.. Copyright ÂŠ 2007 Bruce Frederiksen
.. Permission is hereby granted, free of charge, to any person obtaining a copy
.. of this software and associated documentation files (the "Software"), to deal
.. in the Software without restriction, including without limitation the rights
.. to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
.. copies of the Software, and to permit persons to whom the Software is
.. furnished to do so, subject to the following conditions:
.. The above copyright notice and this permission notice shall be included in
.. all copies or substantial portions of the Software.
.. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
.. IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
.. FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
.. AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
.. LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
.. OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
.. THE SOFTWARE.
crumb: Using Pyke
How your python program uses pyke.
There isn't that much to using pyke. Generally, all the functions you need
are provided directly by the ``pyke`` package:
>>> import pyke
There are two functions to initialize pyke:
load(paths = ('.',), load_fc = True, load_bc = True)
The ``load`` function may only be called once. It expects either a single
directory or a sequence directories. It recursively walks each directory
looking for `.krb files`_. Each `.krb file`_ that it finds is compiled,
if out of date, and then the resulting python modules imported.
This causes all of the `rule bases`_ to be loaded.
add_universal_fact(kb_name, fact_name, args)
The ``add_universal_fact`` function is called once per fact_. These facts_
are never deleted and apply to all *cases*.
Setting up Each Case
Pyke is designed to be run multiple times for multiple *cases*. In
general each case has its own set of starting facts_ and may use different
`rule bases`_, depending on the situation.
Three functions initialize each case:
The ``reset`` function is called once to delete all of the `case specific
facts`_ from the last run. It also deactivates all `rule bases`_.
assert_(kb_name, fact_name, args)
Call ``assert_`` (or the equivalent, ``add_case_specific_fact``,
see `Other Functions`_, below) for each starting fact_ for this case.
Then call ``activate`` to activate the appropriate `rule bases`_. This
may be called more than once, if desired, or it can simply take multiple
Pyke is now ready to prove goals for this case!
Two functions are provided that cover the easy cases. More general
functions are provided in `Other Functions`_, below.
prove_1(kb_name, entity_name, fixed_args, num_returns)
Returns the first proof as two results: a tuple whose length == num_returns
containing the returned values, and a plan_. The plan_ is None if no plan_
was generated. Raises pyke.CanNotProve if no proof is found.
prove_n(kb_name, entity_name, fixed_args, num_returns)
This is a generator yielding two results: a tuple whose length == num_returns,
and a plan_, for each possible proof. Like ``prove_1``, the plan_ is None
if no plan_ was generated.
Running and Pickling Plans
Once you've obtained a plan_ from ``prove_1`` or ``prove_n``, you just call
it like a normal python function. The arguments required are simply those
specified, if any, in the `taking clause`_ of the rule_ proving the top-level
You may call the plan_ function any number of times. You may even pickle
the plan_ for later use. But the plans_ are constructed out of
`functools.partial`_ functions, so you need to register this with copy_reg_
before pickling the plan_:
>>> import copy_reg
>>> import functools
... lambda p: (functools.partial, (p.func,) + p.args))
The program that unpickles the plan_ does not have to import any pyke
modules to be able to run the plans_. Just unpickle and call it.
There are a few more functions that may be useful in special situations.
The first two of these provide more general access to the fact_ lookup and
goal proof mechanisms. The catch is that you must first convert **all**
arguments into patterns_ and create a *context* for these patterns_. This is
lookup(kb_name, entity_name, pattern_context, patterns)
This is a generator that binds patterns_ to successive facts_. Yields None
for each successful match.
prove(kb_name, entity_name, pattern_context, patterns)
A generator that binds patterns_ to successive proofs. Yields a
*prototype_plan* or None for each successful match. To turn the
prototype_plan into a python function, call create_plan() on it.
This returns the plan_ function.
The remaining functions are:
add_case_specific_fact(kb_name, fact_name, args)
This is an alternate to the ``assert_`` function.
Finds and returns the `knowledge base`_ by the name ``kb_name``. Raises
KeyError if not found. Note that for `rule bases`_, this returns the
active `rule base`_ where ``kb_name`` is the `rule base category`_ name.
Thus, not all `rule bases`_ are accessible through this call.
Finds and returns the `rule base`_ by the name ``rb_name``. Raises
KeyError if not found. This works for any `rule base`_, whether it is
active or not.
This is called once automatically for you by the ``load`` function.
You shouldn't have to call it yourself. If you don't have any `rule
bases`_ this isn't necessary anyway.
Creating Your Own Patterns
You'll need two more pyke modules to create your own patterns_ and contexts:
>>> from pyke import pattern, contexts
There are four kinds of patterns_:
This matches the ``data`` provided.
pattern.pattern_tuple((elements), rest_var = None)
This matches a tuple. ``Elements`` must each be a pattern and must
match the first *n* elements of the tuple. ``Rest_var`` must be a
variable (or anonymous). It will match the rest of the tuple and is
always bound to a (possibly empty) tuple.
This will match anything the first time it is encountered. But then
must match the first value each additional time it is encountered.
Calling the constructor twice with the same name produces the same
variable and must match the same value in all of the places that it is
This will match anything each time it is encountered.
Calling the constructor many times is not a problem.
Finally, to create a *pattern context*, you need::
You'll need to save this context to lookup your variable values after each
proof is yielded. This is done by either::
.. _below: `Creating Your Own Patterns`_
.. _case specific facts:
.. _copy_reg: http://docs.python.org/lib/module-copyreg.html
.. _fact: overview/knowledge_bases/fact_bases.html#facts
.. _facts: fact_
.. _functools.partial: http://docs.python.org/lib/module-functools.html
.. _knowledge base: overview/knowledge_bases/index.html
.. _.krb file: krb_syntax/index.html
.. _.krb files: `.krb file`_
.. _pattern: krb_syntax/pattern.html
.. _patterns: pattern_
.. _plan: overview/plans.html
.. _plans: plan_
.. _rule: overview/rules/index.html
.. _rules: rule_
.. _rule base: overview/knowledge_bases/rule_bases.html
.. _rule base category:
.. _rule bases: `rule base`_
.. _taking clause: krb_syntax/bc_rule.html#taking-clause