用于EagleEye3.0 规则集漏报和误报测试的示例项目,项目收集于github和gitee
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

1608 lines
52 KiB

# -*- cperl -*-
# Copyright (c) 2005, 2019, Oracle and/or its affiliates. All rights reserved.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License, version 2.0,
# as published by the Free Software Foundation.
#
# This program is also distributed with certain software (including
# but not limited to OpenSSL) that is licensed under separate terms,
# as designated in a particular file or component or in included license
# documentation. The authors of MySQL hereby grant you an additional
# permission to link the program and your derivative works with the
# separately licensed software that they have included with MySQL.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License, version 2.0, for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
# This is a library file used by the Perl version of mysql-test-run,
# and is part of the translation of the Bourne shell script with the
# same name.
package mtr_cases;
use strict;
use base qw(Exporter);
our @EXPORT = qw(collect_option collect_test_cases init_pattern
$do_test $group_replication);
use File::Basename;
use File::Spec::Functions qw / splitdir /;
use IO::File;
use My::Config;
use My::File::Path qw / get_bld_path /;
use My::Find;
use My::Platform;
use My::SysInfo;
use My::Test;
use mtr_match;
use mtr_report;
require "mtr_misc.pl";
my $threads_support = eval 'use threads; 1';
my $threads_shared_support = eval 'use threads::shared; 1';
# Precompiled regex's for tests to do or skip
my $do_test_reg;
my $skip_test_reg;
# Related to adding InnoDB plugin combinations
my $do_innodb_plugin;
my $lib_innodb_plugin;
# If "Quick collect", set to 1 once a test to run has been found.
my $some_test_found;
# Options used for the collect phase
our $binlog_format;
our $default_storage_engine;
our $defaults_extra_file;
our $defaults_file;
our $do_test;
our $enable_disabled;
our $opt_with_ndbcluster_only;
our $print_testcases;
our $quick_collect;
our $skip_rpl;
our $skip_test;
our $start_from;
# Set to 1 if you want the tests to override default storage engine
# settings, and use MyISAM as default (temporary option used in
# connection with the change of default storage engine to InnoDB).
our $default_myisam = 0;
our $group_replication;
sub collect_option {
my ($opt, $value) = @_;
# Evaluate $opt as string to use "Getopt::Long::Callback legacy API"
my $opt_name = "$opt";
# Convert - to _ in option name
$opt_name =~ s/-/_/g;
no strict 'refs';
${$opt_name} = $value;
}
sub init_pattern {
my ($from, $what) = @_;
return undef unless defined $from;
if ($from =~ /^[a-z0-9\.]*$/) {
# Does not contain any regex (except '.' that we allow as
# separator betwen suite and testname), make the pattern match
# beginning of string.
$from = "^$from";
mtr_verbose("$what='$from'");
}
# Check that pattern is a valid regex
eval { "" =~ /$from/; 1 } or
mtr_error("Invalid regex '$from' passed to $what\nPerl says: $@");
return $from;
}
## Report an error for an invalid disabled test entry in a disabled.def
## file.
##
## Arguments:
## $line An entry from a disabled.def file
## $line_number Line number
## $disabled_def_file disabled.def file location
sub report_disabled_test_format_error($$$) {
my $line = shift;
my $line_number = shift;
my $disabled_def_file = shift;
mtr_error("The format of line '$line' at '$disabled_def_file:$line_number' " .
"is incorrect. The format is '<suitename>.<testcasename> " .
"[\@platform|\@!platform] : <BUG|WL>#<XXXX> [<comment>]'.");
}
## Check if the test name format in a disabled.def file is correct. The
## format is "suite_name.test_name". If the format is incorrect, throw an
## error and abort the test run.
##
## Arguments:
## $test_name Test name
## $line An entry from a disabled.def file
## $line_number Line number
## $disabled_def_file disabled.def file location
sub validate_test_name($$$$) {
my $test_name = shift;
my $line = shift;
my $line_number = shift;
my $disabled_def_file = shift;
# Test name must be in "<suitename>.<testcasename>" format.
my ($suite_name, $tname, $extension) = split_testname($test_name);
if (not defined $suite_name || not defined $tname || defined $extension) {
report_disabled_test_format_error($line, $line_number, $disabled_def_file);
}
# disabled.def should contain only non-ndb suite tests, and disabled_ndb.def
# should contain only any of the ndb suite tests.
my $fname = basename($disabled_def_file);
if (($fname eq 'disabled.def' and $suite_name =~ /ndb/) or
($fname eq 'disabled_ndb.def' and $suite_name !~ /ndb/)) {
mtr_error("'$disabled_def_file' shouldn't contain '$suite_name' " .
"suite test(s).");
}
}
## Check if the test name section in a disabled.def file is correct. The
## format is "suite_name.test_name [@platform|@!platform]". If the format
## is incorrect, throw an error and abort the test run.
##
## Arguments:
## $test_name_part Test name section in a disabled test entry
## $line An entry from a disabled.def file
## $line_number Line number
## $disabled_def_file disabled.def file location
sub validate_test_name_part($$$$) {
my $test_name_part = shift;
my $line = shift;
my $line_number = shift;
my $disabled_def_file = shift;
my $test_name;
if ($test_name_part =~ /\@/) {
# $^O on Windows considered not generic enough.
my $plat = (IS_WINDOWS) ? 'windows' : $^O;
# Test name part contains platform name.
if ($test_name_part =~ /^\s*(\S+)\s+\@$plat\s*$/) {
# Platform name is mentioned on which the test should be disabled
# i.e "suite_name.test_name @platform : <comment>".
$test_name = $1;
validate_test_name($test_name, $line, $line_number, $disabled_def_file);
return $test_name;
} elsif ($test_name_part =~ /^\s*(\S+)\s+\@!(\S*)\s*$/) {
# Platform name is mentioned on which the test shouldn't be
# disabled i.e "suite_name.test_name @!platform : XXXX".
$test_name = $1;
validate_test_name($test_name, $line, $line_number, $disabled_def_file);
return $test_name if ($2 ne $plat);
} elsif ($test_name_part !~ /^\s*(\S+)\s+\@(\S*)\s*$/) {
# Invalid format for a disabled test.
report_disabled_test_format_error($line, $line_number,
$disabled_def_file);
}
} elsif ($test_name_part =~ /^\s*(\S+)\s*$/) {
# No platform specified.
$test_name = $1;
validate_test_name($test_name, $line, $line_number, $disabled_def_file);
return $test_name;
} else {
# Invalid format, throw an error.
report_disabled_test_format_error($line, $line_number, $disabled_def_file);
}
}
## Check that a test specified in disabled.def actually exists
## if the suite it belongs to is collected.
##
## Arguments:
## $test_name Name of the test to be located
sub validate_test_existence($) {
my $test_name = shift;
my ($sname, $tname, $extension) = split_testname($test_name);
# Search if the specified suite is present in the list
# of suites which MTR collects
for my $suite (split(",", $::opt_suites)) {
if ($sname eq $suite or $sname eq "i_" . $suite) {
# Proceed to check if the disabled test exists only if its
# suite is picked up
my $suitedir = get_suite_dir($sname);
mtr_error("Disabled test '$test_name' could not be located.")
