ILIAS  Release_4_1_x_branch Revision 61804
 All Data Structures Namespaces Files Functions Variables Groups Pages
CleanUpTest Class Reference
+ Inheritance diagram for CleanUpTest:
+ Collaboration diagram for CleanUpTest:

Public Member Functions

 setUp ()
 tearDown ()
 testAscii ()
 testNull ()
 testLatin ()
 testLatinNormal ()
 XtestAllChars ()
 This test is very expensive!
 testAllBytes ()
 doTestBytes ($head, $tail)
 testDoubleBytes ()
 doTestDoubleBytes ($head, $tail)
 testTripleBytes ()
 doTestTripleBytes ($head, $tail)
 testChunkRegression ()
 testInterposeRegression ()
 testOverlongRegression ()
 testSurrogateRegression ()
 testBomRegression ()
 testForbiddenRegression ()
 testHangulRegression ()

Detailed Description

Definition at line 45 of file CleanUpTest.php.

Member Function Documentation

CleanUpTest::doTestBytes (   $head,
  $tail 
)
Todo:
document

Definition at line 127 of file CleanUpTest.php.

References UtfNormal\cleanUp(), and UTF8_REPLACEMENT.

Referenced by testAllBytes().

{
for( $i = 0x0; $i < 256; $i++ ) {
$char = $head . chr( $i ) . $tail;
$clean = UtfNormal::cleanUp( $char );
$x = sprintf( "%02X", $i );
if( $i == 0x0009 ||
$i == 0x000a ||
$i == 0x000d ||
($i > 0x001f && $i < 0x80) ) {
$this->assertEquals(
bin2hex( $char ),
bin2hex( $clean ),
"ASCII byte $x should be intact" );
if( $char != $clean ) return;
} else {
$norm = $head . UTF8_REPLACEMENT . $tail;
$this->assertEquals(
bin2hex( $norm ),
bin2hex( $clean ),
"Forbidden byte $x should be rejected" );
if( $norm != $clean ) return;
}
}
}

+ Here is the call graph for this function:

+ Here is the caller graph for this function:

CleanUpTest::doTestDoubleBytes (   $head,
  $tail 
)
Todo:
document

Definition at line 163 of file CleanUpTest.php.

References UtfNormal\cleanUp(), UtfNormal\NFC(), and UTF8_REPLACEMENT.

Referenced by testDoubleBytes().

{
for( $first = 0xc0; $first < 0x100; $first++ ) {
for( $second = 0x80; $second < 0x100; $second++ ) {
$char = $head . chr( $first ) . chr( $second ) . $tail;
$clean = UtfNormal::cleanUp( $char );
$x = sprintf( "%02X,%02X", $first, $second );
if( $first > 0xc1 &&
$first < 0xe0 &&
$second < 0xc0 ) {
$norm = UtfNormal::NFC( $char );
$this->assertEquals(
bin2hex( $norm ),
bin2hex( $clean ),
"Pair $x should be intact" );
if( $norm != $clean ) return;
} elseif( $first > 0xfd || $second > 0xbf ) {
# fe and ff are not legal head bytes -- expect two replacement chars
$norm = $head . UTF8_REPLACEMENT . UTF8_REPLACEMENT . $tail;
$this->assertEquals(
bin2hex( $norm ),
bin2hex( $clean ),
"Forbidden pair $x should be rejected" );
if( $norm != $clean ) return;
} else {
$norm = $head . UTF8_REPLACEMENT . $tail;
$this->assertEquals(
bin2hex( $norm ),
bin2hex( $clean ),
"Forbidden pair $x should be rejected" );
if( $norm != $clean ) return;
}
}
}
}

+ Here is the call graph for this function:

+ Here is the caller graph for this function:

CleanUpTest::doTestTripleBytes (   $head,
  $tail 
)
Todo:
document

Definition at line 207 of file CleanUpTest.php.

References UtfNormal\cleanUp(), UtfNormal\NFC(), UTF8_REPLACEMENT, and UTF8_SURROGATE_FIRST.

Referenced by testTripleBytes().

