fork download
copy 
  1. #include <stdio.h>
  2. #include <string.h> // strncpy
  3.  
  4. #include <stdlib.h>  //exit
  5. #include <ctype.h>  // isspace
  6. #include <stdint.h>  // uint32_t
  7. #include <math.h>  // round
  8.  
  9. #include <string> //std::string
  10. #include <vector> //std::vector
  11.  
  12. //#define BUF_SIZE 15 // 512 bype
  13. //#define BUF_SIZE 512 // 512 bype
  14. //#define BUF_SIZE 1024 // 1KB
  15. //#define BUF_SIZE 1024 * 1024  // 1MB
  16. #define BUF_SIZE 1024 * 1024 * 8 // 8MB
  17. //#define BUF_SIZE 1024 * 1024 *16 // 16MB
  18.  
  19. #define LINE_SIZE 1024 * 2
  20. #define NAME_SIZE 128
  21.  
  22. //#define IS_PRINT true
  23. #define IS_PRINT false
  24.  
  25. #define READ_MODE 0  // 0=fread only, 1=read_data()関数のみ, それ以上=read_data() & 解析
  26.  
  27. typedef enum __TYPE {
  28.     TYPE_x,
  29.     TYPE_y
  30. } TYPE;
  31.  
  32. //--------------------------------------------------------------------------------
  33. typedef struct __tag_qq_db {
  34.     char qq_name[NAME_SIZE];
  35. } t_qq_db;
  36. typedef struct __tag_elem_db {
  37.     TYPE type;
  38.     unsigned int no;
  39.     unsigned int num_line_count;
  40.     t_qq_db qq_db;
  41.     std::vector<double> numbers;
  42. } t_elem_db;
  43. typedef struct __tag_section_db {
  44.     char name[NAME_SIZE];
  45.     unsigned int num_1;
  46.     unsigned int num_2;
  47.     unsigned int text_line_count;
  48.     std::vector<std::string> texts;
  49. } t_section_db;
  50. typedef struct __tag_read_db {
  51.     char header_name[NAME_SIZE];
  52.     unsigned int mag;
  53. } t_read_db;
  54. //--------------------------------------------------------------------------------
  55. struct CCC {
  56.     FILE *fp;
  57.     unsigned char real_data[BUF_SIZE+1];
  58.     unsigned char* data;
  59.     unsigned char* temp_data;
  60.  
  61.     t_section_db* p_current_section_db;
  62.     t_elem_db* p_current_elem_db;
  63.  
  64.     t_read_db read_db;
  65.  
  66.     bool initialize();
  67.     bool read(char* file_path);
  68.     size_t read_data();
  69.     bool get_header();
  70.     bool get_section_name();
  71.     bool get_section_index();
  72.     bool get_text();
  73.     bool get_elem_index();
  74.     bool get_elem_qq();
  75.     bool get_elem_x(unsigned int i);
  76.     bool get_elem_y(unsigned int i);
  77. };
  78. //--------------------------------------------------------------------------------
  79. bool CCC::initialize(){
  80.     real_data[BUF_SIZE] = '\x00';  //最後はかならずNULLになるようにセット
  81.     data = real_data;
  82.     return true;
  83. }
  84. size_t CCC::read_data(){
  85.     size_t size = 0;
  86.     if( *data == '\x00' ){
  87.         size = fread( &real_data, sizeof(real_data[0]), BUF_SIZE, fp);
  88.         data = real_data;
  89.         // 終端処理。最大値で取得されてなければそこを末尾にする
  90.         if( size != BUF_SIZE ) data[size] = '\x00';
  91.         // 取得bufの最後尾が改行でなければfpを改行まで戻す
  92.         if( data[size-1] != '\x0a' ){
  93.             int newline_index = -2;
  94.             for(; data[size+newline_index] != '\x0a'; --newline_index);
  95.             fseek(fp,newline_index+1,SEEK_CUR);
  96.             // bufの最後の改行の次にx00(null)を入れてそれ以降をカット
  97.             data[size+newline_index+1] = '\x00';
  98.         }
  99.     }
  100.     return size;
  101. }
  102. // 32=空白
  103. // 0a=改行
  104. // 00=null
  105.  
  106. bool CCC::get_section_name(){
  107.     read_data();
  108.     temp_data = data;
  109.     for(; *++data != '\x0a'; );
  110.  
  111.     memcpy(p_current_section_db->name, temp_data, data-temp_data);
  112.     p_current_section_db->name[data-temp_data] = '\x00';
  113.  
  114.     ++data;
  115.     IS_PRINT == true && printf("%s\n", p_current_section_db->name);
  116.     return true;
  117. }
  118. bool CCC::get_section_index(){
  119.     read_data();
  120.     t_section_db section_db;
  121.     //数値グループ数を取得
  122.     auto func1 = [&](unsigned int* num) {
  123.         unsigned int n = 0;
  124.         do {
  125.             unsigned char ch = (unsigned char)(*data++ - '0');
  126.             if (ch <= 9) {
  127.                 n = n * 10 + ch;
  128.             } else {
  129.                 break;
  130.             }
  131.         } while (true);
  132.         *num = n;
  133.     };
  134.     func1(&p_current_section_db->num_1);
  135.     func1(&p_current_section_db->num_2);
  136.     func1(&p_current_section_db->text_line_count);
