After writing deployment scripts from within the ISE, we need our continuous integration (CI) server to be able to run them automatically, i.e. from the command line or via a batch file.
I have noticed some significant differences between the following calls:
powershell.exe -File Script.ps1
powershell.exe -Command "& '.\Script.ps1'"
powershell.exe .\Script.ps1
Some simple examples:
-File, errors are handled in the exact same way as the ISE.$ErrorActionPreference variable, and do not catch Write-Error in try/catch blocks.When using pSake:
-File parameter will fail with the following error:The variable '$script:context' cannot be retrieved because it has not been set
What are the implications of each syntax, and why they are behaving differently? I would ideally like to find a syntax that works all the time and behaves like the ISE.
Not an answer, just a note.
I searched for explanation of -file parameter. Most sources say only "Execute a script file.". At http://technet.microsoft.com/en-us/library/dd315276.aspx I read
Runs the specified script in the local scope ("dot-sourced"), so that the functions
and variables that the script creates are available in the current session. Enter
the script file path and any parameters.
After that I tried to call this:
powershell -command ". c:\temp\aa\script.ps1"
powershell -file c:\temp\aa\script.ps1
powershell -command "& c:\temp\aa\script.ps1"
Note that first two stop after Get-Foo, but the last one doesn't.
The problem I describe above is related to modules -- if you define Get-Foo inside script.ps1, all the 3 calls I described stop after call to Get-Foo.
Just try to define it inside the script.ps1 or dotsource the file with Get-Foo and check it. There is a chance it will work :)