archive/php-src
1
0
Fork 0
mirror of https://github.com/php/php-src.git synced 2025-08-20 09:24:05 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

- MFB: Fix bug #46478 (htmlentities() uses obsolete mapping table for character

Browse source 
entity references)
  (this should be gone to r292467)
This commit is contained in:
Moriyoshi Koizumi 2009-12-24 09:32:52 +00:00
parent 016ba1f1ed
commit 139c26f795
3 changed files with 840 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

38
ext/standard/html.c
View file

@ -150,7 +150,7 @@ static entity_table_t ent_uni_greek[] = {
NULL, NULL, NULL, NULL, NULL, NULL, NULL,
"thetasym", "upsih",
NULL, NULL, NULL,
"piv"
"piv"
};
static entity_table_t ent_uni_punct[] = {
@ -160,7 +160,7 @@ static entity_table_t ent_uni_punct[] = {
NULL, NULL, NULL, "ndash", "mdash", NULL, NULL, NULL,
/* 8216 */
"lsquo", "rsquo", "sbquo", NULL, "ldquo", "rdquo", "bdquo", NULL,
"dagger", "Dagger", "bull", NULL, NULL, NULL, "hellip",
"dagger", "Dagger", "bull", NULL, NULL, NULL, "hellip",
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, "permil", NULL,
/* 8242 */
"prime", "Prime", NULL, NULL, NULL, NULL, NULL, "lsaquo", "rsaquo", NULL,
@ -200,39 +200,39 @@ static entity_table_t ent_uni_8592_9002[] = {
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8656 (0x21d0) */
"lArr", "uArr", "rArr", "dArr", "hArr", "vArr", NULL, NULL,
NULL, NULL, "lAarr", "rAarr", NULL, "rarrw", NULL, NULL,
"lArr", "uArr", "rArr", "dArr", "hArr", NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8672 (0x21e0) */
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8704 (0x2200) */
"forall", "comp", "part", "exist", "nexist", "empty", NULL, "nabla",
"isin", "notin", "epsis", "ni", "notni", "bepsi", NULL, "prod",
"forall", NULL, "part", "exist", NULL, "empty", NULL, "nabla",
"isin", "notin", NULL, "ni", NULL, NULL, NULL, "prod",
/* 8720 (0x2210) */
"coprod", "sum", "minus", "mnplus", "plusdo", NULL, "setmn", "lowast",
"compfn", NULL, "radic", NULL, NULL, "prop", "infin", "ang90",
NULL, "sum", "minus", NULL, NULL, NULL, NULL, "lowast",
NULL, NULL, "radic", NULL, NULL, "prop", "infin", NULL,
/* 8736 (0x2220) */
"ang", "angmsd", "angsph", "mid", "nmid", "par", "npar", "and",
"or", "cap", "cup", "int", NULL, NULL, "conint", NULL,
"ang", NULL, NULL, NULL, NULL, NULL, NULL, "and",
"or", "cap", "cup", "int", NULL, NULL, NULL, NULL,
/* 8752 (0x2230) */
NULL, NULL, NULL, NULL, "there4", "becaus", NULL, NULL,
NULL, NULL, NULL, NULL, "sim", "bsim", NULL, NULL,
NULL, NULL, NULL, NULL, "there4", NULL, NULL, NULL,
NULL, NULL, NULL, NULL, "sim", NULL, NULL, NULL,
/* 8768 (0x2240) */
