Grabbing specific sections of a txt file via Powershell - powershell

Question

I am new to Powershell scripting, but I feel I am overlooking a simple answer, hopefully some of you can help.
My company exports files from all of our computers with a section around the middle of Mapped Network Printers. It looks like this:
-------------------------------------------------------------------------
Mapped Network Printers:
NetworkAddress\HP425DN [DEFAULT PRINTER]
-------------------------------------------------------------------------
Local Printers:
What I have been asked to do is copy just the Mapped Network Printers to a new text file.
I tried using Select-String with a context parameter, but I have no way of knowing how many network printers there are, so I can't guess.
I also tried using the following code which I found on this site, but it returns nothing:
$MapPrint = gc C:\Users\User1\Documents\Config.txt
$from = ($MapPrint | Select-String -pattern "Mapped Network Printers:" |
Select-Object LineNumber).LineNumber
$to = ($MapPrint | Select-String -pattern "-------------------------------
--------------------------------------------" | Select-Object
LineNumber).LineNumber
$i = 0
$array = #()
foreach ($line in $MapPrint)
{
foreach-object { $i++ }
if (($i -gt $from) -and ($i -lt $to))
{
$array += $line
}
}
$array
I basically want to start the search at "Mapped Network Printers" and end it at the next row of "------"
Any help would be greatly appreciated.

Answer 1

Select-String has no feature for extracting a range of lines based on content.
The simplest approach is to read the file as a whole and use the -replace operator to extract the range via a regular expression (regex):
$file = 'C:\Users\User1\Documents\Config.txt'
$regex = '(?sm).*^Mapped Network Printers:\r?\n(.*?)\r?\n---------------------.*'
(Get-Content -Raw $file) -replace $regex, '$1'
Reading an input file as a whole can be problematic with files too large to fit into memory, but that's probably not a concern for you.
On the plus side, this approach is much faster than processing the lines in a loop.
Get-Content -Raw (PSv3+) reads the input file as a whole.
Inline regex options (?sm) turn on both the multi-line and the single-line option:
m means that ^ and $ match the start and end of each line rather than the input string as a whole.
s means that metacharacter . matches \n characters too, so that an expression such as .* can be used to match across lines.
\r?\n matches a single line break, both the CRLF and the LF variety.
(.*?) is the capture group that (non-greedily) captures everything between the bounding lines.
Note that the regex matches the entire input string, and then replaces it with just the substring (range) of interest, captured in the 1st (and only) capture group ($1).
Assuming that $file contains:
-------------------------------------------------------------------------
Mapped Network Printers:
NetworkAddress\HP425DN [DEFAULT PRINTER]
NetworkAddress\HP426DN
-------------------------------------------------------------------------
Local Printers:
the above yields:
NetworkAddress\HP425DN [DEFAULT PRINTER]
NetworkAddress\HP426DN

Answer 2

You could use Select-String or Where-Object to look for words with a \. Taking that even further you could look for just the server\printer values with a RegEx like this:
Get-Content C:\Users\User1\Documents\Config.txt -Raw |
Select-String '[A-Z0-9]+\\[A-Z0-9]+' -AllMatches |
ForEach-Object {$_.Matches.Value}
Note that this makes the assumption the Server Names and Printers use only A-Z and 0-9, you may need to look for more characters if that is not a valid assumption.
Here would be an example of using Where-Object to filter for lines with \
Get-Content 'C:\Users\User1\Documents\Config.txt' | Where-Object {$_ -like '*\*'}

Answer 3

$Doc= "C:\temp\test.txt"
$Doc_end ="C:\temp\testfiltered.txt"
$reader = [System.IO.File]::OpenText($Doc)
$cdata=""
while($null -ne ($line = $reader.ReadLine()))
{
if ($line -like ('---*') ) {$Read = 0 }
if ($Read -eq 1) {$cdata+= $line + "`r`n"}
if ($line -like ('Mapped Network Printers:*')) {$Read = 1}
}
$cdata | Out-File $Doc_end -Force

Answer 4

You can do what you are attempting with the foreach-object command and a few additional test conditions. Simply setting a flag when you encounter the Mapped Network Printers: line and then terminating output on the next line -like "---*" will work, e.g.
## positional parameters
param(
[Parameter(Mandatory=$true)][string]$infile
)
$beginprn = 0
get-content $infile | foreach-object {
# terminate condition
if ([int]$beginprn -eq 1 -and $_ -like "---*") {
break
}
# output Mapped printers
if ([int]$beginprn -eq 1) {
write-host $_
}
# begin condition
if ($_ -eq "Mapped Network Printers:") {
$beginprn = 1
}
}
Example Input File
-------------------------------------------------------------------------
Mapped Network Printers:
NetworkAddress\HP425DN [DEFAULT PRINTER]
NetworkAddress\HP4100N
-------------------------------------------------------------------------
Local Printers:
Example Use/Output
PS> parseprn.ps1 .\tmp\prnfile.txt
NetworkAddress\HP425DN [DEFAULT PRINTER]
NetworkAddress\HP4100N