{
for( $first = 0xc0; $first < 0x100; $first++ ) {
for( $second = 0x80; $second < 0x100; $second++ ) {
#for( $third = 0x80; $third < 0x100; $third++ ) {
for( $third = 0x80; $third < 0x81; $third++ ) {
$char = $head . chr( $first ) . chr( $second ) . chr( $third ) . $tail;
$clean = UtfNormal::cleanUp( $char );
$x = sprintf( "%02X,%02X,%02X", $first, $second, $third );
if( $first >= 0xe0 &&
$first < 0xf0 &&
$second < 0xc0 &&
$third < 0xc0 ) {
if( $first == 0xe0 && $second < 0xa0 ) {
$this->assertEquals(
bin2hex( $head . UTF8_REPLACEMENT . $tail ),
bin2hex( $clean ),
"Overlong triplet $x should be rejected" );
} elseif( $first == 0xed &&
( chr( $first ) . chr( $second ) . chr( $third )) >= UTF8_SURROGATE_FIRST ) {
$this->assertEquals(
bin2hex( $head . UTF8_REPLACEMENT . $tail ),
bin2hex( $clean ),
"Surrogate triplet $x should be rejected" );
} else {
$this->assertEquals(
bin2hex( UtfNormal::NFC( $char ) ),
bin2hex( $clean ),
"Triplet $x should be intact" );
}
} elseif( $first > 0xc1 && $first < 0xe0 && $second < 0xc0 ) {
$this->assertEquals(
bin2hex( UtfNormal::NFC( $head . chr( $first ) . chr( $second ) ) . UTF8_REPLACEMENT . $tail ),
bin2hex( $clean ),
"Valid 2-byte $x + broken tail" );
} elseif( $second > 0xc1 && $second < 0xe0 && $third < 0xc0 ) {
$this->assertEquals(
bin2hex( $head . UTF8_REPLACEMENT . UtfNormal::NFC( chr( $second ) . chr( $third ) . $tail ) ),
bin2hex( $clean ),
"Broken head + valid 2-byte $x" );
} elseif( ( $first > 0xfd || $second > 0xfd ) &&
( ( $second > 0xbf && $third > 0xbf ) ||
( $second < 0xc0 && $third < 0xc0 ) ||
( $second > 0xfd ) ||
( $third > 0xfd ) ) ) {
# fe and ff are not legal head bytes -- expect three replacement chars
$this->assertEquals(
bin2hex( $head . UTF8_REPLACEMENT . UTF8_REPLACEMENT . UTF8_REPLACEMENT . $tail ),
bin2hex( $clean ),
"Forbidden triplet $x should be rejected" );
} elseif( $first > 0xc2 && $second < 0xc0 && $third < 0xc0 ) {
$this->assertEquals(
bin2hex( $head . UTF8_REPLACEMENT . $tail ),
bin2hex( $clean ),
"Forbidden triplet $x should be rejected" );
} else {
$this->assertEquals(
bin2hex( $head . UTF8_REPLACEMENT . UTF8_REPLACEMENT . $tail ),
bin2hex( $clean ),
"Forbidden triplet $x should be rejected" );
}
}
}
}
}

+ Here is the call graph for this function:

+ Here is the caller graph for this function:

CleanUpTest::setUp ( )
Todo:
document

Definition at line 47 of file CleanUpTest.php.

{
}
CleanUpTest::tearDown ( )
Todo:
document

Definition at line 51 of file CleanUpTest.php.

{
}
CleanUpTest::testAllBytes ( )
Todo:
document

Definition at line 119 of file CleanUpTest.php.

References doTestBytes().

{
$this->doTestBytes( '', '' );
$this->doTestBytes( 'x', '' );
$this->doTestBytes( '', 'x' );
$this->doTestBytes( 'x', 'x' );
}

+ Here is the call graph for this function:

CleanUpTest::testAscii ( )
Todo:
document

Definition at line 55 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = 'This is plain ASCII text.';
$this->assertEquals( $text, UtfNormal::cleanUp( $text ) );
}

+ Here is the call graph for this function:

CleanUpTest::testBomRegression ( )
Todo:
document

Definition at line 371 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "\xef\xbf\xbe" . # U+FFFE, illegal char
"\xb2" . # bad tail
"\xef" . # bad head
"\x59";
$expect = "\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\x59";
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testChunkRegression ( )
Todo:
document

Definition at line 273 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
# Check for regression against a chunking bug
$text = "\x46\x55\xb8" .
"\xdc\x96" .
"\xee" .
"\xe7" .
"\x44" .
"\xaa" .
"\x2f\x25";
$expect = "\x46\x55\xef\xbf\xbd" .
"\xdc\x96" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\x44" .
"\xef\xbf\xbd" .
"\x2f\x25";
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testDoubleBytes ( )
Todo:
document

Definition at line 153 of file CleanUpTest.php.

References doTestDoubleBytes().