"wreath", "nsim", NULL, "sime", "nsime", "cong", NULL, "ncong",
"asymp", "nap", "ape", NULL, "bcong", "asymp", "bump", "bumpe",
NULL, NULL, NULL, NULL, NULL, "cong", NULL, NULL,
"asymp", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8784 (0x2250) */
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8800 (0x2260) */
"ne", "equiv", NULL, NULL, "le", "ge", "lE", "gE",
"lnE", "gnE", "Lt", "Gt", "twixt", NULL, "nlt", "ngt",
"ne", "equiv", NULL, NULL, "le", "ge", NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8816 (0x2270) */
"nles", "nges", "lsim", "gsim", NULL, NULL, "lg", "gl",
NULL, NULL, "pr", "sc", "cupre", "sscue", "prsim", "scsim",
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8832 (0x2280) */
"npr", "nsc", "sub", "sup", "nsub", "nsup", "sube", "supe",
NULL, NULL, "sub", "sup", "nsub", NULL, "sube", "supe",
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
/* 8848 (0x2290) */
NULL, NULL, NULL, NULL, NULL, "oplus", NULL, "otimes",

516
ext/standard/tests/strings/html_entity_decode_html4.phpt Normal file
View file

@ -0,0 +1,516 @@
--TEST--
html_entity_decode() conformance check (HTML 4)
--FILE--
<?php
$map = array(
"&quot;",
"&amp;",
"&lt;",
"&gt;",
"&nbsp;",
"&iexcl;",
"&cent;",
"&pound;",
"&curren;",
"&yen;",
"&brvbar;",
"&sect;",
"&uml;",
"&copy;",
"&ordf;",
"&laquo;",
"&not;",
"&shy;",
"&reg;",
"&macr;",
"&deg;",
"&plusmn;",
"&sup2;",
"&sup3;",
"&acute;",
"&micro;",
"&para;",
"&middot;",
"&cedil;",
"&sup1;",
"&ordm;",
"&raquo;",
"&frac14;",
"&frac12;",
"&frac34;",
"&iquest;",
"&Agrave;",
"&Aacute;",
"&Acirc;",
"&Atilde;",
"&Auml;",
"&Aring;",
"&AElig;",
"&Ccedil;",
"&Egrave;",
"&Eacute;",
"&Ecirc;",
"&Euml;",
"&Igrave;",
"&Iacute;",
"&Icirc;",
"&Iuml;",
"&ETH;",
"&Ntilde;",
"&Ograve;",
"&Oacute;",
"&Ocirc;",
"&Otilde;",
"&Ouml;",
"&times;",
"&Oslash;",
"&Ugrave;",
"&Uacute;",
"&Ucirc;",
"&Uuml;",
"&Yacute;",
"&THORN;",
"&szlig;",
"&agrave;",
"&aacute;",
"&acirc;",
"&atilde;",
"&auml;",
"&aring;",
"&aelig;",
"&ccedil;",
"&egrave;",
"&eacute;",
"&ecirc;",
"&euml;",
"&igrave;",
"&iacute;",
"&icirc;",
"&iuml;",
"&eth;",
"&ntilde;",
"&ograve;",
"&oacute;",
"&ocirc;",
"&otilde;",
"&ouml;",
"&divide;",
"&oslash;",
"&ugrave;",
"&uacute;",
"&ucirc;",
"&uuml;",
"&yacute;",
"&thorn;",
"&yuml;",
"&OElig;",
"&oelig;",
"&Scaron;",
"&scaron;",
"&Yuml;",
"&fnof;",
"&circ;",
"&tilde;",
"&Alpha;",
"&Beta;",
"&Gamma;",
"&Delta;",
"&Epsilon;",
"&Zeta;",
"&Eta;",
"&Theta;",
"&Iota;",
"&Kappa;",
"&Lambda;",
"&Mu;",
"&Nu;",
"&Xi;",
"&Omicron;",
"&Pi;",
"&Rho;",
"&Sigma;",
"&Tau;",
"&Upsilon;",
"&Phi;",
"&Chi;",
"&Psi;",
"&Omega;",
"&alpha;",
"&beta;",
"&gamma;",
"&delta;",
