RLE的基本思路是，把数据分两种情况对待：

A1.一些连续的重复字节

A2.一些连续的，不相重复的字节



RLE压缩最常见的一种算法思路：



将全部的数据分成很多块，这些块的长度各不一样:

all data = [block] + [block] + ... + [block]

每一块由两部分顺序组成:

a block = [header] + [data]

其中header部分占2字节16位，这16位中的最高位，标志了这个block的属性，是属于上面的A1还是A2。对应于A1和A2，剩下的15位以及后面的Data部分的意义又分为两种：

A1: block的剩下15位记录重复的次数，取值范围[0,32767]；data段仅含一个字节，即重复的那个字节

A2: block的剩下15位记录data段有多少个字节；data段则是一系列不相重复的字节。

-RLE the basic idea is that the two data points of the deal : A1. number of consecutive repeat byte A2. some continuous, do not duplicate byte RLE compression one of the most common algorithm : all of the data into many pieces, the length of these pieces not the same : all data = [block] [block] ... [block] for every part of a two-shun sequence components : a block = [header] [data] header which accounted for some two-byte 16, This 16-the highest mark of the attributes of the block, is above the A1 or A2. Corresponding to the A1 and A2, and the remaining 15 and the back part of the Data significance will be divided into two : A1 : block the remaining 15 the number of duplicate records, the value in the range [0,32767]; with only one of the data byte, which repeat the byte A2 : block the rem