使用Powershell从CSV的列中的字符中删除字符串

Question

I have a CSV file containing two columns:server name with domain and date

servername.domain.domain.com,10/15/2018 6:28
servername1.domain.domain.com,10/13/2018 7:28

I need to remove the fully qualified name so it only has the shortname and I need to keep the second column so it looks as is like below either by sending to a new CSV or somehow removing the domain inplace somehow. Basically I want the second column untouched but I need it to be included when creating a new CSV with the altered column 1.

servername,10/15/2018 6:28
servername1,10/13/2018 7:28

I have this:

Import-Csv "filename.csv" -Header b1,b2 |
    % {$_.b1.Split('.')[0]} |
    Set-Content "filename1.csv"

This works great, but the problem is the new CSV is missing the 2nd column. I need to send the second column to the new CSV file as well.

Answer 1

Use a calculated property to replace the property you want changed, but leave everything else untouched:

Import-Csv 'input.csv' -Header 'b1', 'b2' |
    Select-Object -Property @{n='b1';e={$_.b1.Split('.')[0]}}, * -Exclude b1 |
    Export-Csv 'output.csv' -NoType

Note that you only need to use the parameter -Header if your CSV data doesn't already have a header line. Otherwise you should remove the parameter.

If your input file doesn't have headers and you want to create the output file also without headers you can't use Export-Csv , though. Use ConvertTo-Csv to create the CSV text output, then skip over the first line (to remove the headers) and write the rest to the output file with Set-Content .

Import-Csv 'input.csv' -Header 'b1', 'b2' |
    Select-Object -Property @{n='b1';e={$_.b1.Split('.')[0]}}, * -Exclude b1 |
    ConvertTo-Csv -NoType |
    Select-Object -Skip 1 |
    Set-Content 'output.csv'