"&epsilon;",
"&zeta;",
"&eta;",
"&theta;",
"&iota;",
"&kappa;",
"&lambda;",
"&mu;",
"&nu;",
"&xi;",
"&omicron;",
"&pi;",
"&rho;",
"&sigmaf;",
"&sigma;",
"&tau;",
"&upsilon;",
"&phi;",
"&chi;",
"&psi;",
"&omega;",
"&thetasym;",
"&upsih;",
"&piv;",
"&ensp;",
"&emsp;",
"&thinsp;",
"&zwnj;",
"&zwj;",
"&lrm;",
"&rlm;",
"&ndash;",
"&mdash;",
"&lsquo;",
"&rsquo;",
"&sbquo;",
"&ldquo;",
"&rdquo;",
"&bdquo;",
"&dagger;",
"&Dagger;",
"&bull;",
"&hellip;",
"&permil;",
"&prime;",
"&Prime;",
"&lsaquo;",
"&rsaquo;",
"&oline;",
"&frasl;",
"&euro;",
"&image;",
"&weierp;",
"&real;",
"&trade;",
"&alefsym;",
"&larr;",
"&uarr;",
"&rarr;",
"&darr;",
"&harr;",
"&crarr;",
"&lArr;",
"&uArr;",
"&rArr;",
"&dArr;",
"&hArr;",
"&forall;",
"&part;",
"&exist;",
"&empty;",
"&nabla;",
"&isin;",
"&notin;",
"&ni;",
"&prod;",
"&sum;",
"&minus;",
"&lowast;",
"&radic;",
"&prop;",
"&infin;",
"&ang;",
"&and;",
"&or;",
"&cap;",
"&cup;",
"&int;",
"&there4;",
"&sim;",
"&cong;",
"&asymp;",
"&ne;",
"&equiv;",
"&le;",
"&ge;",
"&sub;",
"&sup;",
"&nsub;",
"&sube;",
"&supe;",
"&oplus;",
"&otimes;",
"&perp;",
"&sdot;",
"&lceil;",
"&rceil;",
"&lfloor;",
"&rfloor;",
"&lang;",
"&rang;",
"&loz;",
"&spades;",
"&clubs;",
"&hearts;",
"&diams;",
);
foreach ($map as $str) {
echo bin2hex(html_entity_decode($str, ENT_QUOTES, "UTF-8")), "\n";
}
?>
--EXPECT--
22
26
3c
3e
c2a0
c2a1
c2a2
c2a3
c2a4
c2a5
c2a6
c2a7
c2a8
c2a9
c2aa
c2ab
c2ac
c2ad
c2ae
c2af
c2b0
c2b1
c2b2
c2b3
c2b4
c2b5
c2b6
c2b7
c2b8
c2b9
c2ba
c2bb
c2bc
c2bd
c2be
c2bf
c380
c381
c382
c383
c384
c385
c386
c387
c388
c389
c38a
c38b
c38c
c38d
c38e
c38f
c390
c391
c392
c393
c394
c395
c396
c397
c398
c399
c39a
c39b
c39c
c39d
c39e
c39f
c3a0
c3a1
c3a2
c3a3
c3a4
c3a5
c3a6
c3a7
c3a8
c3a9
c3aa
c3ab
c3ac
c3ad
c3ae
c3af
c3b0
c3b1
c3b2
c3b3
c3b4
c3b5
c3b6
c3b7
c3b8
c3b9
c3ba
c3bb
c3bc
c3bd
c3be
c3bf
c592
c593
c5a0
c5a1
c5b8
c692
cb86
cb9c
ce91
ce92
ce93
ce94
ce95
ce96
ce97
ce98
ce99
ce9a
ce9b
ce9c
ce9d
ce9e
ce9f
cea0
cea1
cea3
cea4
cea5
cea6
cea7
cea8
cea9
ceb1
ceb2
ceb3
ceb4
ceb5
ceb6
ceb7
ceb8
ceb9
ceba
cebb
cebc
cebd
cebe
cebf
cf80
cf81
cf82
cf83
cf84
cf85
cf86
cf87
cf88
cf89
cf91
cf92
cf96
e28082
e28083
e28089
e2808c
e2808d
e2808e
e2808f
e28093
e28094
e28098
e28099
e2809a
e2809c
e2809d
e2809e
e280a0
e280a1
e280a2
e280a6
e280b0
e280b2
e280b3
e280b9
e280ba
e280be
e28184
e282ac
e28491
e28498
e2849c
e284a2
e284b5
e28690
e28691
e28692
e28693
e28694
e286b5
e28790
e28791
e28792
e28793
e28794
e28880
e28882
e28883
e28885
e28887
e28888
e28889
e2888b
e2888f
e28891
e28892
e28897
e2889a
e2889d
e2889e
e288a0
e288a7
e288a8
e288a9
e288aa
e288ab
e288b4
e288bc
e28985
e28988
e289a0
e289a1
e289a4
e289a5
e28a82
e28a83
