c - 从具有结构数组的 CSV 文件中读取

Question

我是 C 编程的新手，我想尝试读取 CSV 文件并将其打印出来。我的 CSV 文件格式为:

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

等等……

我不确定如何读取我的 CSV 文件的末尾，因此我的代码输出最多只能打印索引 143，而我当前的 CSV 文件最多只能打印 204。有人可以指导我吗？

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

另一种选择是简单地单步执行 fgets 或 getline 使用指针读取的行，并将逗号替换为 \0 字符和随着您的工作，逐步读取每个变量。这是给定您的数据文件的有效实现:

#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，给定一个格式为 .csv 的输入文件:

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

这两种方式都差不多。使用 strtok 或只是单步执行缓冲区并直接读取您需要的内容。如果您花时间仔细阅读代码。你会看到它所做的一切都是读取直到遇到逗号，用 \0 替换逗号，从头(或开始)读取以获取所需的信息，将指针前进 1，设置新的“开始”点，然后一直重复缓冲区。读取“死亡”后，最后一个循环确保已读取该行中的所有字符。 getline 读取下一行并继续，直到无法在 .csv 文件中读取更多行。我希望这有帮助。 (如果不是很明显，您可以在命令行中为其提供 csv 文件名:

$ readcsv dat.csv

您已经拥有数据并将其解析为上面的变量。如果要使用结构，只需将存储类型从普通变量更改为结构数据类型即可。类似于以下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);
}

除了最初声明结构指针的最大数量外，您还需要使用动态数据类型，如链表。 (超出本次讨论的范围)。