My personal understanding is to use an algorithm to find the repeated data sequence in the file and extract the record position. In this way, only the record position and a sequence are used, thus completing the compression.
I think this is what it means: 1W after 10000 compression. Here is the content to be compressed. I repeated it several times. After compression, bula1, just save a copy of the original content. It has the same meaning as a variable and replace it when decompressing.
This should be the simplest idea. There must be NB's algorithm that is more efficient.
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
If you don’t study it in depth first, then Huffman coding is enough.
One more thing, Huffman coding is taught in data structure courses.
"Information Theory and Coding"
My personal understanding is to use an algorithm to find the repeated data sequence in the file and extract the record position. In this way, only the record position and a sequence are used, thus completing the compression.
I think this is what it means: 1W after 10000 compression. Here is the content to be compressed. I repeated it several times. After compression, bula1, just save a copy of the original content. It has the same meaning as a variable and replace it when decompressing.
This should be the simplest idea. There must be NB's algorithm that is more efficient.