use XML::SAX::ParserFactory;
use XML::Validator::Schema;
use XML::Simple qw(:strict);
use Data::Printer;
use Data::Dumper;
use Config::Auto;
use Modern::Perl; # say
use Test::More;
use IO::All;
=begin # tests:
tests the genomics_xml.pl script - generates dummy data in lims_test and provides
overrides for location/name of archive & prevents ftp transfer of file
=cut
BEGIN { require 't/test-lib.pl' }
my $dbix = get_dbix();
my $cfg = get_config(); # p $cfg;
my $db = 'lims_test';
# for validation:
my $xsd = 'genomics.xsd'; # current xsd - synlinked to most recent file
my $xsd_src = $cfg->{path_to_app_root} . '/setup/schema/xsd/' . $xsd; # p $xsd_src;
create_genomics_data(); # drop & recreate tables/data
{ # execute genomics_xml.pl script:
my $script = '/script/crons/multi/genomics_xml.pl';
my @args = ( -d => $db, -t => 10 ); # override dbname, set duration 10 seconds
system ( $cfg->{path_to_app_root} . $script, @args );
}
# for validation tests:
my $xml_src = io("/tmp/${db}.xml")->slurp; # p $xml_src;
# unzip archive and extract to $xml (not using gzip any more):
# gunzip "/tmp/${db}.gz" => \my $xml or die "gunzip failed: $GunzipError\n"; # p $xml;
my $genomics_uat = do {
my $cfg_file = $cfg->{path_to_app_root} . '/script/crons/lib/settings.cfg'; # p $cfg_file;
Config::Auto::parse($cfg_file)->{genomics}->{uat};
}; # p $genomics_uat;
{ # compare src data and parsed xml hashrefs:
my $ref = XMLin("/tmp/${db}.xml", ForceArray => 0, KeyAttr => []); # p $ref->{Record};
# read Data::Dumper txt back into hashref format:
my $src = do "/tmp/${db}.txt"; # p $src; # http://www.perlmonks.org/?node_id=304068
# if Consent / Files exist, convert hashref element in $ref to an arrayref to match $src:
if ( my $r = $ref->{Record}->{Consent}->{Files} ) { # p $ref;
my $str = $r->{file}; # p $str; # scalar value - should be arrayref to match $src
$ref->{Record}->{Consent}->{Files}->{file} = [ $str ]; # scalar value to arrayref
}
is_deeply($src, $ref->{Record}, 'data structures identical');
{ # check Rare disease has expected participant/GeL ID:
my $allocated_id = $ref->{Record}->{Demographics}->{participant_id};
my $expected_id = $genomics_uat->{rare_disease} + 1;
is($allocated_id, $expected_id, 'expected GeL ID for rare disease type');
}
}
# change some xml params and check validation failure:
{
# decimal on xs:short:
validate('edta1_fluidx_vol', 175.5);
# string on xs:int:
validate('family_id', 'string');
# invalid date:
validate('consent_date', '2/4/2016');
# invalid datetime:
validate('edta1_qc_date', '2/4/2016 10.00');
# exceed max value for xs:int (2147483647):
validate('family_id', 2_147_483_647 + 1);
# exceed max value for xs:short (32767):
validate('volume_stored', 32_767 + 1);
# exceed max value for xs:byte (127):
validate('total_samples', 127 + 1);
}
# delete temp files:
io("/tmp/${db}.$_")->unlink for qw(txt xml);
done_testing(9);
sub create_genomics_data { # create new genomics data:
my $dbh = $dbix->dbh;
# drop lab_tests & lab_sections, re-create from genomics db:
for ( qw/lab_tests lab_sections screens screen_category/ ) {
$dbh->do(qq!TRUNCATE TABLE `$_`!) ;
$dbh->do(qq!INSERT INTO `$_` SELECT * FROM `genomics`.`$_`!);
}
$dbh->do(qq!TRUNCATE TABLE `$_`!) for qw(request_lab_test_results
storage_racks request_initial_screen request_history request_storage);
{ # test/results:
my $lab_tests = $dbix->select('lab_tests', [qw(test_name id)])->map; # p $lab_tests;
my @data = grep chomp $_, <DATA>; # p @data; exit;
# only grep lab_tests with a result (comma followed by 1 or more chars):
my %results = map { split ',' } grep /.+,.+/, @data; # p %results; exit;
$dbix->insert('request_lab_test_results',
{
lab_test_id => $lab_tests->{$_},
request_id => 1,
result => $results{$_},
}
) for keys %results;
}
$dbix->select('screens', 'id', { description => 'Rare disease' })->into(my $id);
