File parsing for specific strings and extract data

[clife]

Hi,
I am using the below program to parse a file ,extarct values and store them into an array.
Here i have same values repeated for example IP , so i need to take only instance of it but rest of data store accordingly.

so the correct format is {IP = 192.168.0.1 , DLD = 123451 , ULD = 123454} .. so on.

The contents of my file ( test_ud.txt ) are as follows.

IP:192.168.0.1 DLD:123451
IP:192.168.0.2 DLD:123452
IP:192.168.0.3 DLD:123453
IP:192.168.0.1 ULD:123454
IP:192.168.0.2 ULD:123455
IP:192.168.0.3 ULD:123456 /*No new line after last entry*/

I just want to find out is there any other efficient way to do this using shell script or a c program.

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

struct data_info {
        char ipaddr[20];
        char upload_data[20];
        char download_data[20];
};

struct data_info *df;
void print( void);
int NOE;
int main( void )
{
        FILE *fp;
        char str[100];
        char *tok,*ip,*dd,*ud;
        int noe=0;
        fp = fopen("test_ud.txt", "r");

        while(fgets(str,sizeof (str),fp)){
                noe++;
        }
        NOE = noe / 2;
        printf("NOE = %d\n", NOE);
        fseek(fp,0,SEEK_SET);

        if(fp){
                int cnt = 0;
                int i;
                int skip_dd = 0;
                while(fgets(str,sizeof (str),fp)){
                        cnt++;
                        if( cnt == 1 ){
                                tok = strtok(str,":");
                                printf("tok = %s\n",ip=strtok(NULL," "));
                                df=(struct data_info*)malloc(sizeof(struct data_info));
                                memset(df,0,sizeof(struct data_info));
                                strcpy(df->ipaddr,ip);
                                tok = strtok(NULL,":");
                                printf("tok = %s\n",dd=strtok(NULL,"\n"));
                                strcpy(df->download_data        ,dd);

                        }else {
                                i=0;
                                tok=strtok(str,":");
                                printf("tok = %s\n",ip=strtok(NULL," "));
                                tok=strtok(NULL,":");
                                printf("tok = %s\n",dd=strtok(NULL,"\n"));

                                while( i< (cnt-1) ){
                                if (!strcmp(df[i].ipaddr,ip)){
                                        strcpy(df[i].upload_data,dd);
                                        skip_dd = 1;
                                        break;
                                }
                                printf(" df[%d].ipaddr = %s\n", i, df->ipaddr);
                                i++;
                                }
                                if(!skip_dd) {
                                        struct data_info* tmp;
                                        tmp = (struct data_info*)realloc(df,sizeof(struct data_info)*cnt);
                                        if(tmp){
                                                df = tmp;
                                                strcpy(df[cnt-1].ipaddr,ip);
                                                strcpy(df[cnt-1].download_data,dd);
                                        } else {
                                                printf("Cannot reallocate\n");
                                                return 0;
                                        }
                                }


                        }
                        printf("\n");
                                       }
        }else {
                return 0;
        }
        fclose(fp);

        print();
        return 0;
}
void print(void )
{
        int i=0;
        while(i<NOE){
                printf("ip = %s dd = %s up = %s\n", df[i].ipaddr,df[i].download_data,df[i].upload_data);
                i++;
        }
}

[David W]

You may like to see this ...
(uses some freely available custom C utility files ...
so see the comments for links to files needed in the example program.)

/* parseDataFile2.c */

/*
    NOTE! the include file readLine.h ... IS available at:
    http://developers-heaven.net/forum/index.php/topic,2580.msg2864.html#msg2864
*/

#include "readLine.h"

const char* FNAME = "test_ud.txt";
/*
    IP:192.168.0.1 DLD:123451
    IP:192.168.0.2 DLD:123452
    IP:192.168.0.3 DLD:123453
    IP:192.168.0.1 ULD:123454
    IP:192.168.0.2 ULD:123455
    IP:192.168.0.3 ULD:123456 //No new line after last entry//
*/

/* firstly define these ... */

typedef struct MyData
{
    char* paddr;
    char direction; /* 'u' or 'd' */
    int data;
} Rec;

#define DataInfo Rec

/* NEED to define ... so can free dynamic memory in Cvec */
void freeVrec( DataInfo* p )
{
    free( p->paddr );
}

/*
    NOW can include file Cvec.h ... available at:
    http://developers-heaven.net/forum/index.php/topic,2580.msg2862.html#msg2862
    and file Cvec_func's.h ... avaliable at:
    http://developers-heaven.net/forum/index.php/topic,2580.msg2866.html#msg2866
*/
#include "Cvec.h"
#include "Cvec_func's.h"


void showRec( const DataInfo* di )
{
    printf( "%20s, %c, %10d\n", di->paddr, di->direction, di->data );
}
void print( const Cvec* cv )
{
    int i = 0;
    for( ; i < cv->size; ++ i ) showRec( &cv->ary[i] );
}


void parseLine( const char* line, DataInfo* di )
{
    char* p = strchr( line, ' ' );
    if( p )
    {
        di->paddr = newsubstr( &line[3], 0, p-(line+3) );
        di->direction = *(++p);
        di->data = atoi( &p[4] );
    }
    else
    {
        printf( "Data error in file %s\n", FNAME );
        di->paddr = newsubstr( line, 0, my_strlen( line ) );
        di->direction = '?';
        di->data = 0;
    }
}

/* re. sort */
int myCmp( const DataInfo* a, const DataInfo* b )
{
    return strcmp( a->paddr, b->paddr );
}



int main( void )
{
    FILE* fp = fopen( FNAME, "r" );

    if( fp )
    {
        Cvec cv;
        char* line;
        initCvec( &cv );
        printf( "Reading and showing all lines in file '%s' ...\n", FNAME );
        while( (line = readLine( fp ) ) )
        {
            DataInfo di;
            printf( "%s\n", line );
            parseLine( line, &di );
            free( line );
            push_backCvec( &cv, &di );
        }
        fclose( fp );

        printf( "\nNow showing all records in Cvec 'cv'...\n" );
        print( &cv );

        msortCvec( &cv, myCmp );
        printf( "\nNow showing all (sorted by IP) records in Cvec 'cv'...\n" );
        print( &cv );

        clearCvec( &cv ); /* free all dynamic memory when done ... */
    }
    else
        printf( "There was a problem opening file %s\n", FNAME );

    fputs( "\nPress 'Enter' to continue/exit ... ", stdout ); fflush( stdout );
    getchar();
    return 0;
}