I'm running into an issue where a removal and reinstallation of a particular application is creating duplicate entries in System Variables under Path. So for example, it's showing C:\Apps\folder;C:\Apps\folder;typical entries.
While this is not causing a problem with functionality of the app, I'd prefer to not have the entry in there twice (or more if it requires an additional removal/installation). I want to automate something so I don't have to go into each system and manually remove one of those entries.
Can this be done through either a batch file or a PowerShell script? I'm not able to find a way, but hopefully someone here will know a way. It's fine if the method removes both entries, as I can add something to the script to add one of them back. One important note, I need to make sure everything else under Path is left intact.
Here is a script (taken from a Microsoft repository) that I used to do the exact same thing.
$RegKey = ([Microsoft.Win32.Registry]::LocalMachine).OpenSubKey("SYSTEM\CurrentControlSet\Control\Session Manager\Environment", $True)
$PathValue = $RegKey.GetValue("Path", $Null, "DoNotExpandEnvironmentNames")
Write-host "Original path :" + $PathValue
$PathValues = $PathValue.Split(";", [System.StringSplitOptions]::RemoveEmptyEntries)
$IsDuplicate = $False
$NewValues = #()
ForEach ($Value in $PathValues)
{
if ($NewValues -notcontains $Value)
{
$NewValues += $Value
}
else
{
$IsDuplicate = $True
}
}
if ($IsDuplicate)
{
$NewValue = $NewValues -join ";"
$RegKey.SetValue("Path", $NewValue, [Microsoft.Win32.RegistryValueKind]::ExpandString)
Write-Host "Duplicate PATH entry found and new PATH built removing all duplicates. New Path :" + $NewValue
}
else
{
Write-Host "No Duplicate PATH entries found. The PATH will remain the same."
}
$RegKey.Close()
Related
I've got into a bit of a rut with this and I'm hoping someone can help me nail it:
I've been tasked with dropping a custom settings file into the roaming profile of the logged on user of all workstations.
The file would need to deployed if one of the 2 conditions are correct:
The file is not there.
The size of the file is smaller than that which is being deployed.
The deployment method is good, but the detection method is giving me a headache.
Andy Dawson's blog got me most of the way there, it's the file size (length) bit that's not working for me.
So, my first script went like this:
Part 1: Get logged on user
Function CurrentUser {
$LoggedInUser = get-wmiobject win32_computersystem | select username
$LoggedInUser = [string]$LoggedInUser
$LoggedInUser = $LoggedInUser.split(“=”)
$LoggedInUser = $LoggedInUser[1]
$LoggedInUser = $LoggedInUser.split(“}”)
$LoggedInUser = $LoggedInUser[0]
$LoggedInUser = $LoggedInUser.split(“\”)
$LoggedInUser = $LoggedInUser[1]
Return $LoggedInUser
}
Part 2: Assign User Path and Targetfile variables
$user = CurrentUser
$path = C:\Users\"+$user+"\AppData\Roaming\folder\folder\settings.cfg"
$targetfile = Get-Item $path
Part 3: Detect path exists and check filesize
if (( Test-Path $targetfile) -and ((Get-Item $targetfile).length -ge 26831))
{
Write-Host "installed"
}
else
{
Write-Host "NOT installed"
}
This works if condition 1 is satisfied, but if not the detection method would fail because $path variable and therefore $targetfile could not be populated.
I was using the $targetfile variable because other methods of getting .length seemed to be returning the character count (therefore string length) rather than the intended file size value.
Back to the drawing board. It seems I need to only commence with the rest of the script, assigning the is Condition 1 is NOT satisfied, so...
...My second (current) script goes like this:
Part 1: Get logged on user
UNCHANGED THIS BIT'S FINE, BECAUSE IS STOLE IT :)
Part 2: Assign User and Userpath variables (simplified as this one will always be found)
$user = CurrentUser
$Userpath = "C:\Users\"+$user
Part 3: If path is NOT there state exists and check filesize
if (-not (Test-Path "$Userpath\AppData\Roaming\folder\folder\settings.cfg"))
{
Write-Host "NOT installed"
}
elseif.....
...and this is where the rut sets in! I want the second check to be along the lines of ((Get-Item $targetfile).length -ge 26831)) as before but I just can't seem to get the syntax/logic right here.
Any help greatly appreciated!!!