if $suitedir and
not my_find_file($suitedir, [ "", "t" ], "$tname.test", NOT_REQUIRED);
last;
}
}
}
## Check if the comment section in a disabled.def file is correct. The
## format is "<BUG|WL>#<XXXX> [<comment>]". If the format is incorrect,
## throw an error and abort the test run.
##
## Arguments:
## $comment_part Comment section in a disabled test entry
## $line An entry from a disabled.def file
## $line_number Line number
## $disabled_def_file disabled.def file location
sub validate_comment_part($$$$) {
my $comment_part = shift;
my $line = shift;
my $line_number = shift;
my $disabled_def_file = shift;
if ($comment_part =~ /^\s*(BUG|WL)#\d+\s*(\s+(.*))?$/i) {
my $comment = $3;
# Length of a comment section can't be more than 80 characters.
if (length($comment) > 79) {
mtr_error("Length of a comment section in a disabled test entry can't " .
"be more than 80 characters, " .
"'$line' at '$disabled_def_file:$line_number'.");
}
# Valid format for comment section.
return $comment_part;
} else {
# Invalid format. throw an error.
report_disabled_test_format_error($line, $line_number, $disabled_def_file);
}
}
## Validate the format of an entry in a disabled.def file.
##
## Arguments:
## $line An entry from a disabled.def file
## $line_number Line number
## $disabled_def_file disabled.def file location
##
## Returns:
## Test name and the comment part.
sub validate_disabled_test_entry($$$) {
my $line = shift;
my $line_number = shift;
my $disabled_def_file = shift;
my $comment_part;
my $test_name_part;
if ($line =~ /^(.*?):(.*)$/) {
$test_name_part = $1;
$comment_part = $2;
}
# Check the format of test name part.
my $test_name =
validate_test_name_part($test_name_part, $line, $line_number,
$disabled_def_file);
# Check if the disabled test exists
validate_test_existence($test_name) if $test_name;
# Check the format of comment part.
my $comment =
validate_comment_part($comment_part, $line, $line_number,
$disabled_def_file);
return ($test_name, $comment);
}
## Create a list of disabled tests from disabled.def file. This list
## is used while collecting the test cases. If a test case is disabled,
## disabled flag is enabled for the test and the test run will be
## disabled.
##
## Arguments:
## $disabled List to store the disabled tests
## $opt_skip_test_list File containing list of tests to be disabled
##
## Returns:
## List of disabled tests
sub create_disabled_test_list($$) {
my $disabled = shift;
my $opt_skip_test_list = shift;
if ($opt_skip_test_list) {
$opt_skip_test_list = get_bld_path($opt_skip_test_list);
}
# Array containing files listing tests that should be disabled.
my @disabled_collection = $opt_skip_test_list if $opt_skip_test_list;
# Add 'disabled.def' files.
unshift(@disabled_collection,
"$::glob_mysql_test_dir/collections/disabled.def");
# Add internal 'disabled.def' file only if it exists
my $internal_disabled_def_file =
"$::basedir/internal/mysql-test/collections/disabled.def";
unshift(@disabled_collection, $internal_disabled_def_file)
if -f $internal_disabled_def_file;
# 'disabled.def' file in cloud directory.
$internal_disabled_def_file =
"$::basedir/internal/cloud/mysql-test/collections/disabled.def";
unshift(@disabled_collection, $internal_disabled_def_file)
if -f $internal_disabled_def_file;
# Add 'disabled_ndb.def' to the list of disabled files if ndb is enabled.
unshift(@disabled_collection,
"$::glob_mysql_test_dir/collections/disabled_ndb.def")
if $::ndbcluster_enabled;
# Check for the tests to be skipped in a sanitizer which are listed
# in "mysql-test/collections/disabled-<sanitizer>.list" file.
if ($::opt_sanitize) {
# Check for disabled-asan.list
if ($::mysql_version_extra =~ /asan/i &&
$opt_skip_test_list !~ /disabled-asan\.list$/) {
push(@disabled_collection, "collections/disabled-asan.list");
}
# Check for disabled-ubsan.list
elsif ($::mysql_version_extra =~ /ubsan/i &&
$opt_skip_test_list !~ /disabled-ubsan\.list$/) {
push(@disabled_collection, "collections/disabled-ubsan.list");
}
}
# Check for the tests to be skipped in valgrind which are listed
# in "mysql-test/collections/disabled-valgrind.list" file.
if ($::opt_valgrind) {
# Check for disabled-valgrind.list
if ($opt_skip_test_list !~ /disabled-valgrind\.list$/) {
push(@disabled_collection, "collections/disabled-valgrind.list");
}
}
for my $disabled_def_file (@disabled_collection) {
my $file_handle = IO::File->new($disabled_def_file, '<') or
mtr_error("Can't open '$disabled_def_file' file: $!.");
if (defined $file_handle) {
while (my $line = <$file_handle>) {
# Skip a line if it starts with '#' or is an empty line.
next if ($line =~ /^#/ or $line =~ /^$/);
chomp($line);
my $line_number = $.;
# Check the format of an entry in a disabled.def file.
my ($test_name, $comment) =
validate_disabled_test_entry($line, $line_number, $disabled_def_file);
# Disable the test case if defined.
$disabled->{$test_name} = $comment if defined $test_name;
}
$file_handle->close();
