用C读取CSV文件并处理数据中的逗号

问题描述 投票:0回答:3

我写了一段代码来用c读取csv文件。该文件包含游戏数据,我应该读取它并根据分数对其进行排序并打印排名前 10 的游戏。代码如下:

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#define tablesize 18626

typedef struct
{
    char title[200];
    char platform[20];
    char Score[20];
    char release_year[20];
} dict;

void printValues(dict *values)
{
    for (int i = 0; i < 100; i++)
    {
        printf("title->%s,platform->%s,Score->%s,release->%s\n", values[i].title, values[i].platform, values[i].Score, values[i].release_year);
    }
}

void sort(dict *values)
{
    for (int i = 0; i < tablesize; i++)
    {
        for (int j = i + 1; j < tablesize; j++)
        {
            int a = *values[i].Score - '0';
            int b = *values[j].Score - '0';
            // printf("%d %d\n",values[i].Score,values[j].Score);
            if (a < b)
            {
                dict temp = values[i];
                values[i] = values[j];
                values[j] = temp;
            }
        }
    }
}

int main()
{
    FILE *fp = fopen("t4_ign.csv", "r");
    if (!fp)
    {
        printf("Error");
        return 0;
    }
    char buff[1024];
    int row = 0, column = 0;
    int count = 0;
    dict *values = NULL;
    int i = 0;
    while (fgets(buff, 1024, fp))
    {
        column = 0;
        row++;
        count++;
        values = realloc(values, sizeof(dict) * count);
        if (NULL == values)
        {
            perror("realloc");
            break;
        }
        if (row == 1)
        {
            continue;
        }
        char *field = strtok(buff, ",");
        while (field)
        {
            if (column == 0)
            {
                strcpy(values[i].title, field);
            }
            if (column == 1)
            {
                strcpy(values[i].platform, field);
            }
            if (column == 2)
            {
                strcpy(values[i].Score, field);
            }
            if (column == 3)
            {
                strcpy(values[i].release_year, field);
            }
            field = strtok(NULL, ",");
            column++;
        }
        i++;
    }
    fclose(fp);
    printf("File loaded!\n", fp);
    sort(values);
    printValues(values);
    free(values);
    return 0;
}

我面临的问题是 CSV 文件的标题字段中有逗号,因此它将用逗号分隔的数据区分为不同的列,这在将数据加载到结构中时会出错。

以下是输入文件的两个示例行。当标题包含逗号时使用引号。

"The Chronicles of Narnia: The Lion, The Witch and The Wardrobe",PlayStation 2,8,2005  
The Chronicles of Narnia: Prince Caspian,Wireless,5,2008

有什么建议吗?预先感谢。

c csv file-handling
3个回答
2
投票

由于标题字段包含逗号时会使用引号,因此我建议您检查是否使用了

"
。如果是这样,请使用该分隔符作为第一项。

char *field;
if(buff[0] == '"') {
    field = strtok(buff, "\"");
}
else {
    field = strtok(buff, ",");
}

第一个将保留逗号作为下一个字段的第一个字符,但下一个

strtok
将过滤掉该逗号,因为它不允许“空”字段。


0
投票

函数

strtok
不适合您的需求,因为它将引号视为与其他字符一样的字符。因此,当
strtok
看到逗号时,它不会关心逗号是否在引号内。

此外,正如其他人在评论部分指出的那样,

strtok
的另一个问题是它会跳过空字段。

因此,我不建议使用

strtok
来完成您想做的事情。

为了解决您的问题,我建议您编写自己的函数,其功能与

strtok
strsep
非常相似,但如果第一个非 whitespace 字符是引号,则它会考虑下一个引号作为分隔符而不是下一个逗号。在下面的代码中,我将这个函数命名为
my_strsep

这是一个例子:

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

#define NUM_LINES 2

//this function is equivalent to the POSIX function "strsep", except
//that it always uses "," as a delimiter, unless the first
//non-whitespace character is a quotation mark, in which case it //skips the quotation mark and uses the next quotation mark as a
//delimiter, also consuming the next comma
char *my_strsep( char **restrict stringp )
{
    char *p = *stringp;
    char *start;
    char delimiter = ',';

    //do nothing if *stringp is 
    if ( *stringp == NULL )
        return NULL;

    //skip all whitespace characters
    while ( isspace( (unsigned char)*p ) )
        p++;

    //remember start of field
    start = p;

