> On 12 Jan 2021, at 02:41, Andreas Rheinhardt <andreas.rheinha...@gmail.com>
> wrote:
> Of course I am all ears for how to make it clear that someone who
> modifies the strings also needs to check the array dimensions.
I think I kind of agree with the other comments, this would/should
rather have to be something that can be checked in an automated way.
In principle I also do not much like this solution, it does not
work very well when the strings are of very different sizes.
I’ve not found a solution that is really worth the effort needed,
but I think it would be better to have an approach that is more
along the lines of what PIC does: encode only offsets.
In assembler you can just encode the strings and then have an array
with the offsets to them, but in C that is undefined behaviour…
Where performance doesn’t matter, what you definitely can do
is to just dump one string after the other in an array and then
use a function to scan for the desired index.
The ideal solution from my point of view, would be to have
such an array of all strings together and then an array with
the offsets to each.
With some kind of special preprocessor or code generator that
would be easy, but that makes the code messy.
Doing it in pure C also ends up quite a mess, the below is about as
far as I got, no idea if someone knows some magic to make it actually
bearable…
#include <stdio.h>
#define STRINGS \
X(0, 1, "test1") \
X(1, 2, "test2") \
X(2, 3, "testteststeste2") \
#define X(n, m, x) x "\0"
static const char stringdata[] = STRINGS ;
#undef X
#define X(n, m, x) static const int strpos##m = strpos##n + sizeof(x);
static const int strpos0 = 0;
STRINGS
#undef X
int main()
{
printf("0: %s\n", stringdata + strpos0);
printf("1: %s\n", stringdata + strpos1);
printf("2: %s\n", stringdata + strpos2);
}
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
