Pika

Pika is a dynamic, multi-paradigm programming language.

Features Include

• First class functions and Closures.
• Class based object system.
• Automatic memory managment through an incremental, mark and sweep, garbage collector.
• Packages.
• Properties with get and set accessor functions.
• Native API for binding with C/C++.
• Unified import system works with scripts and native C/C++ modules.
• Exception handling.
• Block finalizers.
• Lambda expressions.
• Anonymous function expressions.
• Default values for functions.
• Variable argument functions.
• Keyword arguments.
• Lexically scoped.
• Multiple return and yield values.
• Coroutines (aka cooperative threads).
• Generators (light weight coroutines).
• Annotations for functions, classes and packages.
• Array comprehension. [ i for x = 1 to 20 if x mod 2 ]

Some Examples

See the samples subdirectory for a complete list of sample scripts.

You can import them once installed by doing the following:

# Install sample file classes
import "samples/classes.pika" 

Or run them from the command line:

pika samples/classes.pika

###Classes

class Person
    function init(name, age)
        self.name = name
        self.age = age
    end
    
    function toString()
        return 'Person {self.name} {self.age}'
    end
end

class Employee: Person
    function init(name, age, occupation)
        super(name, age)
        self.occupation = occupation
    end
    
    function toString()
        return 'Employee {self.name} {self.age}, {self.occupation}'
    end
end

jane = Person.new('Jane Doe', 33)
john = Employee.new('John Smith', 35, 'Programmer')

print jane is Person #=> true
print john is Person #=> true    

print jane is Employee #=> false
print john is Employee #=> true

print jane
print john

###Generators

function fib()
    a, b = 0, 1
    loop
        yield a
        a, b = b, a + b
    end
end

fibIterator = fib()

print fibIterator()
print fibIterator()
print fibIterator()

{* Generators can also created from instance methods. *}
class RangeMultiplier
    function init(scalar)
        self.scalar = scalar
    end
    
    function range(a, b)
        for i = a to b
            yield i * self.scalar
        end
    end
end

# Create a RangeMultiplier instance
rangeMultiplier = RangeMultiplier.new(scalar=2)

# Create the generator bound to the instance
multIter = rangeMultiplier.range(1, 6)

while multIter
    print multIter()
end

Get Pika

This document will explain how to build and install the Pika library, interpreter and modules from source using the CMake build system.

First Get CMake

(If CMake is already installed skip step 1.)

####Step 1

Go to the CMake home page at http://www.cmake.org and follow the instructions for installing CMake for your platform.

• On Windows you will probably want to use the CMake installer.
• On Mac OS X you can use fink, Mac Ports, Homebrew or the CMake installer.
• On a *nix platform if might be preferable to use the system's package manager.

####Step 2

Now open up the terminal and navigate to the project's root directory.

####Step 3

At the terminal type: note: (don't type the > character)

> mkdir build && cd build
> cmake ../../pika
> make
> sudo make install

The CMake option -DCMAKE_INSTALL_PREFIX=INSTALL_PATH can be used to specify a different location. You might want to point it to your user directory if you do not have administrator privileges.

cmake -DCMAKE_INSTALL_PREFIX=/home/myhome ../../pika

To build without modules use the following

cmake -DPIKA_NO_MODULES=1 ../../pika

Modules

In addition to the builtin modules: os, math and ctime; the following can be built and installed.

You can import them by calling the import method with one or more module names.

For example:

math, datetime, os = import "math", "datetime", "os"

Will load and store the math, datetime and os modules in variables of the same name.

Additional Modules

base64

• A base 64 encoder and decoder, meant to be an example module.

bignum

• Arbitrary precision arithmetic for integers and real numbers (floating point numbers).
• Depends on GMP and MPRF for the BigInteger and BigReal classes respectively.

bzip2

• Compression and decompression of data using bzip2 (included).

curses

• Bindings for curses/ncurses library.
• Depends on curses/ncurses, no support for pdcurses at this time.

datetime

• Provides Date, DateTime, Time, Timezone and TimeSpan classes and date manipulation.

event

• Bindings for LibEvent for use with the Socket and File classes.
• Depends on LibEvent version 2.0.1 alpha and up.

json

• JSON encoder and decoder.

random

• Pseudorandom number generator based on Mersenne Twister (included).
• Currently part of the math module.

re

• Regular expressions module.
• Depends on PCRE which is now included in the code base. The project's home page is http://www.pcre.org/. I use PCRE version 8.10, so any version compatible with that release should work. Specify RE_USE_EXTERNAL_PCRE to search for the system's PCRE.

socket

• Berkley Sockets library.
• Depends on Posix Socket support, not ported to Winsock yet.

unittest

• Unittest module.
• Depends on the re module.

zlib

• Compression and decompression of data using zlib (included).