e28a84
e28a86
e28a87
e28a95
e28a97
e28aa5
e28b85
e28c88
e28c89
e28c8a
e28c8b
e28ca9
e28caa
e2978a
e299a0
e299a3
e299a5
e299a6

305
ext/standard/tests/strings/htmlentities_html4.phpt Normal file
View file

@ -0,0 +1,305 @@
--TEST--
htmlentities() conformance check (HTML 4)
--FILE--
<?php
function utf32_utf8($k) {
if ($k < 0x80) {
$retval = pack('C', $k);
} else if ($k < 0x800) {
$retval = pack('C2',
0xc0 | ($k >> 6),
0x80 | ($k & 0x3f));
} else if ($k < 0x10000) {
$retval = pack('C3',
0xe0 | ($k >> 12),
0x80 | (($k >> 6) & 0x3f),
0x80 | ($k & 0x3f));
} else if ($k < 0x200000) {
$retval = pack('C4',
0xf0 | ($k >> 18),
0x80 | (($k >> 12) & 0x3f),
0x80 | (($k >> 6) & 0x3f),
0x80 | ($k & 0x3f));
} else if ($k < 0x4000000) {
$retval = pack('C5',
0xf8 | ($k >> 24),
0x80 | (($k >> 18) & 0x3f),
0x80 | (($k >> 12) & 0x3f),
0x80 | (($k >> 6) & 0x3f),
0x80 | ($k & 0x3f));
} else {
$retval = pack('C6',
0xfc | ($k >> 30),
0x80 | (($k >> 24) & 0x3f),
0x80 | (($k >> 18) & 0x3f),
0x80 | (($k >> 12) & 0x3f),
0x80 | (($k >> 6) & 0x3f),
0x80 | ($k & 0x3f));
}
return $retval;
}
for ($i = 0; $i < 0x110000; $i++) {
if ($i >= 0xd800 && $i < 0xe000)
continue;
$str = utf32_utf8($i);
$result = htmlentities($str, ENT_QUOTES, 'UTF-8');
if ($str != $result) {
printf("%s\tU+%05X\n", $result, $i);
}
}
?>
--EXPECT--
&quot; U+00022
&amp; U+00026
&#039; U+00027
&lt; U+0003C
&gt; U+0003E
&nbsp; U+000A0
&iexcl; U+000A1
&cent; U+000A2
&pound; U+000A3
&curren; U+000A4
&yen; U+000A5
&brvbar; U+000A6
&sect; U+000A7
&uml; U+000A8
&copy; U+000A9
&ordf; U+000AA
&laquo; U+000AB
&not; U+000AC
&shy; U+000AD
&reg; U+000AE
&macr; U+000AF
&deg; U+000B0
&plusmn; U+000B1
&sup2; U+000B2
&sup3; U+000B3
&acute; U+000B4
&micro; U+000B5
&para; U+000B6
&middot; U+000B7
&cedil; U+000B8
&sup1; U+000B9
&ordm; U+000BA
&raquo; U+000BB
&frac14; U+000BC
&frac12; U+000BD
&frac34; U+000BE
&iquest; U+000BF
&Agrave; U+000C0
&Aacute; U+000C1
&Acirc; U+000C2
&Atilde; U+000C3
&Auml; U+000C4
&Aring; U+000C5
&AElig; U+000C6
&Ccedil; U+000C7
&Egrave; U+000C8
&Eacute; U+000C9
&Ecirc; U+000CA
&Euml; U+000CB
&Igrave; U+000CC
&Iacute; U+000CD
&Icirc; U+000CE
&Iuml; U+000CF
&ETH; U+000D0
&Ntilde; U+000D1
&Ograve; U+000D2
&Oacute; U+000D3
&Ocirc; U+000D4
&Otilde; U+000D5
&Ouml; U+000D6
&times; U+000D7
&Oslash; U+000D8
&Ugrave; U+000D9
&Uacute; U+000DA
&Ucirc; U+000DB
&Uuml; U+000DC
&Yacute; U+000DD
&THORN; U+000DE
&szlig; U+000DF
&agrave; U+000E0
&aacute; U+000E1
&acirc; U+000E2
&atilde; U+000E3
&auml; U+000E4
&aring; U+000E5
&aelig; U+000E6
&ccedil; U+000E7
&egrave; U+000E8
&eacute; U+000E9
&ecirc; U+000EA
&euml; U+000EB
&igrave; U+000EC
&iacute; U+000ED