  137.     if( p_current_section_db->num_1 != p_current_section_db->num_2 ){
  138.         printf("num_1, num_2の数が違う！！解析する！！ %d %d\n", p_current_section_db->num_1, p_current_section_db->num_2);
  139.         printf("%s\n", data);
  140.         exit(1);
  141.     }
  142.  
  143.     if( IS_PRINT == false ){
  144.         // 日付を捨てる
  145.         for(; *data++ != '\x0a'; );
  146.     }else{
  147.         printf("%d %d %d ", p_current_section_db->num_1, p_current_section_db->num_2, p_current_section_db->text_line_count);
  148.         printf("%c", data[0]);
  149.         for(; *data++ != '\x0a'; ){
  150.             printf("%c", data[0]);
  151.         }
  152.     }
  153.     return true;
  154. }
  155.  
  156. bool CCC::get_text(){
  157.     read_data();
  158.     temp_data = data;
  159.     for(; *++data != '\x0a'; );
  160.  
  161.     char line[LINE_SIZE];
  162.  
  163.     memcpy(line, temp_data, data-temp_data);
  164.     line[data-temp_data] = '\x00';
  165.     IS_PRINT == true && printf("%s\n", line);
  166.  
  167.     ++data;
  168.     return true;
  169. }
  170. bool CCC::get_elem_index(){
  171.     read_data();
  172.  
  173.     //数値グループ数を取得
  174.     auto func1 = [&](unsigned int* num) {
  175.         unsigned int n = 0;
  176.         do {
  177.             unsigned char ch = (unsigned char)(*data++ - '0');
  178.             if (ch <= 9) {
  179.                 n = n * 10 + ch;
  180.             } else {
  181.                 break;
  182.             }
  183.         } while (true);
  184.         *num = n;
  185.     };
  186.  
  187.     // 数値グループ取得開始
  188.     // 78=x, 79=y
  189.     if( *data == '\x78' ){
  190.         p_current_elem_db->type = TYPE_x;
  191.     }else if ( *data == '\x79' ){
  192.         p_current_elem_db->type = TYPE_y;
  193.     }else{
  194.         printf("## ERROR 102 ##\n");
  195.         printf("%s\n", data);
  196.         exit(1);
  197.     }
  198.     data+=2;
  199.     func1(&p_current_elem_db->no);
  200.     func1(&p_current_elem_db->num_line_count);
  201.  
  202.     IS_PRINT == true && printf("%c %d %d\n", p_current_elem_db->type == TYPE_x ? 'x' : 'y', p_current_elem_db->no, p_current_elem_db->num_line_count);
  203.     return true;
  204. }
  205. bool CCC::get_elem_qq(){
  206.     read_data();
  207.     // 51=Q,
  208.     if( *data == '\x51' ){
  209.         // テストなのでとりあえずQQ行はスキップ
  210.         if( IS_PRINT == false ){
  211.             for(; *data++ != '\x0a'; );
  212.         }else{
  213.             printf("%c", data[0]);
  214.             for(; *data++ != '\x0a'; ){
  215.                 printf("%c", data[0]);
  216.             }
  217.         }
  218.     }else if( *data == '\x2d' || *data == '\x30' || *data == '\x31' || *data == '\x32' || *data == '\x33' || *data == '\x34' || *data == '\x35' || *data == '\x36' || *data == '\x37' || *data == '\x38' || *data == '\x39' || *data == '\x00'){
  219.         // 2d=マイナス記号(-), 30~39=数字, 00=NULL, の時はスルー
  220.     }else{
  221.         printf("????\n");
  222.         printf("%02x\n", *data);
  223.         exit(1);
  224.     }
  225.     return true;
  226. }
  227. bool CCC::get_elem_x(unsigned int i){
  228.     read_data();
  229.     //係数掛けで数値取得
  230. //    auto func2 = [&](double* num) {
  231.     auto func2 = [&](unsigned int i) {
  232.         uint64_t n = 0;
  233.         bool is_minus = false;
  234.         do {
  235.             // 2d=マイナス記号(-)
  236.             if( *data == '\x2d' ){
  237.                 ++data;
  238.                 is_minus = true;
  239.             }
  240.             unsigned char ch = (unsigned char)(*data++ - '0');
  241.             if (ch <= 9) {
  242.                 n = n * 10 + ch;
  243.             } else {
  244.                 break;
  245.             }
  246.         } while (true);
  247.         double num;
  248.         if( is_minus == false ){
  249.             num = (double)n / (double)read_db.mag;
  250.         }else{
  251.             num = -(double)n / (double)read_db.mag;
  252.         }
  253.         IS_PRINT == true && printf("%ld", lround(num*read_db.mag));
  254.     };
  255.     func2(i); IS_PRINT == true && printf(" ");
  256.     func2(i+1); IS_PRINT == true && printf(" ");
  257.     func2(i+2); IS_PRINT == true && printf(" ");
  258.     func2(i+3); IS_PRINT == true && printf("\n");
  259.     return true;
  260. }
  261.  
  262. bool CCC::get_elem_y(unsigned int i){
  263.     read_data();
  264.     //係数掛けで数値取得
