解析函数解析文件，但仅部分解析

Question

我正在处理一种必须用C解析的文件格式，如下所示：

5 3 1 3

1 4

1 5

3 2 3

2 4

2 5

2 3 1

3 2

3 4 1

4 2

4 5

3 5 1

5 2

5 4

除了第一行，它告诉我们顶点的数量，每一行有一个数字，表示有两个数字的行数。在这两个数字线上，我们可以看到两个顶点ID，它们是在一个图中连接的。我的目标是获取这些信息并将int转换成一个图。我编写了以下函数来读取单个顶点：

graph_vertex_t* read_vertex(FILE *fp, graph_vertex_t *last_vertex) 
{
    graph_vertex_t *current_vertex;
    
    graph_edge_t *tmp_edge;
    
    graph_edge_t *last_edge;
    
    char *token;
    
    boolean_t error = false;
    
    current_vertex = (graph_vertex_t *)malloc(sizeof(graph_vertex_t));
    
    int tmp_number_edges = 1;
    /*Temporary character string used to read line in file. */
    char *tmp;
    /*Allocating memory for string. Again I'm assuming 64 charcters is*/
    /*all I need. */
    tmp = malloc(sizeof(char) * 64);
    while (fgets(tmp, 
                 63, 
                 fp) != NULL 
                 &&
                 tmp_number_edges > 0)
    {   
        tmp[strcspn(tmp, "\n")] = 0;
        /* Check if the line contains a vertex number. */
        if (is_number(tmp))
        {
            /* Set the line to the vertex number. */
            tmp_number_edges = atoi(tmp);
        }
        else
        {           
            /* check if line has the right format. */
            token = strtok(tmp, " \t");
            if (is_number(token))
            {
                /* Set the identifier of the vertex. */
                current_vertex->identifier = atoi(token);
                current_vertex->next_vertex_p = last_vertex;
            }
            else
            {
                error = true;
            }
            
            token = strtok(NULL, " \t");
            if (is_number(token))
            {
                /* Make new edge. */
                tmp_edge = (graph_edge_t *)malloc(sizeof(graph_edge_t));
                tmp_edge->adjac_vertex_p = (graph_vertex_t *)malloc(sizeof(graph_vertex_t));
                tmp_edge->adjac_vertex_p->identifier = atoi(token);
                tmp_edge->next_edge_p = last_edge;              
                last_edge = tmp_edge;
                printf(" %d", tmp_edge->adjac_vertex_p->identifier);
            }
            else
            {
                error = true;
            }
            tmp_number_edges--;
        }
    }
    current_vertex->edge_list_p = tmp_edge;
    if(error)
    {
        current_vertex = NULL;
    }
    return(current_vertex);
}

这个函数是在一个for循环中调用的，这个循环是对顶点数的循环。然后打印一些信息：

printf("connecting to: ");
            
current_vertex = read_vertex(fp,
                             last_vertex);
printf(" found vertex ID: %d \n", current_vertex->identifier);

因此，我希望得到一个类似于以下内容的输出：

连接到:3 4 5找到顶点ID: 1

连接到:3 4 5找到顶点ID: 2

连接到:12找到顶点ID: 3连接到:1 2 5找到顶点ID: 4

连接到:12 4找到的顶点ID: 5

但实际上看起来是这样的：

连接到:3 4 5找到顶点ID: 1

连接到:3个已找到的顶点ID: 2个连接到:5个已找到的顶点ID: 2个已找到的顶点ID: 3个已找到的顶点ID: 3个已找到的顶点ID:4个

连接到:1个找到的顶点ID: 5

我们可以看到，对于ID 1和4，输出是正确的，但对于其余的输出，则不是。是什么导致了这种行为？我怎么才能解决这个问题？

编辑：

以下是graph_vertex_t和graph_edge_t的定义。

* Structure used to hold a graph vertex information. */
typedef struct graph_vertex
{
    int identifier;
    struct graph_vertex *next_vertex_p;
    struct graph_edge *edge_list_p;
    boolean_t visited;
} graph_vertex_t;

/* Structure used to hold a graph edge information. */
typedef struct graph_edge
{
    struct graph_vertex *adjac_vertex_p;
    struct graph_edge *next_edge_p;
} graph_edge_t;

Answer 1

从你的话

i首先需要理解sscanf

帮助你阅读这里的一些建议，从你的结构中的任何记忆中选出。

不检查数字的第一项建议是否在分隔线上：

#include <stdio.h>

int main()
{
  int nVertices, verticesRank;
  
  if ((scanf("%d", &nVertices) != 1) ||
      (nVertices < 0)) {
    fputs("wrong number of vertices\n", stderr);
    return -1;
  }

  for (verticesRank = 0; verticesRank < nVertices; verticesRank += 1) {
    int nGroups, groupRank, vertex;
    
    if ((scanf("%d", &nGroups) != 1) ||
        (nGroups < 0)) {
      fprintf(stderr, "wrong number of groups in vertices #%d\n", verticesRank);
      return -1;
    }
    
    for (groupRank = 0; groupRank < nGroups; groupRank += 1) {
      int a, b;
      
      if ((scanf("%d%d", &a, &b) != 2) ||
          (a < 1) || (b < 1)) {
        fprintf(stderr, "wrong vertex IDs group #%d vertices #%d\n",
               groupRank, verticesRank);
        return -1;
      }
      
      if (groupRank == 0) {
        vertex = a;
        printf("vertex %d connecting", a);
      }
      else if (a != vertex) {
        fprintf(stderr, "wrong vertex ID group #%d vertices #%d, %d rather than %d\n",
                groupRank, verticesRank, a, vertex);
        return -1;
      }
      printf(" %d", b);
    }
    putchar('\n');
  }
  
  return 0;
}

汇编和执行：

pi@raspberrypi:/tmp $ gcc -Wall c.c
pi@raspberrypi:/tmp $ cat > v
5
3
1 3
1 4
1 5
3
2 3
2 4
2 5
2
3 1
3 2
3
4 1
4 2
4 5
3
5 1
5 2
5 4
pi@raspberrypi:/tmp $ ./a.out <v
vertex 1 connecting 3 4 5
vertex 2 connecting 3 4 5
vertex 3 connecting 1 2
vertex 4 connecting 1 2 5
vertex 5 connecting 1 2 4
pi@raspberrypi:/tmp $ 

