Linux处理文本时,有些文本是有固定的列,可以很方便地使用awk或者cut获取某一列,但在某些情况下,文本可能没有规律可言。比如某数据库的log.txt中,每访问一次表,就会记录一条信息"TableId = x",其中x为1-99999之间的数字,但其他部分的文本没有任何规律。我们需要统计出每个表访问的次数,并按照访问次数排序:
下面是一段文本示例:
qingsong@db2a:~$ cat log.txt
Hello, world, TableId = 19039 jl
This is a long story TableId = 1909 jlsfd
TableId = 39 sdlf where
PID : 28703 TID : TableId =
he TableId = 9 EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, Self tuning memory manager, stmm
### TableId = 29099 ::: ::
SSAGE : ADM7513W TableId = 1 SSAGE : ADM7513W
Are you TableId = 9
2017-08-16-23.36.17.920719 Hello, world, TableId = 29099 xxl
Hello, world, TableId = 19039 jl
This is a long story TableId = 1909 jlsfd
TableId = 393 sdlf where
PID : 2 TableId = 8703 TID :
he TableId = 19 EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, Self tuning memory manager, stmm
anager, stmm ### TableId = 29099 ::: ::
Sanager, stmmSAGE : ADM7513W TableId = 1 SSAGE : ADM7513W
Are you TableId = 9 anager, stmm
719 Hello, world, TableId = 29099 xxl2017-08-16-23.36.1
可以使用grep命令的-o选项,只打印匹配部分
-o, --only-matching
Print only the matched (non-empty) parts of a matching line, with each such part on a separate output line.
qingsong@db2a:~$ grep -o 'TableId = [1-9][0-9]\{0,4\}\>' log.txt
TableId = 19039
TableId = 1909
TableId = 39
TableId = 9
TableId = 29099
TableId = 1
TableId = 9
TableId = 29099
TableId = 19039
TableId = 1909
TableId = 393
TableId = 8703
TableId = 19
TableId = 29099
TableId = 1
TableId = 9
TableId = 29099
qingsong@db2a:~$ grep -o 'TableId = [1-9][0-9]\{0,4\}\>' log.txt | cut -d ' ' -f 3 | sort -n | uniq -c | sort -nr
4 29099
3 9
2 1909
2 19039
2 1
1 8703
1 393
1 39
1 19
另外,除了grep之外,应该还有其他命令,比如sed能完成这个任务。
下面是一段文本示例:
qingsong@db2a:~$ cat log.txt
Hello, world, TableId = 19039 jl
This is a long story TableId = 1909 jlsfd
TableId = 39 sdlf where
PID : 28703 TID : TableId =
he TableId = 9 EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, Self tuning memory manager, stmm
### TableId = 29099 ::: ::
SSAGE : ADM7513W TableId = 1 SSAGE : ADM7513W
Are you TableId = 9
2017-08-16-23.36.17.920719 Hello, world, TableId = 29099 xxl
Hello, world, TableId = 19039 jl
This is a long story TableId = 1909 jlsfd
TableId = 393 sdlf where
PID : 2 TableId = 8703 TID :
he TableId = 19 EDUNAME: db2sysc 0
FUNCTION: DB2 UDB, Self tuning memory manager, stmm
anager, stmm ### TableId = 29099 ::: ::
Sanager, stmmSAGE : ADM7513W TableId = 1 SSAGE : ADM7513W
Are you TableId = 9 anager, stmm
719 Hello, world, TableId = 29099 xxl2017-08-16-23.36.1
可以使用grep命令的-o选项,只打印匹配部分
-o, --only-matching
Print only the matched (non-empty) parts of a matching line, with each such part on a separate output line.
qingsong@db2a:~$ grep -o 'TableId = [1-9][0-9]\{0,4\}\>' log.txt
TableId = 19039
TableId = 1909
TableId = 39
TableId = 9
TableId = 29099
TableId = 1
TableId = 9
TableId = 29099
TableId = 19039
TableId = 1909
TableId = 393
TableId = 8703
TableId = 19
TableId = 29099
TableId = 1
TableId = 9
TableId = 29099
qingsong@db2a:~$ grep -o 'TableId = [1-9][0-9]\{0,4\}\>' log.txt | cut -d ' ' -f 3 | sort -n | uniq -c | sort -nr
4 29099
3 9
2 1909
2 19039
2 1
1 8703
1 393
1 39
1 19
另外,除了grep之外,应该还有其他命令,比如sed能完成这个任务。