| 1 |
723 |
jeremybenn |
# Library of functions for dealing with DejaGNU, or which are otherwise
|
| 2 |
|
|
# generally useful for the DejaGNU tool stack.
|
| 3 |
|
|
#
|
| 4 |
|
|
# Author: Matthew Sachs <msachs@apple.com>
|
| 5 |
|
|
#
|
| 6 |
|
|
# Functions:
|
| 7 |
|
|
# parseLogFile: See "sub parseLogFile" below for details. This function
|
| 8 |
|
|
# returns a detailed parse of a DejaGNU log or sum file.
|
| 9 |
|
|
# ispass: Takes a DejaGNU result (e.g. "PASS", "XPASS") and returns
|
| 10 |
|
|
# true if and only if it is a passing result (PASS, XFAIL, or
|
| 11 |
|
|
# KFAIL.)
|
| 12 |
|
|
#
|
| 13 |
|
|
# Copyright (c) 2006 Free Software Foundation.
|
| 14 |
|
|
#
|
| 15 |
|
|
# This file is part of GCC.
|
| 16 |
|
|
#
|
| 17 |
|
|
# GCC is free software; you can redistribute it and/or modify
|
| 18 |
|
|
# it under the terms of the GNU General Public License as published by
|
| 19 |
|
|
# the Free Software Foundation; either version 3, or (at your option)
|
| 20 |
|
|
# any later version.
|
| 21 |
|
|
#
|
| 22 |
|
|
# GCC is distributed in the hope that it will be useful,
|
| 23 |
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
| 24 |
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
| 25 |
|
|
# GNU General Public License for more details.
|
| 26 |
|
|
#
|
| 27 |
|
|
# You should have received a copy of the GNU General Public License
|
| 28 |
|
|
# along with GCC; see the file COPYING. If not, write to
|
| 29 |
|
|
# the Free Software Foundation, 51 Franklin Street, Fifth Floor,
|
| 30 |
|
|
# Boston, MA 02110-1301, USA.
|
| 31 |
|
|
|
| 32 |
|
|
package dglib;
|
| 33 |
|
|
|
| 34 |
|
|
use strict;
|
| 35 |
|
|
use warnings;
|
| 36 |
|
|
use Exporter;
|
| 37 |
|
|
|
| 38 |
|
|
our @ISA = qw(Exporter);
|
| 39 |
|
|
our @EXPORT = qw(ispass parseLogFile);
|
| 40 |
|
|
|
| 41 |
|
|
use File::Basename;
|
| 42 |
|
|
use POSIX qw(mktime);
|
| 43 |
|
|
|
| 44 |
|
|
|
| 45 |
|
|
# Create a group hierarchy, returning the leaf node
|
| 46 |
|
|
sub mkGroupPath {
|
| 47 |
|
|
my($root, $groups, @newgroups) = @_;
|
| 48 |
|
|
|
| 49 |
|
|
my $parent = $root;
|
| 50 |
|
|
my $fullname = "";
|
| 51 |
|
|
foreach my $group(@newgroups) {
|
| 52 |
|
|
$fullname .= "/" if $fullname;
|
| 53 |
|
|
$fullname .= $group;
|
| 54 |
|
|
if(exists($groups->{$fullname})) {
|
| 55 |
|
|
$parent = $groups->{$fullname};
|
| 56 |
|
|
} else {
|
| 57 |
|
|
my $newgroup = {name => $group, parent => $parent};
|
| 58 |
|
|
$groups->{$fullname} = $newgroup;
|
| 59 |
|
|
$parent->{testgroup} ||= [];
|
| 60 |
|
|
push @{$parent->{testgroup}}, $newgroup;
|
| 61 |
|
|
$parent = $newgroup;
|
| 62 |
|
|
}
|
| 63 |
|
|
}
|
| 64 |
|
|
|
| 65 |
|
|
return $parent;
