bndit
asked on
Powershell regex
Hello,
I'm trying to understand regular expressions in powershell, but seem to be having a hard time. I have a CSV file and I need to extract a value (IP address) from one of the columns. The column looks like this:
kdajfakldjfa.en.jlkjdaflsa
kdajfakldjfa.en.jlkjdaflsa
kdajfakldjfa.en.jlkjdaflsa
kdajfakldjfa.en.jlkjdaflsa
Obviously IP address are gonna differ so my regex filter would say find the first "[" and extract everything inside until you find the closing bracket "]"
I've tried doing this, but not getting me anywhere
$ips = import-csv c:\myfile.csv
foreach ($line in $lines){
$line = regex::($line,"^\]")
}
can't seem to figure out how to say "extract everything until you find the next bracket"
I'm trying to understand regular expressions in powershell, but seem to be having a hard time. I have a CSV file and I need to extract a value (IP address) from one of the columns. The column looks like this:
kdajfakldjfa.en.jlkjdaflsa
kdajfakldjfa.en.jlkjdaflsa
kdajfakldjfa.en.jlkjdaflsa
kdajfakldjfa.en.jlkjdaflsa
Obviously IP address are gonna differ so my regex filter would say find the first "[" and extract everything inside until you find the closing bracket "]"
I've tried doing this, but not getting me anywhere
$ips = import-csv c:\myfile.csv
foreach ($line in $lines){
$line = regex::($line,"^\]")
}
can't seem to figure out how to say "extract everything until you find the next bracket"
Sorry, make that $matches[1]
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Explanation: -match works in the way that it applies the right-hand RegExp against the left-hand value. Anything grouped (in parens) in the RegExp will result into an own group in the $matches array, with $matches[0] being the whole string.
The RegExp used will catch all contents between the first and the last (!) square bracket - default mode for RegExp is "lazy", which means the widest fit. You will need to make sure you do not have more than one square bracket'd value per line in your CSV file, but I reckon that is ok.
The RegExp used will catch all contents between the first and the last (!) square bracket - default mode for RegExp is "lazy", which means the widest fit. You will need to make sure you do not have more than one square bracket'd value per line in your CSV file, but I reckon that is ok.
ASKER
Ok, so that works...now I'm trying to send the output to a text file but it's lumping everything together as opposed to placing each IP address on its own line such as:
192.168.0.2
192.168.0.3
my current output:
192.168.0.2192.168.0.3
192.168.0.2
192.168.0.3
my current output:
192.168.0.2192.168.0.3
$lines = Import-Csv C:\file.csv
$ips = @()
foreach($line in $lines){
If ($line -match "\[(.+)\]") {$ips += $matches[1]}
}
$ips | Set-Content c:\cleanips.txtASKER
Nevermind...it was a simple fix
If ($line -match "\[(.+)\]") {$ips += $matches[1] + "`\n"}
If ($line -match "\[(.+)\]") {$ips += $matches[1] + "`\n"}
ASKER
Spoke too soon.....it's "`n" instead of "`\n" and it only works when I run the script from within the PowerGui editor and send the output to the screen...but not when the output is sent to a text file...any ideas why?
ASKER
Strange....added "`r`n" and it worked when I sent the output to a text file.....why would `n not work?
Windows uses a line feed and carriage return at the end of each line, whereas linux systems just use a carriage return - sometimes there is compatibility between the 2 methods, but it often causes trouble.
Strange, because you should not need to insert newline and/or carriage return when outputting array content. It works for me in the plain PowerShell prompt:
$ip = @()
$ip += "192.168.1.1"
$ip += "192.168.1.2"
$ip | set-content c:\cleanips.txtget-content C:\file.csv | % {if ($_ -match "\[(.+)\]") {$matches[1]} } | Set-Content c:\cleanips.txt
My help was not worth some points?
$line -match "\[(.+)\]"
$ip = $matches[0]