Reputation:
Any idea why this program will fail in downloading images from page #12 while it did pages 1-9 perfectly? I really don't know how can I debug. Maybe there's a problem with wget when it doesn't find the first image? http://img717.imageshack.us/img717/7954/white2u.png
#include <stdio.h>
#include <stdlib.h> // for using system calls
#include <unistd.h> // for sleep
int main ()
{
char body[] = "forum-post-body-content", notes[] = "p-comment-notes", img[] = "img src=", link[200], cmd[200]={0}, file[10];
int c, pos = 0, pos2 = 0, fin = 0, i, j, num = 0, found = 0;
FILE *fp;
for (i = 12; i <= 149; ++i)
{
sprintf(cmd,"wget -O page%d.txt 'http://www.mtgsalvation.com/forums/creativity/artwork/340782-official-digital-rendering-thread?page=%d'",i,i);
system(cmd);
sprintf(file, "page%d.txt", i);
fp = fopen (file, "r");
while ((c = fgetc(fp)) != EOF)
{
if (body[pos] == c)
{
if (pos == 22)
{
pos = 0;
while (fin == 0)
{
c = fgetc (fp);
if (feof (fp))
break;
if (notes[pos] == c)
{
if (pos == 14)
{
fin = 1;
pos = -1;
}
++pos;
}
else
{
if(pos > 0)
pos = 0;
}
if (img[pos2] == c)
{
if (pos2 == 7)
{
pos2 = 0;
while (found == 0)
{
c = fgetc (fp); // get char from file
link[pos2] = c;
if (pos2 > 0)
{
if(link[pos2-1] == 'g' && link[pos2] == '\"')
{
found = 1;
}
}
++pos2;
}
--pos2;
found = 0;
char link2[pos2];
for (j = 1; j < pos2; ++j)
{
link2[j - 1] = link[j];
}
link2[j - 1] = '\0';
sprintf(cmd, "wget -O /home/arturo/Dropbox/Digital_Renders/%d \'%s\'", ++num, link2);
system(cmd);
pos2 = -1;
}
++pos2;
}
else
{
if(pos2 > 0)
pos2 = 0;
}
}
fin = 0;
}
++pos;
}
else
pos = 0;
}
// closing file
fclose (fp);
if (remove (file))
fprintf(stderr, "Can't remove file\n");
}
}
Upvotes: 0
Views: 160
Reputation: 511
You have an overflow.
file[10];
page1.txt = 10 characters including the null terminator
page12.txt = 11 characters
Look into to the safe functions like snprintf()
Upvotes: 0
Reputation: 4767
char file[10];
"page12.txt"
has 11 characters in it including the null character. Please just do something like char file[128]
. Memory is cheap. Time spent debugging is expensive.
Upvotes: 2