【发布时间】：2021-08-03 06:27:28
【问题描述】：
我对两者都使用了相同的代码，但我在 Debian 中遇到了分段错误。为什么会出现这个错误？
注意：我知道代码没有意义，这是学校的作业。下面添加了错误图像。我调查了类似的主题，但没有找到错误解决方案。
代码如下：
#include <stdio.h>
#include <string.h>
struct record {
char name[64]; //utf16
char surname[32]; //utf8
char gender[1];
char email[32];
char phone_number[16];
char address[32];
char level_of_education[8];
unsigned int income_level; // given little-endian
unsigned int expenditure; // given big-endian
char currency_unit[16];
char currentMood[32];
float height;
unsigned int weight;
};
struct tag {
char name[5];
char surname[8];
char gender[7];
char email[10];
char phone_number[13];
char address[8];
char level_of_education[19];
char income_level[13];
char expenditure[12];
char currency_unit[14];
char currentMood[12];
char height[7];
char weight[7];
};
/*
* Converts big endian to little or little endian to big endian
*/
unsigned int endian_swap(unsigned int x)
{
x = (x>>24) |
((x<<8) & 0x00FF0000) |
((x>>8) & 0x0000FF00) |
(x<<24);
return x;
}
/*
* Find uncorrect tag headers and fixed
*/
struct tag rearrange_tag_names(struct tag t){
struct tag *tagPtr;
tagPtr = &t;
// Check name header
if(strcmp(t.name,"name") != 0) {
strncpy(tagPtr->name,"name",5);
}
// Check surname header
if(strcmp(t.surname,"surname") != 0) {
strncpy(tagPtr->surname,"surname",8);
}
// Check gender header
if(strcmp(t.gender,"gender") != 0) {
strncpy(tagPtr->gender,"gender",7);
}
// Check email header
if(strcmp(t.email,"email") != 0) {
strncpy(tagPtr->email,"email",10);
}
// Check phone_number header
if(strcmp(t.phone_number,"phone_number") != 0) {
strncpy(tagPtr->phone_number,"phone_number",13);
}
// Check address header
if(strcmp(t.address,"address") != 0) {
strncpy(tagPtr->address,"address",8);
}
// Check level_of_education header
if(strcmp(t.level_of_education,"level_of_education") != 0) {
strncpy(tagPtr->level_of_education,"level_of_education",20);
}
// Check income_level header
if(strcmp(t.income_level,"income_level") != 0) {
strncpy(tagPtr->income_level,"income_level",14);
}
// Check expenditure header
if(strcmp(t.expenditure,"expenditure") != 0) {
strncpy(tagPtr->expenditure,"expenditure",12);
}
// Check currency_unit header
if(strcmp(t.currency_unit,"currency_unit") != 0) {
strncpy(tagPtr->currency_unit,"currency_unit",14);
}
// Check currentMood header
if(strcmp(t.currentMood,"currentMood") != 0) {
strncpy(tagPtr->currentMood,"currentMood",12);
}
// Check height header
if(strcmp(t.height,"height") != 0) {
strncpy(tagPtr->height,"height",7);
}
// Check weight header
if(strcmp(t.weight,"weight") != 0) {
strncpy(tagPtr->weight,"weight",7);
}
// print to test
//printf("%s \t %s \t %s \t %s \t %s \t %s \t %s \t %s \t %s \t %s \t %s \t %s",t.name,t.surname,t.gender,t.email,t.phone_number,t.address,t.level_of_education,t.income_level,t.expenditure,t.currency_unit,t.currentMood,t.height,t.weight);
return t;
}
/*
* Creates XML from given records
*/
void createXML(struct record* arr,struct tag tag,char *xml_name)
{
// Create file pointer
FILE *fxml;
//Check if error is occured
if ((fxml = fopen(xml_name,"wt")) == NULL) {
printf("%s","Some error is occured.");
}
fprintf (fxml,"<?xml version=\"1.0\" encoding=\"utf-8\"?>\n");
fprintf(fxml,"<records>\n");
int i;
// Print tag to test
//printf("%s",tag.email);
//printf("%s",tag.level_of_education);
// printf("%s",tag.level_of_education);
// printf("%s",tag.income_level);
// printf("%s",tag.expenditure);
// printf("%s",tag.height);
// printf("%s",tag.weight);
// Crate xml rows
for (i = 1; i < 51; i++) {
fprintf(fxml," <row id='%d'>\n",i);
fprintf(fxml," <%s>%s</%s>\n",tag.name,arr[i].name,tag.name);
fprintf(fxml," <%s>%s</%s>\n",tag.surname,arr[i].surname,tag.surname);
fprintf(fxml," <%s>%s</%s>\n",tag.gender,arr[i].gender,tag.gender);
fprintf(fxml," <%s>%s</%s>\n",tag.email,arr[i].email,tag.email);