}
}
}
# This is the top level collection routine. If tests are explicitly
# named on the command line, it collects suite name information from
# the test names if all of them are qualified with suite name.
#
# Then it loops over all the suites to use (which is either the
# default or an explicit list), collecting tests using
# collect_one_suite.
#
# After collecting all the tests, it goes through all of them and
# constructs a sort criteria string for each, which is finally used
# to reorder the list in a way which is intended to reduce the number
# of restarts if 'reorder' option is enabled.
sub collect_test_cases ($$$$) {
my $opt_reorder = shift; # True if we're reordering tests
my $suites = shift; # Semicolon separated list of test suites
my $opt_cases = shift;
my $opt_skip_test_list = shift;
my $cases = []; # Array of hash(one hash for each testcase)
# Unit tests off by default also if using --do-test or --start-from
$::opt_ctest = 0 if $::opt_ctest == -1 && ($do_test || $start_from);
$do_test_reg = init_pattern($do_test, "--do-test");
$skip_test_reg = init_pattern($skip_test, "--skip-test");
$lib_innodb_plugin =
my_find_file($::basedir,
[ "storage/innodb_plugin", "storage/innodb_plugin/.libs",
"lib/mysql/plugin", "lib/plugin"
],
[ "ha_innodb_plugin.dll", "ha_innodb_plugin.so",
"ha_innodb_plugin.sl"
],
NOT_REQUIRED);
$do_innodb_plugin =
($::mysql_version_id >= 50100 && !(IS_WINDOWS) && $lib_innodb_plugin);
# Build a hash of disabled testcases
my %disabled;
create_disabled_test_list(\%disabled, $opt_skip_test_list);
# If not reordering, we also shouldn't group by suites, unless no
# test cases were named. This also effects some logic in the loop
# following this.
if ($opt_reorder or !@$opt_cases) {
my $parallel = $ENV{NUMBER_OF_CPUS};
$parallel = $::opt_parallel if ($::opt_parallel < $parallel);
$parallel = 1 if $quick_collect;
$parallel = 1 if @$opt_cases;
if ($parallel == 1 or !$threads_support or !$threads_shared_support) {
foreach my $suite (split(",", $suites)) {
push(@$cases,
collect_one_suite($suite, $opt_cases,
$opt_skip_test_list, \%disabled
));
last if $some_test_found;
push(@$cases,
collect_one_suite("i_" . $suite, $opt_cases,
$opt_skip_test_list, \%disabled
));
}
} else {
share(\$group_replication);
share(\$some_test_found);
# Array containing thread id of all the threads used for
# collecting test cases from different test suites.
my @collect_test_cases_thrds;
foreach my $suite (split(",", $suites)) {
push(@collect_test_cases_thrds,
threads->create("collect_one_suite", $suite,
$opt_cases, $opt_skip_test_list,
\%disabled
));
while ($parallel <= scalar @collect_test_cases_thrds) {
mtr_milli_sleep(100);
@collect_test_cases_thrds = threads->list(threads::running());
}
last if $some_test_found;
push(@collect_test_cases_thrds,
threads->create("collect_one_suite", "i_" . $suite,
$opt_cases, $opt_skip_test_list,
\%disabled
));
while ($parallel <= scalar @collect_test_cases_thrds) {
mtr_milli_sleep(100);
@collect_test_cases_thrds = threads->list(threads::running());
}
}
foreach my $collect_thrd (threads->list()) {
my @suite_cases = $collect_thrd->join();
push(@$cases, @suite_cases);
}
}
}
if (@$opt_cases) {
# A list of tests was specified on the command line. Among those,
# the tests which are not already collected will be collected and
# stored temporarily in an array of hashes pointed by the below
# reference. This array is eventually appeneded to the one having
# all collected test cases.
my $cmdline_cases;
# Check that the tests specified were found in at least one suite.
foreach my $test_name_spec (@$opt_cases) {
my $found = 0;
my ($sname, $tname, $extension) = split_testname($test_name_spec);
foreach my $test (@$cases) {
last unless $opt_reorder;
# 'test->{name}' value is always in suite.name format
if ($test->{name} =~ /^$sname.*\.$tname$/) {
$found = 1;
last;
}
}
if (not $found) {
if ($sname) {
# If suite was part of name, find it there, may come with combinations
my @this_case = collect_one_suite($sname, [$tname]);
# If a test is specified multiple times on the command line, all
# instances of the test need to be picked. Hence, such tests are
# stored in the temporary array instead of adding them to $cases
# directly so that repeated tests are not run only once
if (@this_case) {
push(@$cmdline_cases, @this_case);
} else {
mtr_error("Could not find '$tname' in '$sname' suite");
}
} else {
if (!$opt_reorder) {
# If --no-reorder is passed and if suite was not part of name,
# search in all the suites
foreach my $suite (split(",", $suites)) {
my @this_case = collect_one_suite($suite, [$tname]);
if (@this_case) {
push(@$cmdline_cases, @this_case);
$found = 1;
}
@this_case = collect_one_suite("i_" . $suite, [$tname]);
if (@this_case) {
push(@$cmdline_cases, @this_case);
$found = 1;
}
}
}
if (!$found) {
mtr_error("Could not find '$tname' in '$suites' suite(s)");
}
}
}
}
# Add test cases collected in the temporary array to the one
# containing all previously collected test cases
push(@$cases, @$cmdline_cases) if $cmdline_cases;
}
if ($opt_reorder && !$quick_collect) {
# Reorder the test cases in an order that will make them faster to
# run. Make a mapping of test name to a string that represents how
# that test should be sorted among the other tests. Put the most
# important criterion first, then a sub-criterion, then
# sub-sub-criterion, etc.
foreach my $tinfo (@$cases) {
my @criteria = ();
# Append the criteria for sorting, in order of importance.
push(@criteria, "ndb=" . ($tinfo->{'ndb_test'} ? "A" : "B"));
push(@criteria, $tinfo->{template_path});
# Group test with equal options together. Ending with "~" makes
# empty sort later than filled.
my $opts = $tinfo->{'master_opt'} ? $tinfo->{'master_opt'} : [];
push(@criteria, join("!", sort @{$opts}) . "~");
# Add slave opts if any
if ($tinfo->{'slave_opt'}) {
push(@criteria, join("!", sort @{ $tinfo->{'slave_opt'} }));
}
# This sorts tests with force-restart *before* identical tests
push(@criteria, $tinfo->{force_restart} ? "force-restart" : "no-restart");
$tinfo->{criteria} = join(" ", @criteria);
}
@$cases = sort { $a->{criteria} cmp $b->{criteria}; } @$cases;
