I am trying to find a simple way to change multiple blank lines into one.  I've got multiple ways of deleting all blank lines, but I can't see a way to do this without writing code to set a switch on blank line and delete everything till next non-blank line, reset the switch and etc.  Testing for "\n\n" never matches anything, at least using get-content.

kct

that is where I started to get rid of the blank lines, but decided a little white space wouldn't hurt and found the example in the PS Cookbook for multiple line searches.  However that left me with the requirement to delete any trailing blank lines.  It looks so simple, IF ($file[-1] = "\s*") then $file.length = $file.length-1, only it does not work.  I understand what the scripting guys did in vbscript, just drop the EOF marker before the CRLF and the blank line is gone.  What I would really like to understand is how to get line 53 removed from an array of 100.  If I can identify the element I should be able to remove it and close up ranks.  I guess part of the problem is I'm trying to find a "neat" way of doing this in PowerShell rather than something along the lines of  for x = 1 to $fileLength - 1 ; $newfile[x] = $file[x].

Another thing I seemed to have found, but not understood, is that an array object and an arraylist object are different, or have different properties.  I had been trying to figure out remove and removerange, but never got anything working.

Thanks for asking

kct

See the show notes for my podcast ep 23.  First item under the Tips section.

I ended up going down the lines of solution #3 and never quite getting it to work.  I had thought I had already tried the first solution as well, but my guess is that I did the test using "" instead of $null.  One of the joys of mulitple languages is that you use the wrong syntax, NULL, and then move on to something else while muttering how stupid the language is.  Then PS books have index of $null, not NULL so once you know what to do, it is all obvious. 

I had to enhance the solution once it worked by marching backwards through multiple lines.  Today's fight has been with match and contains working with "\s" whether * neccesary and why empty line returned false to both \s and \S.  Are those not opposites?

    foreach ($OrigFile in Get-ChildItem -name) 
    {
    $filename = Get-Item $OrigFile
    $singleLine = [System.IO.File]::ReadAllText($filename.fullname)
    $singleLine = $singleLine -creplace "(\r\n){3,}", "`r`n`r`n"
    $TempName = [System.IO.Path]::GetRandomFileName()
    $singleLine | out-file -filepath $TempName
    $NewFile = get-content $TempName
    $loop = -1
    $answer = Read-Host "processing $newFile"
    while ($loop -gt -10)
        {
        if ($NewFile[$loop] -notmatch "\s") {
            $NewFile[$loop] = $null
            }
        else {
            Write-Host "Not empty"
            }
        write-host "$loop `t $NewFile[$loop]"
        $loop = $loop - 1
        }
    $ProcName = $OrigFile -replace ("`.prc", "")
    $NewFile = $NewFile | 
            foreach {$_ -replace ( "sp_", "spWW_Conv_")
        } | foreach {$_ -replace ( "QUOTED_IDENTIFIER OFF", "QUOTED_IDENTIFIER ON")
        } | foreach {$_ -replace ( 'numeric\(9,0\)', "dbo.seq_nbr")
        } | foreach {$_ -replace ( "@clnt_nbr", "dbo.clnt_nbr")
        } | foreach {$_ -replace ( "\bchar\b", "nchar")        
        } | foreach {$_ -replace ( "\bvarchar\b", "nvarchar")
        } | foreach {$_ -replace ( "\bas\b", "$CommentBlock")
        } | foreach {$_ -replace ("\t", ( " " * 4 ))
        }
    $ConvertedName = $OrigFile -replace ( "sp_", "spWW_GApps_")
    $NewFile | out-file -filepath $ConvertedName
    }

Option 1 - Set the item to $null.  This does not actually remove the item, but for most purposes it serves well. $servers[3] = $null

$null seems to work on display, but outputing text file has all the lines set to null back again.

Option 2 - Create a new collection which is a subset of the first.  Drawback here is double the memory as the collection is copied in place.   $servers = $servers -ne “itemthatyouwantremoved”

Problem here is that I want multiple blank lines changed to one blank line and then remove all blank lines from end of file.

Option 3 - Use system.collection.arraylist instead of a generic array.  More steps, but the item or items are removed, and it much more efficient than option 2.  The Scripting Guys explain it well in one of their PowerShell tips of the week. $servers = new-object system.collection.arraylist; $servers.Remove(”item”)

Created the arraylist, but then get-content converted it back to system.object which did not have a remove method.  removerange would be preferred as I need to distinguish by place in file.

As a side issue, how would I read a line and then write the line to a different array?  Will $NewArray = $_ or $NewArray.add($_) work in conjunction of test for current line is blank, last line was blank, skip, else copy line.