You'll want to test whether the file doesn't exist OR is to small:
if(-not(Test-Path $targetFile) -or (Get-Item $targetFile).Length -lt $newFileSize){
# File either doesn't exist or is incomplete - go ahead and copy the new config file to $targetFile
Copy-Item .\path\to\reference\config.cfg -Destination $targetFile -Force
}
I'm having issue with my foreach method. I am checking in the registry whether a good amount of programs are installed. How would I write it to say something is not installed one time versus it saying something's not installed for each key it checks? Now, If I place a ElseIf it executes "PowerBroker not installed." about 16 times. This is due to it checking every key and writing it out for each key it does not find a match to the displayname. How do I go about it checking the key and only writing it out one time if it's not installed?? Thanks!
$UninstallKeys = Get-ChildItem 'HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Uninstall'
foreach($Key in $UninstallKeys){
if($Key.GetValue("DisplayName") -Match "BeyondTrust"){
$PBW = $Key.GetValue("DisplayName")
$PBWV = $Key.GetValue("DisplayVersion")
if ($PBW) {
$PBW = $PBW, $PBWV
}
else {
$PBW = "PowerBroker not installed."
$installsmissing = "True"
}
}
Give this script a whirl. If I've understood the requirement correctly it should give you what you need.
$displayName = "BeyondTrust"
$uninstallKeys = Get-ChildItem -Path "HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Uninstall"
# Filter the keys down by their display name property
$specificUninstallKeys = $uninstallKeys |
Where-Object {
$_.GetValue("DisplayName") -eq $displayName
}
# Did we find any keys of that name?
if ($specificUninstallKeys) {
Write-Output "Keys found: $($specificUninstallKeys.Length)"
}
else {
Write-Output "Sorry pal, no keys by that name here!"
}
# There may be more than one; hence the loop-y requirement here.
foreach ($specificUninstallKey in $specificUninstallKey) {
Write-Output $displayName
Write-Output $specificUninstallKey.GetValue("DisplayVersion")
}
I'm working on a TFS build with a pre-build PowerShell script that (in addition to building my app) automatically checks out a file where we maintain version, increments the build number, then checks the file in.
I have been able to do this, except that I get an error from the script which results in a partially successful build (orange). I need to have a fully successful (green) build.
Here's the check-in line (using TFS Power Tools for VS 2013):
New-TfsChangeset -Item $versionFile -Override "Automated" -Notes "Code Reviewer=tfs" -Comment "Automated"
The error I receive is that the changeset is not associated with a work item, but the -Override should handle that. The funny thing is that it checks in anyway.
Running locally on my machine instead of the build server, I get the same thing, except that I also see a line that says The policies have been overridden. This tells me that the override is working, but it still outputs the error.
I've tried adding -ErrorAction SilentlyContinue, but it has no effect.
I need one of three options:
A way to suppress output of the checkin error,
A way to create a work item and associate it to the checkin, or
Some other third option that will result in a green build.
Any ideas?
Credit goes to Eddie - MSFT for leading me the right direction, but I wanted to consolidate everything here.
WARNING This will check in all pending changes in the workspace.
Creating a new work item (source)
I did modify it quite a bit to support automation. It connects to TFS and generates a new work item.
function New-WorkItem()
{
# These *should* be registered in the GAC.
# The version numbers will likely have to change as new versions of Visual Studio are installed on the server.
Add-Type -Assembly "Microsoft.TeamFoundation.Client, Version=12.0.0.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a"
Add-Type -Assembly "Microsoft.TeamFoundation.WorkItemTracking.Client, Version=12.0.0.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a"
$server = "http://YOURTFSSERVER:8080/tfs"
$tfs = [Microsoft.TeamFoundation.Client.TeamFoundationServerFactory]::GetServer($server)
$type = [Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItemStore]
$store = [Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItemStore] $tfs.GetService($type)
$workItem = New-Object Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItem($store.Projects[0].WorkItemTypes[0])
$workItem.Title = "Automated work item"
$workItem
}
Associating the work item and checking in
Slight modifications to the code from the link given by Eddie, we get the following:
function CheckIn()
{
param([Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItem] $workItem)
$col = Get-TfsCollection("http://YOURTFSSERVER:8080/tfs/YOURCOLLECTION")
$vcs = Get-TfsVersionControlServer($col)
$ws = $vcs.GetWorkspace([System.IO.Path]::GetDirectoryName($anyPathInWorkspace))
$pc = $ws.GetPendingChanges()
$wici = Get-TfsWorkItemCheckinInfo($workItem)
$changeset = $ws.CheckIn($pc, "Automated check in", $null, $wici, $null)
}
That post doesn't tell you that Get-TfsCollection, Get-TfsVersionControlServer, and Get-TfsWorkItemCheckinInfo aren't defined. I had to find them.