}
# When $opt_repeat > 1 and $opt_parallel > 1, duplicate each test
# $opt_repeat number of times to allow them running in parallel.
if ($::opt_repeat > 1 and $::opt_parallel > 1) {
$cases = duplicate_test_cases($cases);
}
if (defined $print_testcases) {
print_testcases(@$cases);
exit(1);
}
return $cases;
}
# Duplicate each test $opt_repeat number of times
sub duplicate_test_cases($) {
my $tests = shift;
my $new_tests;
foreach my $test (@$tests) {
# Don't repeat the test if 'skip' flag is enabled.
if ($test->{'skip'}) {
push(@{$new_tests}, $test);
} else {
for (my $i = 1 ; $i <= $::opt_repeat ; $i++) {
# Create a duplicate test object
push(@{$new_tests}, create_duplicate_test($test));
}
}
}
return $new_tests;
}
# Create a new test object identical to the original one.
sub create_duplicate_test($) {
my $test = shift;
my $new_test = My::Test->new();
while (my ($key, $value) = each(%$test)) {
if (ref $value eq "ARRAY") {
push(@{ $new_test->{$key} }, @$value);
} else {
$new_test->{$key} = $value;
}
}
return $new_test;
}
# Returns (suitename, testname, extension)
sub split_testname {
my ($test_name) = @_;
# If .test file name is used, get rid of directory part
$test_name = basename($test_name) if $test_name =~ /\.test$/;
# Now split name on .'s
my @parts = split(/\./, $test_name);
if (@parts == 1) {
# Only testname given, ex: alias
return (undef, $parts[0], undef);
} elsif (@parts == 2) {
# Either testname.test or suite.testname given.
# E.g. main.alias or alias.test
if ($parts[1] eq "test") {
return (undef, $parts[0], $parts[1]);
} else {
return ($parts[0], $parts[1], undef);
}
} elsif (@parts == 3) {
# Fully specified suitename.testname.test, E.g. main.alias.test
return ($parts[0], $parts[1], $parts[2]);
}
mtr_error("Illegal format of test name: $test_name");
}
# Read a combination file and return an array of all possible
# combinations.
sub combinations_from_file($) {
my $combination_file = shift;
my @combinations;
# Read combinations file in my.cnf format
mtr_verbose("Read combinations file $combination_file.");
my $config = My::Config->new($combination_file);
foreach my $group ($config->groups()) {
my $comb = {};
$comb->{name} = $group->name();
foreach my $option ($group->options()) {
push(@{ $comb->{comb_opt} }, $option->option());
}
push(@combinations, $comb);
}
return @combinations;
}
# Fetch different combinations specified on command line using
# --combination option.
sub combinations_from_command_line($) {
my @opt_combinations = shift;
# Take the combination from command-line
mtr_verbose("Take the combination from command line");
# Collect all combinations
my @combinations;
foreach my $combination (@::opt_combinations) {
my $comb = {};
$comb->{name} = $combination;
push(@{ $comb->{comb_opt} }, $combination);
push(@combinations, $comb);
}
return @combinations;
}
# Create a new test object for each combination and return an array
# containing all new tests.
sub create_test_combinations($$) {
my $test = shift;
my $combinations = shift;
my @new_cases;
foreach my $comb (@{$combinations}) {
# Skip this combination if the values it provides already are set
# in master_opt or slave_opt.
if (My::Options::is_set($test->{master_opt}, $comb->{comb_opt}) ||
My::Options::is_set($test->{slave_opt}, $comb->{comb_opt}) ||
My::Options::is_set(\@::opt_extra_bootstrap_opt, $comb->{comb_opt}) ||
My::Options::is_set(\@::opt_extra_mysqld_opt, $comb->{comb_opt})) {
next;
}
# Create a new test object.
my $new_test = My::Test->new();
# Copy the original test options.
while (my ($key, $value) = each(%$test)) {
if (ref $value eq "ARRAY") {
push(@{ $new_test->{$key} }, @$value);
} else {
$new_test->{$key} = $value;
}
}
# Append the combination options to master_opt and slave_opt
push(@{ $new_test->{master_opt} }, @{ $comb->{comb_opt} })
if defined $comb->{comb_opt};
push(@{ $new_test->{slave_opt} }, @{ $comb->{comb_opt} })
if defined $comb->{comb_opt};
# Add combination name
$new_test->{combination} = $comb->{name};
# Add the new test to list of new test cases
push(@new_cases, $new_test);
}
return @new_cases;
}
# Look through one test suite for tests named in the second parameter,
# or all tests in the suite if that list is empty.
#
# It starts by finding the proper directory, then handle any
# disabled.def file found. Then loop over either the list of test names
# or all files in the suite directory ending with ".test", adding them
# using collect_one_test_case().
#
# After the list is complete (in local variable @cases), check if
# combinations are being used, from command line or from a combination
# file. If so, start building a new list of test cases in @new_cases
# using those combinations, then assigns that over to @cases.
sub collect_one_suite($$$$) {
my $suite = shift; # Test suite name
my $opt_cases = shift;
my $opt_skip_test_list = shift;
my $disabled = shift;
my @cases; # Array of hash
mtr_verbose("Collecting: $suite");
# Default suite(i.e main suite) directory location
my $suitedir = "$::glob_mysql_test_dir";
if ($suite ne "main") {
# Allow suite to be path to "some dir" if $suite has at least
# one directory part
if (-d $suite and splitdir($suite) > 1) {
$suitedir = $suite;
mtr_report(" - from '$suitedir'");
} else {
$suitedir = get_suite_dir($suite);
return unless $suitedir;
}
mtr_verbose("suitedir: $suitedir");
}
my $resdir = "$suitedir/r";
my $testdir = "$suitedir/t";
# Check if test directory ('t/') exists
if (-d $testdir) {
# Test directory exists
if (-d $resdir) {
# Result directory ('r/') exists
} else {
# Result directory doesn't exist, use test dorectory as
# result directory
$resdir = $testdir;
