.... If a bottom-level machine is missing data,.... It can ask the outside world directly for information....

The fact that data flows up and anti-data flows downwards means that, in a certain sense, a Trellis can run either forwards or backwards, or both at the same time....

A Trellis, it turns out, is a lot like a crystal.... When you turn it on, it vibrates at a certain frequency.

Meaning? In concept, each Trellis element is an infomachine. All these infomachines run separately and simultaneously.

In practice, we do things somewhat differently....

We run the Trellis in a series of sweeps. During the first sweep, each machine gets a chance to [produce one output value]. During the second, each [produces a second value], and so on. No machine [produces] a second [value] until every [machine] has [produced] a first [value]."

While Dr. Gelernter's Trellis was designed to be run by an arbitary number of parallel processors, our Trellis is scaled down to run in a single Python thread. But on the plus side, our Trellis automatically connects its "tethers" as it goes, so we don't have to explicitly plot out an entire network of dependencies, either!

The Implementation

Ken Tilton's "Cells" library for Common Lisp inspired the implementation of the Trellis. While Tilton had never heard of Gelernter's Trellis, he independently discovered the value of having synchronous updates, like the "sweeps" of Gelernter's design, and combined them with automatic dependency detection to create his "Cells" library.

I heard about this library only because Google sponsored a "Summer of Code" project to port Cells to Python - a project that produced the PyCells implementation. My implementation, however, is not a port but a re-visioning based on native Python idioms and extended to handle mutually recursive rules, side-effects, rollback, and various other features that do not precisely map onto the features of Cells, PyCells, or other Python frameworks inspired by Cells (such as "Cellulose").

While the first very rough drafts of this package were done in 2006 on my own time, virtually all of the work since has been generously funded by OSAF, the Open Source Applications Foundation.


Open Issues
  • Debugging code that does modifications can be difficult because it can be hard to know which cells are which. There should be a way to give cells an identifier, so you know what you're looking at.
  • Currently, there's no protection against accessing Cells from other threads, nor support for having different logical tasks in the same thread with their own contexts, services, etc. This should be fixed by using the "Contextual" library to manage thread-local (and task-local) state for the Trellis, and by switching to the appropriate context.State whenever non-rule/non-modifier code tries to read or write a cell.
  • There should probably be an easier way to reference cells directly, instead of using Cells(ob)['name'] -- perhaps a .link property, similar to the .future of "todo" cells, would make this easier.
  • The poll() and repeat() functions are undocumented in this release.
  • It's a bad idea to use on_commit() for user-level operations
  • A system for processing relational-like records and "active queries" mapped from zero or more backend storage mechanism.
  • Framework for mapping application components to UI views.
  • Widget specification, styling, and layout system that's backend-agnostic, ala Adobe's "Eve2" layout constraint system. Should be equally capable of spitting out text-mode drawings of a UI, as it is of managing complex wx "GridBagSizer" layouts.
  • IO events
  • Cross-thread bridge cells
  • signal() events

EditText of this page (last modified 2008-05-23 12:49:02)
FindPage by browsing, title search , text search or an index
Or try one of these actions: AttachFile, DeletePage, LikePages, LocalSiteMap, SpellCheck