I found the first two on http://nkdagility.com/powershell-tfs-2013-api-1-get-tfscollection-and-tfs-services/. I didn't have to change anything.
function Get-TfsCollection
{
param([string] $CollectionUrl)
if ($CollectionUrl -ne "")
{
#if collection is passed then use it and select all projects
$tfs = [Microsoft.TeamFoundation.Client.TfsTeamProjectCollectionFactory]::GetTeamProjectCollection($CollectionUrl)
}
else
{
#if no collection specified, open project picker to select it via gui
$picker = New-Object Microsoft.TeamFoundation.Client.TeamProjectPicker([Microsoft.TeamFoundation.Client.TeamProjectPickerMode]::NoProject, $false)
$dialogResult = $picker.ShowDialog()
if ($dialogResult -ne "OK")
{
#exit
}
$tfs = $picker.SelectedTeamProjectCollection
}
$tfs
}
function Get-TfsVersionControlServer
{
param([Microsoft.TeamFoundation.Client.TfsTeamProjectCollection] $TfsCollection)
$TfsCollection.GetService("Microsoft.TeamFoundation.VersionControl.Client.VersionControlServer")
}
But I couldn't find Get-TfsWorkItemCheckinInfo. The only Google hit was the kinook link from Eddie (and soon probably this answer). Here's what I came up with:
function Get-TfsWorkItemCheckinInfo
{
param([Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItem] $workItem)
$wi = New-Object Microsoft.TeamFoundation.VersionControl.Client.WorkItemCheckinInfo($workItem, [Microsoft.TeamFoundation.VersionControl.Client.WorkItemCheckinAction]::Resolve)
$wi
}
Now we can use it
CheckIn (New-WorkItem)
That's it!
You can create a work item from PowerShell by following this article: http://halanstevens.com/blog/powershell-script-to-create-a-workitem/
Quote the code here for reference:
$key = Get-ItemProperty HKLM:\SOFTWARE\Microsoft\VisualStudio\8.0
$dir = [string] (Get-ItemProperty $key.InstallDir)
$dir += "PrivateAssemblies\"
$lib = $dir + "Microsoft.TeamFoundation.WorkItemTracking.Client.dll"
[Reflection.Assembly]::LoadFrom($lib)
$lib = $dir + "Microsoft.TeamFoundation.Client.dll"
[Reflection.Assembly]::LoadFrom($lib)
"Please enter your Team Foundation Server Name:"
$server = [Console]::ReadLine()
$server = $server.Trim()
"Connecting to " + $server + "..."
$tfs = [Microsoft.TeamFoundation.Client.TeamFoundationServerFactory]::GetServer($server)
$type = [Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItemStore]
$store = [Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItemStore] $tfs.GetService($type)
$workItem = new-object Microsoft.TeamFoundation.WorkItemTracking.Client.WorkItem($store.Projects[0].WorkItemTypes[0])
"Created a new work item of type " + $workItem.Type.Name
$workItem.Title = "Created by Windows PowerShell!"
$workItem.Save()
And then refer to this article to associate the work item to changeset: http://www.kinook.com/Forum/showthread.php?t=4502
I've been working on this Powershell script for a good week now, and it almost works as expected.
Essentially, the script reaches into the specified directory which we have another script dropping .CSV files into, grabs the .CSV file(s) and pushes the information found into a Sharepoint list, well, that's the intention anyway. I've gotten the script to work perfectly if I manually specify the file, the issue I am having is actually getting all the .CSV files into a group, and then looping through each .CSV to pull the information out and push it into a Sharepoint list. Once done, it renames the file from .CSV to .ARCHIVED for another script to come in and re-locate after we're done with it.
I think I have, through selective (creative) troubleshooting, figured out what I am doing wrong, I just don't know how to proceed after identifying the issue.
I declare the string $Filecsv like so:
$Filecsv = get-childitem "Z:\" -recurse | where {$_.extension -eq ".csv"}
So, this reaches into my 'Z:\' directory, and pulls all the files with .CSV extension and combines them into a table...
ForEach ($items in $Filecsv) {
And this says for each item, perform logic...
foreach($row in $Filecsv)
The only problem is, when I call $Filecsv, it is returning the list of each .CSV file in the directory like such:
And as such, when I execute the bit of code that says 'put the information into my list', only the file name is added to my Sharepoint list....
Now, I can see what's going on here, it's pulling the 'Name' from the $Filecsv table, and pushing that up to Sharepoint, however, I am not sure how to re-construct my logic so that it operates as expected because as it exists now, it should (to me anyway) work as I think it does, but I am still new to Sharepoint and am certainly missing something here.
