|This article relies too much on references to primary sources. (June 2015) (Learn how and when to remove this template message)|
|Paradigm||functional, procedural, object-oriented, declarative, reflective, meta|
|Designed by||Alexander Burger|
17.6 / June 29, 2017
|Typing discipline||duck, dynamic, strong|
Its most prominent features are simplicity and minimalism. It is built on top of a single internal data type (cell), without giving up flexibility and expressive power. On the language level the programmer can use three different data types (numbers, symbols and lists) being represented by cells and differentiated by bits at the end of the cell.
Because the only non-atomic data type is the linked list, many interoperable functions exist that concentrate on list processing. As a result, PicoLisp programs are often more succinct - and at the same time faster - than those of other interpreted languages (see examples from Rosetta Code). Functions are free from the restrictions that would be imposed by a compiler, and can so accept arbitrary types and numbers of arguments. Macros are needed only in rare cases and are implemented using the quote function. PicoLisp breaks a tradition by dropping the lambda function seen in many other Lisps, like Common Lisp and Emacs Lisp. This is caused by the fact that the quote function is changed to return all its arguments unevaluated, not only the car of the first. This results in a more readable syntax, less run-time and no necessity to have a lambda function.
A special feature is the intrinsic CRUD functionality. Persistent symbols are first-class objects, they are loaded from database files automatically when accessed, and written back when modified. Applications are written using a class hierarchy of entities and relations. Additional features include: Prolog engine and database queries, distributed databases, inlining of C language functions and native C function calls, child process management, interprocess communication, browser GUI, and internationalization.
Originally developed on the Apple Macintosh in the 1980s, and used in commercial application development since then. It was soon ported to MS-DOS and SCO Unix, and used mainly on Linux since 1993. Database functionality was added in the mid-1990s.
While the first versions were written in a mix of C and assembly language, a first rewrite from scratch was done in 1999 completely in C. That version was released 2002 under the GNU GPL license, and changed to a MIT/X11 license in 2010.
In September 2014, Burger announced the PilMCU project on the PicoLisp development listserv, an effort with George Orais to implement PicoLisp directly in hardware in a spirit reminiscent of Lisp Machines.
- Burger, Alexander. "Internal structures". http://software-lab.de. External link in
- "Category:PicoLisp Rosetta Code". http://rosettacode.org. External link in
- "Even small details make a difference!".
- Burger, Alexander (8 May 2013). "Ersatz PicoLisp". http://software-lab.de. External link in
- "Announce: PicoLisp in Hardware".
- PicoLisp Home
- Latest Version (immediate download)
- Download page
- Reference manual
- Frequently asked questions
- Paper A Radical Approach to Application Development (PDF)
- #picolisp connect IRC channel
- Mailing list archive
- ohloh project page
- Lambda the Ultimate 2007 A Case for Minimalist Interpreters?
- John Fremlin's blog PicoLisp takes the performance crown?
- First Glance
- c't Magazine Database contest results (in German)
- Macropis, a web development framework in Picolisp
- MIT-licensed PicoLisp libraries (bcrypt, json, nanomsg, unit testing, https)
|Lisp 1.5||Lisp 1.5|
|ZetaLisp||Lisp Machine Lisp|
|Common Lisp||Common Lisp|
|Emacs Lisp||Emacs Lisp|