此代码用%20
替换了空格,char数组应该包含足够的空格来添加%20
。
你能帮助我理解为什么我会在*str = *tmp; // segfault at this location..
得到段错误吗?
#include <stdio.h>
#include <string.h>
char* encodeSpace(char* str) {
char *orig = str;
while (*str++);
str--;
char *tmp = --str;
while (*tmp-- == ' ');
tmp++;
while (tmp != orig) {
while(*tmp != ' ' && tmp != orig) {
*str = *tmp; // segfault at this location..
tmp--;
str--;
}
*str-- = '0';
*str-- = '2';
*str-- = '%';
tmp--;
}
return tmp;
}
main()
{
printf("output is %s", encodeSpace("My Name is john "));
}
答案 0 :(得分:3)
您遇到分段错误,因为您尝试修改字符串文字(函数的参数)。
字符串文字通常位于只读内存中,可能无法合法修改。尝试这样做会导致未定义的行为。分段错误是一种合法的未定义行为。
如果你写的话,你会没事的。
int main(void)
{
char str[] = "My Name is john ";
printf("output is %s", encodeSpace(str));
return 0;
}
或甚至(使用C99复合文字):
int main(void)
{
printf("output is %s", encodeSpace((char[]){"My Name is john "}));
return 0;
}
因为你可以在一个循环中递减tmp
两次,所以你不能保证它何时到达orig
指针。此代码在assert()
:
#include <assert.h>
#include <inttypes.h>
#include <stdio.h>
#include <string.h>
static
char *encodeSpace(char *str)
{
char *orig = str;
printf("-->> str = %p <<%s>>\n", (void *)str, str);
while (*str++)
;
str--;
char *tmp = --str;
while (*tmp-- == ' ')
;
tmp++;
printf("tmp = %p <<%s>>\n", (void *)tmp, tmp);
printf("str = %p <<%s>>\n", (void *)str, str);
while (tmp != orig)
{
while (*tmp != ' ' && tmp != orig)
{
printf("--1: tmp = %p <<%s>>; str = %p <<%s>>\n", (void *)tmp, tmp, (void *)str, str);
*str = *tmp; // segfault at this location..
tmp--;
str--;
printf("--2: tmp = %p <<%s>>; str = %p <<%s>>\n", (void *)tmp, tmp, (void *)str, str);
}
//if (tmp != orig)
//{
*str-- = '0';
*str-- = '2';
*str-- = '%';
tmp--;
//}
printf("--3: tmp = %p <<%s>>; str = %p <<%s>>\n", (void *)tmp, tmp, (void *)str, str);
assert(tmp >= orig);
}
printf("<<-- tmp = <<%s>>\n", tmp);
return tmp;
}
int main(void)
{
printf("output is <<%s>>\n", encodeSpace((char[]){"My name is John "}));
return 0;
}
删除注释,它不再崩溃(tmp
的第二次减少受到保护)。但输出并不是你想要的。
-->> str = 0x7fff54cb44c0 <<My name is John >>
tmp = 0x7fff54cb44ce <<n >>
str = 0x7fff54cb44d7 << >>
--1: tmp = 0x7fff54cb44ce <<n >>; str = 0x7fff54cb44d7 << >>
--2: tmp = 0x7fff54cb44cd <<hn n>>; str = 0x7fff54cb44d6 << n>>
--1: tmp = 0x7fff54cb44cd <<hn n>>; str = 0x7fff54cb44d6 << n>>
--2: tmp = 0x7fff54cb44cc <<ohn hn>>; str = 0x7fff54cb44d5 << hn>>
--1: tmp = 0x7fff54cb44cc <<ohn hn>>; str = 0x7fff54cb44d5 << hn>>
--2: tmp = 0x7fff54cb44cb <<John ohn>>; str = 0x7fff54cb44d4 << ohn>>
--1: tmp = 0x7fff54cb44cb <<John ohn>>; str = 0x7fff54cb44d4 << ohn>>
--2: tmp = 0x7fff54cb44ca << John John>>; str = 0x7fff54cb44d3 << John>>
--3: tmp = 0x7fff54cb44c9 <<s John %20John>>; str = 0x7fff54cb44d0 << %20John>>
--1: tmp = 0x7fff54cb44c9 <<s John %20John>>; str = 0x7fff54cb44d0 << %20John>>