|
| 66 |
|
|
}
|
| 67 |
|
|
|
| 68 |
|
|
# Extract information from DejaGNU log or sum files.
|
| 69 |
|
|
# Options, if provided, should be a hashref with zero or more of the following keys:
|
| 70 |
|
|
# gccdir:
|
| 71 |
|
|
# Passing in the full path to the root of the gcc/testsuite directory
|
| 72 |
|
|
# will help in the parsing, but if it isn't provided, it will be guessed.
|
| 73 |
|
|
# diagnostics:
|
| 74 |
|
|
# If set to 0, diagnostics will not be returned. This can save a lot
|
| 75 |
|
|
# of memory if you are not using this information.
|
| 76 |
|
|
# fullname:
|
| 77 |
|
|
# If set to 0, the fullname key will not be included in tests.
|
| 78 |
|
|
# Returns a hash with the following keys:
|
| 79 |
|
|
# incomplete: 1 if the summary file appears truncated, otherwise 0
|
| 80 |
|
|
# diagnostics: List of (type, value) for any errors detected. Type can be ERROR, WARNING, or NOTE.
|
| 81 |
|
|
# test: Array of root-level tests, with keys:
|
| 82 |
|
|
# name: Name of the test, relative to the enclosing test group.
|
| 83 |
|
|
# fullname: Fully-qualified name of the test.
|
| 84 |
|
|
# result: DejaGNU result (PASS, FAIL, XPASS, &c)
|
| 85 |
|
|
# detail: For multi-phase (e.g. compile/link/execute), this will be
|
| 86 |
|
|
# the furthest phase which the test was able to attempt,
|
| 87 |
|
|
# so if the result is FAIL and this is "link phase", the test
|
| 88 |
|
|
# compiled but failed to link. This key may contain other
|
| 89 |
|
|
# auxiliary data.
|
| 90 |
|
|
# pseudotest: If 1, this test may not really exist; see "pseudotest" below.
|
| 91 |
|
|
# testgroup: Array of root-level testgroups, with keys:
|
| 92 |
|
|
# name: Name of the group.
|
| 93 |
|
|
# parent: Parent test group.
|
| 94 |
|
|
# test: As per above.
|
| 95 |
|
|
# testgroup: Child test groups.
|
| 96 |
|
|
# compiler: Version string from compiler used to run the tests (if detected)
|
| 97 |
|
|
sub parseLogFile($;$) {
|
| 98 |
|
|
my($logfile, $options) = @_;
|
| 99 |
|
|
$options ||= {};
|
| 100 |
|
|
my $gccdir = $options->{gccdir} || "";
|
| 101 |
|
|
my $return_diags = exists($options->{diagnostics}) ? $options->{diagnostics} : 1;
|
| 102 |
|
|
my $emit_fullname = exists($options->{fullname}) ? $options->{fullname} : 1;
|
| 103 |
|
|
my $is_gdb = 0;
|
| 104 |
|
|
my $gdbhack = "";
|
| 105 |
|
|
|
| 106 |
|
|
my %ret = (incomplete => 1, diagnostics => [], testgroup => []);
|
| 107 |
|
|
my(%testindex, %groupindex);
|
| 108 |
|
|
|
| 109 |
|
|
open(LOGFILE, $logfile) or die "Couldn't open log file $logfile: $!\n";
|
| 110 |
|
|
|
| 111 |
|
|
my($currgroup, $currtest, $lastrun);
|
| 112 |
|
|
$currgroup = \%ret;
|
| 113 |
|
|
|
| 114 |
|
|
my %monmap = (Jan => 0, Feb => 1, Mar => 2, Apr => 3, May => 4, Jun => 5, Jul => 6, Aug => 7, Sep => 8, Oct => 9, Nov => 10, Dec => 11);
