How to read machine.config file to check whether it contains a key?
Example: oracle.managedaccess.client.
Below is the code:
$configFilePath = [System.Runtime.InteropServices.RuntimeEnvironment]::SystemConfigurationFile
Write-Host $configFilePath
$cnf = . $configFilePath
echo name
Write-Host $cnf
#check machine.config contains oracle.managedaccess.client
The . operator is for running PowerShell scripts into the context of the calling script (as opposed to running them in a child context via the & operator). It's not for reading arbitrary files.
For reading plain text files use Get-Content. For reading XML files (which your machine.config most likely is) create an XML object and use its Load() method:
$xml = New-Object Xml
$xml.Load('C:\path\to\your\machne.config')
You can check for the presence of a node or attribute by using the SelectNodes() method with an XPath expression. An empty result means the node doesn't exist.
Ansgar Wiechers' helpful answer explains the problem with your code well and offers helpful background information.
Here's a concrete solution that shows some related techniques:
$file = [System.Runtime.InteropServices.RuntimeEnvironment]::SystemConfigurationFile
[xml] $xml = Get-Content -Raw $file
# Get the <oracle.manageddataaccess.client> element
# that, in this example, is assumed to be a child element of the
# top-level <configuration> element.
$elem = $xml.configuration.'oracle.manageddataaccess.client'
# Determine if the element exists.
$exists = $null -ne $elem
Explanation:
[xml] $xml = Get-Content -Raw $file is a more PowerShell-idiomatic way of reading an XML document from a file; using the [xml] type accelerator automatically turns the file's content, obtained with Get-Content, into an XML document (System.Xml.XmlDocument).
$xml.configuration.'oracle.manageddataaccess.client' uses PowerShell's adaptation of the XML DOM to provide convenient access to the elements in the document hierarchy with dot notation (.)
Note that this assumes that you know the precise location (path to) the target element; if you don't, use .SelectNodes() or .SelectSingleNode() as recommended in Ansgar's answer; see example at the bottom.
Note the need to enclose oracle.manageddataaccess.client in '...' (quotes), because the element name itself contains . characters.
For more information about how PowerShell adapts the XML DOM, see this answer.
If the targeted element exists, a System.Xml.XmlElement instances is returned, if the target element has child elements of its own (otherwise, its .InnerText property would be returned).
If no such element exists, the return value is $null.
$exists = $null -eq $elem creates a Boolean variable that indicates whether the element of interest exists or not.
Example uses of .SelectSingleNode() with XPath queries:
The following returns a <oracle.manageddataaccess.client> element found anywhere in the document, if present, and $null otherwise:
$xml.SelectSingleNode('//oracle.manageddataaccess.client')
The following locates a <section> element that is a child element of elements <configuration> and <configSections> and that contains a name attribute containing the string appSettings
$xml.SelectSingleNode('/configuration/configSections/section[@name="appSettings"]')