    //determine whether this field uses quotation marks
    if ( *p == '"' )
    {
        //set delimiter to quotation mark instead of comma
        delimiter = '\"';

        //skip the first quotation mark
        p++;
    }

    //remember the start of the string
    start = p;

    while ( *p != delimiter )
    {
        if ( *p == '\0' )
        {
            if ( delimiter == '\"' )
            {
                fprintf( stderr,
                    "Warning: Encountered end of string before the "
                    "second quotation mark!\n"
                );
            }

            //pass information back to calling function
            *stringp = NULL;
            return start;
        }

        p++;
    }

    //overwrite the delimiter with a null character
    *p = '\0';

    //go past the delimiter
    p++;

    //skip the comma too, if quotation marks are being used
    if ( delimiter == '\"' )
    {
        //skip all whitespace characters
        while ( isspace( (unsigned char)*p ) )
            p++;

        //skip the comma
        if ( *p == ',' )
            p++;
    }

    //pass information back to calling function
    *stringp = p;
    return start;
}

int main( void )
{
    char lines[NUM_LINES][200] = {
        "\"The Chronicles of Narnia: The Lion, The Witch and The Wardrobe\",PlayStation 2,8,2005",
        "The Chronicles of Narnia: Prince Caspian,Wireless,5,2008"
    };

    for ( int i = 0; i < NUM_LINES; i++ )
    {
        char *p, *q;

        printf( "Processing line #%d:\n", i + 1 );

        p = lines[i];

        while ( ( q = my_strsep( &p ) ) != NULL )
        {
            printf( "Found field: %s\n", q );
        }

        printf( "\n" );
    }
}

该程序有以下输出:

Processing line #1:
Found field: The Chronicles of Narnia: The Lion, The Witch and The Wardrobe
Found field: PlayStation 2
Found field: 8
Found field: 2005

Processing line #2:
Found field: The Chronicles of Narnia: Prince Caspian
Found field: Wireless
Found field: 5
Found field: 2008

如您所见,函数

my_strsep
可以处理带引号和不带引号的字段。


0
投票
#include <iostream>
#include <string>
#include <fstream>
#include <iomanip>
#include <limits>
#include <algorithm>
using namespace std;

const int MAX_TOPPINGS = 2;
const int MAX_DONUTS = 50;

struct donutType {
    string name;
    bool type;
    double price;
    string filling;
    string toppings[MAX_TOPPINGS];
};

ifstream getFileStream(string);
int getDonuts(ifstream&, donutType[]);
bool continueMenu(string);
void sortByPrice(donutType[], int);
int searchByName(const donutType[], int, string);
void removeDonutFromList(donutType[], int&, int);
int getCheapestDonut(const donutType[], int);
void soldDonut(donutType, donutType[], int&);
void outputSoldDonuts(ofstream&, const donutType[], int);
void displayAvailableDonuts(const donutType[], int);
string allCaps(string);

/**
 * The main function to manage the donut ordering program.
 *
 * @return The exit status of the program.
 */
int main() {
    ifstream infile;
    ofstream outfile("sold.csv");
    string filename;
    string request;
    donutType donuts[MAX_DONUTS];
    donutType soldDonuts[MAX_DONUTS];
    int amtDonuts = 0;
    int amtSold = 0;
    double total = 0;

    cout << fixed << setprecision(2);

    // gets step 1 the input stream
    infile = getFileStream("Enter a donut file: ");

    // step 2 get total donuts
    amtDonuts = getDonuts(infile, donuts);

    // step 3 print to console
    cout << "Welcome to Hank's Donut World!\n\n";
    while (true) {

        // step 4 print the available donuts to console
        displayAvailableDonuts(donuts, amtDonuts);

        //step 5 ask for user input
        string nameEnteredByUser;
        cout << "Enter donut name or cheapest: ";
        getline(cin, nameEnteredByUser);

        // step 6 check the user input choice and call function accordingly
        int donutIndex;
        transform(nameEnteredByUser.begin(), nameEnteredByUser.end(),
                nameEnteredByUser.begin(), ::tolower);

        if (nameEnteredByUser.compare("cheapest") == 0)
            donutIndex = getCheapestDonut(donuts, amtDonuts);
        else
            donutIndex = searchByName(donuts, amtDonuts, nameEnteredByUser);

        // step 7 check if the search results was empty
        if (donutIndex == -1) {
            cout << "Donut not found!\n";
            continue;
        }

