C[omp]ute

Welcome to my blog, which was once a mailing list of the same name and is still generated by mail. Please reply via the "comment" links.

Always interested in offers/projects/new ideas. Eclectic experience in fields like: numerical computing; Python web; Java enterprise; functional languages; GPGPU; SQL databases; etc. Based in Santiago, Chile; telecommute worldwide. CV; email.

Personal Projects

Lepl parser for Python.

Colorless Green.

Photography around Santiago.

SVG experiment.

Professional Portfolio

Calibration of seismometers.

Data access via web services.

Cache rewrite.

Extending OpenSSH.

Last 100 entries

Re: Python's sad, unimaginative Enum; Re: Some explanation; Some explanation; Printing binary trees sideways; Atoms in python; About "Python's sad, unimaginative Enum"; Frustration Understood; Some good feedback here; this is fucking useless; I agree with you #nt; What would be imaginative?; Re: Enum; Enum; Python's sad, unimaginative Enum; Possible Fix; Work, Exhaustion, Vacation; VirtualBox with Centos 6.3 to 6.4, client; Matasano - Programming Lessons Learned; PDF to HTML; Alternate Substitution; Why RSA Works; Trigger; Dreaming of Death; Example: Tracing; Using Coroutines In Protocol Simulations; Python 3.3 Only; Pure Python SHA1 and MD4 Implementations; Ubuntu on VirtualBox; Starting TOR as a service on OpenSuse 12.3; 1001 Albums; Using fail2ban on OpenSuse 12.3; PPPoE on OpenSuse 12.3; Good Article on Unified Physics; It's Police (Carabineros); Linux Software for Listening to and Exploring Music; Android is Pretty Bad; Lucky Number; 3D Printing for Casting; Cover Art for MPDroid; Who'd a thought the French were so bigoted?; PS Input Signal; Small Problem with Roksan K2 Amp; Roksan K2 Amp + ATC SCM7 Speakers; Do What Makes Sense; Re: Arguing About Tests, Still; Arguing About Tests, Still; Images; Good Article on NY Drummers; Related Bug Report; Getting Python 3.3 and Virtualenv Working in OpenSuse 12.3; How I Am; Awesome video about digital audio; The Difference Between Dimensional and Normalized Databases; The rise of the new Chinese bogeyman; Updated Syntax; Very First Steps to C-ORM; The Ideal User Interface For Music Exploration; Can The Republicans Be Saved?; Rate Limiting Calls to EchoNest; Mods to Cache; Comparing UYKFG and UYKFD/E/F; Someone Else is Concerned; EchoNest-based Playlist Generator for MPD; Example Voting Results; A Heavyweight Python Cache; Identifying Artists with EchoNest; Notes on Pregalex / Pregabalina / Lyrica; The Neil Cowley Trio; Drake - Make for Data; A Reliable Python Web Service; Useful Python Date/Time Library?; Need to Sleep, But this is Good; Command Line Set Difference; Little Details...; Linux Command Line Tricks; AutoTools Tutorial; Hangman Tactics; A Tor Proxy Embedded In A Web Page; Tree (Nested Dicts) in Python; Sleeping at Parties; I Know Someone Who Hurts Other People; Light and Tea; Description of the LCS35 Time Capsule Crypto-Puzzle; Re: I can relate to that ...; I can relate to that ...; Re: It's 2012 Why Does My IDE Suck?; My Own Alternative Medicine; Nice explanation of SVM; Why and How Writing Crypto is Hard; Re: It's 2012 Why Does My IDE Suck?; Incremental Regular Expressions; BBC Map Confused at Pole; Social Media: Ground Zero in the Culture War; My Visit to the Psycho Doc; Learning Modern 3D Graphics Programming; Hope you got some crackers to go with the cheese; Re: But how easy would it be ...; But how easy would it be ...; Powerline Freq Fingerprinting of Audio; The Folly of Scientism; Cheese - Because You're Going to Die Anyway

