文件压缩系统 [英] file compression system

问题描述

数据压缩pro

data compression pro

#include<stdio.h>
#include<conio.h>
#include<string.h>
#include<iostream.h>
void fsize(FILE *fp);
void fcal(FILE *fp);
void fa_z(FILE *fp);
char r;
long char c;
void main()
{


    FILE *fp;
    clrscr();

    fp=fopen("file name.txt","r");
    fcal(fp);
    rewind(fp);
    fsize(fp);
    rewind(fp);
    fa_z(fp);
    fclose(fp);
    getch();
}

void fsize(FILE *fp)/*this calculate file size*/
{
    for(;fscanf(fp,"%c",&r)!=EOF;);
    long size;
    size=ftell(fp);
    printf("\n\n the size of file=%d\n",size);
}

void fcal(FILE *fp)/*this calculate character,word,line*/
{
    int ch=0,word=1,line=1;
    for(;fscanf(fp,"%c",&r)!=EOF;)
    {
        ch++;
        printf("%c",r);
        c=r;
        if(c=='\n'||c=='\t'||c==' ')
        ch--;
        if(c=='\n')
        line++;
        if(c=='\n'||c=='\t'||c==' '||c=='.'||c==',')
        word++;

    }
    printf("\n\n total character=%d",ch);
    printf("\n\n total word =%d",word);
    printf("\n\n total line=%d",line);

}
void fa_z(FILE *fp)/*this cal how many a-z*/
{
    char a[27];
    int i;
    int n[26];
    for(i=0;i<26;i++)
    {
        n[i]=0;
    }
    while((c=fgetc(fp))!=EOF)
    {
        if(c=='a')
        n[0]=n[0]+1;
        else if(c=='b')
        n[1]=n[1]+1;
        else if(c=='c')
        n[2]=n[2]+1;
        else if(c=='d')
        n[3]=n[3]+1;
        else if(c=='e')
        n[4]=n[4]+1;
        else if(c=='f')
        n[5]=n[5]+1;
        else if(c=='g')
        n[6]=n[6]+1;
        else if(c=='h')
        n[7]=n[7]+1;
        else if(c=='i')
        n[8]=n[8]+1;
        else if(c=='j')
        n[9]=n[9]+1;
        else if(c=='k')
        n[10]=n[10]+1;
        else if(c=='l')
        n[11]=n[11]+1;
        else if(c=='m')
        n[12]=n[12]+1;
        else if(c=='n')
        n[13]=n[13]+1;
        else if(c=='o')
        n[14]=n[14]+1;
        else if(c=='p')
        n[15]=n[15]+1;
        else if(c=='q')
        n[16]=n[16]+1;
        else if(c=='r')
        n[17]=n[17]+1;
        else if(c=='s')
        n[18]=n[18]+1;
        else if(c=='t')
        n[19]=n[19]+1;
        else if(c=='u')
        n[20]=n[20]+1;
        else if(c=='v')
        n[21]=n[21]+1;
        else if(c=='w')
        n[22]=n[22]+1;
        else if(c=='x')
        n[23]=n[23]+1;
        else if(c=='y')
        n[24]=n[24]+1;
        else if(c=='z')
        n[25]=n[25]+1;
    }
    char j;
    printf("\n");
    for(i=0,j='a';i<26,j<='z';i++,j++)
    {
        printf("%c=%d\t",j,n[i]);
    }
}

程序告诉我们字符行和文件大小

并且还显示程序中使用了多少个az字

但这不能减少文件大小的损失..



如何更好地解决这个问题...

that program tell us character word line and size of file
and also show that how many a-z word use in program
but this couldn't be loss less the file size..

how to better work on this...

推荐答案

1）在 void fa_z（FILE * fp）中替换你的if-else-cascade / code>带有switch语句。

2）用 void fa_z（FILE * fp）替换switch语句数组或字典 * 并且完全没有区别。

3）对<$ c $的结果进行排序c> void fa_z（FILE * fp）按频率。

4）根据频率结果构建一个huffman-tree。

5）逐字节迭代输入文件，并通过将输入字节替换为将霍夫曼树移动到相应的输入字节值所产生的位表示来构建压缩输出流。

6）随机保存霍夫曼树h压缩的输出流在文件中。

7）构建解压缩功能。



编辑： < br $>


* ：并使用输入字节的值作为数组 - index /作为字典键而不是为所有不同的输入字节值手动分配索引。



点1 + 2的重复替换建议不是意思是认真的但是，每当谓词适用于switch语句时，你应该使用switch语句（不是在这种情况下！）而不是很长的if-else if语句。



进一步阅读：

http://en.wikipedia.org/wiki/Huffman_coding [ ^ ]

https://www.siggraph.org/education/materials/HyperGraph/video/mpeg/mpegfaq/huffman_tutorial .html [ ^ ]



另一个编辑：

解决方案2演示了我在上面的注释中的意思标有 ^*

1) replace your if-else-cascade in void fa_z(FILE *fp) with a switch-statement.
2) replace the switch-statement in void fa_z(FILE *fp) with an array or dictionary^* and no case-distinction at all.
3) sort the result of void fa_z(FILE *fp) by frequency.
4) build a huffman-tree from the frequency-results.
5) iterate the input-file byte-by-byte and build a compressed outputstream by replacing the input-bytes by their bit-representation that results from walking the huffman-tree to the according input-byte-value.
6) save the huffman-tree along with the compressed outputstream in a file.
7) build the decompression-function.



^* : and use the value of the input-byte as the array-index / as the dictionary-key instead of having manually assigned indices for all different input-byte-values.

The duplicate replacement suggestion of points 1+2 isn't meant seriously. But you should use switch-statements (not in this case!) instead of very long if-else if-statements whenever the predicate is suitable for the switch-statement.

Further reading:
http://en.wikipedia.org/wiki/Huffman_coding[^]
https://www.siggraph.org/education/materials/HyperGraph/video/mpeg/mpegfaq/huffman_tutorial.html[^]

Another
Solution 2 demonstrates what I meant by the note above marked with ^*

你可以从替换while循环i开始n

You can start with replacing the while loop in

void fa_z(FILE *fp)

with

with

int index = 0;
while ((c = fgetc(fp)) != EOF)
{
    if (c >= 'a' && c <= 'z')
    {
        index = (int)(c - 'a');
        n[index]++;
    }
}

假设 c 是一个 int 并且数组 n 已初始化为0。

Assuming c is an int and the array n has been initialized to 0.

这篇关于文件压缩系统的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！