|
| 115 |
|
|
|
| 116 |
|
|
# We don't want gccdir matching on an empty string.
|
| 117 |
|
|
$gccdir ||= "this will never match, or my name isn't Reginald St. Croix";
|
| 118 |
|
|
|
| 119 |
|
|
my $line = 1;
|
| 120 |
|
|
while(<LOGFILE>) {
|
| 121 |
|
|
chomp;
|
| 122 |
|
|
s/\x{d}$//; #^M
|
| 123 |
|
|
next if $_ eq "";
|
| 124 |
|
|
|
| 125 |
|
|
if(/^gcc version/) {
|
| 126 |
|
|
$ret{compiler} = $_;
|
| 127 |
|
|
} elsif(/^got a .* signal, interrupted by user /) {
|
| 128 |
|
|
$ret{incomplete} = 2;
|
| 129 |
|
|
} elsif(/^\s*=== gdb/) {
|
| 130 |
|
|
$is_gdb = 1;
|
| 131 |
|
|
# The log file from the GDB test suite is prone to have random crap
|
| 132 |
|
|
# in front of test result lines, so we need to be looser about how
|
| 133 |
|
|
# we parse those for GDB.
|
| 134 |
|
|
$gdbhack = ".*";
|
| 135 |
|
|
} elsif(/^(Test Run By \S+ on|runtest completed at) ... (.{3}) (\d{1,2}) (\d{1,2}):(\d{1,2}):(\d{1,2}) (\d{4})/) {
|
| 136 |
|
|
my $time = mktime($6, $5, $4, $3, $monmap{$2}, $7 - 1900);
|
| 137 |
|
|
if($1 eq "runtest completed at") {
|
| 138 |
|
|
$ret{end_time} = $time;
|
| 139 |
|
|
} else {
|
| 140 |
|
|
$ret{start_time} = $time;
|
| 141 |
|
|
}
|
| 142 |
|
|
} elsif(m<^Running (?!target )\Q$gccdir\E/?(\S+)> or m<^Running (?!target )\S*?((?:gcc|gdb|libstdc\+\+-v3)/testsuite/\S+)>) {
|
| 143 |
|
|
# We keep track of the last "Running foo/bar/baz.exp" line because
|
| 144 |
|
|
# some tests don't bother printing out the full paths of their files,
|
| 145 |
|
|
# and this gives us the directory information.
|
| 146 |
|
|
|
| 147 |
|
|
$lastrun = $1;
|
| 148 |
|
|
$lastrun =~ s!/[^/]*/\.\.!!; # foo/bar/../baz -> foo/baz
|
| 149 |
|
|
$currgroup = mkGroupPath(\%ret, \%groupindex, split(m!/!, $lastrun));
|
| 150 |
|
|
#$currgroup->{testfile} = $lastrun;
|
| 151 |
|
|
} elsif(/^Executing on (.*?):(.*)/) {
|
| 152 |
|
|
# Okay, if it's "Executing on host", it's a new
|
| 153 |
|
|
# file. If it's "Executing on unix", it's probably
|
| 154 |
|
|
# a test within the file -- an execution test, specifically --
|
| 155 |
|
|
# (makes sense, no?) But not always, sometimes we
|
| 156 |
|
|
# see "Executing on unix" outside the context of a
|
| 157 |
|
|
# file.
|
| 158 |
|
|
|
| 159 |
|
|
# Try to pick out the gccdir-relative filename.
|
| 160 |
|
|
# If we can't find it, it isn't really a new testfile,
|
| 161 |
|
|
# but a derived file.
|
| 162 |
|
|
my($exectype, $execwhat) = ($1, $2);
|
| 163 |
|
|
next if $execwhat =~ /^dsymutil/;
|
| 164 |
|
|
$execwhat =~
|
| 165 |
|
|
s!.*?\s\Q$gccdir\E/?(\S+).*!$1! or
|
| 166 |
|
|
s!.*?/((?:gcc|gdb|libstdc\+\+-v3)/testsuite/\S+).*!$1! or
|
| 167 |
|
|
$exectype = "unix";
|
| 168 |
|
|
|
| 169 |
|
|
if($exectype eq "host" or !$currgroup) {
|
| 170 |
|
|
# New file
|
| 171 |
|
|
|
| 172 |
|
|
my $nogroup = 0;