Insomniac Typed Programming in Python

From: andrew cooke <andrew@...>

Date: Tue, 12 Apr 2011 02:51:49 -0300

Just got the following code to run (it will be in the next version of pytyp).
Note how there is dispatch by type defined in the TypedProperty subclass
(instances of which occur in both Tree and Node) and how functors are
naturally supported.


  def tree_functor(leaf_type):

      tree_type = Delayed()

      class TreeProperty(TypedProperty):

	  def __init__(self, value):
	      super(TreeProperty, self).__init__(value, tree_type)

	  @staticmethod
	  def size(value, spec):
	      return spec.on(value,
			     none=lambda _: 0,
			     leaf=lambda l: 1,
			     node=lambda n: len(n))

	  @staticmethod
	  def set_add(value, spec, leaf:leaf_type):
	      return spec.on(value,
			     none=lambda _: leaf,
			     leaf=lambda l: Node(l).add(leaf),
			     node=lambda n: n.add(leaf))

      class Node(Typed):

	  value = TypedProperty(leaf_type)
	  left = TreeProperty(None)
	  right = TreeProperty(None)

	  @checked
	  def __init__(self, value:leaf_type):
	      super(Node, self).__init__()
	      self.value = value

	  @checked
	  def add(self, value:leaf_type):
	      if value < self.value:
		  self.p.left.set_add(value)
	      else:
		  self.p.right.set_add(value)
	      return self

	  def __len__(self):
	      return 1 + self.p.left.size() + self.p.right.size()

      class Tree(Typed):

	  root = TreeProperty(None)

	  def add(self, value:leaf_type):
	      self.p.root.set_add(value)

	  def __len__(self):
	      return self.p.root.size()

      tree_type += Alt(none=None, leaf=leaf_type, node=Node)

      return Tree


  # create a functor that supports trees of integers
  Tree = tree_functor(int)
  t1 = Tree()
  t1.add(1)
  try:
      t1.add('bad')
      assert False, 'Expected error'
  except TypeError:
      pass
  for n in [8,3,6,5,9,2]:
      t1.add(n)
  assert len(t1) == 7, len(t1)

A Little More Detail

From: andrew cooke <andrew@...>

Date: Tue, 12 Apr 2011 03:18:55 -0300

To expand on the above, the Tree class contains a tree_type value called
root.  From the definition of tree_type we can see that root is either None
(empty tree), leaf_type (a single value), or Node.

Similarly, Node contains both a value (leaf_type) and two sub-nodes of
tree_type.

Because the add and size methods are defined on the typed parameter they can
be used (to extend the tree and count the number of nodes) from within both
classes.

And although the logic for those methods is quite complex (because of the
different types involved), the dispatch by type gives a clear defintion.

To explain displatch by types, here is the size function:

        @staticmethod
        def size(value, spec):
            return spec.on(value,
                           none=lambda _: 0,
                           leaf=lambda l: 1,
                           node=lambda n: len(n))

and the tree_type:

        Alt(none=None, leaf=leaf_type, node=Node)

The arguments in the "on" call match the names for the alternate types that
may be present.  When size() is invoked, the value is matched against the spec
and the correct expression evaluated.  So, for example, if value is None, the
result is 0.

The only other important detail is how TypedProperties are instantiated within
a class.  They are defined at the class level, but installed into each
instance via the Typed suprrclass.  When installed they are visible in two
ways.  First, as a type-verified value.  Second, under the ".p" attribute,
with the methods defined in the TypedParameter class (value and spec
automatically substituted).

So code like

    class Tree(Typed):
        root = TreeProperty(None)
        ...

supports the creation of tree instances for which self.root returns a value of
type tree_type (and which is verified when assigned to).  Those instances also
have self.p.root which has size() and set_add() methods (the "set_" prefix
means that the result of the method is used to update the attribute).

Andrew

Comment on this post