Debug v20 - fix dol_escape_htmltag

This commit is contained in:
Laurent Destailleur
2024-07-22 20:12:35 +02:00
parent a735aaeb36
commit 1cb57b03ae
2 changed files with 43 additions and 23 deletions

View File

@@ -1962,6 +1962,7 @@ function dol_escape_htmltag($stringtoescape, $keepb = 0, $keepn = 0, $noescapeta
$tmparrayoftags = explode(',', $noescapetags); $tmparrayoftags = explode(',', $noescapetags);
} }
if (count($tmparrayoftags)) { if (count($tmparrayoftags)) {
$reg = array();
$tmp = str_ireplace('__DOUBLEQUOTE', '', $tmp); // The keyword DOUBLEQUOTE is forbidden. Reserved, so we removed it if we find it. $tmp = str_ireplace('__DOUBLEQUOTE', '', $tmp); // The keyword DOUBLEQUOTE is forbidden. Reserved, so we removed it if we find it.
foreach ($tmparrayoftags as $tagtoreplace) { foreach ($tmparrayoftags as $tagtoreplace) {
@@ -1970,30 +1971,37 @@ function dol_escape_htmltag($stringtoescape, $keepb = 0, $keepn = 0, $noescapeta
$tmp = preg_replace('/<'.preg_quote($tagtoreplace, '/').' \/>/', '__BEGINENDTAGTOREPLACE'.$tagtoreplace.'__', $tmp); $tmp = preg_replace('/<'.preg_quote($tagtoreplace, '/').' \/>/', '__BEGINENDTAGTOREPLACE'.$tagtoreplace.'__', $tmp);
// For case of tag with attribute // For case of tag with attribute
$reg = array(); do {
if (preg_match('/<'.preg_quote($tagtoreplace, '/').'\s+([^>]+)>/', $tmp, $reg)) { $tmpold = $tmp;
$tmpattributes = str_ireplace(array('[', ']'), '_', $reg[1]); // We must never have [ ] inside the attribute string
$tmpattributes = str_ireplace('href="http:', '__HREFHTTPA', $tmpattributes); if (preg_match('/<'.preg_quote($tagtoreplace, '/').'\s+([^>]+)>/', $tmp, $reg)) {
$tmpattributes = str_ireplace('href="https:', '__HREFHTTPSA', $tmpattributes); $tmpattributes = str_ireplace(array('[', ']'), '_', $reg[1]); // We must never have [ ] inside the attribute string
$tmpattributes = str_ireplace('src="http:', '__SRCHTTPIMG', $tmpattributes); $tmpattributes = str_ireplace('href="http:', '__HREFHTTPA', $tmpattributes);
$tmpattributes = str_ireplace('src="https:', '__SRCHTTPSIMG', $tmpattributes); $tmpattributes = str_ireplace('href="https:', '__HREFHTTPSA', $tmpattributes);
$tmpattributes = str_ireplace('"', '__DOUBLEQUOTE', $tmpattributes); $tmpattributes = str_ireplace('src="http:', '__SRCHTTPIMG', $tmpattributes);
$tmpattributes = preg_replace('/[^a-z0-9_\/\?\;:\s=&\.-]/i', '', $tmpattributes); $tmpattributes = str_ireplace('src="https:', '__SRCHTTPSIMG', $tmpattributes);
//$tmpattributes = preg_replace("/float:\s*(left|right)/", "", $tmpattributes); // Disabled: we must avoid escaping but not remove content $tmpattributes = str_ireplace('"', '__DOUBLEQUOTE', $tmpattributes);
$tmp = preg_replace('/<'.preg_quote($tagtoreplace, '/').'\s+([^>]+)>/', '__BEGINTAGTOREPLACE'.$tagtoreplace.'['.$tmpattributes.']__', $tmp); $tmpattributes = preg_replace('/[^a-z0-9_\/\?\;\s=&\.\-@:\.#\+]/i', '', $tmpattributes);
} //$tmpattributes = preg_replace("/float:\s*(left|right)/", "", $tmpattributes); // Disabled: we must not remove content
if (preg_match('/<'.preg_quote($tagtoreplace, '/').'\s+([^>]+)> \/>/', $tmp, $reg)) { $tmp = preg_replace('/<'.preg_quote($tagtoreplace, '/').'\s+'.preg_quote($reg[1], '/').'>/', '__BEGINTAGTOREPLACE'.$tagtoreplace.'['.$tmpattributes.']__', $tmp);
$tmpattributes = str_ireplace(array('[', ']'), '_', $reg[1]); // We must not have [ ] inside the attribute string }
$tmpattributes = str_ireplace('"', '__DOUBLEQUOTE', $tmpattributes); if (preg_match('/<'.preg_quote($tagtoreplace, '/').'\s+([^>]+)\s+\/>/', $tmp, $reg)) {
$tmpattributes = preg_replace('/[^a-z0-9_\/\?\;:\s=&]/i', '', $tmpattributes); $tmpattributes = str_ireplace(array('[', ']'), '_', $reg[1]); // We must not have [ ] inside the attribute string
//$tmpattributes = preg_replace("/float:\s*(left|right)/", "", $tmpattributes); // Disabled: we must avoid escaping but not remove content $tmpattributes = str_ireplace('"', '__DOUBLEQUOTE', $tmpattributes);
$tmp = preg_replace('/<'.preg_quote($tagtoreplace, '/').'\s+([^>]+) \/>/', '__BEGINENDTAGTOREPLACE'.$tagtoreplace.'['.$tmpattributes.']__', $tmp); $tmpattributes = preg_replace('/[^a-z0-9_\/\?\;\s=&\.\-@:\.#\+]/i', '', $tmpattributes);
} //$tmpattributes = preg_replace("/float:\s*(left|right)/", "", $tmpattributes); // Disabled: we must not remove content.
$tmp = preg_replace('/<'.preg_quote($tagtoreplace, '/').'\s+'.preg_quote($reg[1], '/').'\s+\/>/', '__BEGINENDTAGTOREPLACE'.$tagtoreplace.'['.$tmpattributes.']__', $tmp);
}
$diff = strcmp($tmpold, $tmp);
} while ($diff);
} }
} }
$result = htmlentities($tmp, ENT_COMPAT, 'UTF-8'); // Convert & into &amp; and more... $result = htmlentities($tmp, ENT_COMPAT, 'UTF-8'); // Convert & into &amp; and more...
//print $result;
if (count($tmparrayoftags)) { if (count($tmparrayoftags)) {
foreach ($tmparrayoftags as $tagtoreplace) { foreach ($tmparrayoftags as $tagtoreplace) {
$result = str_ireplace('__BEGINTAGTOREPLACE'.$tagtoreplace.'__', '<'.$tagtoreplace.'>', $result); $result = str_ireplace('__BEGINTAGTOREPLACE'.$tagtoreplace.'__', '<'.$tagtoreplace.'>', $result);