fprintf(fxml," <%s>%s</%s>\n",tag.phone_number,arr[i].phone_number,tag.phone_number);
fprintf(fxml," <%s>%s</%s>\n",tag.address,arr[i].address,tag.address);
fprintf(fxml," <%s>%s</%s>\n",tag.level_of_education,arr[i].level_of_education,tag.level_of_education);
fprintf(fxml," <%s>%d</%s>\n",tag.income_level,arr[i].income_level,tag.income_level);
fprintf(fxml," <%s>%d</%s>\n",tag.expenditure,arr[i].expenditure,tag.expenditure);
fprintf(fxml," <%s>%s</%s>\n",tag.currency_unit,arr[i].currency_unit,tag.currency_unit);
fprintf(fxml," <%s>%s</%s>\n",tag.currentMood,arr[i].currentMood,tag.currentMood);
fprintf(fxml," <%s>%.2f</%s>\n",tag.height,arr[i].height,tag.height);
fprintf(fxml," <%s>%d</%s>\n",tag.weight,arr[i].weight,tag.weight);
fprintf(fxml," </row>\n");
}
fprintf(fxml,"</records>\n");
fclose(fxml); //Close file
}
/*
* Converts bin file to XML
*/
int Bin2XML(char* rafname,char* xml_name)
{
// Create necessary variables
FILE *fraf;
struct record rec;
struct tag tag;
struct record arr_rec[50];
int number_of_record = 51;
// Encode file in utf-8 format
if ((fraf = fopen(rafname,"rb, ccs=UTF-8")) == NULL)
return 0; // failure
fread(&arr_rec, sizeof(struct record), number_of_record, fraf);
// Read first line into tag and
// make necessary convertion and arrangement
strcpy(tag.name,arr_rec[0].name);
strcpy(tag.surname,arr_rec[0].surname);
strcpy(tag.email,arr_rec[0].email);
strcpy(tag.phone_number,arr_rec[0].phone_number);
strcpy(tag.address,arr_rec[0].address);
strcpy(tag.level_of_education,arr_rec[0].level_of_education);
// convert int income level to string format
char converted_income_level[10];
sprintf(converted_income_level,"%d",arr_rec[0].income_level);
strcpy(tag.income_level,converted_income_level);
// convert int expenditure to string format
char converted_expenditure[10];
sprintf(converted_expenditure,"%d",arr_rec[0].expenditure);
strcpy(tag.expenditure,converted_expenditure);
strcpy(tag.currency_unit,arr_rec[0].currency_unit);
strcpy(tag.currentMood,arr_rec[0].currentMood);
// convert float height to string format
char buf[10];
snprintf(buf,5,"%s",arr_rec[0].height);
strcpy(tag.height,buf);
// convert unsigned weight to string format
char converted_weight[10];
sprintf(converted_weight,"%d",arr_rec[0].weight);
strcpy(tag.weight,converted_weight);
tag = rearrange_tag_names(tag);
// Rearrange formats
int i;
for(i=1;i<number_of_record;i++){
arr_rec[i].expenditure = endian_swap(arr_rec[i].expenditure); // Convert big endian to little endian
// Income level is already little endian but if it was not little endian we can read as little endian
// with endian_swap function.
//arr_rec[i].income_level =endian_swap(arr_rec[i].income_level);
}
// Create an XML file with readed records
// First record readed into tag,so we can decrease number of record by one
createXML(arr_rec,tag,xml_name);
// close file
fclose(fraf);
return 1; // success
}
int main(void)
{
if( Bin2XML("records.dat","test.xml") == 1 ){
printf("Bin file is readed and XML file is created successfully.");
} else{
printf("(!) An error occured please check your input file name. (!)");
}
return 0;
}
有人可以帮我吗？
【问题讨论】：
-
像 gdb 这样的调试器可以轻松地向您准确显示段错误发生的位置，而无需我们通过几百行代码来尝试找到问题。
-
我去看看
-
我对两者都使用了相同的代码，但我在 Debian 中遇到了分段错误。为什么会出现此错误？ 这是调用 undefined behavior 的代码的典型行为 - 它有时或在某处工作，但在其他条件下或在另一个系统上运行时会失败。我看到不少
strcpy()调用可能导致缓冲区溢出，还有不少strncpy()calls that could result in a non-terminated string。 -
我会尝试摆脱这些 strcpy() 和 strncpy()。
-
你有大量的代码复制。您还有 UB：
strncpy(tagPtr->income_level,"income_level",14)-->strncpy(tagPtr->income_level,"income_level",13)（因为：char income_level[13];）。您的记录字段不确定是否需要 EOS 的空间（例如）char hello[5] = "hello";没有空间用于'\0'但char hello[6] = "hello";需要。但是，由于%s格式，您的打印代码 假定 为 EOS 空间。您的文件/磁盘数据（来自fread）是否为 all 字段提供 EOS 空间？
标签： c segmentation-fault debian