Thread: Ignoring most information from File.

It looks like to me that all you need to do is parse line by line and look in between the pipe characters (|) for tokens. The strcspn() function makes tasks like that easy. Since the function returns the number of characters before a pipe was found, you can print or copy a string of exactly the right length. Try this in a loop:

Code:

        tokenLength = strcspn(string, "|\n");
        printf("\t%.*s\n", (int)tokenLength, string);

With some pointer arithmetic, you can get it done in a few more lines. You need to skip the delimiter (++string; ) and know when you've reached the end of the line you are parsing.

Also note that ignoring input is basically reading input that you never used. You are going to have to slog through data that you don't need to get to what you do need. Count columns to realize where you are in relation to something useful on each line of data.

I don't think magically slapping on wide characters is going to make the output look better, if that is still a problem. If the file is ANSI encoded there is no point in putting in more difficult code when a change of the command prompt window code page would make the output look correct. Does the OP even know the file's encoding?

Siggi, sadly I must admit that the language you are using for your variables and for your user interface doesn't help me to understand the requirements of the program you are writing. Keep in mind that I'm and Italian guy who's reading/writing in (poor) English, and that I have no idea of what sentences like "Pokud soubor existuje" mean (of course I try to guess, but guessing is not a good communication technique, I fear).

Indeed, I'm enjoying what we are doing here a lot, but I'm stuck and can't go on in an efficient way unless you use plain/basic English. Or Italian, if you prefer.

Dai, che è divertente! (Come on, it's fun!)

Anyway, if I got it right, the problem you are trying to solve has a rather simple solution via the strcmp() or strncmp() standard functions from the string.h include file. One should get the relevant data while parsing the data file, store them in a couple of proper variable, break out of the cicle and show the solution to the user.

Waiting for your code in a language that we can share (English, I bet).

This is my code for now. Tryed translate to english.

Code:

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


#define endingChar 5


void convert(float summ){
    printf("\nHow much do you want convert: ");
    scanf("%f", &summ);
    if(summ < 0){
        printf("Error: negative number.\n");
    }else printf("Summ: %.3f", summ);
}


int main(void)
{
    //printf("Dom. ukol c.4.\n");


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


    if(mena == NULL){
        printf("Error: file wasn't found.\nIf exist: rename on mena.txt.\n");
        return 1;
    }


    char line[50], *code, codeState[4], instruction; //rod[50]; testing what it did
    float summ=0, exchangeRate, meziResult=0, result=0;
    int quantity;




start:
    convert(summ);
    printf("\nDefault currency: ");
    scanf("%3s", codeState);


    for(int i = 0; fgets(line, 50, mena) != NULL; ++i){
        if(i > 1){
            char *data = line;
            char *dataPtrs[endingChar];
            for(dataPtrs[0] = data, i = 1; i < endingChar; ++i){
                dataPtrs[i] = strchr(dataPtrs[i-1], '|')+1;
            }
            for(int i = 0; data[i] != '\0'; ++i){
                if(data[i] == '|' || data[i] == '\n')
                   data[i] = '\0';
                if(data[i] == ',')
                   data[i] = '.';
            }


            exchangeRate = atof(dataPtrs[4]);
            quantity = atol(dataPtrs[2]);
            code = dataPtrs[3];


           printf("%4ld %s = %.3f\n", quantity , code, exchangeRate);
             if(strcmp(code, codeState) == NULL){ //only working if is exactly in form, but i want testing upper and lowercase, it must working same if i'll get codeState = gbP or GBP or Gbp
        *rod = kod[i]; // testing what it do
    }
        }
    }
    puts("\n");


    printf("%4ld %s = %.3f\n", quantity , code, exchangeRate);
    printf("***%s***", rod); // doing wierd things






operation:
    printf("\nContinue typing: a/n\n");
    scanf("%c", &instruction);
    scanf("%c", &instruction);


    switch(instruction){
        case 'n': case 'N':{
            goto end;
        }
        case 'a': case 'A':{
            goto start;
        }
        default:{
            printf("Negative opration.\n\n");
            goto operation;
        }
    }
end:


    fclose(mena);


    puts("\nEnd programm.");
    return 0;
}

I'm thinking about it but don't know how it realize. I try it explain.