{
$this->doTestDoubleBytes( '', '' );
$this->doTestDoubleBytes( 'x', '' );
$this->doTestDoubleBytes( '', 'x' );
$this->doTestDoubleBytes( 'x', 'x' );
}

+ Here is the call graph for this function:

CleanUpTest::testForbiddenRegression ( )
Todo:
document

Definition at line 386 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "\xef\xbf\xbf"; # U+FFFF, illegal char
$expect = "\xef\xbf\xbd";
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testHangulRegression ( )
Todo:
document

Definition at line 395 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "\xed\x9c\xaf" . # Hangul char
"\xe1\x87\x81"; # followed by another final jamo
$expect = $text; # Should *not* change.
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testInterposeRegression ( )
Todo:
document

Definition at line 296 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "\x4e\x30" .
"\xb1" . # bad tail
"\x3a" .
"\x92" . # bad tail
"\x62\x3a" .
"\x84" . # bad tail
"\x43" .
"\xc6" . # bad head
"\x3f" .
"\x92" . # bad tail
"\xad" . # bad tail
"\x7d" .
"\xd9\x95";
$expect = "\x4e\x30" .
"\xef\xbf\xbd" .
"\x3a" .
"\xef\xbf\xbd" .
"\x62\x3a" .
"\xef\xbf\xbd" .
"\x43" .
"\xef\xbf\xbd" .
"\x3f" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\x7d" .
"\xd9\x95";
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testLatin ( )
Todo:
document

Definition at line 70 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "L'\xc3\xa9cole";
$this->assertEquals( $text, UtfNormal::cleanUp( $text ) );
}

+ Here is the call graph for this function:

CleanUpTest::testLatinNormal ( )
Todo:
document

Definition at line 76 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "L'e\xcc\x81cole";
$expect = "L'\xc3\xa9cole";
$this->assertEquals( $expect, UtfNormal::cleanUp( $text ) );
}

+ Here is the call graph for this function:

CleanUpTest::testNull ( )
Todo:
document

Definition at line 61 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "a \x00 null";
$expect = "a \xef\xbf\xbd null";
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testOverlongRegression ( )
Todo:
document

Definition at line 331 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "\x67" .
"\x1a" . # forbidden ascii
"\xea" . # bad head
"\xc1\xa6" . # overlong sequence
"\xad" . # bad tail
"\x1c" . # forbidden ascii
"\xb0" . # bad tail
"\x3c" .
"\x9e"; # bad tail
$expect = "\x67" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\x3c" .
"\xef\xbf\xbd";
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testSurrogateRegression ( )
Todo:
document

Definition at line 356 of file CleanUpTest.php.

References UtfNormal\cleanUp().

{
$text = "\xed\xb4\x96" . # surrogate 0xDD16
"\x83" . # bad tail
"\xb4" . # bad tail
"\xac"; # bad head
$expect = "\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd" .
"\xef\xbf\xbd";
$this->assertEquals(
bin2hex( $expect ),
bin2hex( UtfNormal::cleanUp( $text ) ) );
}

+ Here is the call graph for this function:

CleanUpTest::testTripleBytes ( )
Todo:
document

Definition at line 199 of file CleanUpTest.php.

References doTestTripleBytes().

{
$this->doTestTripleBytes( '', '' );
$this->doTestTripleBytes( 'x', '' );
$this->doTestTripleBytes( '', 'x' );
$this->doTestTripleBytes( 'x', 'x' );
}

+ Here is the call graph for this function:

CleanUpTest::XtestAllChars ( )

This test is very expensive!

Todo:
document

Definition at line 86 of file CleanUpTest.php.

References $utfCanonicalComp, $utfCanonicalDecomp, UtfNormal\cleanUp(), codepointToUtf8(), UtfNormal\NFC(), UNICODE_MAX, UNICODE_SURROGATE_FIRST, UNICODE_SURROGATE_LAST, and UTF8_REPLACEMENT.

{
for( $i = 0x0; $i < UNICODE_MAX; $i++ ) {
$char = codepointToUtf8( $i );
$clean = UtfNormal::cleanUp( $char );
$x = sprintf( "%04X", $i );
if( $i % 0x1000 == 0 ) echo "U+$x\n";
if( $i == 0x0009 ||
$i == 0x000a ||
$i == 0x000d ||
($i > 0x001f && $i < UNICODE_SURROGATE_FIRST) ||
($i > UNICODE_SURROGATE_LAST && $i < 0xfffe ) ||
($i > 0xffff && $i <= UNICODE_MAX ) ) {
if( isset( $utfCanonicalComp[$char] ) || isset( $utfCanonicalDecomp[$char] ) ) {
$comp = UtfNormal::NFC( $char );
$this->assertEquals(
bin2hex( $comp ),
bin2hex( $clean ),
"U+$x should be decomposed" );
} else {
$this->assertEquals(
bin2hex( $char ),
bin2hex( $clean ),
"U+$x should be intact" );
}
} else {
$this->assertEquals( bin2hex( $rep ), bin2hex( $clean ), $x );
}
}
}

+ Here is the call graph for this function:


The documentation for this class was generated from the following file: