🚀 go-pugleaf

Wolfgang Denk replied to my discussion of the following problem 
within mined:

> >Mined runs fine on SunOS, Micro-Vax Ultrix, Iris IRIX, VMS, and MSDOS including the 
> >following assignment you feel disturbed by and its later uses:
> >char * (* fnamv) [];
> >  ...
> >  char * argv [];
> >  fnamv = argv;	/* Why does this produce a warning? C is such a stupid language! */
> >I really don't think it's my fault that compilers produce warnings like 
> >"illegal pointer combination" on this. 
> 
> But this _IS_ an illegal pointer combination, and  a  C  compiler
> _is_ expected to complain about it. 
I was actually not complaining about C compilers, but the C language.
Since I am dealing with a pointer to an array (of pointers), the solution 
you and others proposed and which works, where the declaration is 
  char * * fnamv  , does not look very logical, does it? Nor does the fact 
that every attempt to include [] in the declaration yielded an error or 
warning.

> >On the one hand, type inconsistencies are quite common in C, aren't they?, and the 
> 
> Bad argument - never follow bad examples!
That was not meant to be an argument, of course, but sarcasm.


Thank you for your further proposals, I comment on some of them below.

Thomas Wolff
wolff@inf.fu-berlin.de
-----------------------------------------------------------------------------

> ! #if	(defined(vms) || defined(msdos) || defined(USG))
I don't think defined(xx) would work with every C compiler.
Can you tell exactly what USG stands for? Is it a symbol that will always 
be predefined on the appropriate systems or will it have to be configured?

> BTW - ANSI C does not allow text other than comments following an 
> #endif. After "#ifndef msdos" a  "ifdef  msdos"  will  always  be
> evaluate  to  false  (I hope, at least :-). 
At several places, I just left some (not too) old code in the source 
in case it might be necessary to reactivate it.

> getwd is usually only
> available in BSD-like environments, not all flavour of  System  V
> provide  it. Etc. - the code has really lots of (mostly minor and
> easy to fix) problems. 
Whereever there is interest to adapt the editor, I would like to hear of 
all such problems.

> I don't call that a Makefile.
Well, it works for a single compilation. I don't use it for development.

> *** minedmp.c!	Wed Jul 29 08:53:17 1992
> --- minedmp.c	Wed Jul 29 12:54:15 1992
> ***************
> *** 98,104 ****
>   	case 'P' : return '^';
>   	case 'p' : return '~';
>   	case '|' : return '&';
> ! 	case '&' : return ''';
>   	case '\\' : return ',';
>   	case '_' : return '/';
>   	case '-' : return '-';
> --- 98,104 ----
>   	case 'P' : return '^';
>   	case 'p' : return '~';
>   	case '|' : return '&';
> ! 	case '&' : return '\'';
>   	case '\\' : return ',';
>   	case '_' : return '/';
>   	case '-' : return '-';
  etc.
These are problems with the Internet news system. At some sites, the files 
will come with bit 8 stripped off (or even only be saved without it, e.g. 
nn has a switch for that). All the characters transformed by code similar 
to the above in minedmp.c have their 8th bit set. Unfortunately, some 
characters get stripped so that a syntax error results. Maybe I should 
post a uuencoded version.