}
} else {
# Test directory doesn't exists, so there can't be a resullt
# directory as well.
mtr_error("Can't have r/ dir without t/") if -d $resdir;
# No t/ or r/ => use suitedir
$resdir = $testdir = $suitedir;
}
mtr_verbose("testdir: $testdir");
mtr_verbose("resdir: $resdir");
# Read suite.opt file
my $suite_opt_file = "$testdir/suite.opt";
if ($::opt_suite_opt) {
$suite_opt_file = "$testdir/$::opt_suite_opt";
}
my $suite_opts = [];
if (-f $suite_opt_file) {
$suite_opts = opts_from_file($suite_opt_file);
}
if (@$opt_cases) {
# Collect in specified order
foreach my $test_name_spec (@$opt_cases) {
my ($sname, $tname, $extension) = split_testname($test_name_spec);
# Check correct suite if suitename is defined
next if (defined $sname and $suite ne $sname);
if (defined $extension) {
my $full_name = "$testdir/$tname.$extension";
# Extension was specified, check if the test exists
if (!-f $full_name) {
# This is only an error if suite was specified, otherwise it
# could exist in another suite
mtr_error("Test '$full_name' was not found in suite '$sname'")
if $sname;
next;
}
} else {
# No extension was specified, use default
$extension = "test";
my $full_name = "$testdir/$tname.$extension";
# Test not found here, could exist in other suite
next if (!-f $full_name);
}
push(@cases,
collect_one_test_case($suitedir, $testdir,
$resdir, $suite,
$tname, "$tname.$extension",
$disabled, $suite_opts
));
}
} else {
opendir(TESTDIR, $testdir) or mtr_error("Can't open dir \"$testdir\": $!");
foreach my $elem (sort readdir(TESTDIR)) {
my $tname = mtr_match_extension($elem, 'test');
next unless defined $tname;
# Skip tests that does not match the --do-test= filter
next if ($do_test_reg and not $tname =~ /$do_test_reg/o);
push(@cases,
collect_one_test_case($suitedir, $testdir, $resdir,
$suite, $tname, $elem,
$disabled, $suite_opts
));
}
closedir TESTDIR;
}
# Return empty list if no testcases found
return if (@cases == 0);
# Read combinations for this suite and build testcases x
# combinations if any combinations exists.
if (!$::opt_skip_combinations && !$quick_collect) {
my @combinations;
if (@::opt_combinations) {
@combinations = combinations_from_command_line(@::opt_combinations);
} else {
my $combination_file = "$suitedir/combinations";
@combinations = combinations_from_file($combination_file)
if -f $combination_file;
}
if (@combinations) {
my @new_cases;
mtr_report(" - Adding combinations for $suite");
foreach my $test (@cases) {
next if ($test->{'skip'} or defined $test->{'combination'});
push(@new_cases, create_test_combinations($test, \@combinations));
}
# Add the plain test if it was not already added as part
# of a combination.
my %added;
foreach my $new_test (@new_cases) {
$added{ $new_test->{name} } = 1;
}
foreach my $test (@cases) {
push(@new_cases, $test) unless $added{ $test->{name} };
}
@cases = @new_cases;
}
}
optimize_cases(\@cases);
return @cases;
}
# Find location of the specified suite
sub get_suite_dir($) {
my $suite = shift;
return $::glob_mysql_test_dir if ($suite eq "main");
return my_find_dir(
$::basedir,
[ "internal/cloud/mysql-test/suite/",
"internal/mysql-test/suite/",
"internal/plugin/$suite/tests",
"lib/mysql-test/suite",
"mysql-test/suite",
# Look in plugin specific suite dir
"plugin/$suite/tests",
"share/mysql-test/suite",
],
[ $suite, "mtr" ],
# Allow reference to non-existing suite in PB2
($suite =~ /^i_/ || defined $ENV{PB2WORKDIR}));
}
# Set the skip flag in test object and add the skip comments.
sub skip_test {
my $tinfo = shift;
my $comment = shift;
$tinfo->{'skip'} = 1;
$tinfo->{'comment'} = $comment;
}
# Loop through the list of test cases and marks those that should be
# skipped if incompatible with what the server supports etc or update
# the test settings if necessary.
sub optimize_cases {
my ($cases) = @_;
foreach my $tinfo (@$cases) {
# Skip processing if already marked as skipped
next if $tinfo->{skip};
# If a binlog format was set with '--mysqld=--binlog-format=x',
# skip all tests that doesn't support it.
if (defined $binlog_format) {
# Fixed '--binlog-format=x' specified on command line
if (defined $tinfo->{'binlog_formats'}) {
# The test supports different binlog formats check if the
# selected one is ok.
my $supported =
grep { $_ eq lc $binlog_format } @{ $tinfo->{'binlog_formats'} };
if (!$supported) {
$tinfo->{'skip'} = 1;
$tinfo->{'comment'} =
"Doesn't support --binlog-format='$binlog_format'";
}
}
} else {
# =======================================================
# Use dynamic switching of binlog format
# =======================================================
# Get binlog-format used by this test from master_opt
my $test_binlog_format;
foreach my $opt (@{ $tinfo->{master_opt} }) {
(my $dash_opt = $opt) =~ s/_/-/g;
$test_binlog_format =
mtr_match_prefix($dash_opt, "--binlog-format=") ||
$test_binlog_format;
}
if (defined $test_binlog_format and
defined $tinfo->{binlog_formats}) {
my $supported =
grep { My::Options::option_equals($_, lc $test_binlog_format) }
@{ $tinfo->{'binlog_formats'} };
if (!$supported) {
skip_test($tinfo,
"Doesn't support --binlog-format = '$test_binlog_format'");
next;
}
}
}
# Check that engine set by '--default-storage-engine=<engine>' is supported
my %builtin_engines = ('myisam' => 1, 'memory' => 1, 'csv' => 1);
foreach my $opt (@{ $tinfo->{master_opt} }) {
(my $dash_opt = $opt) =~ s/_/-/g;
my $default_engine =
mtr_match_prefix($dash_opt, "--default-storage-engine=");
my $default_tmp_engine =
mtr_match_prefix($dash_opt, "--default-tmp-storage-engine=");
# Allow use of uppercase, convert to all lower case
$default_engine =~ tr/A-Z/a-z/;
$default_tmp_engine =~ tr/A-Z/a-z/;
if (defined $default_engine) {
if (!exists $::mysqld_variables{$default_engine} and
!exists $builtin_engines{$default_engine}) {
skip_test($tinfo, "'$default_engine' not supported");
}
$tinfo->{'ndb_test'} = 1
if ($default_engine =~ /^ndb/i);
$tinfo->{'myisam_test'} = 1
if ($default_engine =~ /^myisam/i);
}
if (defined $default_tmp_engine) {
if (!exists $::mysqld_variables{$default_tmp_engine} and
!exists $builtin_engines{$default_tmp_engine}) {
skip_test($tinfo, "'$default_tmp_engine' not supported");
}
$tinfo->{'ndb_test'} = 1
if ($default_tmp_engine =~ /^ndb/i);
$tinfo->{'myisam_test'} = 1
if ($default_tmp_engine =~ /^myisam/i);
}
}
if ($quick_collect && !$tinfo->{'skip'}) {
$some_test_found = 1;
return;
