Pike 0.6 BETA
Pike 0.6 BETA is now available for public beta testing, though
unfortunately it is not quite complete. Due to Swedish export
restrictions we are unable to supply the cryptographic toolkit and the
full library of algorithms.
Pike version 0.6 BETA does not come with a manual yet, there is a manual
for Pike 0.5, which will be updated to cover all new features in 0.6 as
soon as possible. There are many changes from Pike 0.5 to Pike 0.6, but
here are some of the most important ones you should be aware of:
o The compiler is now two-pass, which means that you can access
variables and functions declared later in the source without
having to use prototypes. The two-pass compiler also makes
it possible to use the name of a class from inside itself.
o It is now possible to access variables and functions in the
'parent' class. This greatly simplifies coding in many
o The process handling has been changed to being object-
oriented. For this purpose, a new function has been created:
Process.create_process. The returned object can be used
as an argument to kill, and has methods which will report
the status of the child process.
o Similar to process handling, threads have been changed to use
an object-oriented approach.
o The 'varargs' keyword has been removed.
o You can now 'import "dirname"'.
o The module system has been modified so that the current
directory is no longer searched for modules. To use a module
in the current directory, you must either use 'import "."'
or specify the module as '.module' instead of just 'module'.
o Lots of bugs have been fixed. :)
o Implicit blocks have been implemented around if, switch, for,
do-while and foreach, this means that the scope of variables
such as 'x' in 'if(string x)' ends after the if construct.
o popen, system and spawn in the Process module have been changed
to *NOT* execute /bin/sh on the given string. This is to ensure
compatibility with Win32 systems.
While still young, Pike matured over the last few years, and
development is now co-ordinated by a team at Idonex AB. Pike is still
released under the GNU General Public License (GPL). It is, and will
continue to be, freely distributed. Briefly, this means that you are
free to use and modify the source code in any way you please, see the
file COPYING for details. Idonex maintains the right to the brand name
and the copyright, guaranteeing that Pike will be continuously
developed and enhanced, and remain free software.
Pike is easily learned, and just as easily used to develop powerful
applications. Pike is designed to be useful and powerful. As the
syntax of Pike is similar to that of C, most programmers will find it
easy to use. As Pike is a high-level, interpreted and modular
object-oriented language, powerful applications can be rapidly
developed. Pike has evolved rather than been designed. The changes to
Pike that have been made have been guided by its users' needs. In
general, the better you get to know Pike, the more you will appreciate
it, from a user's perspective.
Pike is just as suitable for intricate and large applications as for
creating use-once scripts. When needed, you can also write
Pike modules in C, to gain better performance or access C
libraries. This is transparent to the module's users; the modules
written in C and those written in Pike are accessed in exactly the
A list of pros and cons with Pike follows:
+ General Purpose Programming Language
+ Freely distributed under GPL
+ Interpreted - No compilation and linking needed to run code
+ Object Oriented - Straightforward and powerful object
orientation, including multiple inheritance
+ Modular - Extendible with modules written in C or Pike
+ Powerful - High-level language, concise code
+ Incremental - Allows on-the-fly modifications and recompilations
+ Source Code Available - To promote sharing of useful code
+ Portable - Platform independent and available on most flavors of Unix
At present the following platforms are supported:
Solaris 2.5 & 2.6 (Sun SPARC & Intel x86)
Digital UNIX 4.0
Linux Red Hat 4.2 (Intel x86)
+ Powerful, Flexible and Efficient Data Types
+ Automatic Memory Management
+ Scalable - as useful for small scripts as for powerful and
+ Text Processing - e.g. for filtering
+ Exception Handling
+ Multi-Threaded Application Development Possible
How Pike relates to other programming languages:
+ Optimized - Faster run-times than Perl, Python, tcl and Java
+ Faster Development than using C or C++
+ C-like Syntax - Easy to learn
Below you will find a list of some desired but missing features, or
cons, of Pike. As indicated by the version number 0.5, Pike is still
under development and the goal is to incorporate those in future
- No Pike native compiler or debugger available
- Limited image format import and export support
- Not optimal for heavy numeric processing jobs
- Not for GUI based applications
- Limited range of special effect image filters for image
Some successful applications using Pike include:
+ Network Applications - as Client/Server applications
e.g. the Roxen Challenger Server and a range of server extensions
Client/server Protocol Implementations
+ System Administration Tools & Utilities - Even as quick one-shots
e.g. adduser or adddomain scripts that setup users/domains
Real-time Client/Server Status Monitor
+ Assorted Text Processing Utilities
+ CGI Programs
Pike is distributed accompanied by a set of modules, some of which are
Image - Image Processing Module
The Image module lets you perform common image manipulation and
processing operations from Pike programs. The supported operations
include cutting and pasting, scaling, rotation, linear filtering of
most kinds, rendering text, etc. Anti-aliasing can be used when
performing any kind of drawing operation. Using the Image module, you
can automate jobs that would otherwise have required tedious work in
an interactive image processing program such as Adobe PhotoShop.
Currently, the only input format is ppm. Output formats are ppm and
Sql - Interface to SQL Databases
The SQL interface makes it possible to use common SQL databases from
within your applications. It includes support for mSQL, MySQL and
MIME - Multimedia Internet Mail Extension Module
This module makes it easy to create and decode e-mail, including
Gz - Data Compression Module
With this module you can compress data. Its name is slightly
misleading, as it uses ZLIB. ZLIB uses the same compression algorithm
as gzip, but the format for the compressed data is not compatible with
Crypto - Cryptographic Toolkit
The crypto toolkit is a set of objects implementing various
cryptographic algorithms, as well as objects used to glue the other
objects together. The toolkit includes block ciphers IDEA, DES
(including tripple DES) and CAST, the RC4 stream cipher, the hash
functions MD5 and SHA, and the public key cryptography algorithm
RSA. It also includes a some randomness sources.
The toolkit can be used for most kinds of cryptographic applications,
including encrypted network connections, as well as encryption of
files or other data.
Unfortunately, the cryptograpy toolkit in this release has been
stripped of all real cryptography. It includes only the hash functions
and the glue objects. The full cryptographic toolkit will be released
as soon as the export issues have been resolved.