I want to ask how codeState i want convert: Euro on lira or rupie on kuna etc..
I know formula:
meziResult = summ * [(exchangeRate / quantity) /* default code */]
Result = ( meziResult * [quantity) /* on what we want convert = end Code */] / [exchangeRate /* end code */]

when i asking on stateCode i'm going get only code but i need his quantity and exchangeRate on operating with it

Oh, you want work with the values. Actually, you only read the file line by line and don't hold the datas in memory.
Every time you read a line, the prevous line is gone. So you must hold all lines, or better the extracted datas from the lines.
There are 2 capabilities:

1) you input the code and the programm search for the correct line in file and give back the values.
Then a second time for the second code. If you have the values, you can work with it.

2) you read in all datas and take it from memory. With this method, you read the file only once.
If you input the code, you run only through the datas in memory and return the right values.

But i don't know if you have learned about arrays or structures and memory allocating (malloc, calloc, realloc).
I have an idea with structures, but if you don't have learned about it, you will be more confused like before.

There are mistakes here and there, but I suppose many of them are translation mistakes.

Apart from those, you are changing ',' to '.' in the whole line of data. In my opinion it would be safer (and faster) changing just the string representing the numeric data, that is dataPtrs[2] and dataPtrs[4]. Just a minor improvement.

At some point you ask the user to tell you the source currency code, and he gave you a three characters string. You stored it in "codeState". Ok. You say that you want to make both uppercase and lowercase valid. Well, just convert codeState to all uppercase with something like this:

Code:

for( counter=0; counter<3; ++counter )
    codeState[counter] = toupper(codeState[counter]);

(toupper() is a standard function defined in ctype.h, so include ctype.h)

Just to be sure, you could convert to uppercase your "code" variable, too.
Now you can use strcmp() safely, as both "codeState" and "code" contain only uppercase characters.

If you realize that "codeState" equals to "code", you can save the relevant data taken from the current line and LEAVE THE "for" LOOP WITH A "break" INSTRUCTION (if you don't, you are going to overwrite the data you found with other data that you don't need). Once outside of the "for" loop, you can perform your conversions.

Ah, don't forget that you have to go back to the beginning of the data file each time you perform a new conversion. Use rewind(mena) for that task, just after "start:" (we are trying to implement the solution that WoodSTokk described at his point #1, so we have to read the file over and over).

Further revisions could be needed in order to get exactly what you need.

edit:
thanks aldo_baldo
You describe #1 of the list.
The following is a description of #2

Okay, not dynamic.
You can define a structure that hold the information of one country.
In the main()-function, you define an array with the typ of the structure that is big enough to hold all countries.

Code:

# define MAX_CURRENCY 50
# define MAX_LEN_COUNTRY 30
# define MAX_LEN_CURRENCY 10
# define MAX_LEN_CODE 10

typedef struct currency_t {
    char country[MAX_LEN_COUNTRY];
    char currency[MAX_LEN_CURRENCY];
    int quantity;
    char code[MAX_LEN_CODE];
    float course;
} Currency;

int main(void) {

Currency currency_list[MAX_CURRENCY];
int currency_now = 0;
…

In the loop where you extract the information out of line, you assign the information to the members of the struct-array.
Don't forget to increment the counter 'currency_now' at end.

Code:

…
        strncpy (currency_list[currency_now].country, dataPtrs[0], MAX_LEN_COUNTRY);
        strncpy (currency_list[currency_now].currency, dataPtrs[1], MAX_LEN_CURRENCY);
        currency_list[currency_now].quantity = atol(dataPtrs[2]);
        strncpy (currency_list[currency_now].code, dataPtrs[3], MAX_LEN_CODE);
        currency_list[currency_now].course = atof(dataPtrs[4]);

// make sure to terminate the strings
        currency_list[currency_now].country[MAX_LEN_COUNTRY - 1] = '\0';
        currency_list[currency_now].currency[MAX_LEN_CURRENCY - 1] = '\0';
        currency_list[currency_now].code[MAX_LEN_CODE - 1] = '\0';

        if (++currency_now >= MAX_CURRENCY) {
            printf("Too many currencies! exit.\n");
            return -1;
        }
…

At the end of the loop, you can close the file.
All informations are now in the struct-array and the counter hold the number of used entries.
You can every time loop through the array and search for specific information.

Originally Posted by aldo_baldo

One could decide to make the array as large as needed to include the currency used in the whole world. Consider that there are less than 200 Countries on the planet. Indeed, we could define MAX_CURRENCY as 200 and be safe just with a few KB in memory.

Yes, i only looked at the file from OP and set MAX_CURRENCY a bit larger and have a security break build in.
Normaly, i use a dynamic array or linked list, but the OP has not learned about that until now.

Code:

    printf("\nVychozi mena: ");    
for(int i = 0; i < 3 ; ++i){
        scanf("%3s", vychoziKod);
        vychoziKod[i] = toupper(vychoziKod[i]);
    }

It's possible do same thing but print whole array and not printing for letters?