Below, is the full code, if it helps:
# Add SharePoint PowerShell Snapin which adds SharePoint specific cmdlets
Add-PSSnapin Microsoft.SharePoint.PowerShell -EA SilentlyContinue
#start the counter at 1 to track times script has looped
$iterations = 1
# set the location where the .CSV files will be pulled from and define the
# file extension we are concerned with
$filecsv = get-childitem "Z:\" -recurse | where {$_.extension -eq ".csv"}
# for each file found in the directory
ForEach ($items in $Filecsv) {
# check to see if files exist, if not exit cleanly
if ($Filecsv) {"File exists" + $Filecsv} else {exit}
# count the times we've looped through
"Iterations : $iterations"
# specify variables needed. The webURL should be the site URL, not including the list
# the listName should be the list name
$WebURL = "http://SHAREPOINTURL/"
$listName = "test"
# Get the SPWeb object and save it to a variable
$web = Get-SPWeb -identity $WebURL
# Get the SPList object to retrieve the list
$list = $web.Lists[$listName]
# START deletes all items. code shows the number of items in a list, then deletes all items
# If you don't want your script to delete items, then remove this
$site = new-object Microsoft.SharePoint.SPSite ( $WebURL )
$web = $site.OpenWeb()
"Web is : " + $web.Title
# Enter name of the List below instead of
$oList = $web.Lists["test"];
"List is :" + $oList.Title
"List Item Count: " + $oList.ItemCount
#delete existing contents and replace with new stuff
$collListItems = $oList.Items;
$count = $collListItems.Count - 1
for($intIndex = $count; $intIndex -gt -1; $intIndex--) {
"Deleting record: " + $intIndex
$collListItems.Delete($intIndex);
}
# END Deletes all items
# goes through the CSV file and performs action for each row
foreach($row in $Filecsv)
{
$newItem = $list.items.Add()
$item = $list.items.add()
# Check if cell value is not null in excel
if ($row."Name" -ne $null)
# Add item to sharepoint list. for this one, I had to use the internal column name.
#You don't always have to, but I had trouble with one SharePoint column, so I did
{$newItem["Name"] = $row."Name"}
else{$newItem["Name"] = $row."Not Provided"}
if ($row."Description" -ne $null)
{$newItem["Description"] = $row."Description"}
else{$newItem["Description"] = $row."No Description"}
if ($row."NetworkID" -ne $null)
{$newItem["Network ID"] = $row."NetworkID"}
else{$newItem["Network ID"] = $row."No NetworkID"}
if ($row."Nested" -ne $null)
{$newItem["Nested"] = $row."Nested"}
else{$newItem["Nested"] = $row."Not Nested"}
# Commit the update, then loop again until end of file
$newItem.Update()
}
# get the date and time from the system
$datetime = get-date -f MMddyy-hhmmtt
# rename the file
$NewName = $items.fullname -replace ".csv$","$datetime.csv.archived"
$Items.MoveTo($NewName)
# +1 the counter to count the number of files we've looped through
$iterations ++
}
a very cursory look would suggest that you need to use $items not $filecsv in your main loop.
essentially you are looping over the contents of the $filecsv collection, so you need to look at $items.
Your ForEach loops look redundant since they are both looping through a list of FileInfo objects. I think you want to find all the files, and for each file load it into memory and process it's contents. We'll go that route.
I have moved your SharePoint object creation out of the loop since I don't see any point to creating the object over and over for each file processed since it never references anything based on the file or it's contents. It simply makes the same object over, and over, and over.
# Add SharePoint PowerShell Snapin which adds SharePoint specific cmdlets
Add-PSSnapin Microsoft.SharePoint.PowerShell -EA SilentlyContinue
#start the counter at 1 to track times script has looped
$iterations = 1
# specify variables needed. The webURL should be the site URL, not including the list
# the listName should be the list name
#Setup SP object
$WebURL = "http://SHAREPOINTURL/"
$listName = "test"
# Get the SPWeb object and save it to a variable
$web = Get-SPWeb -identity $WebURL
# Get the SPList object to retrieve the list
$list = $web.Lists[$listName]
# START deletes all items. code shows the number of items in a list, then deletes all items
# If you don't want your script to delete items, then remove this
$site = new-object Microsoft.SharePoint.SPSite ( $WebURL )
$web = $site.OpenWeb()
"Web is : " + $web.Title
# Enter name of the List below instead of
$oList = $web.Lists["test"];
"List is : " + $oList.Title
"List Item Count: " + $oList.ItemCount
#delete existing contents and replace with new stuff
$collListItems = $oList.Items;
$count = $collListItems.Count - 1
for($intIndex = $count; $intIndex -gt -1; $intIndex--) {
"Deleting record: " + $intIndex
$collListItems.Delete($intIndex);
}
# END Deletes all items
Find all the CSV files, and start looping through the list of them. I removed the check to see if the file exists. You just pulled a directory listing to find these files, they really should exist.
