In article <1992Aug6.081846.22050@axion.bt.co.uk> pknighto@axion.bt.co.uk writes:
>Hello People,
>2 novice queries:-
>
>i) Could someone provide me with/refer me to a reasonably concise summary of
> the issues that make different C++ language/compilers incompatible.
>
> Am I right in my understanding that it is not that the language
> definition that has not been standardised,

 Nope, the language has NOT been standardised. Yet :-)

> but that the internal
> representation of the class/objects has not, which leads to an
> inability to link object files from different compilers. What moves
> are afoot to resolve existing incompatibilities ?

 None. On the contrary, the inability to link modules from different
 compilers is more or less considered an ADVANTAGE. Since C and thus
 C++ is close to machine hardware, internal layouts will
 always be different, and different name mangling schemes an
 linker layouts protect you from porting unportable binaries.

--
;----------------------------------------------------------------------
        JOHN (MAX) SKALLER,         maxtal@extro.ucc.su.oz.au
 Maxtal Pty Ltd, 6 MacKay St ASHFIELD, NSW 2131, AUSTRALIA
;--------------- SCIENTIFIC AND ENGINEERING SOFTWARE ------------------

Hello People,
2 novice queries:-

i) Could someone provide me with/refer me to a reasonably concise summary of
 the issues that make different C++ language/compilers incompatible.

 Am I right in my understanding that it is not that the language
 definition that has not been standardised, but that the internal
 representation of the class/objects has not, which leads to an
 inability to link object files from different compilers. What moves
 are afoot to resolve existing incompatibilities ? What compilers are
 'recommended' as being the most reliable/ robust/ efficient/
 interoperable (mainly on Sun-SPARC, HP-UX & PC platforms) ? If this
 understanding is completely false then please make criticism
 instructive, not destructive (:->).  Thanks.

ii) I am aware of a few static analyzers which have been "extended" from
 parsing C to parsing C++. Could someone provide me with/refer me to a
 more complete list of scanning/measurement tools which are considered
 USEFUL.  If no/few such tools exists is there at least a 'de facto'
 standard/consensus on just what metrics/measurements are useful to
 apply to C++ either from academics or 'real programmers'.

Hope these don't insult anyone's intelligence.

Thanks in advance for any help forthcoming,

Paul Knighton
-----------------------------------------------------------------------------
+  Paul A. M. Knighton,                  BoaT/Office Automation Group       +
+  Software Development Div.,            Tel : (+44) (0)473 649159          +
+  Tray 2b G61 B81,                      FAX : (+44) (0)473 644605          +
+  BT Laboratories,                                                         +
+  Martlesham Heath,            Email (Internet) : pknighto@axion.bt.co.uk  +
+  Ipswich, IP5 7RE, ENGLAND.                                               +
-----------------------------------------------------------------------------