        // step 8 print the selection to console
        cout << "You selected " << donuts[donutIndex].name
                << ".\nExcellent choice!\n";

        // Step 9 inserting the sold donut into solddonuts array
        total += donuts[donutIndex].price;
        soldDonuts[amtSold] = donuts[donutIndex];
        amtSold++;

        // Step 10 removing the donut from donuts array
        removeDonutFromList(donuts, amtDonuts, donutIndex);

        // Step 11 choice to continue with more purchase
        bool complete = continueMenu("Will this complete your order? ");
        if (!complete) {
            if (amtDonuts > 0)
                continue; // go to Step 4
            else
                break; // End the program if there are no available donuts left

        }

        // Step 13 sort the donuts sold by price using bubble sort
        sortByPrice(soldDonuts, amtSold);

        // Step 14
        if (outfile.is_open()) {
            outfile << "Sold," << fixed << setprecision(2) << total << "\n";
            outputSoldDonuts(outfile, soldDonuts, amtSold);
        } else {
            cerr << "Error opening output file.\n";
        }
        break; // End the program
    }

    return 0;
}

/**
 * Retrieves an input file stream for a given filename after prompting the user.
 *
 * @param msg The message to prompt the user for the filename.
 * @return An ifstream object for the specified filename.
 */
ifstream getFileStream(string msg) {
    ifstream fileStream;
    string fileName;

    while (true) {
        // Prompt user for a filename
        cout << msg;
        getline(cin, fileName);

        // Open the file stream
        fileStream.open(fileName);

        // Check if the file stream is open
        if (fileStream.is_open()) {
            break;
        } else {
            fileStream.clear(); // Clear any error flags
            cin.ignore(numeric_limits<streamsize>::max(), '\n'); // Clear input buffer
        }
    }

    return fileStream;
}

/**
 * Reads donut data from an input file stream into a struct array of donutType.
 *
 * @param infile An input file stream containing donut data.
 * @param donuts An array of donutType to store the read data.
 * @return The number of donuts read from the file.
 */
int getDonuts(ifstream &infile, donutType donuts[]) {
    const int MAX_RECORDS = 50;
    size_t MAX_CHARS = 100;
    char line[MAX_CHARS];
    int count = 0;

    // Read and ignore the header line
    infile.getline(line, MAX_CHARS);

    // Read the CSV file line by line
    while (!infile.fail() && count < MAX_RECORDS) {
        // Read the entire line into 'line'
        infile.getline(line, MAX_CHARS);

        // Create a strings to parse the line
        string lineString(line);
        if(lineString.size() < 2)
            break;
        // Tokenize the line based on commas
        size_t start = 0;
        size_t end = lineString.find(',');

        // Read name
        donuts[count].name = lineString.substr(0, end);
        //add plus one to skip the comma
        lineString = lineString.substr(end+1);

        // Read type
        end = lineString.find(',');
        donuts[count].type = (lineString.substr(0, end) == "Cake");
        lineString = lineString.substr(end+1);

        // Read filling
        end = lineString.find(',');
        donuts[count].filling = lineString.substr(0, end);
        lineString = lineString.substr(end+1);

        // Read toppings
        end = lineString.find(',');
        donuts[count].toppings[0] = lineString.substr(0, end);
        lineString = lineString.substr(end+1);

        end = lineString.find(',') ;
        donuts[count].toppings[1] = lineString.substr(0, end);
        lineString = lineString.substr(end+1);

        // Read price
        donuts[count].price = stod(lineString);

        count++;
    }
    return count;
}

/**
 * Displays a prompt and waits for user input to continue or exit.
 *
 * @param prompt The message to display as a prompt.
 * @return True if the user chooses to continue, false if the user chooses to exit.
 */
bool continueMenu(string prompt) {
    string input;

    while (true) {
        cout << prompt;
        getline(cin, input);

        // Convert the input to lowercase for case-insensitive comparison
        transform(input.begin(), input.end(), input.begin(), ::tolower);

        if (input == "no")
            return false;
        else if (input == "yes")
            return true;
        else
            cerr << "Invalid input. Please enter 'Yes' or 'No' (case insensitive).\n";