}
}
}
# Read options from a test options file and append them as an array
# to $tinfo->{$opt_name} member. Some options however have a different
# semantics for MTR instead of just being passed to the server.
# These are being handled here as well.
sub process_opts_file {
my ($tinfo, $opt_file, $opt_name) = @_;
if (-f $opt_file) {
my $opts = opts_from_file($opt_file);
foreach my $opt (@$opts) {
my $value;
# The opt file is used both to send special options to the
# mysqld as well as pass special test case specific options to
# this script.
$value = mtr_match_prefix($opt, "--timezone=");
if (defined $value) {
$tinfo->{'timezone'} = $value;
next;
}
$value = mtr_match_prefix($opt, "--result-file=");
if (defined $value) {
# Specifies the file mysqltest should compare output against.
$tinfo->{'result_file'} = "r/$value.result";
next;
}
$value = mtr_match_prefix($opt, "--config-file-template=");
if (defined $value) {
# Specifies the configuration file to use for this test
$tinfo->{'template_path'} = dirname($tinfo->{path}) . "/$value";
next;
}
# If we set default time zone, remove the one we have
$value = mtr_match_prefix($opt, "--default-time-zone=");
if (defined $value) {
# Set timezone for this test case to something different
$tinfo->{'timezone'} = "GMT-8";
# Fallthrough, add the --default-time-zone option
}
# The --restart option forces a restart even if no special option
# is set. If the options are the same as next testcase there is no
# need to restart after the testcase has completed.
if ($opt eq "--force-restart") {
$tinfo->{'force_restart'} = 1;
next;
}
$value = mtr_match_prefix($opt, "--testcase-timeout=");
if (defined $value) {
# Overrides test case timeout for this test
$tinfo->{'case-timeout'} = $value;
next;
}
if ($opt eq "--nowarnings") {
$tinfo->{'skip_check_warnings'} = 1;
next;
}
# Ok, this was a real option, add it
push(@{ $tinfo->{$opt_name} }, $opt);
}
}
}
# Collect all necessary information about a single test and returns
# a new My::Test instance to be added to the global list.
#
# Most of the code is about finding and setting various test specific
# options or properties. Subroutine tags_from_test_file() uses a list
# of fixed patterns called "tags" to set the properties for a test
# case.
sub collect_one_test_case {
my $suitedir = shift;
my $testdir = shift;
my $resdir = shift;
my $suitename = shift;
my $tname = shift;
my $filename = shift;
my $disabled = shift;
my $suite_opts = shift;
# Test file name should consist of only alpha-numeric characters, dash (-)
# or underscore (_), but should not start with dash or underscore.
if ($tname !~ /^[^_\W][\w-]*$/) {
die("Invalid test file name '$suitename.$tname'. Test file " .
"name should consist of only alpha-numeric characters, " .
"dash (-) or underscore (_), but should not start with " .
"dash or underscore.");
}
# Check --start-from
if ($start_from) {
# start_from can be specified as [suite.].testname_prefix
my ($suite, $test, $ext) = split_testname($start_from);
# Skip silently
return if (($suite and $suitename lt $suite) || ($tname lt $test));
}
# Set defaults
my $tinfo = My::Test->new(name => "$suitename.$tname",
path => "$testdir/$filename",
shortname => $tname,);
my $result_file = "$resdir/$tname.result";
if (-f $result_file) {
$tinfo->{result_file} = $result_file;
} else {
# Result file doesn't exist
if ($::opt_check_testcases and !$::opt_record) {
# Set 'no_result_file' flag if check-testcases is enabled.
$tinfo->{'no_result_file'} = $result_file;
} else {
# No .result file exist, remember the path where it should
# be saved in case of --record.
$tinfo->{record_file} = $result_file;
}
}
# Disable quiet output when a test file doesn't contain a result file
# and --record option is disabled.
if ($::opt_quiet and defined $tinfo->{record_file} and !$::opt_record) {
$::opt_quiet = 0;
mtr_report("Turning off '--quiet' option since the MTR run contains " .
"a test without a result file.");
}
# Skip some tests but include in list, just mark them as skipped
if ($skip_test_reg and $tname =~ /$skip_test_reg/o) {
$tinfo->{'skip'} = 1;
return $tinfo;
}
# Check for replicaton tests
$tinfo->{'rpl_test'} = 1 if ($suitename =~ 'rpl');
# Check for replication tests which need GTID mode OFF
$tinfo->{'rpl_nogtid_test'} = 1 if ($suitename =~ 'rpl_nogtid');
# Check for replication tests which need GTID mode ON
$tinfo->{'rpl_gtid_test'} = 1 if ($suitename =~ 'rpl_gtid');
# Check for group replication tests
$tinfo->{'grp_rpl_test'} = 1 if ($suitename =~ 'group_replication');
# Check for disabled tests
if ($disabled->{"$suitename.$tname"}) {
# Test was marked as disabled in disabled.def file
$tinfo->{'comment'} = $disabled->{"$suitename.$tname"};
if ($enable_disabled or @::opt_cases) {
# User has selected to run all disabled tests
mtr_report(" - Running test $tinfo->{name} even though it's been",
"disabled due to '$tinfo->{comment}'.");
} else {
$tinfo->{'skip'} = 1;
# Disable the test case
$tinfo->{'disable'} = 1;
return $tinfo;
}
}
# Append suite extra options to both master and slave
push(@{ $tinfo->{'master_opt'} }, @$suite_opts);
push(@{ $tinfo->{'slave_opt'} }, @$suite_opts);
# Check for test specific config file
my $test_cnf_file = "$testdir/$tname.cnf";
if (-f $test_cnf_file) {
# Specifies the configuration file to use for this test
$tinfo->{'template_path'} = $test_cnf_file;
}
# master sh
my $master_sh = "$testdir/$tname-master.sh";
if (-f $master_sh) {
if (IS_WIN32PERL) {
$tinfo->{'skip'} = 1;
$tinfo->{'comment'} = "No tests with sh scripts on Windows";
return $tinfo;
} else {
$tinfo->{'master_sh'} = $master_sh;
}
}
# slave sh
my $slave_sh = "$testdir/$tname-slave.sh";
if (-f $slave_sh) {
if (IS_WIN32PERL) {
$tinfo->{'skip'} = 1;
$tinfo->{'comment'} = "No tests with sh scripts on Windows";
return $tinfo;
} else {
$tinfo->{'slave_sh'} = $slave_sh;
}
}
# <tname>.slave-mi
mtr_error("$tname: slave-mi not supported anymore")
if (-f "$testdir/$tname.slave-mi");
tags_from_test_file($tinfo, "$testdir/${tname}.test");
# Check that test wth "ndb" in their suite name
# have been tagged as 'ndb_test', this is normally fixed
# by adding a "source include/have_ndb.inc" to the file
if ($suitename =~ /ndb/) {
mtr_error("The test '$tinfo->{name}' is not tagged as 'ndb_test'")
unless ($tinfo->{'ndb_test'});