--2: tmp = 0x7fff54cb44c8 <<is John s%20John>>; str = 0x7fff54cb44cf << s%20John>>
--1: tmp = 0x7fff54cb44c8 <<is John s%20John>>; str = 0x7fff54cb44cf << s%20John>>
--2: tmp = 0x7fff54cb44c7 << is Johnis%20John>>; str = 0x7fff54cb44ce <<nis%20John>>
--3: tmp = 0x7fff54cb44c6 <<e is J%20is%20John>>; str = 0x7fff54cb44cb <<J%20is%20John>>
--1: tmp = 0x7fff54cb44c6 <<e is J%20is%20John>>; str = 0x7fff54cb44cb <<J%20is%20John>>
--2: tmp = 0x7fff54cb44c5 <<me is e%20is%20John>>; str = 0x7fff54cb44ca << e%20is%20John>>
--1: tmp = 0x7fff54cb44c5 <<me is e%20is%20John>>; str = 0x7fff54cb44ca << e%20is%20John>>
--2: tmp = 0x7fff54cb44c4 <<ame isme%20is%20John>>; str = 0x7fff54cb44c9 <<sme%20is%20John>>
--1: tmp = 0x7fff54cb44c4 <<ame isme%20is%20John>>; str = 0x7fff54cb44c9 <<sme%20is%20John>>
--2: tmp = 0x7fff54cb44c3 <<name iame%20is%20John>>; str = 0x7fff54cb44c8 <<iame%20is%20John>>
--1: tmp = 0x7fff54cb44c3 <<name iame%20is%20John>>; str = 0x7fff54cb44c8 <<iame%20is%20John>>
--2: tmp = 0x7fff54cb44c2 << name name%20is%20John>>; str = 0x7fff54cb44c7 << name%20is%20John>>
--3: tmp = 0x7fff54cb44c1 <<y na%20name%20is%20John>>; str = 0x7fff54cb44c4 <<a%20name%20is%20John>>
--1: tmp = 0x7fff54cb44c1 <<y na%20name%20is%20John>>; str = 0x7fff54cb44c4 <<a%20name%20is%20John>>
--2: tmp = 0x7fff54cb44c0 <<My ny%20name%20is%20John>>; str = 0x7fff54cb44c3 <<ny%20name%20is%20John>>
--3: tmp = 0x7fff54cb44c0 <<My ny%20name%20is%20John>>; str = 0x7fff54cb44c3 <<ny%20name%20is%20John>>
<<-- tmp = <<My ny%20name%20is%20John>>
output is <<My ny%20name%20is%20John>>
我认为您的算法很复杂,因为您正在尝试进行原位更改。我无法解决如何解决它。传入一个输入字符串(可能是一个字符串文字,并且不需要空白填充)和一个输出缓冲区会更简单,你可以在调用中指定它的大小:
#include <stdio.h>
static int encodeSpace(char const *src, char *buffer, size_t buflen)
{
char *end = buffer + buflen;
char c;
char *dst = buffer;
while ((c = *src++) != '\0' && dst < end)
{
if (c != ' ')
*dst++ = c;
else if (dst < end - 3)
{
*dst++ = '%';
*dst++ = '2';
*dst++ = '0';
}
else
return -1;
}
*dst = '\0';
return 0;
}
int main(void)
{
char buffer[64];
if (encodeSpace("My name is John", buffer, sizeof(buffer)) == 0)
printf("output is <<%s>>\n", buffer);
return 0;
}
输出:
output is <<My%20name%20is%20John>>
答案 1 :(得分:0)
我试图修复原始算法,但在原始字符缓冲区中看起来很难更改因为%20
占用三个字符位置,如果space
字符不够,代码仍然出错。< / p>
如果空间足够,
代码跟随效果很好。
#include <stdio.h>
#include <string.h>
char* encodeSpace(char* str) {
char *orig = str;
while (*str++);
str--;
char *tmp = --str;
while (*tmp == ' ') tmp--;
tmp++;
while (tmp != orig) {
while(*tmp != ' ') {
*str = *tmp;
str--;
if (tmp == orig)
break;
else
tmp--;
}
if (*tmp == ' ') {
*str-- = '0';
*str-- = '2';
*str-- = '%';
while (*tmp == ' ') tmp--;
}
}
return ++str;
}
int main()
{
char str[] = "My Name is john ";
printf("output is %s\n", encodeSpace(str));
return 0;
}