ILIAS  release_5-2 Revision v5.2.25-18-g3f80b828510
Utf8Test.php
Go to the documentation of this file.
1<?php
2# Copyright (C) 2004 Brion Vibber <brion@pobox.com>
3# http://www.mediawiki.org/
4#
5# This program is free software; you can redistribute it and/or modify
6# it under the terms of the GNU General Public License as published by
7# the Free Software Foundation; either version 2 of the License, or
8# (at your option) any later version.
9#
10# This program is distributed in the hope that it will be useful,
11# but WITHOUT ANY WARRANTY; without even the implied warranty of
12# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13# GNU General Public License for more details.
14#
15# You should have received a copy of the GNU General Public License along
16# with this program; if not, write to the Free Software Foundation, Inc.,
17# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18# http://www.gnu.org/copyleft/gpl.html
19
29require_once 'include/Unicode/UtfNormal.php';
30mb_internal_encoding( "utf-8" );
31
32#$verbose = true;
33if( php_sapi_name() != 'cli' ) {
34 die( "Run me from the command line please.\n" );
35}
36
37$in = fopen( "UTF-8-test.txt", "rt" );
38if( !$in ) {
39 print "Couldn't open UTF-8-test.txt -- can't run tests.\n";
40 print "If necessary, manually download this file. It can be obtained at\n";
41 print "http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt";
42 exit(-1);
43}
44
46while( false !== ( $line = fgets( $in ) ) ) {
47 $matches = array();
48 if( preg_match( '/^(Here come the tests:\s*)\|$/', $line, $matches ) ) {
49 $columns = strpos( $line, '|' );
50 break;
51 }
52}
53
54if( !$columns ) {
55 print "Something seems to be wrong; couldn't extract line length.\n";
56 print "Check that UTF-8-test.txt was downloaded correctly from\n";
57 print "http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt";
58 exit(-1);
59}
60
61# print "$columns\n";
62
63$ignore = array(
64 # These two lines actually seem to be corrupt
65 '2.1.1', '2.2.1' );
66
68 # Tests that should mark invalid characters due to using long
69 # sequences beyond what is now considered legal.
70 '2.1.5', '2.1.6', '2.2.4', '2.2.5', '2.2.6', '2.3.5',
71
72 # Literal 0xffff, which is illegal
73 '2.2.3' );
74
75$longTests = array(
76 # These tests span multiple lines
77 '3.1.9', '3.2.1', '3.2.2', '3.2.3', '3.2.4', '3.2.5',
78 '3.4' );
79
80# These tests are not in proper subsections
81$sectionTests = array( '3.4' );
82
83$section = NULL;
84$test = '';
88while( false !== ( $line = fgets( $in ) ) ) {
89 $matches = array();
90 if( preg_match( '/^(\d+)\s+(.*?)\s*\|/', $line, $matches ) ) {
91 $section = $matches[1];
92 print $line;
93 continue;
94 }
95 if( preg_match( '/^(\d+\.\d+\.\d+)\s*/', $line, $matches ) ) {
96 $test = $matches[1];
97
98 if( in_array( $test, $ignore ) ) {
99 continue;
100 }
101 if( in_array( $test, $longTests ) ) {
102 $line = fgets( $in );
103 for( $line = fgets( $in ); !preg_match( '/^\s+\|/', $line ); $line = fgets( $in ) ) {
104 testLine( $test, $line, $total, $success, $failed );
105 }
106 } else {
107 testLine( $test, $line, $total, $success, $failed );
108 }
109 }
110}
111
112if( $failed ) {
113 echo "\nFailed $failed tests.\n";
114 echo "UTF-8 DECODER TEST FAILED\n";
115 exit (-1);
116}
117
118echo "UTF-8 DECODER TEST SUCCESS!\n";
119exit (0);
120
121
122function testLine( $test, $line, &$total, &$success, &$failed ) {
123 $stripped = $line;
124 UtfNormal::quickisNFCVerify( $stripped );
125
126 $same = ( $line == $stripped );
127 $len = mb_strlen( substr( $stripped, 0, strpos( $stripped, '|' ) ) );
128 if( $len == 0 ) {
129 $len = strlen( substr( $stripped, 0, strpos( $stripped, '|' ) ) );
130 }
131
132 global $columns;
133 $ok = $same ^ ($test >= 3 );
134
135 global $exceptions;
136 $ok ^= in_array( $test, $exceptions );
137
138 $ok &= ($columns == $len);
139
140 $total++;
141 if( $ok ) {
142 $success++;
143 } else {
144 $failed++;
145 }
146 global $verbose;
147 if( $verbose || !$ok ) {
148 print str_replace( "\n", "$len\n", $stripped );
149 }
150}
151
152?>
$failed
Definition: Utf8Test.php:85
$total
Definition: Utf8Test.php:87
$test
Definition: Utf8Test.php:84
testLine( $test, $line, &$total, &$success, &$failed)
Definition: Utf8Test.php:122
$longTests
Definition: Utf8Test.php:75
$exceptions
Definition: Utf8Test.php:67
if(! $in) $columns
Definition: Utf8Test.php:45
$success
Definition: Utf8Test.php:86
if(php_sapi_name() !='cli') $in
Definition: Utf8Test.php:37
$sectionTests
Definition: Utf8Test.php:81
$section
Definition: Utf8Test.php:83
while(false !==($line=fgets($in))) if(! $columns) $ignore
Definition: Utf8Test.php:63
if(! $in) print
$verbose
An exception for terminatinating execution or to throw for unit testing.