PostgreSQL
 sql >> база данни >  >> RDS >> PostgreSQL

PostgreSQL Запазване и избиране на файлове с помощта на двоични данни Bytea с C libpq

Това е резултат от добрата работа на кода. Имам малко време да го публикувам сега, в случай че може да помогне на някого. Беше ми малко трудно да разбера как работи bytea, но най-накрая е решен. Сърдечни поздрави към общността

#include <string>
#include <fstream>
#include <iostream>
#include <sstream>

#include "libpq/libpq-fs.h"
#include "libpq-fe.h"

#pragma comment(lib, "libpq.lib")   /*!< Only for windows compilation */

int main(int argc, char* argv[])
{
    ManageDB manager;       

    manager.conn = manager.ConnectDB();  // My manager, connects ok

    const char* fileName = {"test.png"};
    const char* fileNameOut = { "testOut.png" };

    FILE* file = fopen(fileName, "rb");
    if (file == NULL) cout << endl << "FILE WAS UNABLE TO BE READED" << endl;
    fseek(file, 0, SEEK_END);
    long int fileSize = ftell(file);

    rewind(file);

    unsigned char* buffRead = (unsigned char*)malloc(fileSize);
    size_t bytes_read = fread(buffRead, 1, fileSize, file);
    if (bytes_read != fileSize) cout << endl << "fread reading Error";
    fclose(file);
    
    size_t* sizeP = new size_t(fileSize);

    const char* paramValues[3];
    paramValues[0] = "1";
    paramValues[1] = fileName;
    paramValues[2] = reinterpret_cast<const char*>(buffRead);   // type cast required to PQexeParams

    const int paramFormats[3]{ 0,0,1 };   //0 = text, 1 = binary
    const int paramLenghts[3]{ strlen(paramValues[0]), strlen(paramValues[1]), fileSize};

    PGresult *res = PQexecParams(manager.conn, "INSERT INTO filebyte (id, filename, file) VALUES($1::text, $2::text, $3::bytea)",
        3,              /* params num */
        NULL,           /* let the backend deduce param type */
        paramValues,
        paramLenghts,
        paramFormats,
        1);

    if (res && PQresultStatus(res) == PGRES_COMMAND_OK) cout << endl << "Inserted data to filebyte OK";

    PQfreemem(res);

    //************************download from DB*************************
    
    const char* bytesFromDB = new const char[fileSize];
    int sizeR;
    // Table columns = id(text) | filename(text) | file(bytea)
    const char* sentenceEx = "SELECT file FROM filebyte WHERE id='1'";

    res = PQexec(manager.conn, sentenceEx);

    if (res && PQresultStatus(res) == PGRES_TUPLES_OK)
    {
        sizeR = PQgetlength(res, 0, 0);
        bytesFromDB = PQgetvalue(res, 0, 0);
    }
    else cout << endl << "Error at inserting file data in filebyte table";

    string hex(bytesFromDB, 2, sizeR-2);  //removing the first '\x' characters of the result.

    // vars for converting to real bytes process
    std::basic_string<uint8_t> bytes;
    bytes.clear();
    std::string nextbyte; 
    nextbyte.clear();
    uint16_t byte;

    // Iterate over every pair of hex values in the input string (e.g. "18", "0f", ...)
    for (size_t i = 0; i < hex.length(); i += 2)
    {
        // Get current pair and store in nextbyte
        nextbyte = hex.substr(i, 2);

        // Put the pair into an istringstream and stream it through std::hex for
        // conversion into an integer value.
        // This will calculate the byte value of your string-represented hex value.
        std::istringstream(nextbyte) >> std::hex >> byte;

        // As the stream above does not work with uint8 directly, we have to cast it now.
        // As every pair can have a maximum value of "ff",
        // which is "11111111" (8 bits), we will not lose any information during this cast.
        bytes.push_back(static_cast<uint8_t>(byte));
    }

    // string obj from bytes-"array"
    std::string result(begin(bytes), end(bytes));

    ofstream myFile(fileNameOut, ios::out | ios::ios_base::binary);
    
    if (myFile.is_open())
    {
        myFile << result;
        myFile.close();
    }
    else cout << endl << "Impossible to writte the file " << fileNameOut;
    
    manager.CloseConn(manager.conn);   //closes connection with database internally ok

    PQfreemem(res);
    free(buffRead);
    delete sizeP;
    free((char*)fileName[8]);
    free((char*)fileNameOut[11]);
    
    return true;
}



  1. Database
  2.   
  3. Mysql
  4.   
  5. Oracle
  6.   
  7. Sqlserver
  8.   
  9. PostgreSQL
  10.   
  11. Access
  12.   
  13. SQLite
  14.   
  15. MariaDB
  1. Как да изброявате потребителски типове с помощта на Postgres information_schema

  2. Как да използвате една и съща стойност няколко пъти в подготвения оператор на jdbc postgresql

  3. CSV копие в Postgres с масив от потребителски тип с помощта на JDBC

  4. to_sql + sqlalchemy + копиране от + postgresql двигател?

  5. Защо Postgres казва, че колоната не съществува?