sed文本处理工具------切换域

[root@station1 ~]# cat data

#[ERROR] [fxmon] [path] attr {vscsi_err_recov} should be , current is

#[ERROR] [fxmon] [path] attr {vscsi_err_recov} should be [fast_fail], current is

#[ERROR] [fxmon] [path] [backing-device] <${_bd}> is an unknow type disk, please check"

#[ERROR] [dwn2] [path] attr {vscsi_err_recov} should be , current is

#[ERROR] [dwn2] [path] attr {vscsi_path_to} should be <30>, current is <0>

#[ERROR] [dwn2] [disk] attr {hcheck_interval} should be <60>, current is <0>

这是文本处理结果：

fxmon|error|path|vscsi0| attr {vscsi_err_recov} should be , current is

fxmon|error|path|vscsi0| attr {vscsi_err_recov} should be [fast_fail], current is

fxmon|error|path|vscsi0| [backing-device] <${_bd}> is an unknow type disk, please check"

这个问题是用sed比是用awk要简单的多，因为其中很多的[],而且不止前三个字段中包含[],后面的数据也有。所以想到是用sed去处理，代码如下：

 sed -e 's/^#\[\([^[]*\)\] \[\([^[]*\)\] \[\([^[]*\)\] <\(.\{,10\}\)>/\2|\L\1\E|\3|\4|/' data

perl的代码如下

[root@station1 ~]# cat data | perl -ne 's/^#\[(.*)\] \[(.*)\] \[(.*)\] <(.*?[0-9])>/\2|\L\1\E|\3|\4|/;print '

fxmon|error|path|vscsi0| attr {vscsi_err_recov} should be , current is

fxmon|error|path|vscsi0| attr {vscsi_err_recov} should be [fast_fail], current is

fxmon|error|path|vscsi0| [backing-device] <${_bd}> is an unknow type disk, please check"

dwn2|error|path|vscsi1| attr {vscsi_err_recov} should be , current is

dwn2|error|path|vscsi1| attr {vscsi_path_to} should be <30>, current is <0>

dwn2|error|disk|hdisk0| attr {hcheck_interval} should be <60>, current is <0>