Code:

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


#define ukoncovaciZnak 5


typedef struct registrMeny_t {
    char zeme[30];
    char bankovka[10];
    int mnozstvi;
    char kod[10];
    float kurz;
} RegistrMeny;


int main(void)
{
    printf("Dom. ukol c.4.\n");


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


    if(mena == NULL){
        printf("Error: soubor nebyl nalezen.\nPokud soubor existuje: prejmenujejte jej na mena.txt.\n");
        return 1;
    }


    char radek[50], *kod, vychoziKod[4], cilovyKod[4], instrukce;
    float suma=0, kurz, meziVysl=0, vysl=0;
    int mnozstvi;




zacatek:
    rewind(mena);


    printf("\nKolik si prejete prevest: ");
    scanf("%f", &suma);


    printf("\nVychozi mena: ");
    for(int i = 0; i < 3 ; ++i){
        scanf("%3s", vychoziKod);
        vychoziKod[i] = toupper(vychoziKod[i]);
    }


    printf("\nCilova mena: ");
    scanf("%3s", cilovyKod);


    for(int i = 0; fgets(radek, 50, mena) != NULL; ++i){
        if(i > 1){
            char *data = radek;
            char *dataPtrs[ukoncovaciZnak];


            RegistrMeny seznamMeny[50];
            int aktualniMena = 0;


            for(dataPtrs[0] = data, i = 1; i < ukoncovaciZnak; ++i){
                dataPtrs[i] = strchr(dataPtrs[i-1], '|')+1;
            }
            for(int i = 0; data[i] != '\0'; ++i){
                if(data[i] == '|' || data[i] == '\n')
                   data[i] = '\0';
                if(data[i] == ',')
                   data[i] = '.';
            }


            strncpy(seznamMeny[aktualniMena].zeme, dataPtrs[0], 30);
            strncpy(seznamMeny[aktualniMena].bankovka, dataPtrs[1], 10);
            seznamMeny[aktualniMena].mnozstvi = atol(dataPtrs[2]);
            strncpy (seznamMeny[aktualniMena].kod, dataPtrs[3], 10);
            seznamMeny[aktualniMena].kurz = atof(dataPtrs[4]);


            printf("%4ld %s = %.3f\n", seznamMeny[aktualniMena].mnozstvi , seznamMeny[aktualniMena].kod, seznamMeny[aktualniMena].kurz);


            if(strcmp(seznamMeny[aktualniMena].kod, vychoziKod) ==  NULL){
                meziVysl = suma * (seznamMeny[aktualniMena].kurz / (float)seznamMeny[aktualniMena].mnozstvi);
           /*         printf("\n%.3f", suma);
                    printf("\n%.3f", seznamMeny[aktualniMena].kurz);
                    printf("\n%d", seznamMeny[aktualniMena].mnozstvi);   <--- testing
                    printf("\n%.3f\n", meziVysl);
            }*/

/** Question **/
  /* when i had there other if (i will paste)
  if(strcmp(seznamMeny[aktualniMena].kod, cilovyKod) == NULL){
  vysl = (meziVysl /  seznamMeny[aktualniMena].kurz) * (float)seznamMeny[aktualniMena].mnozstvi);

so there theese two if fighting each other ... i must have twice that cyclo? or i can somehow save place?

*/

            seznamMeny[aktualniMena].zeme[30 - 1] = '\0';
            seznamMeny[aktualniMena].bankovka[10 - 1] = '\0';
            seznamMeny[aktualniMena].kod[10 - 1] = '\0';


            if(++aktualniMena >= 50){
                printf("Error: prilis mnoho informaci!\n");
                return -1;
            }


            ++aktualniMena;
        }
    }
    puts("\n");


operace:
    printf("\nPokracovat v zadavani: a/n\n");
    scanf("%c", &instrukce);
    scanf("%c", &instrukce);


    switch(instrukce){
        case 'n': case 'N':{
            goto konec;
        }
        case 'a': case 'A':{
            goto zacatek;
        }
        default:{
            printf("Neplatna instrukce.\n\n");
            goto operace;
        }
    }
konec:


    fclose(mena);


    puts("\nKonec programu.");
    return 0;
}

I'm a bit confusion, but it'll be okay. I have few questions in code.
(Don't be angry, again in my langauge, but i hope u will not need theese words.)

If you want to use a label and goto for cleanup, that could work. However, what I see here is that you're using goto where you should be using loops or separating your code into functions.