awk没有选择字符串和管道作为分隔符

Question

I have a file where each record has a word "TST_BI|" and I need to use as a delimiter and populate the value after this string to a file . There is only one occurrence of this string in each record.

It is working fine in AIX environment with below command. awk -F "TST_BI|" '{print $2}' file.txt awk -F "TST_BI|" '{print $2}' file.txt .

But when I migrated the code to Linux and tried the same, command is NOT working, where the value "|" is also getting populated. Below are the output from both AIX and Linux

Input :

<14>1 2017-08-31T04:13:47.2345839+00:00 loggregator ecsdasc0985-cs656-4asdsds-asds-asdasg6ds73 [DEV/2] - - TST_BI|DATE~2017-08-31 04:13:47,095|TIMESTAMP~04:13:47|TEST_ID~biTestExecutor-2|COUNTRY_CODE~XX|GROUP_TESTS~BZAG

OutPut from AIX :

DATE~2017-08-31 04:13:47,095|TIMESTAMP~04:13:47|TEST_ID~biTestExecutor-2|COUNTRY_CODE~XX|GROUP_TESTS~BZAG

With same command, Linux Output is

|DATE~2017-08-31 04:13:47,095|TIMESTAMP~04:13:47|TEST_ID~biTestExecutor-2|COUNTRY_CODE~XX|GROUP_TESTS~BZAG

A pipe is getting populated and which is not treating as delimiter.

Can anyone please help?

Answer 1

vertical bar char | specified with adjacent characters is treated as regex alternation operator/alternation group. In such case, to treat | literally - it should be escaped \\\\| or put into a character class [|] :

awk -F'TST_BI[|]' '{print $2}' file.txt

The output:

DATE~2017-08-31 04:13:47,095|TIMESTAMP~04:13:47|TEST_ID~biTestExecutor-2|COUNTRY_CODE~XX|GROUP_TESTS~BZAG