|
| 173 |
|
|
if($execwhat =~ / /) {
|
| 174 |
|
|
# We probably haven't parsed the file correctly.
|
| 175 |
|
|
# Try getting it from $lastrun.
|
| 176 |
|
|
|
| 177 |
|
|
$execwhat = dirname($lastrun) . "/" . basename($execwhat) if $lastrun and $execwhat;
|
| 178 |
|
|
$execwhat =~ s/\s.*//;
|
| 179 |
|
|
|
| 180 |
|
|
# At the end of each tool, it invokes "gcc -v" or "c++ -v"
|
| 181 |
|
|
# as a test. We don't really want to treat this as a test.
|
| 182 |
|
|
if($execwhat =~ m!/(gcc|c\+\+)$!) {
|
| 183 |
|
|
undef $currtest;
|
| 184 |
|
|
undef $currgroup;
|
| 185 |
|
|
$nogroup = 1;
|
| 186 |
|
|
}
|
| 187 |
|
|
}
|
| 188 |
|
|
|
| 189 |
|
|
if(!$nogroup) {
|
| 190 |
|
|
undef $currtest;
|
| 191 |
|
|
$execwhat =~ s!/[^/]*/\.\.!!; # foo/bar/../baz -> foo/baz
|
| 192 |
|
|
|
| 193 |
|
|
if($lastrun) {
|
| 194 |
|
|
my $lastbase = dirname($lastrun);
|
| 195 |
|
|
my $basegroup = $execwhat;
|
| 196 |
|
|
$basegroup =~ s!^\Q$lastbase\E/!!;
|
| 197 |
|
|
$execwhat = "$lastrun/$basegroup";
|
| 198 |
|
|
}
|
| 199 |
|
|
|
| 200 |
|
|
$currgroup = mkGroupPath(\%ret, \%groupindex, split(m!/!, $execwhat));
|
| 201 |
|
|
#$currgroup->{testfile} = $execwhat;
|
| 202 |
|
|
}
|
| 203 |
|
|
} else {
|
| 204 |
|
|
# New test within current file
|
| 205 |
|
|
|
| 206 |
|
|
$currtest = {};
|
| 207 |
|
|
}
|
| 208 |
|
|
} elsif(/^# of/) {
|
| 209 |
|
|
# This line appears should appear near the end of summary files.
|
| 210 |
|
|
# If it doesn't, something went wrong.
|
| 211 |
|
|
|
| 212 |
|
|
if($ret{incomplete} == 2) {
|
| 213 |
|
|
#Ah, but we previously saw indication that we were killed via a signal.
|
| 214 |
|
|
$ret{incomplete} = 1;
|
| 215 |
|
|
} else {
|
| 216 |
|
|
$ret{incomplete} = 0;
|
| 217 |
|
|
}
|
| 218 |
|
|
} elsif(/^testcase .* completed/) {
|
| 219 |
|
|
# End of a .exp file
|
| 220 |
|
|
undef $currtest;
|
| 221 |
|
|
undef $currgroup;
