Home/ Questions/Q 9252655
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-18T11:00:06+00:00

I have a ~ delimited text file with about 20 nullable columns. I am

  • 0

I have a ~ delimited text file with about 20 nullable columns.

I am trying to use SED (from cygwin) to “blank out” the value in column 11 if the following conditions are met…

  1. Column 3 is a zero (0)

  2. Column 11 is in date format mm/dd/yy (I’m not really concerned if it’s a valid date)

Here’s what I’m trying…

s/\([^~]*~[^~]*~0~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~\)\(\d{2}\/\d{2}\/\d{2}~\)\(.*$\)/\1~\3/

Here’s a sample from the file:

Test A~7~1~~~~72742050~~~Z370~10/25/11~~~0~8.58563698~6.40910452~4.59198764~3.18239469~1.72955975~.23345372~-1.30891113~-2.89971394~1~0
Test B~7~0~~~~72742060~~~Z351~05/15/12~05/14/12~~0~18.88910518~12.69425528~9.96182381~6.76077612~6.76077612~3.86279298~.22449489~-.91021010~0~0
Test C~7~0~~~~72742060~~~Z352~06/12/12~ABC~~0~20.60845679~17.54889351~15.52912556~12.43279217~12.43279217~10.32033576~9.35296144~8.09245899~0~0

…and here’s what I expect to get back

Test A~7~1~~~~72742050~~~Z370~10/25/11~~~0~8.58563698~6.40910452~4.59198764~3.18239469~1.72955975~.23345372~-1.30891113~-2.89971394~1~0
Test B~7~0~~~~72742060~~~Z351~05/15/12~~~0~18.88910518~12.69425528~9.96182381~6.76077612~6.76077612~3.86279298~.22449489~-.91021010~0~0
Test C~7~0~~~~72742060~~~Z352~06/12/12~ABC~~0~20.60845679~17.54889351~15.52912556~12.43279217~12.43279217~10.32033576~9.35296144~8.09245899~0~0

but the file comes through with line 2 completely unchanged.

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    You are trying to replace column 12 instead of 11:

    \([^~]*~[^~]*~0~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~\)\(\d{2}\/\d{2}\/\d{2}~\)\(.*$\)
    1     2   3   4     5     6     7     8     9    10    11              12

    If just removing one of the [^~]*~ from the end of the first group doesn’t fix it, it could be because your version of sed doesn’t support either \d or repetition with {2} (although escaping the curly brackets would probably fix that).

    Here is a version that should work everywhere which replaces each \d{2} with [0-9][0-9] (and fixes the incorrect column issue mentioned above):

    s/\([^~]*~[^~]*~0~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~[^~]*~\)\([0-9][0-9]\/[0-9][0-9]\/[0-9][0-9]~\)\(.*$\)/\1~\3/
    • 0