&icirc; U+000EE
&iuml; U+000EF
&eth; U+000F0
&ntilde; U+000F1
&ograve; U+000F2
&oacute; U+000F3
&ocirc; U+000F4
&otilde; U+000F5
&ouml; U+000F6
&divide; U+000F7
&oslash; U+000F8
&ugrave; U+000F9
&uacute; U+000FA
&ucirc; U+000FB
&uuml; U+000FC
&yacute; U+000FD
&thorn; U+000FE
&yuml; U+000FF
&OElig; U+00152
&oelig; U+00153
&Scaron; U+00160
&scaron; U+00161
&Yuml; U+00178
&fnof; U+00192
&circ; U+002C6
&tilde; U+002DC
&Alpha; U+00391
&Beta; U+00392
&Gamma; U+00393
&Delta; U+00394
&Epsilon; U+00395
&Zeta; U+00396
&Eta; U+00397
&Theta; U+00398
&Iota; U+00399
&Kappa; U+0039A
&Lambda; U+0039B
&Mu; U+0039C
&Nu; U+0039D
&Xi; U+0039E
&Omicron; U+0039F
&Pi; U+003A0
&Rho; U+003A1
&Sigma; U+003A3
&Tau; U+003A4
&Upsilon; U+003A5
&Phi; U+003A6
&Chi; U+003A7
&Psi; U+003A8
&Omega; U+003A9
&alpha; U+003B1
&beta; U+003B2
&gamma; U+003B3
&delta; U+003B4
&epsilon; U+003B5
&zeta; U+003B6
&eta; U+003B7
&theta; U+003B8
&iota; U+003B9
&kappa; U+003BA
&lambda; U+003BB
&mu; U+003BC
&nu; U+003BD
&xi; U+003BE
&omicron; U+003BF
&pi; U+003C0
&rho; U+003C1
&sigmaf; U+003C2
&sigma; U+003C3
&tau; U+003C4
&upsilon; U+003C5
&phi; U+003C6
&chi; U+003C7
&psi; U+003C8
&omega; U+003C9
&thetasym; U+003D1
&upsih; U+003D2
&piv; U+003D6
&ensp; U+02002
&emsp; U+02003
&thinsp; U+02009
&zwnj; U+0200C
&zwj; U+0200D
&lrm; U+0200E
&rlm; U+0200F
&ndash; U+02013
&mdash; U+02014
&lsquo; U+02018
&rsquo; U+02019
&sbquo; U+0201A
&ldquo; U+0201C
&rdquo; U+0201D
&bdquo; U+0201E
&dagger; U+02020
&Dagger; U+02021
&bull; U+02022
&hellip; U+02026
&permil; U+02030
&prime; U+02032
&Prime; U+02033
&lsaquo; U+02039
&rsaquo; U+0203A
&oline; U+0203E
&frasl; U+02044
&euro; U+020AC
&image; U+02111
&weierp; U+02118
&real; U+0211C
&trade; U+02122
&alefsym; U+02135
&larr; U+02190
&uarr; U+02191
&rarr; U+02192
&darr; U+02193
&harr; U+02194
&crarr; U+021B5
&lArr; U+021D0
&uArr; U+021D1
&rArr; U+021D2
&dArr; U+021D3
&hArr; U+021D4
&forall; U+02200
&part; U+02202
&exist; U+02203
&empty; U+02205
&nabla; U+02207
&isin; U+02208
&notin; U+02209
&ni; U+0220B
&prod; U+0220F
&sum; U+02211
&minus; U+02212
&lowast; U+02217
&radic; U+0221A
&prop; U+0221D
&infin; U+0221E
&ang; U+02220
&and; U+02227
&or; U+02228
&cap; U+02229
&cup; U+0222A
&int; U+0222B
&there4; U+02234
&sim; U+0223C
&cong; U+02245
&asymp; U+02248
&ne; U+02260
&equiv; U+02261
&le; U+02264
&ge; U+02265
&sub; U+02282
&sup; U+02283
&nsub; U+02284
&sube; U+02286
&supe; U+02287
&oplus; U+02295
&otimes; U+02297
&perp; U+022A5
&sdot; U+022C5
&lceil; U+02308
&rceil; U+02309
&lfloor; U+0230A
&rfloor; U+0230B
&lang; U+02329
&rang; U+0232A
&loz; U+025CA
&spades; U+02660
&clubs; U+02663
&hearts; U+02665
&diams; U+02666