}
# Disable the result file check for NDB tests not having its
# corresponding result file.
if ($tinfo->{'ndb_test'} and $tinfo->{'ndb_no_result_file_test'}) {
delete $tinfo->{'no_result_file'} if $tinfo->{'no_result_file'};
}
if (defined $default_storage_engine) {
# Different default engine is used tag test to require that engine.
$tinfo->{'ndb_test'} = 1
if ($default_storage_engine =~ /^ndb/i);
$tinfo->{'mysiam_test'} = 1
if ($default_storage_engine =~ /^mysiam/i);
}
# Skip non-parallel tests if 'non-parallel-test' option is disabled
if ($tinfo->{'not_parallel'} and !$::opt_non_parallel_test) {
skip_test($tinfo, "Test needs 'non-parallel-test' option");
return $tinfo;
}
# Except the tests which need big-test or only-big-test option to run
# in valgrind environment(i.e tests having no_valgrind_without_big.inc
# include file), other normal/non-big tests shouldn't run with
# only-big-test option.
if ($::opt_only_big_test) {
if ((!$tinfo->{'no_valgrind_without_big'} and !$tinfo->{'big_test'}) or
($tinfo->{'no_valgrind_without_big'} and !$::opt_valgrind)) {
skip_test($tinfo, "Not a big test");
return $tinfo;
}
}
# Check for big test
if ($tinfo->{'big_test'} and !($::opt_big_test or $::opt_only_big_test)) {
skip_test($tinfo, "Test needs 'big-test' or 'only-big-test' option.");
return $tinfo;
}
# Tests having no_valgrind_without_big.inc include file needs either
# big-test or only-big-test option to run in valgrind environment.
if ($tinfo->{'no_valgrind_without_big'} and $::opt_valgrind) {
if (!$::opt_big_test and !$::opt_only_big_test) {
skip_test($tinfo,
"Need '--big-test' or '--only-big-test' when running " .
"with Valgrind.");
return $tinfo;
}
}
if ($tinfo->{'need_debug'} && !$::debug_compiled_binaries) {
skip_test($tinfo, "Test needs debug binaries.");
return $tinfo;
}
if ($tinfo->{'ndb_test'}) {
# This is a NDB test
if ($::ndbcluster_enabled == 0) {
# ndbcluster is disabled
skip_test($tinfo, "ndbcluster disabled");
return $tinfo;
}
} else {
# This is not a ndb test
if ($opt_with_ndbcluster_only) {
# Only the ndb test should be run, all other should be skipped
skip_test($tinfo, "Only ndbcluster tests");
return $tinfo;
}
}
if ($tinfo->{'federated_test'}) {
# This is a test that needs federated, enable it
push(@{ $tinfo->{'master_opt'} }, "--loose-federated");
push(@{ $tinfo->{'slave_opt'} }, "--loose-federated");
}
if ($tinfo->{'myisam_test'}) {
# This is a temporary fix to allow non-innodb tests to run even if
# the default storage engine is innodb.
push(@{ $tinfo->{'master_opt'} }, "--default-storage-engine=MyISAM");
push(@{ $tinfo->{'slave_opt'} }, "--default-storage-engine=MyISAM");
push(@{ $tinfo->{'master_opt'} }, "--default-tmp-storage-engine=MyISAM");
push(@{ $tinfo->{'slave_opt'} }, "--default-tmp-storage-engine=MyISAM");
}
if ($tinfo->{'need_binlog'}) {
if (grep(/^--skip[-_]log[-_]bin/, @::opt_extra_mysqld_opt)) {
skip_test($tinfo, "Test needs binlog");
return $tinfo;
}
}
if ($tinfo->{'rpl_test'} or $tinfo->{'grp_rpl_test'}) {
if ($skip_rpl) {
skip_test($tinfo, "No replication tests, --skip-rpl is enabled.");
return $tinfo;
}
}
# Check for group replication tests
$group_replication = 1 if ($tinfo->{'grp_rpl_test'});
if ($tinfo->{'not_windows'} && IS_WINDOWS) {
skip_test($tinfo, "Test not supported on Windows");
return $tinfo;
}
# Find config file to use if not already selected in <testname>.opt file
if (defined $defaults_file) {
# Using same config file for all tests
$defaults_file = get_bld_path($defaults_file);
$tinfo->{template_path} = $defaults_file;
} elsif (!$tinfo->{template_path}) {
my $config = "$suitedir/my.cnf";
if (!-f $config) {
# Assume default.cnf will be used
$config = "include/default_my.cnf";
# rpl_gtid tests must use their suite's cnf file having gtid mode on.
if ($tinfo->{rpl_gtid_test}) {
$config = "suite/rpl_gtid/my.cnf";
}
# rpl_nogtid tests must use their suite's cnf file having gtid mode off.
elsif ($tinfo->{rpl_nogtid_test}) {
$config = "suite/rpl_nogtid/my.cnf";
}
# rpl tests must use their suite's cnf file.
elsif ($tinfo->{rpl_test}) {
$config = "suite/rpl/my.cnf";
}
}
$tinfo->{template_path} = $config;
}
# Set extra config file to use
if (defined $defaults_extra_file) {
$defaults_extra_file = get_bld_path($defaults_extra_file);
$tinfo->{extra_template_path} = $defaults_extra_file;
}
if (!$::start_only or @::opt_cases) {
# Add master opts, extra options only for master
process_opts_file($tinfo, "$testdir/$tname-master.opt", 'master_opt');
# Add slave opts, list of extra option only for slave
process_opts_file($tinfo, "$testdir/$tname-slave.opt", 'slave_opt');
}
if (!$::start_only) {
# Add client opts, extra options only for mysqltest client
process_opts_file($tinfo, "$testdir/$tname-client.opt", 'client_opt');
}
my @new_tests;
