[20230428]bash实现xor计算.txt

[20230428]bash实现xor计算.txt --//昨天使用自己以前写的脚本做xor计算,仔细看了自己以前写的源代码,这样写主要问题调用bc次数太多,计算效率很低. --//http://blog.itpub.net/267265/viewspace-2134945/ => [20170308]bc做xor(异或)计算.txt $ cat xor.sh #! /bin/bash # just play , calc xor!! s='0' for i in $(cat $1| tr 'a-f' 'A-F') do     #echo "obase=16;ibase=16; xor($s,$i)"     echo $i         [ $i != '0000' ] &&  s=$(echo "obase=16;ibase=16; xor($s,$i)" | bc -l ~/bc/logic.bc) done echo -e "\nxor result: $s \n" --//理论讲bash 应该支持xor,以前的学习不仔细,仔细看一些文档.发现bash本身就支持xor的计算,例子如下: $ echo $(( 0x5 ^ 0x9 )) 12 --//自己尝试修改看看. $ cat  xor2.sh #! /bin/bash # just play , calc xor!! s='0' for i in $(cat $1| tr 'a-f' 'A-F') do     #echo "obase=16;ibase=16; xor($s,$i)"     echo $i     #[ $i != '0000' ] &&  s=$(echo "obase=16;ibase=16; xor($s,$i)" | bc -l ~/bc/logic.bc)     [ $i != '0000' ] &&  s=$((0x$s ^ 0x$i )) && s=$(printf '%04x\n' $s) done echo -e "\nxor result: $s \n" --//测试看看: 1.环境: SCOTT@book> @ ver1 PORT_STRING                    VERSION        BANNER ------------------------------ -------------- -------------------------------------------------------------------------------- x86_64/Linux 2.4.xx            11.2.0.4.0     Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production 2.测试: SCOTT@book> select rowid from dept where rownum=1; ROWID ------------------ AAAVRCAAEAAAACHAAA SCOTT@book> @ rowid AAAVRCAAEAAAACHAAA     OBJECT       FILE      BLOCK        ROW ROWID_DBA            DBA                  TEXT ---------- ---------- ---------- ---------- -------------------- -------------------- ----------------------------------------      87106          4        135          0  0x1000087           4,135                alter system dump datafile 4 block 135 ; SCOTT@book> @ bbvi 4 135 BVI_COMMAND ---------------------------------------------------------------------------------------------------- bvi -b 1105920 -s 8192 /mnt/ramdisk/book/users01.dbf xxd -c16 -g 2 -s 1105920 -l 8192 /mnt/ramdisk/book/users01.dbf dd if=/mnt/ramdisk/book/users01.dbf bs=8192 skip=135 count=1 of=4_135.dd conv=notrunc 2>/dev/null od -j 1105920 -N 8192 -t x1 -v /mnt/ramdisk/book/users01.dbf hexdump -s 1105920 -n 8192 -C -v /mnt/ramdisk/book/users01.dbf alter system dump datafile '/mnt/ramdisk/book/users01.dbf' block 135; alter session set events 'immediate trace name set_tsn_p1 level 5'; alter session set events 'immediate trace name buffer level 16777351'; 9 rows selected. $ time xxd -c16 -g 2 -s 1105920 -l 8192 /mnt/ramdisk/book/users01.dbf | cut -c10-48| xor.sh | grep result xor result: 0 real    0m18.801s user    0m9.147s sys     0m8.431s --//使用旧的xor.sh脚本计算需要18秒. $ time xxd -c16 -g 2 -s 1105920 -l 8192 /mnt/ramdisk/book/users01.dbf | cut -c10-48| xor2.sh | grep result xor result: 0000 real    0m3.853s user    0m0.539s sys     0m1.258s --//而新的xor2.sh脚本计算需要4秒完成,快了不少.