I have a text file; within the file I have some square brackets that may include a comma. I want to replace/remove this comma. Say the text =

##logissue type=warning;Column 'Employee'[Address, Post]. Do not use floating point data types"

What I want is:

##logissue type=warning;Column 'Employee'[Address Post]. Do not use floating point data types"

I am using this PowerShell script, but this one replacing everything within the bracket squares.

$string -replace '\[.*\]', ''

CodePudding user response：

You can capture everything inside the square brackets and do the replacement with the help of a script block.

Within the script block you can use the $_ automatic variable to access the input text being replaced. Also note that this variable class type is System.Text.RegularExpressions.Match, Hence you need to use .Value to gets the captured substring.

$string -replace "\[(.*?)\]", { $_.Value.replace(",", "") }

Please note that this is available In PowerShell 6 and later.

For older versions you can use the Replace function from .NET regex class.

[regex]::Replace($string, "\[(.*?)\]", { $args[0].Value.replace(',', "") })

CodePudding user response：

Powershell supports infinite quantifiers in a lookbehind assertion and you can assert [ to the left and ] to the right while matching a comma.

(?<=\[[^][]*),(?=[^][]*])

Explanation

• (?<=\[[^][]*) Assert [ to the left matching any char except [ and ] in between
• , Match the comma
• (?=[^][]*]) Assert ] to the right matching any char except [ and ] in between

.NET regex demo

$string = "##logissue type=warning;Column 'Employee'[Address, Post]. Do not use floating point data types"
$string -replace '(?<=\[[^][]*),(?=[^][]*])', ''

Output

##logissue type=warning;Column 'Employee'[Address Post]. Do not use floating point data types