MediaWiki
REL1_22
|
00001 <?php 00030 if ( PHP_SAPI != 'cli' ) { 00031 die( "Run me from the command line please.\n" ); 00032 } 00033 00034 require_once 'UtfNormalDefines.php'; 00035 require_once 'UtfNormalUtil.php'; 00036 require_once 'UtfNormal.php'; 00037 mb_internal_encoding( "utf-8" ); 00038 00039 $verbose = false; 00040 #$verbose = true; 00041 00042 $in = fopen( "UTF-8-test.txt", "rt" ); 00043 if( !$in ) { 00044 print "Couldn't open UTF-8-test.txt -- can't run tests.\n"; 00045 print "If necessary, manually download this file. It can be obtained at\n"; 00046 print "http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt\n"; 00047 exit(-1); 00048 } 00049 00050 $columns = 0; 00051 while( false !== ( $line = fgets( $in ) ) ) { 00052 $matches = array(); 00053 if( preg_match( '/^(Here come the tests:\s*)\|$/', $line, $matches ) ) { 00054 $columns = strpos( $line, '|' ); 00055 break; 00056 } 00057 } 00058 00059 if( !$columns ) { 00060 print "Something seems to be wrong; couldn't extract line length.\n"; 00061 print "Check that UTF-8-test.txt was downloaded correctly from\n"; 00062 print "http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt\n"; 00063 exit(-1); 00064 } 00065 00066 # print "$columns\n"; 00067 00068 $ignore = array( 00069 # These two lines actually seem to be corrupt 00070 '2.1.1', '2.2.1' ); 00071 00072 $exceptions = array( 00073 # Tests that should mark invalid characters due to using long 00074 # sequences beyond what is now considered legal. 00075 '2.1.5', '2.1.6', '2.2.4', '2.2.5', '2.2.6', '2.3.5', 00076 00077 # Literal 0xffff, which is illegal 00078 '2.2.3' ); 00079 00080 $longTests = array( 00081 # These tests span multiple lines 00082 '3.1.9', '3.2.1', '3.2.2', '3.2.3', '3.2.4', '3.2.5', 00083 '3.4' ); 00084 00085 # These tests are not in proper subsections 00086 $sectionTests = array( '3.4' ); 00087 00088 $section = null; 00089 $test = ''; 00090 $failed = 0; 00091 $success = 0; 00092 $total = 0; 00093 while( false !== ( $line = fgets( $in ) ) ) { 00094 $matches = array(); 00095 if( preg_match( '/^(\d+)\s+(.*?)\s*\|/', $line, $matches ) ) { 00096 $section = $matches[1]; 00097 print $line; 00098 continue; 00099 } 00100 if( preg_match( '/^(\d+\.\d+\.\d+)\s*/', $line, $matches ) ) { 00101 $test = $matches[1]; 00102 00103 if( in_array( $test, $ignore ) ) { 00104 continue; 00105 } 00106 if( in_array( $test, $longTests ) ) { 00107 $line = fgets( $in ); 00108 for( $line = fgets( $in ); !preg_match( '/^\s+\|/', $line ); $line = fgets( $in ) ) { 00109 testLine( $test, $line, $total, $success, $failed, $columns, $exceptions, $verbose ); 00110 } 00111 } else { 00112 testLine( $test, $line, $total, $success, $failed, $columns, $exceptions, $verbose ); 00113 } 00114 } 00115 } 00116 00117 if( $failed ) { 00118 echo "\nFailed $failed tests.\n"; 00119 echo "UTF-8 DECODER TEST FAILED\n"; 00120 exit (-1); 00121 } 00122 00123 echo "UTF-8 DECODER TEST SUCCESS!\n"; 00124 exit (0); 00125 00126 00127 function testLine( $test, $line, &$total, &$success, &$failed, $columns, $exceptions, $verbose ) { 00128 $stripped = $line; 00129 UtfNormal::quickisNFCVerify( $stripped ); 00130 00131 $same = ( $line == $stripped ); 00132 $len = mb_strlen( substr( $stripped, 0, strpos( $stripped, '|' ) ) ); 00133 if( $len == 0 ) { 00134 $len = strlen( substr( $stripped, 0, strpos( $stripped, '|' ) ) ); 00135 } 00136 00137 $ok = $same ^ ($test >= 3 ); 00138 00139 $ok ^= in_array( $test, $exceptions ); 00140 00141 $ok &= ($columns == $len); 00142 00143 $total++; 00144 if( $ok ) { 00145 $success++; 00146 } else { 00147 $failed++; 00148 } 00149 00150 if( $verbose || !$ok ) { 00151 print str_replace( "\n", "$len\n", $stripped ); 00152 } 00153 }