30-TestScript-reliable-exe.t

Enable keyboard shortcuts

++ #!/usr/bin/env perl

++ use strict;

++ use warnings;

++

++ package TestReliableExe;

++

++ use Capture::Tiny qw( capture );

++ use English qw( -no_match_vars );

++ use Env::Path;

++ use File::Spec::Functions;

++ use File::Temp qw( tempdir );

++ use FindBin;

++ use Readonly;

++ use Test::Exception;

++ use Test::More;

++ use File::Basename qw( basename );

++

++ use lib catfile( $FindBin::Bin, qw(..) x 2 );

++ use QtQA::Proc::Reliable::TESTDATA qw( %TESTDATA );

++ use QtQA::Test::More qw( :all );

++

++ use base 'QtQA::TestScript';

++

++ =head1 NAME

++

++ 30-testscript-reliable-exe.t - system test of exe's `reliable' feature

++

++ =head1 DESCRIPTION

++

++ This test runs various commands with simulated error cases and verifies that

++ exe automatically recovers from certain errors.

++

++ =cut

++

++ sub new

++ {

++     my ($class) = @_;

++

++     my $self = $class->SUPER::new( );

++     bless( $self, $class );

++     return $self;

++ }

++

++ sub run_one_test

++ {

++     my ($self, $testname, $testdata) = @_;

++

++     my $tempdir = tempdir( basename($0).'.XXXXXX', TMPDIR => 1, CLEANUP => 1 );

++

++     my $command             =   $testdata->{ command };

++     my %mock_command        = %{$testdata->{ mock_command }};

++     my $reliable            =   $testdata->{ reliable }         // 1;

++     my $expected_stderr     =   $testdata->{ expected_exe_stderr }

++                             //  $testdata->{ expected_raw_stderr }

++                             //  q{};

++     my $expected_stdout     =   $testdata->{ expected_exe_stdout }

++                             //  $testdata->{ expected_raw_stdout }

++                             //  q{};

++     my $expected_status     =   $testdata->{ expected_status }  // 0;

++

++     $mock_command{ directory } = $tempdir;

++

++     if (! exists $mock_command{ name }) {

++         $mock_command{ name } = $command->[0];

++     }

++

++     create_mock_command( %mock_command );

++

++     # Put our mock command first in PATH

++     local $ENV{ PATH } = $ENV{ PATH };

++     Env::Path->PATH->Prepend( $tempdir );

++

++     # Usually, we just pass our command to exe ...

++     my @exe_args = @{$command};

++

++     # ... but, for a non-auto `reliable', we pass options too.

++     if (ref($reliable) || $reliable != 1) {

++         @exe_args = ( { reliable => $reliable }, @exe_args );

++     }

++

++     my ($lives_or_dies_text, $lives_or_dies_ok)

++         = ($expected_status == 0) ? ('exits successfully', \&lives_ok)

++         :                           ('fails as expected',  \&dies_ok )

++     ;

++

++     my ($stdout, $stderr) = capture {

++         $lives_or_dies_ok->(

++             sub { $self->exe( @exe_args ) },

++             "$testname $lives_or_dies_text",

++         );

++     };

++

++     # exe() should have set $? to the expected value

++     is( $?, $expected_status, "$testname \$? set as expected" );

++

++     # Immediately prior to stdout should always be the command;

++     # Prior to that may be either CWD or PATH lines.

++     my $prefix = quotemeta( '+ '.join(' ', @{$command}) );

++     $prefix = qr{

++         \A

++         (?:\+\ CWD:\ [^\n]+\n)?

++         (?:\+\ PATH:\ [^\n]+\n)?

++         $prefix \n

++     }xms;

++     like( $stdout, $prefix, "$testname stdout first line looks correct" );

++

++     # Remove prefix for subsequent comparison

++     $stdout =~ s{$prefix}{}xms;

++

++     is_or_like( $stdout,  $expected_stdout,  "$testname stdout looks correct" );

++     is_or_like( $stderr,  $expected_stderr,  "$testname stderr looks correct" );

++

++     return;

++ }

++

++ sub run

++ {

++     my ($self) = @_;

++

++     while (my ($testname, $testdata) = each %TESTDATA) {

++         $self->run_one_test($testname, $testdata);

++     }

++

++     done_testing( );

++

++     return;

++ }

++

++ TestReliableExe->new( )->run( ) if (!caller);

++ 1;

++