Question

How would I find a multiline pattern in files, such as the contents of an XML node, using Powershell?

i.e. if I were looking for the word “green” within the deviceDescription node, but the XML node text may span multiple lines, this doesn’t work:

dir -r -i *.xml | select-string -Pattern "<deviceDescription>.*green.*</deviceDescription>"

Answer 1

First of all, if is xml, extract the device description string treating it as such and then match for the string you want, in this case, green.

$x = [xml] (gc .\test.xml)
$x.deviceDescription -match "green"

If you don’t want to go that way, you will have to use the ?s – singleline or dotall flag which makes * match newlines:

$x = [IO.File]::ReadAllText("C:\test.xml")
$x -match "(?s)<deviceDescription>.*green.*</deviceDescription>"

Note that you probably want to use .*? or this may span across multiple deviceDescription tags. Edge cases like this are reasons why you should avoid using regex for such things.