  265. //    auto func2 = [&](double* num) {
  266.     auto func2 = [&](unsigned int i) {
  267.         uint64_t n = 0;
  268.         bool is_minus = false;
  269.         do {
  270.             // 2d=マイナス記号(-)
  271.             if( *data == '\x2d' ){
  272.                 ++data;
  273.                 is_minus = true;
  274.             }
  275.             unsigned char ch = (unsigned char)(*data++ - '0');
  276.             if (ch <= 9) {
  277.                 n = n * 10 + ch;
  278.             } else {
  279.                 break;
  280.             }
  281.         } while (true);
  282.         double num;
  283.         if( is_minus == false ){
  284.             num = (double)n / (double)read_db.mag;
  285.         }else{
  286.             num = -(double)n / (double)read_db.mag;
  287.         }
  288.         IS_PRINT == true && printf("%ld", lround(num*read_db.mag));
  289.     };
  290.     func2(i); IS_PRINT == true && printf(" ");
  291.     func2(i+1); IS_PRINT == true && printf("\n");
  292.  
  293.     return true;
  294. }
  295.  
  296. bool CCC::read(char file_path[]){
  297.     if((fp=fopen(file_path, "rb"))==NULL){
  298.         printf("ファイルを開けません。%s",file_path);
  299.         return 0;
  300.     }
  301.  
  302.     // テスト用。
  303.     if( READ_MODE == 0 ){
  304.         // freadだけだとどれくらいの速度？
  305.         size_t size;
  306.         do{
  307.             if((size = fread( data, sizeof(data[0]), BUF_SIZE, fp)) == 0){
  308.                 break;
  309.             }
  310.             if( size != BUF_SIZE ) data[size] = '\x00';
  311.             IS_PRINT == true && printf("%s", data);
  312.         } while(size == BUF_SIZE);
  313.         return true;
  314.     }else if( READ_MODE == 1 ){
  315.         // read_data()だけだとどれくらいの速度？
  316.         while( true ){
  317.             size_t size = read_data();
  318.             IS_PRINT == true && printf("%s", data);
  319.             for(; *++data != '\x00'; );
  320.             if( feof( fp ) && *data == '\x00' ) { break; }
  321.         }
  322.         return true;
  323.     }
  324.  
  325.     // ファイルコピー
  326.     while( true ){
  327.         t_section_db section_db;
  328.         p_current_section_db = &section_db;
  329.             get_section_name();
  330.             get_section_index();
  331.             for(unsigned int i=0; i<p_current_section_db->text_line_count; ++i){
  332.                 get_text();
  333.             }
  334.             for(unsigned int i=0; i<p_current_section_db->num_1; ++i){
  335.                 t_elem_db elem_db;
  336.                 p_current_elem_db = &elem_db;
  337.                 get_elem_index();
  338.                 get_elem_qq();
  339.                 switch( p_current_elem_db->type ){
  340.                     case TYPE_x:
  341.                         // 要素数を最初に広げる
  342.                         p_current_elem_db->numbers.resize(p_current_elem_db->num_line_count*4);
  343.                         for(unsigned int i=0; i<p_current_elem_db->num_line_count; ++i){
  344.                             get_elem_x(i);
  345.                             }
  346.                         break;
  347.                     case TYPE_y:
  348.                         p_current_elem_db->numbers.resize(p_current_elem_db->num_line_count*2);
  349.                         for(unsigned int i=0; i<p_current_elem_db->num_line_count; ++i){
  350.                             get_elem_y(i);
  351.                         }
  352.                         break;
  353.                     default:
  354.                         printf("## ERROR 101 ##\n");
  355.                         printf("%s\n", data);
  356.                         exit(1);
  357.                 }
  358.             }
  359.           if( feof( fp ) && *data == '\x00' ) { break; }
  360.     }
  361.     fclose( fp );
  362.     return 0;
  363. }
  364.  
  365. int main(void){
  366.     char file_path[] = "./aaa_13gb";
  367.  
  368.     CCC ccc;
  369.     ccc.initialize();
  370.     ccc.read(file_path);
  371.  
  372.     return 0;
  373. }
Success #stdin #stdout 0s 4400KB
comments (?)
stdin
copy 
Standard input is empty
stdout
copy 
ファイルを開けません。./aaa_13gb
https://ideone.com/1YhaLR
language:
C++ (gcc 8.3)
created:
5 years ago
visibility:
public

Share or Embed source code

Discover > Sphere Engine API

The brand new service which powers Ideone!

Discover > IDE Widget

Widget for compiling and running the source code in a web browser!