Re: [Plplot-devel] Proposed code change for plshade.c

[Alan W. I. <ir...@be...>]

On 2005-10-25 21:29+0200 Werner Smekal wrote:

> Hi,
>
> I would like to propose a change for plshade.c:
>
> I would like to rename the function poly() in plshade.c and all calls to that 
> function to plpoly() or something like that. The reason for this is, that I 
> made a makefile for the borland compiler 5.5 (free edition) and poly() is 
> already defined which leads to compiler errors. Since this function is only 
> used in this file, I see no problems in renaming this function, or? This 
> would make it possible to have the plplot library compiling also with 
> different compilers on Windows (using the wxwidgets/ps/png driver).

I don't mind a bit changing the name (since this function is completely
internal to plshade.c).  However, I don't understand why there is a name
clash since this function is declared static and therefore it should be
completely internal to plshade.c and not interfere with anything else.

I am still trying to learn some of the exotica of C so this may be a dumb
question for the C experts here, but why doesn't the static declaration of
poly(...) in plshade.c protect against such name clashes? If the answer is
"it should" (i.e., the borland compiler 5.5 is broken) that's fine and we
can change the name anyhow to avoid the name clash for that compiler, but
it's best to know why such name-clash avoiding changes are required.

To make the example even more obvious, in C aren't you free to define your
own static sin function for our own version of the sine routine or is "sin"
disallowed for static function names because there is a math library
function of that name?

In fact I just answered that question with the following code:

cat test.c
#include <stdio.h>
#include <math.h>
static double sin(double x);
double sin(double x) {
    return(x);
}
int main(void) {
    double x;
    int i;
    for(i=0; i < 100; i++) {
       x = ((double)i)*0.01;
       printf("x = %10.2f, sin(x) = %10.2f\n", x, sin(x));
    }
    return(0);
}

"gcc -pedantic -ansi -Wall test.c -lm" didn't complain a bit, and on
execution it gave the "correct" results (sin(x) = x as expected from the
local function rather than what you get from the math library).  If I remove
the static declaration and the definition of the local sin function, then
the math library version is used as expected.

Werner, if you try the above test code with your Borland compiler does
it complain?  If not, what is the exact error message you get with plshade.c?

Alan
__________________________
Alan W. Irwin
email: ir...@be...
phone: 250-727-2902

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________