读取具有结构数组的CSV文件

Question

我是一个C编程新手，我想尝试从CSV文件中读取并打印出来。我的CSV文件的格式为：

指数、国家、年份、死亡率

1个世界

         50s     36
         60s     29
         70s     22

2非洲

         50s     34

209汤加

         50s     49
         60s     67

以此类推。

我不确定如何读取我的CSV文件的结尾，所以我的代码输出只打印到索引143，而我当前的CSV文件打印到204。有没有人能指导我做这个？

void readRecords(){
    FILE *fptr;
    int i,j,k;
    char line[200], hold[10];

    if((fptr=fopen("data.csv","r"))==NULL){
        printf("Cannot open input file data.csv\n");
    }

    else{
        for(i=0;i<999;i++){
            fgets(line, 200, fptr);

            j=0;
            k=0;
            while (line[j] != ','){     // This while loop extract the country name until a comma is detected
                hold[k++] = line[j++];  // Simply copy all character from each line to the country
            }
            hold[k]='\0';
            myRecords[i].index=atoi(hold);
            j++;


            k=0;
            while (line[j] != ','){     // This while loop extract the country name until a comma is detected
                myRecords[i].country[k++] = line[j++];  // Simply copy all character from each line to the country
            }
            myRecords[i].country[k]='\0';
            j++;


            k=0;
            while (line[j] != ','){     // This while loop extract the country name until a comma is detected
                hold[k++] = line[j++];  // Simply copy all character from each line to the country
            }
            hold[k]='\0';
            myRecords[i].year=atoi(hold);
            j++;


            k=0;
            while (line[j] != ','){     // This while loop extract the country name until a comma is detected
                hold[k++] = line[j++];  // Simply copy all character from each line to the country
            }
            hold[k]='\0';
            myRecords[i].deathrate=atoi(hold);
            j++;

        }

        }
        fclose(fptr);
    }

Answer 1

另一种选择是用指针简单地遍历fgets或getline读取的行，并用\0字符替换逗号，然后在一行中逐步读取每个变量。以下是给定您的数据文件的一个工作实现：

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

int main (int argc, char *argv[]) {

    if (argc < 2) {
        fprintf (stderr, "error: insufficient input, usage: %s <filename>\n", argv[0]);
        return 1;
    }
    FILE *fptr;
    char *line = NULL;
    size_t len = 0;
    ssize_t read;
    char *p = NULL;
    char *start = NULL;

    int index = 0;
    char *country;
    int years = 0;
    int deaths = 0;

    if ((fptr=fopen (argv[1],"r")) == NULL) {
        fprintf(stderr, "Cannot open input file %s\n", argv[1]);
    }

    while ((read = getline(&line, &len, fptr)) != -1)
    {
        p = line;

        while (*p != ',' && *p != 0) p++;
        *p = '\0';
        index = atoi (line);
        start = ++p;
        while (*p != ',' && *p != 0) p++;
        *p = '\0';
        country = strdup (start);
        start = ++p;
        while (*p != ',' && *p != 0) p++;
        *p = '\0';
        years = atoi (start);
        start = ++p;
        deaths = atoi (start);

        printf ("idx: %3d   country: %12s  years: %3d  deaths: %3d\n",
                index, country, years, deaths);
    }

    free(line);

    return 0;
}

现在，假设输入.csv文件的格式为：

1,world,50s,36
1,world,60s,65
1,world,70s,58
2,africa,50s,37
2,africa,60s,66
2,africa,70s,59
207,tonga,50s,37
207,tonga,60s,66
207,tonga,70s,59

您的输出将是：

idx:   1   country:        world  years:  50  deaths:  36
idx:   1   country:        world  years:  60  deaths:  65
idx:   1   country:        world  years:  70  deaths:  58
idx:   2   country:       africa  years:  50  deaths:  37
idx:   2   country:       africa  years:  60  deaths:  66
idx:   2   country:       africa  years:  70  deaths:  59
idx: 207   country:        tonga  years:  50  deaths:  37
idx: 207   country:        tonga  years:  60  deaths:  66
idx: 207   country:        tonga  years:  70  deaths:  59

无论哪种方式，它都是一样短的。使用strtok或者只是单步执行缓冲区并直接读取您需要的内容。如果你花时间仔细阅读代码的话。你会看到，它所做的就是读到一个逗号，用\0替换逗号，从开头(或开始)读取所需的信息，将指针前进1，设置新的'start‘点，然后一直向下重复缓冲区。在读取'deaths‘之后，最后一个循环确保行中的所有字符都已被读取。getline将读取下一行，并继续执行，直到在.csv文件中无法读取更多行为止。我希望这能帮到你。(如果不明显，可以在命令行中指定csv文件名：

$ readcsv dat.csv

您已经有了数据，并在上面将其解析为变量。如果要使用结构，只需将存储类型从公共变量更改为结构数据类型即可。下面类似的内容适用于结构数组。

int recs = 0;

struct _stats { 
    int index;
    char *country;
    int years;
    int deaths;
};

struct _stats *stats[100];
struct _stats *statsptr = NULL;

if ((fptr=fopen (argv[1],"r")) == NULL) {
    fprintf(stderr, "Cannot open input file %s\n", argv[1]);
}

while ((read = getline(&line, &len, fptr)) != -1) {

    p = line;
    statsptr = malloc (sizeof (struct _stats));

    while (*p != ',' && *p != 0) p++;
    *p = '\0';
    statsptr->index = atoi (line);
    start = ++p;

    while (*p != ',' && *p != 0) p++;
    *p = '\0';
    statsptr->country = strdup (start);
    start = ++p;

    while (*p != ',' && *p != 0) p++;
    *p = '\0';
    statsptr->years = atoi (start);
    start = ++p;

    statsptr->deaths = atoi (start);

    stats[recs] = statsptr;
    recs++;

}

int iter = 0;
for (iter = 0; iter < recs; iter++) {
    printf ("idx: %3d   country: %12s  years: %3d  deaths: %3d\n",
            stats[iter]->index,
            stats[iter]->country,
            stats[iter]->years,
            stats[iter]->deaths);
}

除了最初声明指向结构的最大数量的指针之外，您还需要使用动态数据类型，如链表。(超出了本讨论的范围)。