|
| 222 |
|
|
} elsif(/^$gdbhack(FAIL|PASS|UNRESOLVED|UNSUPPORTED|UNTESTED|XFAIL|XPASS|KFAIL|KPASS): (.*)/) {
|
| 223 |
|
|
# If the currtest already has a name, that means we've already seen
|
| 224 |
|
|
# its results, so what we have now is a new test. However, if we
|
| 225 |
|
|
# haven't seen results for currtest yet, that means currtest just
|
| 226 |
|
|
# has some diagnostics associated with it but no actual results,
|
| 227 |
|
|
# so just use that one.
|
| 228 |
|
|
undef $currtest if $currtest->{name};
|
| 229 |
|
|
|
| 230 |
|
|
my $phase = ""; # compile/link/execute
|
| 231 |
|
|
my($test, $result) = ($2, $1);
|
| 232 |
|
|
|
| 233 |
|
|
# Compile/(link/)execute combining
|
| 234 |
|
|
if($test =~ /^(.*) compile\s*$/) {
|
| 235 |
|
|
$test = "$1 compile,link,execute";
|
| 236 |
|
|
$phase = "compile";
|
| 237 |
|
|
} elsif($test =~ /^(.*)-(.*) (link|execute)\s*$/) {
|
| 238 |
|
|
$test = "$1 compile,link,execute";
|
| 239 |
|
|
if($3 eq "link") {
|
| 240 |
|
|
$phase = "link";
|
| 241 |
|
|
} else {
|
| 242 |
|
|
$phase = "execute";
|
| 243 |
|
|
}
|
| 244 |
|
|
} elsif($test =~ /(compile|compilation|execute|execution)/) {
|
| 245 |
|
|
my $phasematch = $1;
|
| 246 |
|
|
if($test =~ /^com/) {
|
| 247 |
|
|
$phase = "compile";
|
| 248 |
|
|
} else {
|
| 249 |
|
|
$phase = "execute";
|
| 250 |
|
|
}
|
| 251 |
|
|
$test =~ s!\Q$phasematch\E!compile,execute!;
|
| 252 |
|
|
}
|
| 253 |
|
|
|
| 254 |
|
|
# gcov tests behave in non-standard fashion.
|
| 255 |
|
|
my $failwhy = "";
|
| 256 |
|
|
$test =~ s/ gcov failed: (.*)// and $failwhy = $1;
|
| 257 |
|
|
|
| 258 |
|
|
# And some other tests have random information after a colon :(
|
| 259 |
|
|
# But for scan-assembler, this really is part of the name.
|
| 260 |
|
|
if(!$is_gdb and $test !~ /scan-assembler/ and $test =~ s/:\s*(.+)//) {
|
| 261 |
|
|
$failwhy = $1;
|
| 262 |
|
|
}
|
| 263 |
|
|
|
| 264 |
|
|
$test =~ s/\s*$//;
|
| 265 |
|
|
$test =~ s/^\s*$//;
|
| 266 |
|
|
|
| 267 |
|
|
# Sometimes there's a test which shows up as:
|
| 268 |
|
|
# foo (test for excess errors)
|
| 269 |
|
|
# foo (something else)
|
| 270 |
|
|
# foo: error executing dg-final
|
| 271 |
|
|
# if it runs, but just:
|
| 272 |
|
|
# foo
|
| 273 |
|
|
# if it doesn't. When we see the top form, we create a
|
| 274 |
|
|
# "pseudotest" in the bottom form, so that comparisons
|
| 275 |
|
|
# can be made.
|
| 276 |
|
|
my $basetest = $test;
|
| 277 |
|
|
$basetest =~ s/:? .*//;
|
| 278 |
|
|
|
| 279 |
|
|
if(exists($testindex{$test}) and !$testindex{$test}->{pseudotest}) {
|
| 280 |
|
|
$currtest = $testindex{$test};
|
| 281 |
|
|
if(ispass($currtest->{result})) {
|
| 282 |
|
|
$currtest->{result} = $result;
|
| 283 |
|
|
$currtest->{detail} = "$phase phase";
|
| 284 |
|
|
$currtest->{detail} .= "; $failwhy" if $failwhy;