    }
}

/**
 * Sorts an array of donuts based on their prices in ascending order using bubble sort.
 *
 * @param donuts An array of donuts to be sorted.
 * @param amtDonuts The number of donuts in the array.
 */
void sortByPrice(donutType donuts[], int amtDonuts) {
    for (int i = 0; i < amtDonuts - 1; ++i) {
        for (int j = 0; j < amtDonuts - i - 1; ++j) {
            // Compare prices and swap if needed
            if (donuts[j].price > donuts[j + 1].price) {
                // Swap
                donutType temp = donuts[j];
                donuts[j] = donuts[j + 1];
                donuts[j + 1] = temp;
            }
        }
    }
}

/**
 * Searches for a donut by name in an array of donuts.
 *
 * @param donuts An array of donuts to be searched.
 * @param amtDonuts The number of donuts in the array.
 * @param name The name of the donut to be searched.
 * @return The index of the found donut if present, otherwise -1.
 */
int searchByName(const donutType donuts[], int amtDonuts, string name) {

    transform(name.begin(), name.end(), name.begin(), ::tolower);

    for (int i = 0; i < amtDonuts; ++i) {
        string donutName = donuts[i].name;
        transform(donutName.begin(), donutName.end(), donutName.begin(),
                ::tolower);
        if (donutName.compare(name) == 0)
            return i; // Return the index if the name is found
    }
    return -1; // Return -1 if the name is not found
}

/**
 * Finds the index of the cheapest donut in an array of donuts.
 *
 * @param donuts An array of donuts to be searched.
 * @param amtDonuts The number of donuts in the array.
 * @return The index of the cheapest donut if the array is not empty, otherwise -1.
 */
int getCheapestDonut(const donutType donuts[], int amtDonuts) {
    if (amtDonuts <= 0)
        return -1; // Return -1 if the array is empty

    // Initialize to maximum possible value
    double minPrice = donuts[0].price;
    // Index of the cheapest donut
    int minIndex = 0;

    for (int i = 1; i < amtDonuts; ++i) {
        if (donuts[i].price < minPrice) {
            minPrice = donuts[i].price;
            minIndex = i;
        }
    }

    return minIndex;
}

/**
 * Converts a given string to uppercase.
 *
 * @param s The input string to be converted to uppercase.
 * @return The uppercase version of the input string.
 */
string allCaps(string s) {
    string upper = s;

    for (char &c : upper)
        c = toupper(static_cast<unsigned char>(c));

    return upper;
}

/**
 * Removes a donut from the list based on the provided index.
 *
 * @param donuts The array of donuts.
 * @param amtDonuts The current number of donuts in the array.
 * @param removeIndex The index of the donut to be removed.
 */
void removeDonutFromList(donutType donuts[], int &amtDonuts, int removeIndex) {
    if (removeIndex < 0 || removeIndex >= amtDonuts) {
        cerr << "Invalid index to remove. Index out of range." << endl;
        return;
    }

    // Shift elements to fill the gap
    for (int i = removeIndex; i < amtDonuts - 1; ++i) {
        donuts[i] = donuts[i + 1];
    }

    // Decrement the amount of donuts
    amtDonuts--;
}

/**
 * Outputs information about sold donuts to an output file.
 *
 * @param outfile The output file stream.
 * @param soldDonuts The array of sold donuts.
 * @param amtSold The current number of sold donuts in the array.
 */
void outputSoldDonuts(ofstream &outfile, const donutType soldDonuts[],
        int amtSold) {

    if (!outfile.is_open()) {
        cerr << "Error: Output file is not open." << endl;
        return;
    }

    outfile << fixed << setprecision(2);
    outfile << "Name,Type,Filling,Topping1,Topping2,Price" << endl;

    for (int i = 0; i < amtSold; ++i) {
        outfile << soldDonuts[i].name << ",";
        outfile << (soldDonuts[i].type ? "Cake" : "Dough") << ",";
        outfile << soldDonuts[i].filling << ",";
        outfile << soldDonuts[i].toppings[0] << ",";
        outfile << soldDonuts[i].toppings[1] << ",";
        outfile << soldDonuts[i].price << "\n";
    }
}

/**
 * Displays the list of available donuts with their names and prices.
 *
 * @param donuts The array of available donuts.
 * @param amtDonuts The current number of available donuts in the array.
 */
void displayAvailableDonuts(const donutType donuts[], int amtDonuts) {
    cout << "List of donuts" << endl;
    cout << "---------------------------" << endl;
    int count = 0;
    for (int i = 0; i < amtDonuts; ++i)
        cout << donuts[i].name << " " << donuts[i].price << endl;

    cout << endl;
}
© www.soinside.com 2019 - 2024. All rights reserved.