Tests/Write-SAS.Tests.ps1
|
<#
.SYNOPSIS Write-SAS test cases .DESCRIPTION This script contains Pester 5.3 test cases for the powerSAS modules. There is one Test 'Description' per exported function, and each test description contains a number of contexts which contain the individual test case assertions. .EXAMPLE PS C:\> Invoke-Pester Run all the tests and produce summary .EXAMPLE PS C:\> Invoke-Pester -Output Detailed To run all tests and get detail on each test run. #> # # Pre-requisite is to load the powerSAS module to be tested # BeforeAll { Import-Module $PSScriptRoot\..\powerSAS -Force } ############################################################### ## Write-SAS ############################################################### Describe "Write-SAS" { <# - listorlog : output LST results if there are any, if not output the LOG - logandlist : output both, first the LOG then the LST are output - log : only the log is output - list : only the LST is output #> Context "One liner" { <# # Test write-sas in the context of a single line of sas code that # can be executed 'in one submit'. The different -method parameters # are tested to ensure that only the log or list (or both) is returned. #> BeforeEach { {Connect-SAS -local} | should -not -throw # define a SAS one liner that generates LST and LOG output $oneliner = "proc contents data=sashelp.shoes varnum; run;" } It "default method" { # single line that will generate LST and LOG output $result = ($oneliner | write-SAS) # check that we actually get some results $result | should -not -BeNullOrEmpty # default should be listorlog so check that we get LIST output $result -match "Number of Stores" | should -be $true # and check that there is no LOG output (because there was LIST) $result -match "NOTE: PROCEDURE CONTENTS used" | should -be $false } It "specified method (listorlog)" { # single line that will generate LST and LOG output $result = ($oneliner | write-SAS -method listorlog) # check that we actually get some results $result | should -not -BeNullOrEmpty # default should be listorlog so check that we get LIST output $result -match "Number of Stores" | should -be $true # and check that there is no LOG output (because there was LIST) $result -match "NOTE: PROCEDURE CONTENTS used" | should -be $false # now try again with only log output $result2 = ("%put %str(lo)gok;" | write-SAS -method listorlog) $result2 -match "logok" | should -be $true } It "specified method (logandlist)" { # single line that will generate LST and LOG output $result = ($oneliner | write-SAS -method logandlist) # check that we actually get some results $result | should -not -BeNullOrEmpty # default should be listorlog so check that we get LIST output $result -match "Number of Stores" | should -be $true # and check that there is LOG output too $result -match "NOTE: PROCEDURE CONTENTS used" | should -be $true } It "specified method (log)" { # single line that will generate LST and LOG output $result = ($oneliner | write-SAS -method log) $result | should -not -BeNullOrEmpty # check there is no list output $result -match "Number of Stores" | should -be $false # and check that there is LOG output $result -match "NOTE: PROCEDURE CONTENTS used" | should -be $true } It "specified method (list)" { # single line that will generate LST and LOG output $result = ($oneliner | write-SAS -method list) $result | should -not -BeNullOrEmpty # check that we get LIST output $result -match "Number of Stores" | should -be $true # and check that there is no LOG output $result -match "NOTE: PROCEDURE CONTENTS used" | should -be $false } It "specified method (none)" { # single line that will generate LST and LOG output { $oneliner | write-SAS -method } | should -throw "Missing an argument*" } It "specified method (invalid)" { # single line that will generate LST and LOG output { $oneliner | write-SAS -method loglist } | should -throw "Unknown method*" } AfterEach { {Disconnect-SAS} | should -not -throw } } # end context one liner Context "Multi-Liner" { <# # These tests are in the context of multiple calls to write-SAS that # add up to a completed program. The calls are to 1) setup a libname # 2) use the libname, and 3) create some output from data in 2 #> BeforeEach { {Connect-SAS -local} | should -not -throw } It "Runs as expected" -ForEach @( @{ method = "listorlog"; expectlist = $true; expectlog = $false; } @{ method = "logandlist"; expectlist = $true; expectlog = $true; } @{ method = "log"; expectlist = $false; expectlog = $true; } @{ method = "list"; expectlist = $true; expectlog = $false; } ){ # empty the results var that will be used to accumulate all results $result = "" # write individual sas fragments to SAS split over multiple write-SAS foreach ($prog in @( "libname foo(sashelp);" "data " " test;" "set foo.shoes;" "run;" "proc contents data=test varnum;" "run;" )) { # accumulate all the results $result += ($prog | write-SAS -method $method) } # check that there are no errors $result -match "ERROR" | should -be $false # check list output is as expected $result -match "Number of Stores" | should -be $expectlist # check log output is as expected $result -match "NOTE: PROCEDURE CONTENTS used" | should -be $expectlog } AfterEach { {Disconnect-SAS} | should -not -throw } } } |