|
| 285 |
|
|
}
|
| 286 |
|
|
} else {
|
| 287 |
|
|
# This might have been created earlier as a pseudotest.
|
| 288 |
|
|
# If so, overwrite it.
|
| 289 |
|
|
$currtest ||= $testindex{$test} || {};
|
| 290 |
|
|
|
| 291 |
|
|
$currtest->{name} = basename($test);
|
| 292 |
|
|
if($emit_fullname) {
|
| 293 |
|
|
$currtest->{fullname} = ($currgroup->{name} || dirname($test)) . "/$currtest->{name}";
|
| 294 |
|
|
}
|
| 295 |
|
|
my $grpname = $currgroup->{name} || "";
|
| 296 |
|
|
$currtest->{name} =~ s/^\s*\Q$grpname\E\s*//;
|
| 297 |
|
|
$currtest->{name} =~ s/^: // if $is_gdb;
|
| 298 |
|
|
# Sometimes there's a test at the root of the group.
|
| 299 |
|
|
# For instance, you'll have:
|
| 300 |
|
|
# FAIL: foo/bar.c (test for excess errors)
|
| 301 |
|
|
# UNRESOLVED: foo/bar.c: couldn't open "bar.s": no such file or directory
|
| 302 |
|
|
# In this case, groupname *is* the entire name, so the regex above will make the test name empty.
|
| 303 |
|
|
# In this case, we actually want to use the parent group and make this a test within that group.
|
| 304 |
|
|
my $orig_currgroup = $currgroup;
|
| 305 |
|
|
if(!$currtest->{name}) {
|
| 306 |
|
|
$currtest->{name} = $grpname;
|
| 307 |
|
|
$currgroup = $currgroup->{parent};
|
| 308 |
|
|
$grpname = $currgroup->{name} || "";
|
| 309 |
|
|
}
|
| 310 |
|
|
|
| 311 |
|
|
$currtest->{result} = $result;
|
| 312 |
|
|
if($phase and $failwhy) {
|
| 313 |
|
|
$currtest->{detail} = "$phase phase; $failwhy" if $phase;
|
| 314 |
|
|
} elsif($phase) {
|
| 315 |
|
|
$currtest->{detail} = "$phase phase";
|
| 316 |
|
|
} elsif($failwhy) {
|
| 317 |
|
|
$currtest->{detail} = $failwhy;
|
| 318 |
|
|
}
|
| 319 |
|
|
|
| 320 |
|
|
$currgroup->{test} ||= [];
|
| 321 |
|
|
push @{$currgroup->{test}}, $currtest;
|
| 322 |
|
|
$testindex{$test} = $currtest;
|
| 323 |
|
|
$currgroup = $orig_currgroup;
|
| 324 |
|
|
|
| 325 |
|
|
if($basetest ne $test) {
|
| 326 |
|
|
if(!exists($testindex{$basetest}) ) {
|
| 327 |
|
|
my $btbase = basename($basetest);
|
| 328 |
|
|
$testindex{$basetest} = {
|
| 329 |
|
|
name => $btbase,
|
| 330 |
|
|
result => $result,
|
| 331 |
|
|
pseudotest => 1,
|
| 332 |
|
|
fullname => $btbase
|
| 333 |
|
|
};
|
| 334 |
|
|
if($emit_fullname) {
|
| 335 |
|
|
$testindex{basetest}->{fullname} = ($currgroup->{name} || dirname($basetest)) . "/$btbase";
|
| 336 |
|
|
}
|
| 337 |
|
|
push @{$currgroup->{parent}->{test}}, $testindex{$basetest};
|
| 338 |
|
|
} else {
|
| 339 |
|
|
# Only let the base test pass if all the sub-tests pass
|
| 340 |
|
|
$testindex{$basetest}->{result} = $result if !ispass($result);
