我一直在努力解决这个问题,但无济于事。在下面的代码中,我使用了一个char指针(all_seq),但是当我释放它时,在代码的末尾,有时它会出现段错误。我有时说,因为它显然取决于输入文件。使用小输入文件,它很好。有一个大的,它也没关系。对于中等大小的文件,它是段错误的。我只是无法弄清楚为什么......
任何想法都将不胜感激。感谢。
/* SeqProb (Sequence Probability) - 2012/03/13 */
/* Given a series of aligned sequences -in one file, one per line- this program calculates the probability of having, in each location, the letter in the first sequence. */
#include <stdio.h>
#include <math.h>
#include <stdlib.h>
#include <string.h>
FILE *input_file;
FILE *output_file;
///////////////////////////////
// DECLARATION OF FUNCTIONS: //
///////////////////////////////
int GetSize(FILE *file, int *length, int *lines);
double CalcProb(char *one_seq);
//////////////////////////////////////////////////////////////////
// ************************************************************ //
// ************** START MAIN BLOCK ************** //
// ************************************************************ //
//////////////////////////////////////////////////////////////////
int main( int argc, char *argv[] ) // We expect three command-line arguments: 1)program name, 2)file to read and 3)file to write.
{
int i=0,j=0;
int length, lines, size;
double prob=0;
char *all_seq, *one_seq;
////////////////////////////
// CHECK INPUT ARGUMENTS: //
////////////////////////////
// Check if arguments in command line are right and if file can be opened for reading:
if ( argc != 3 ) { // There should be 3 arguments for correct execution
printf("\n Usage: %s <input file> <output file>\n\n ", argv[0]); // We print argv[0] assuming it is the program name
exit(1);
}
if (!(input_file=fopen(argv[1],"r"))){
printf("\n Error: can't open input file \n\n");
exit(2);
}
if (!(output_file=fopen(argv[2],"w"))){
printf("\n Error: can't open output file \n\n");
exit(3);
}
/////////////////////////////////////////
// GET LENGTH AND NUMBER OF SEQUENCES: //
/////////////////////////////////////////
GetSize(input_file,&length,&lines);
size=length*lines;
printf("\n length of sequences = %d\n",length);
printf(" number of sequences = %d\n",lines);
//////////////////////////////////////////////////////
// ALLOCATE MEMORY FOR all_seq AND one_seq VECTORS: //
//////////////////////////////////////////////////////
all_seq = (char *) malloc((length*lines+10)*sizeof(char));
one_seq = (char *) malloc((lines+10)*sizeof(char));
if (all_seq == NULL){ // In case of memory allocation failure, exit.
printf ("Error: Can't allocate memory \n");
exit(4);
}
if (one_seq == NULL){ // In case of memory allocation failure, exit.
printf ("Error: Can't allocate memory \n");
exit(5);
}
/////////////////////////////////////////////
// INITIALIZE VECTORS all_seq and one_seq: //
/////////////////////////////////////////////
for (i=0;i<=size;i++) {
all_seq[i]=0;
}
for (i=0;i<=length;i++) {
one_seq[i]=0;
}
///////////////////////////////////////////////////////
// LOAD SEQUENCES FROM INPUT FILE IN VECTOR all_seq: //
///////////////////////////////////////////////////////
// Read input file, containing sequences, and put every letter into the all_seq vector, one sequence right after the other (we'll also include the escape character (\n) at the end of each line).
// Note that, in this way, the letter at location (row,col) in the input file will be given by all_seq[(length+1)*row+col]:
printf("\n Processing file: %s\n\n", argv[1]);
rewind(input_file);
for ( i=0; i<=size; ++i) {
fscanf(input_file,"%c",&all_seq[i]);
// printf("%d: %c\n",i, all_seq[i]);
}
// printf("%c",all_seq[2]);
///////////////////////////////////////////////////////////////
// CALCUALTE PROBABILITIES OF EACH LETTER IN FIRST SEQUENCE: //
///////////////////////////////////////////////////////////////
// Read one column at a time, write it to the one_seq vetor, and calculate the probability of occurrance of the first letter in that column; repeat for all columns:
for ( j=0; j<length; j++) {
for ( i=0; i<lines; i++) {
one_seq[i]=all_seq[(length+1)*i+j];
}
prob = CalcProb(one_seq);
// printf("col=%d: prob=%f \n\n",j,prob);
// Write output file: the first column is just an enumeration, from 1 to the length of the sequences, and the second column gives the probability for finding the first letter in the column, anywhere along the column (a measure of its conservation).
fprintf(output_file,"%d %f\n",j+1,prob);
}
////////////////////////////////////
/// CLOSE FILES AND FREE MEMORY: ///
////////////////////////////////////
fclose(input_file);
fclose(output_file);
free( all_seq );
free( one_seq );
return(0);
}
//////////////////////////////////////////////////////////////////
// ************************************************************ //
// *************** END MAIN BLOCK *************** //
// ************************************************************ //
//////////////////////////////////////////////////////////////////
//////////////////////////////////////////////////////
// FUNCTION GetSize: LENGTH AND NUMBER OF SEQUENCES //
//////////////////////////////////////////////////////
int GetSize(FILE *file, int *length, int *lines)
{
char c;
int i=0;
int *len;
// Count how many characters in first line:
while ((c=fgetc(file)) != '\n') {
++i;
}
*length=i;
*lines=0;
*len=0;
// Now count many lines, and check that all lines have same length:
i=0;
rewind(file);
while ( (c=fgetc(file)) != EOF ) {
++i;
if (c=='\n') {
*lines=*lines+1;
*len=i-1;
i=0;
// printf("length=%d\n",*len);
if ( *len != *length && *len != 0) {
printf("ERROR: not all lines in input file have the same length.\n");
exit(6);
}
if ( *len == 0) {
*lines=*lines-1;
}
}
}
// printf("lines=%d\n",*lines);
return(0);
}
///////////////////////////////////////////////////////////////////////////////
// FUNCTION CalcProb: PROBABILITY OF THE FIRST CHARACTER IN THE GIVEN VECTOR //
///////////////////////////////////////////////////////////////////////////////
double CalcProb(char *one_seq)
{
int i=0;
double count=0, prob=0;
while ( one_seq[i] != '\0' ) {
// printf("one_seq[%d]= %d\n",i,one_seq[i]);
if ( toupper(one_seq[i]) == toupper(one_seq[0]) ) {
count=count+1;
}
i=i+1;
}
prob=count/i;
// printf("count=%f, i=%d, prob=%f: \n",count,i,prob);
return(prob);
}
答案 0 :(得分:5)
释放内存中的错误通常是由于释放相同的指针两次(此处不太可能),释放无效指针,或先前的缓冲区欠载(同一块)或溢出(前面的块)导致管理数据变得腐败。既然你说它是以大文件作为输入而不是小的,你很可能会超越缓冲区。
当你只分配length
个元素时,看起来你通过在0和lines
之间迭代来超越one_seq []。
另请注意,您应该可以删除+10
安全区,并且还应将for循环修复为<
最后一个索引,而不是<=
。