读取 2kb txt 文件时出现分段错误 (c)

Question

以下是从文件中读取并将所述文件的每个字符串保存到二叉搜索树的代码。它适用于 1KB 的 txt 文件，但是当尝试使用更大的文件 (2kb) 时，我会遇到分段错误。

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

typedef struct listNode ListNode;


struct listNode {
  int id;
  ListNode *next;
};

typedef struct treeNode TreeNode;


struct treeNode {
  char *word;
  char *key;
  int freq;
  ListNode *head;   
  TreeNode *left;
  TreeNode *right;
};

TreeNode* insertItem(TreeNode *root, char *gword);
void printTreeInorder(TreeNode *v);
void searchforexist(TreeNode *root, char *key);

#define MAX 25
int main()
{
   
    char word[MAX];
    TreeNode *root = NULL;
    
    
   
    FILE *fp=fopen("input.txt","r");

     if (fp!=NULL)
    {
      
        while (fscanf(fp,"%s \n",word) != EOF)
         {
              
            root = insertItem(root,word); 

          if (strcmp(word, "eof")==0)
            break;
          }
     }
    fclose(fp);


    printTreeInorder(root);
    printf("\n");
   
    return 0;
}


TreeNode* insertItem(TreeNode *root, char *gword)
{
   TreeNode *v = root;  
   TreeNode *pv = NULL; 
   while (v != NULL)
   {
      pv = v;
      int comp = strcmp(gword, v->word);
      if (comp < 0) v=v->left;
      else if (comp > 0) v=v->right;
      else
      {
         
          
         char *key=v->word;
         searchforexist(root,key);
          return root;
      }
   }

   TreeNode *tmp = (TreeNode *) malloc(sizeof(TreeNode));
   tmp->word=strdup(gword); 
   tmp->left=tmp->right=NULL;
   tmp->freq=1;
   

   if (root != NULL)
   {
      if (strcmp(gword, pv->word) < 0) pv->left=tmp;
         else pv->right=tmp;
   } else root=tmp;

   return root;

}



void searchforexist(TreeNode *root, char *key)
{
   if (root == NULL || root->key == key) 
       root->freq ++; 
     
    
    if (root->key < key) 
       searchforexist(root->right, key); 
  
    
     searchforexist(root->left, key); 
 } 




void printTreeInorder(TreeNode *v)
{
    if (v==NULL) return;

    printf("(");
    printTreeInorder(v->left);
    printf(")");

    printf(" %.4s ", v->word);

    printf("(");
    printTreeInorder(v->right);
    printf(")");
}

按预期运行的 Txt 文件： { 单词在此测试文件中用于测试目的 将字符串转换为数据结构。 }

但是，如果我将其更改为此，则会出现分段错误： { 单词在此测试文件中用于测试目的 将字符串转换为数据结构。 & 单词在此测试文件中用于测试目的 将字符串转换为数据结构。 }

Answer 1

一种简单的方法是使用 gdb/lldb。

我编译了你的代码，并调试了：

$ g++ -g -ggdb test.cc
$ lldb a.out
(lldb) r
Process 83386 launched: '/path/to/a.out' (x86_64)
Process 83386 stopped
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x10)
    frame #0: 0x0000000100000ec1 a.out`searchforexist(root=0x0000000000000000, key="<link") at test.cc:104
   101  void searchforexist(TreeNode *root, char *key)
   102  {
   103     if (root == NULL || root->key == key)
-> 104         root->freq ++;
   105
   106
   107      if (root->key < key)
Target 0: (a.out) stopped.
(lldb) bt
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x10)
  * frame #0: 0x0000000100000ec1 a.out`searchforexist(root=0x0000000000000000, key="<link") at test.cc:104
    frame #1: 0x0000000100000eed a.out`searchforexist(root=0x0000000100102e50, key="<link") at test.cc:108
    frame #2: 0x0000000100000efe a.out`searchforexist(root=0x0000000100102b00, key="<link") at test.cc:111
    frame #3: 0x0000000100000d34 a.out`insertItem(root=0x0000000100102b00, gword="<link") at test.cc:78
    frame #4: 0x0000000100000c28 a.out`main at test.cc:47
(lldb)

我认为这个错误信息代表了足够的信息。

---

我做了一些修改。让它发挥作用。

除非每个单词都简短而独特，否则我无法想象以前的单词会起作用。

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

typedef struct listNode {
  int id;
  struct listNode *next;
} ListNode;

typedef struct treeNode {
  char *word;
  char *key;
  int freq;
  ListNode *head;
  struct treeNode *left;
  struct treeNode *right;
} TreeNode;

TreeNode *insertItem(TreeNode *root, char *gword);
void printTreeInorder(TreeNode *v);
void searchforexist(TreeNode *root, char *key);
void freeNodes(TreeNode *root);

#define MAX 25
int main() {
  char word[MAX];
  TreeNode *root = NULL;

  FILE *fp = fopen("input.txt", "r");

  memset(word, 0, MAX);
  if (fp != NULL) {
    // why fsanf("%s \n") ? is this a very special format?
    while (fscanf(fp, "%24s \n", word) != EOF) {
      // fprintf(stderr, "got: [%s]\n", word);
      root = insertItem(root, word);

      if (strcmp(word, "eof") == 0) break;
    }

    fclose(fp);
  }

  printTreeInorder(root);
  printf("\n");

  freeNodes(root);
  return 0;
}

TreeNode *insertItem(TreeNode *root, char *gword) {
  TreeNode *v = root;
  TreeNode *pv = NULL;
  while (v != NULL) {
    pv = v;
    int comp = strcmp(gword, v->word);
    if (comp < 0) {
      v = v->left;
    } else if (comp > 0) {
      v = v->right;
    } else {
      // char *key = v->word;
      char *word = v->word;
      searchforexist(root, word);
      return root;
    }
  }

  TreeNode *tmp = (TreeNode *)malloc(sizeof(TreeNode));
  // why both key and word?
  tmp->word = strdup(gword);
  tmp->left = tmp->right = NULL;
  tmp->freq = 1;

  if (root != NULL) {
    if (strcmp(gword, pv->word) < 0) {
      pv->left = tmp;
    } else {
      pv->right = tmp;
    }
  } else
    root = tmp;

  return root;
}

void searchforexist(TreeNode *root, char *word) {
  if(root == NULL) {
    return;
  }

  int comp = strcmp(word, root->word);

  if(comp == 0) {
    root->freq++;
  } else {
    searchforexist(comp < 0 ? root->left : root->right , word);
  }

}

void printTreeInorder(TreeNode *v) {
    if (v==NULL) return;

    printf("(");
    printTreeInorder(v->left);
    printf(")");

    printf(" %.4s ", v->word);

    printf("(");
    printTreeInorder(v->right);
    printf(")");
}

void freeNodes(TreeNode *root) {
  if (root == NULL) {
    return;
  }
  freeNodes(root->left);
  freeNodes(root->right);

  if(root->word != NULL) free(root->word);
  if(root->key != NULL) free(root->key);
  free(root);
  return;
}