|
| 341 |
|
|
}
|
| 342 |
|
|
}
|
| 343 |
|
|
|
| 344 |
|
|
}
|
| 345 |
|
|
} elsif(/^\s+=== .* Summary ===\s*$/) {
|
| 346 |
|
|
undef $currgroup;
|
| 347 |
|
|
undef $currtest;
|
| 348 |
|
|
}
|
| 349 |
|
|
|
| 350 |
|
|
my $severity;
|
| 351 |
|
|
if(/^(ERROR|WARNING|NOTE): (.*)/) {
|
| 352 |
|
|
$severity = $1;
|
| 353 |
|
|
my $message = $2;
|
| 354 |
|
|
|
| 355 |
|
|
if($message eq "program timed out.") {
|
| 356 |
|
|
$currtest->{result} = "TIMEDOUT";
|
| 357 |
|
|
} elsif(
|
| 358 |
|
|
$message =~ /can't read "(HOSTCC|libiconv)": no such variable/ or
|
| 359 |
|
|
$message =~ /no files matched glob pattern/ or
|
| 360 |
|
|
$message =~ /error executing dg-final: .*: no such file/
|
| 361 |
|
|
) {
|
| 362 |
|
|
$severity = "NOTE";
|
| 363 |
|
|
}
|
| 364 |
|
|
} else {
|
| 365 |
|
|
$severity = "logline";
|
| 366 |
|
|
}
|
| 367 |
|
|
|
| 368 |
|
|
if($return_diags) {
|
| 369 |
|
|
my $dobj;
|
| 370 |
|
|
if($currtest) {
|
| 371 |
|
|
$currtest->{diagnostics} ||= [];
|
| 372 |
|
|
$dobj = $currtest->{diagnostics};
|
| 373 |
|
|
} elsif($currgroup) {
|
| 374 |
|
|
$currgroup->{diagnostics} ||= [];
|
| 375 |
|
|
$dobj = $currgroup->{diagnostics};
|
| 376 |
|
|
} else {
|
| 377 |
|
|
$dobj = $ret{diagnostics};
|
| 378 |
|
|
}
|
| 379 |
|
|
|
| 380 |
|
|
push @$dobj, {message => $_, severity => $severity, line => $line};
|
| 381 |
|
|
}
|
| 382 |
|
|
} continue {
|
| 383 |
|
|
$line++;
|
| 384 |
|
|
}
|
| 385 |
|
|
close LOGFILE;
|
| 386 |
|
|
|
| 387 |
|
|
return %ret;
|
| 388 |
|
|
}
|
| 389 |
|
|
|
| 390 |
|
|
# Split a test into testdivs
|
| 391 |
|
|
sub splitTest($$) {
|
| 392 |
|
|
my($root, $test) = @_;
|
| 393 |
|
|
|
| 394 |
|
|
$test->{fullname} =~ /^(\S+)\s*(.*)/;
|
| 395 |
|
|
my($path, $descriptor) = ($1, $2);
|
| 396 |
|
|
my @nodes = split(m!/!, $path);
|
| 397 |
|
|
push @nodes, $descriptor if $descriptor;
|
| 398 |
|
|
my $lastnode = pop @nodes;
|
| 399 |
|
|
|
| 400 |
|
|
my $hash = $root;
|
| 401 |
|
|
foreach (@nodes) {
|
| 402 |
|
|
$hash->{testdiv} ||= {};
|
| 403 |
|
|
$hash = $hash->{testdiv}->{$_} ||= {};
|
| 404 |
|
|
}
|
| 405 |
|
|
|
| 406 |
|
|
|
| 407 |
|
|
$hash->{test} ||= {};
|
| 408 |
|
|
$hash->{test}->{$lastnode} = $test;
|
| 409 |
|
|
}
|
| 410 |
|
|
|
| 411 |
|
|
|
| 412 |
|
|
# ==== Comparison ====
|
| 413 |
|
|
|
| 414 |
|
|
sub ispass($) {
|
| 415 |
|
|
my $result = shift;
|
| 416 |
|
|
|
| 417 |
|
|
if($result eq "PASS" or $result eq "XFAIL" or $result eq "KFAIL") {
|
| 418 |
|
|
return 1;
|
| 419 |
|
|
} else {
|
| 420 |
|
|
return 0;
|
| 421 |
|
|
}
|
| 422 |
|
|
}
|
| 423 |
|
|
|
| 424 |
|
|
1;
|
