Jump to content

Edinburgh IMP

From Wikipedia, the free encyclopedia

This is the current revision of this page, as edited by 70.124.38.160 (talk) at 22:21, 2 May 2024 (Sources: replaced archive.org link to skimp doc with current link (maintained copy, has corrections)). The present address (URL) is a permanent link to this version.

(diff) ← Previous revision | Latest revision (diff) | Newer revision → (diff)

Edinburgh IMP
ParadigmsProcedural, imperative, structured
FamilyALGOL
Designed byBratley, Whitfield, M. M. Barritt, David Rees, Peter D. Schofield, Roderick McLeod, Hamish Dewar, Peter D. Stephens, Peter Robertson
DeveloperUniversity of Edinburgh
First appeared1966; 58 years ago (1966)
Typing disciplineStatic, strong
ScopeLexical
Implementation languageAtlas Autocode, IMP
PlatformEnglish Electric KDF9,
ICL System 4, UNIVAC 1108, IBM System/360, DEC PDP-9, DEC PDP-10, DEC PDP-11, DEC PDP-15, CTL Modular One, x86, Interdata 7-16, Interdata 7-32, Data General Nova
OSBOS/360, DOS, Windows, Linux
Major implementations
Edinburgh IMP
Dialects
IMP77, IMP80
Influenced by
ALGOL 60, Atlas Autocode

Edinburgh IMP is a development of Atlas Autocode, initially developed around 1966-1969 at the University of Edinburgh, Scotland. It is a general-purpose programming language which was used heavily for systems programming.

Expressively, IMP is highly similar to ALGOL and includes all the ALGOL-style block structure, reserved words (keywords), and data types such as arrays, and records. It adds to ALGOL-style languages a string type (an array of characters, although these have a predeclared size) and built-in operators for string manipulation and character handling. One significant difference from ALGOL is that IMP does not support parameters passed by name, although it does support parameters passed by reference.

IMP provides significant control over the storage mapping of data, plus commands for addressing within parts of words. Most IMP compilers offer compiler-generated runtime checks and a stack trace (backtrace) facility by default, even in production code. IMP allows inline assembler machine language instructions in source code.

The ERCC Implementation of IMP for the ICL System 4 (known as IMP9) offered a syntax-driven macro facility (designed by Alan Freeman) that was similar to the Compiler Compiler features offered by IMP's predecessor, Atlas Autocode.

Early IMP compilers were developed for the English Electric KDF9, ICL System 4, UNIVAC 1108, IBM System/360, DEC PDP-9, DEC PDP-15 and CTL Modular One computers. IMP was used to implement the Edinburgh Multiple Access System (EMAS) operating system, and a compiler was written for the ICL 2900 series to allow porting of EMAS to that platform. In later years, a version of IMP named IMP77 was developed by Peter Robertson within the Computer Science department at Edinburgh which was a portable compiler that brought IMP to even more platforms. In 2002, the IMP77 language was resurrected by the Edinburgh Computer History Project for Intel x86 hardware running DOS, Windows, and Linux, and is once again in use by Edinburgh graduates and ex-pats.

The diverged IMP and IMP77 were later consolidated into one language with the introduction of the IMP80 standard, supported by implementations from the Edinburgh Regional Computer Centre. IMP80 has also been ported to several platforms including Intel and was actively in use into the 1990s.

Edinburgh IMP is unrelated to the later IMP syntax-extensible programming language developed by Edgar T. Irons, for the CDC 6600, which was the main language used by the National Security Agency (NSA) for many years.

See also

[edit]

Sources

[edit]

References

[edit]