# set the location where the .CSV files will be pulled from and define the
# file extension we are concerned with
$CSVList = get-childitem "Z:\" -recurse | where {$_.extension -eq ".csv"}
ForEach ($CSVFile in $CSVList) {
# count the times we've looped through
"Iterations : $iterations"
Now, this is different. It loads the CSV file, and processes each row in it as $row. I'm pretty sure this is what you intended to do from the start. I also changed it from If(Something -ne $null) to check for either null, or empty since either can actually exist and the later can cause you some issues. It's just a safer method in general.
foreach($row in (Import-CSV $CSVFile.FullName))
{
$newItem = $list.items.Add()
$item = $list.items.add()
# Check if cell value is not null in excel
if (![string]::IsNullOrEmpty($row."Name"))
# Add item to sharepoint list. for this one, I had to use the internal column name.
#You don't always have to, but I had trouble with one SharePoint column, so I did
{$newItem["Name"] = $row."Name"}
else{$newItem["Name"] = $row."Not Provided"}
if (![string]::IsNullOrEmpty($row."Description"))
{$newItem["Description"] = $row."Description"}
else{$newItem["Description"] = $row."No Description"}
if (![string]::IsNullOrEmpty($row."NetworkID"))
{$newItem["Network ID"] = $row."NetworkID"}
else{$newItem["Network ID"] = $row."No NetworkID"}
if (![string]::IsNullOrEmpty($row."Nested"))
{$newItem["Nested"] = $row."Nested"}
else{$newItem["Nested"] = $row."Not Nested"}
# Commit the update, then loop again until end of file
$newItem.Update()
}
I don't really understand why you are adding a new item twice, but if it works then more power to you. Then your bit to rename files when you're done with them (hey, this looks familiar):
# get the date and time from the system
$datetime = get-date -f MMddyy-hhmmtt
# rename the file
$NewName = $CSVFile.fullname -replace ".csv$","$datetime.csv.archived"
$CSVFile.MoveTo($NewName)
# +1 the counter to count the number of files we've looped through
$iterations ++
}
I did rename a few things to make them more indicative of what they represent ($Items to $CSVFile and what not). See if this works for you. If you have questions or concerns let me know.
Edit: Ok, to fix the loop trying to pull each item from the current folder we reference the FullName property of it. One line changed:
foreach($row in (Import-CSV $CSVFile.FullName))
so i am working with a bit of code that after setting a static location for the CSV i am using for import. When i run the code it seems to run until i get the Windows can't open this File pop. you know the one with what do you want to do options, like user the web services to find the correct program. i am copying the code so hopefully someone can point on where i made this fubar. just for note before i made the CSV static the Script asked you to type the location in every time so maybe i missed a setting there
if ($args[0] -eq $null)
{
$userNameFile = D:\lync_creation\userlog.csv
$userNameFile = $usernamefile -replace '"',""}
else
{$usernamefile = $args[0]}
if ($userNameFile -ne "")
{$csv=import-csv $userNameFile}
else
{"Could not find a valid .csv with the user information."
exit}
foreach($c in $csv)
# enable for lync
{
"Enabling " + $c.Identity + " for Lync 2010"
Enable-csuser -identity $c.Identity -registrarpool pool01.west.com –sipaddresstype EmailAddress
}
write-host "Press any key to continue..."
$x = $host.UI.RawUI.ReadKey("NoEcho,IncludeKeyDown,AllowCtrlC")
You are seriously over complicating things if you're just going to use a static location for your CSV file.
$csv = Import-CSV D:\lync_creation\userlog.csv
foreach($c in $csv){
"Enabling $($c.Identity) for Lync 2010"
Enable-csuser -identity $c.Identity -registrarpool pool01.west.com –sipaddresstype EmailAddress
}
write-host "Press any key to continue..."
$x = $host.UI.RawUI.ReadKey("NoEcho,IncludeKeyDown,AllowCtrlC")
The first line imports the CSV file into a variable.
The next 4 loops through all entries in that variable, write the host who it's enabling and then enables the person.
The last 2 lines give a Press any key to continue message and then waits for a key press before continuing.