View File

@@ -1106,10 +1106,10 @@ class FunctionsLibTest extends CommonClassTest
/** /**
* testDolEscapeHtmlTag * testDolEscapeHtmlTag
* *
* @return void * @return void
*/ */
public function testDolEscapeHtmlTag() public function testDolEscapeHtmlTag()
{ {
$input = 'x&<b>#</b>,"'; // & and " are converted into html entities, <b> are removed $input = 'x&<b>#</b>,"'; // & and " are converted into html entities, <b> are removed
@@ -1123,6 +1123,18 @@ class FunctionsLibTest extends CommonClassTest
$input = '<img alt="" src="https://github.githubassets.com/assets/GitHub-Mark-ea2971cee799.png">'; // & and " are converted into html entities, <b> are not removed $input = '<img alt="" src="https://github.githubassets.com/assets/GitHub-Mark-ea2971cee799.png">'; // & and " are converted into html entities, <b> are not removed
$result = dol_escape_htmltag($input, 1, 1, 'common', 0, 1); $result = dol_escape_htmltag($input, 1, 1, 'common', 0, 1);
$this->assertEquals('<img alt="" src="https://github.githubassets.com/assets/GitHub-Mark-ea2971cee799.png">', $result); $this->assertEquals('<img alt="" src="https://github.githubassets.com/assets/GitHub-Mark-ea2971cee799.png">', $result);
$input = '<div style="float:left; margin-left:0px; margin-right:5px">
<img id="sigPhoto" src="https://www.domain.com/aaa.png" style="height:65px; width:65px" />
</div>
<div style="margin-left:74px"><strong>A text here</strong> and more<br>
<a href="mailto:abc+def@domain.com" id="sigEmail" style="color:#428BCA;">abc+def@domain.com</a><br>
<a href="https://www.another-domain.com" id="sigWebsite" style="color:#428BCA;">https://www.another-domain.com</a><br>
</div>';
$result = dol_escape_htmltag($input, 1, 1, 'common');
$this->assertEquals($input, $result);
} }