Re: This I simply can't swallow

From: Alwyn (alwyn_at_blueyonder.co.uk)
Date: 02/17/05

  • Next message: Alwyn: "Re: optimised" 
    Date: Thu, 17 Feb 2005 19:32:55 +0000

    On Thu, 17 Feb 2005 17:43:19 +0000, Francis Glassborow wrote:

    > In article <pan.2005.02.17.16.23.33.969450@blueyonder.co.uk>, Alwyn
    > <alwyn@blueyonder.co.uk> writes
    >>A concrete type is a type the representation of whose values is visible to
    >>the user. C nuilt-in types and most user-defined types are generally
    >>regarded in this light.
    >
    > And there-in lies a common misconception. In C and C++ most of the
    > fundamental types are abstract in the sense that neither languages
    > specifies how the values shall be represented. About the only exception
    > are the unsigned types which have to be a binary representation of the
    > values (note that the signed types allow several distinct
    > representations even though the behaviour is specified)

    I would agree that efforts towards abstraction has been made in the
    standards, but it is also quite easy to determine properties of the
    representation of types on any implementation by means provided by the C
    or C++ languages; very often you do need this information in order to
    write programs that will work on multiple platforms.

    Alwyn

  • Next message: Alwyn: "Re: optimised"

    Relevant Pages

    • Re: Default rep specs for record types - documented??
      ... representation that numbers the first element as zero, ... "The position number of the value of the first listed enumeration literal is zero; the position number of the value of each subsequent ... this representation is specified by the language. ... If component_clauses are given for all components, the record_representation_clause completely specifies the representation of the type and will be obeyed exactly by the implementation." ...
      (comp.lang.ada)
    • Re: Three questions about signed/unsigned type representations
      ... *internal* representation of values, but it specifies *this* (let me call it ... necessary to perform bitwise operations. ... performing the same operations on unsigned ints, ...
      (comp.lang.cpp)
    • Re: value bits
      ... The definition of pure binary representation specifies what ... combination of bits is a valid value or a trap representation. ... allowed by the standard. ...
      (comp.lang.c)
    • Re: some tricky questions
      ... specifies nothing about the bit order. ... On a conforming implementation, it ... known as the value representation." ... Mark McIntyre ...
      (comp.lang.c)
    • Re: Default rep specs for record types - documented??
      ... > Steve wrote: ... >> relation between the representation values, ... > ARM 3.5.1 defines the position numbers, and ARM 13.4 specifies that the ... Compiler A may choose to use 8 bits to represent an instance of Color, ...
      (comp.lang.ada)