$dbix->insert('request_initial_screen', { request_id => 1, screen_id => $id });
$dbix->insert('request_lab_test_history',
{
request_id => 1,
user_id => 1,
action => 'auto-set Consent given status to complete',
}
);
$dbix->insert('request_lab_test_history',
{ request_id => 1, action => 'screened', user_id => 1 });
$dbix->insert('storage_racks', { plateId => 'SR12345' });
for (1..2) {
$dbix->insert('request_storage',
{
vial_location => 'C'.$_,
part_number => '123456789'.$_,
request_id => 1,
vialId => 'FR1122334'.$_,
sample => 'tissue',
source => 'PST',
rack_id => 1,
}
);
}
{ # new crontab user for participant id event history:
my %h = (
user_location_id => 1,
designation => 'crontab',
first_name => 'cron',
last_name => 'tab',
password => 'not_required',
username => 'crontab',
group_id => 2,
email => 'cron@nowhere.com',
);
$dbix->insert('users', \%h);
}
# revert unit number to UNKNOWN to test participant id generation:
$dbix->update('patient_case', { unit_number => 'UNKNOWN'}, { id => 1 });
# delete 2nd request - gets included in query, only able to handle 1 request:
$dbix->delete('requests', { id => 2});
}
sub validate {
my ($param, $val) = @_; # warn $param;
my $xml = $xml_src; # clone it so text substitution localised
$xml =~ s!<$param>(.*)</$param>!<$param>$val</$param>!;
my $result = validate_xml_against_xsd($xml); # p $result;
my $msg = join ' ', $param, 'validated OK';
like ( $result, qr/Illegal value '$val' in element <$param>/, $msg );
}
sub validate_xml_against_xsd {
my $xml = shift;
# switch debug on to show data structure:
my $validator = XML::Validator::Schema->new(file => $xsd_src, debug => 0);
my $parser = XML::SAX::ParserFactory->parser(Handler => $validator);
eval { $parser->parse_string($xml); };
return $@;
}
__DATA__
appointment_date,2016-06-15
approach_date,2016-05-15
approach_followup_date,
approach_method,Letter
biopsy_gauge,
ccft_dna,
clinic_sample_type,
consent_date,2016-06-15
consent_form_version,R1/R2
consent_given,yes
consent_q1,yes
consent_q10,yes
consent_q11,yes
consent_q12,yes
consent_q13,yes
consent_q14,yes
consent_q2,yes
consent_q3,yes
consent_q4,yes
consent_q5,yes
consent_q6,yes
consent_q7,yes
consent_q8,yes
consent_q9,yes
dna_arrived_ccp,2016-10-21 13:20:12
dna_consignment_number,RR813-21-43-16-1
dna_dispatched,2016-10-19 10:00:12
dna_extraction_protocol,
dna_lab_number,16.06016
dna_sent_ccp,2016-10-20 10:22:12
edta1_fluidx_rack_id,SA00165521
edta1_fluidx_rack_well,A5
edta1_fluidx_vol,200
edta1_fluidx,1023912430
edta1_qc_date,2016-10-12 14:20:38
edta1_qc_result,309
edta1_qc_source,
edta1_qc_type,Qubit
edta1,122000059.1
edta2_qc_source,
edta2,122000059.2
excision_margin,
family_id,122000058
ffpe_dissection_details,
ffpe_macrodissected,
first_outcome,Interested
fixation_comments,
fixation_end,
fixation_start,
fixative_type,
formalin_duration,
frozen_dissection_details,
frozen_macrodissected,
handling_protocol,v2
info_sheet_version,R2
local_lab_identifier,
number_of_biopsies,
number_of_blocks,
number_of_sections,
omics_arrived_ccp,
omics_consignment_number,
omics_dispatched,2016-10-21 10:08:12
omics_sent_ccp,
patient_info_given,
paxgene_rna,122000059.5
pre_invasive_elements,
processing_schedule,
prolonged_storage_method,
pst,122000059.4
received_by_cytogenetics,
sample_taken,2016-06-15 13:30:25
second_outcome,
section_thickness,
sent_from_histopathology,
snap_freezing_start,
sst,122000059.3
tissue_source,
total_samples,3
tracking_number,
tumour_fluidx,
tumour_fluidx_rack_id,
tumour_fluidx_rack_well,
tumour_fluidx_vol,
tumour_lab_number,
tumour_qc_date,
tumour_qc_result,
tumour_qc_type,
tumour_sample_id,
tumour_sample_taken,
tumour_sample_type,
tumour_size,
tumour_type,
tumour_volume_stored,
unsent_sample_reason,
volume_stored,267
withdrawal_date,
withdrawal_form,
withdrawal_option,