第二个版本检查每行值的正确数目：

#include <stdio.h>

int main()
{
  char line[64];
  int nVertices, verticesRank;
  int dummy;
  
  if ((fgets(line, sizeof(line), stdin) == NULL) ||
      (sscanf(line, "%d%d", &nVertices, &dummy) != 1) ||
      (nVertices < 0)) {
    fputs("wrong number of vertices\n", stderr);
    return -1;
  }

  for (verticesRank = 0; verticesRank < nVertices; verticesRank += 1) {
    int nGroups, groupRank, vertex;
    
    if ((fgets(line, sizeof(line), stdin) == NULL) ||
        (sscanf(line, "%d%d", &nGroups, &dummy) != 1) ||
        (nGroups < 0)) {
      fprintf(stderr, "wrong number of groups in vertices #%d\n", verticesRank);
      return -1;
    }
    
    for (groupRank = 0; groupRank < nGroups; groupRank += 1) {
      int a, b;
      
      if ((fgets(line, sizeof(line), stdin) == NULL) ||
          (sscanf(line, "%d%d%d", &a, &b, &dummy) != 2) ||
          (a < 1) || (b < 1)) {
        fprintf(stderr, "wrong vertex IDs group #%d vertices #%d\n",
               groupRank, verticesRank);
        return -1;
      }
      
      if (groupRank == 0) {
        vertex = a;
        printf("vertex %d connecting", a);
      }
      else if (a != vertex) {
        fprintf(stderr, "wrong vertex ID group #%d vertices #%d, %d rather than %d\n",
                groupRank, verticesRank, a, vertex);
        return -1;
      }
      printf(" %d", b);
    }
    putchar('\n');
  }
  
  return 0;
}

汇编和执行：

pi@raspberrypi:/tmp $ gcc -Wall cc.c
pi@raspberrypi:/tmp $ ./a.out <v
vertex 1 connecting 3 4 5
vertex 2 connecting 3 4 5
vertex 3 connecting 1 2
vertex 4 connecting 1 2 5
vertex 5 connecting 1 2 4
pi@raspberrypi:/tmp $ 

注意，我不检查是否有数字以外的其他内容，因此第一行被认为是正确的，例如，如果您需要检查：

#include <stdio.h>

int main()
{
  char line[64];
  int nVertices, verticesRank;
  char dummy;
  
  if ((fgets(line, sizeof(line), stdin) == NULL) ||
      (sscanf(line, "%d%c", &nVertices, &dummy) != 2) ||
      (nVertices < 0) ||
      (dummy != '\n')) {
    fputs("wrong number of vertices\n", stderr);
    return -1;
  }

  for (verticesRank = 0; verticesRank < nVertices; verticesRank += 1) {
    int nGroups, groupRank, vertex;
    
    if ((fgets(line, sizeof(line), stdin) == NULL) ||
        (sscanf(line, "%d%c", &nGroups, &dummy) != 2) ||
        (nGroups < 0) ||
        (dummy != '\n')) {
      fprintf(stderr, "wrong number of groups in vertices #%d\n", verticesRank);
      return -1;
    }
    
    for (groupRank = 0; groupRank < nGroups; groupRank += 1) {
      int a, b;
      
      if ((fgets(line, sizeof(line), stdin) == NULL) ||
          (sscanf(line, "%d%d%c", &a, &b, &dummy) != 3) ||
          (a < 1) || (b < 1) ||
          (dummy != '\n')) {
        fprintf(stderr, "wrong vertex IDs group #%d vertices #%d\n",
               groupRank, verticesRank);
        return -1;
      }
      
      if (groupRank == 0) {
        vertex = a;
        printf("vertex %d connecting", a);
      }
      else if (a != vertex) {
        fprintf(stderr, "wrong vertex ID group #%d vertices #%d, %d rather than %d\n",
                groupRank, verticesRank, a, vertex);
        return -1;
      }
      printf(" %d", b);
    }
    putchar('\n');
  }
  
  return 0;
}

汇编和执行：

pi@raspberrypi:/tmp $ gcc -Wall cc.c
pi@raspberrypi:/tmp $ ./a.out <v
vertex 1 connecting 3 4 5
vertex 2 connecting 3 4 5
vertex 3 connecting 1 2
vertex 4 connecting 1 2 5
vertex 5 connecting 1 2 4
pi@raspberrypi:/tmp $ 

注意，在这种情况下，即使在最后一个期望的数字被认为是一个错误之后，也有一个额外的空格(在换行符之外)，最后一行也必须用换行符来完成。

这并不是说在理论结束之后有什么东西，你可以用几种方式去做。

在您的示例中，顶点ID是每个配对中的第一个数字，是连续数字1、2.在我之前的建议中，我不检查，即使检查vertex == verticesRank+1也很容易。如果这是一条规则，那么在只有第二个数字就足够的情况下，为什么要给出这对值呢？