# Check if test specific combination file (<test_name>.combinations)
# exists, if yes read different combinations from it.
if (!$::opt_skip_combinations && !$quick_collect) {
if (!$tinfo->{'skip'}) {
my $combination_file = "$testdir/$tname.combinations";
# Check for test specific combination file
my @combinations = combinations_from_file($combination_file)
if -f $combination_file;
# Create a new test object for each combination
if (@combinations) {
mtr_report(" - Adding combinations for test '$tname'");
push(@new_tests, create_test_combinations($tinfo, \@combinations));
}
}
}
@new_tests ? return @new_tests : return $tinfo;
}
# List of tags in the .test files that if found should set
# the specified value in "tinfo"
my @tags = (
[ "include/have_binlog_format_row.inc", "binlog_formats", ["row"] ],
[ "include/have_binlog_format_statement.inc", "binlog_formats",
["statement"]
],
[ "include/have_binlog_format_mixed.inc", "binlog_formats",
[ "mixed", "mix" ]
],
[ "include/have_binlog_format_mixed_or_row.inc", "binlog_formats",
[ "mixed", "mix", "row" ]
],
[ "include/have_binlog_format_mixed_or_statement.inc", "binlog_formats",
[ "mixed", "mix", "statement" ]
],
[ "include/have_binlog_format_row_or_statement.inc", "binlog_formats",
[ "row", "statement" ]
],
[ "include/have_log_bin.inc", "need_binlog", 1 ],
# An empty file to use test that needs myisam engine.
[ "include/force_myisam_default.inc", "myisam_test", 1 ],
[ "include/big_test.inc", "big_test", 1 ],
[ "include/have_debug.inc", "need_debug", 1 ],
[ "include/have_ndb.inc", "ndb_test", 1 ],
[ "include/have_multi_ndb.inc", "ndb_test", 1 ],
# Any test sourcing the below inc file is considered to be an NDB
# test not having its corresponding result file.
[ "include/ndb_no_result_file.inc", "ndb_no_result_file_test", 1 ],
# The tests with below four .inc files are considered to be rpl tests.
[ "include/rpl_init.inc", "rpl_test", 1 ],
[ "include/rpl_ip_mix.inc", "rpl_test", 1 ],
[ "include/rpl_ip_mix2.inc", "rpl_test", 1 ],
[ "include/rpl_ipv6.inc", "rpl_test", 1 ],
[ "include/ndb_master-slave.inc", "ndb_test", 1 ],
[ "federated.inc", "federated_test", 1 ],
[ "include/not_windows.inc", "not_windows", 1 ],
[ "include/not_parallel.inc", "not_parallel", 1 ],
# Tests with below .inc file are considered to be group replication tests
[ "have_group_replication_plugin_base.inc", "grp_rpl_test", 1 ],
[ "have_group_replication_plugin.inc", "grp_rpl_test", 1 ],
# Tests with below .inc file needs either big-test or only-big-test
# option along with valgrind option.
[ "include/no_valgrind_without_big.inc", "no_valgrind_without_big", 1 ]);
sub tags_from_test_file {
my $tinfo = shift;
my $file = shift;
my $F = IO::File->new($file) or mtr_error("can't open file \"$file\": $!");
while (my $line = <$F>) {
# Skip line if it start's with '#'
next if ($line =~ /^#/);
# Match this line against tag in "tags" array
foreach my $tag (@tags) {
if (index($line, $tag->[0]) >= 0) {
# Tag matched, assign value to "tinfo"
$tinfo->{"$tag->[1]"} = $tag->[2];
}
}
# If test sources another file, open it as well
if ($line =~ /^\-\-([[:space:]]*)source(.*)$/ or
$line =~ /^([[:space:]]*)source(.*);$/) {
my $value = $2;
$value =~ s/^\s+//; # Remove leading space
$value =~ s/[[:space:]]+$//; # Remove ending space
# Sourced file may exist relative to test or in global location
foreach my $sourced_file (dirname($file) . "/$value",
"$::glob_mysql_test_dir/$value") {
if (-f $sourced_file) {
# Only source the file if it exists, we may get false
# positives in the regexes above if someone writes
# "source nnnn;" in a test case(such as mysqltest.test).
tags_from_test_file($tinfo, $sourced_file);
last;
}
}
}
}
}
sub unspace {
my $string = shift;
my $quote = shift;
$string =~ s/[ \t]/\x11/g;
return "$quote$string$quote";
}
sub opts_from_file ($) {
my $file = shift;
my $file_handle = IO::File->new($file, '<') or
mtr_error("Can't open file '$file': $!");
my @args;
while (<$file_handle>) {
# Skip a line if it starts with '#' (i.e comments)
next if /^\s*#/;
chomp;
s/^\s+//; # Remove leading space
s/\s+$//; # Remove ending space
# This is strange, but we need to fill whitespace inside quotes
# with something, to remove later. We do this to be able to split
# on space. Else, we have trouble with options like
# --someopt="--insideopt1 --insideopt2". But still with this, we
# are not 100% sure it is right, we need a shell to do it right.
s/\'([^\'\"]*)\'/unspace($1,"\x0a")/ge;
s/\"([^\'\"]*)\"/unspace($1,"\x0b")/ge;
s/\'([^\'\"]*)\'/unspace($1,"\x0a")/ge;
s/\"([^\'\"]*)\"/unspace($1,"\x0b")/ge;
foreach my $arg (split(/[ \t]+/)) {
$arg =~ tr/\x11\x0a\x0b/ \'\"/;
# Put back real chars, the outermost quotes has to go
$arg =~ s/^([^\'\"]*)\'(.*)\'([^\'\"]*)$/$1$2$3/ or
$arg =~ s/^([^\'\"]*)\"(.*)\"([^\'\"]*)$/$1$2$3/;
$arg =~ s/\\\\/\\/g;
# Do not pass empty string since my_getopt is not capable to
# handle it.
push(@args, $arg) if (length($arg));
}
}
$file_handle->close();
return \@args;
}
sub print_testcases {
my (@cases) = @_;
print "=" x 60, "\n";
foreach my $test (@cases) {
$test->print_test();
}
print "=" x 60, "\n";
}
1;