[SOLVED] Select certain lines for stripslashes

Kommius · August 17, 2009

Hi everyone, :mrgreen:

I'm trying to code a relatively simple script.

I want to first read the contents of the file, and then apply a

stripslashes();

on all lines starting with the

echo "

tag and ending with

";

.

For now, here's what I was able to come up with :

$file_contents = file_get_contents($fichier);

if(preg_match_all('`^echo"(.*?)"', $file_contents, $matches)){

               foreach ($matches as $val) {
                         $processed .= stripslashes($val);
               } 
} else {
      $processed .= $file_contents; //-- What should i put here if the line doesn't correspond?
}

file_put_contents($fichier, $processed);

Are my heading towards the right direction? Can anybody help me out? Thx

DarkendSoul · August 17, 2009

This may work for you...

<?php

ob_start();

include($fichier);

$output = stripslashes(ob_get_contents());

ob_end_clean();

echo $output;

?>

ignace · August 17, 2009

What are you actually trying to accomplish? Do you want to read a php file, then search for all echo statements and add slashes to strings? You probably need to extend it as ' is also valid.

thebadbad · August 17, 2009

You can use preg_replace() with the e pattern modifier, treating the replacement as PHP. Or use preg_replace_callback():

<?php
$file_contents = file_get_contents($fichier);

$file_contents = preg_replace_callback(
'~^echo\s*([\'"])(.*?)\1;~sm',
create_function(
	'$matches',
	'return stripslashes($matches[2]);'
),
$file_contents
);
?>

The pattern matches echo at the start of the string or at the start of a line (m modifier). Then zero or more whitespace characters followed by either a single or a double quote. Then anything including new lines (s modifier) are matched and captured, until the opening quote (\1) followed by a semi colon is found.

But this will yield unexpected results with code like

echo "testing with a $var" . 'the end';

DarkendSoul · August 17, 2009

thebadbad: I'm curious... could that problem be solves some how with something like... (([\'"])(.*?)\1.*)+

thebadbad · August 17, 2009

In order for it to work flawlessly, we would have to write our own parser, taking every thing into account. And that's not easy, as you might have guessed.

Kommius · August 17, 2009

What are you actually trying to accomplish? Do you want to read a php file, then search for all echo statements and add slashes to strings? You probably need to extend it as ' is also valid.

The goal is to parse a file such as this :

#==================================================
# function HTTPReferer
# Liste des référants
#==================================================
function HTTPReferer() {
global $bgcolor1, $bgcolor2, $bgcolor3, $db, $admin_file;
include_once("header.php");
NCGraphicAdmin();

OpenTable();
echo "<center><b>"._NCWHOLINKS."</b></center><br /><br />";
echo "<table width=\"100%\" border=\"0\" cellspacing=\"1\" bgcolor=\"$bgcolor2\">";
echo "<tr>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCHITS."<b></td>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCURL."</b></td>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCUPDATED."</b></td>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCDELETE."</b></td>";
echo "</tr>";

$req_refs = $db->sql_query("SELECT r_id, r_url, r_count, r_lasttime FROM ".TABLE_REFERER." ORDER BY r_count DESC");
$num_refs = $db->sql_numrows($req_refs);

if (!empty($num_refs)) {
	while ($row = $db->sql_fetchrow($req_refs)) {
			$r_id = intval($row['r_id']);
			$r_url = filter($row['r_url'], nohtml);
			$r_url2 = urlencode($r_url);
			$r_count = intval($row['r_count']);
			$r_lasttime = formatTimestamp($row['r_lasttime']);

			echo "<tr><td bgcolor=\"$bgcolor1\" align=\"center\"><font class=\"content\">$r_count</font></td>";
			echo "<td bgcolor=\"$bgcolor1\" width=\"100%\"><font class=\"content\"><a href=\"index.php?url=$r_url2\" target=\"_blank\">$r_url</a></font></td>";
			echo "<td bgcolor=\"$bgcolor1\" nowrap> <font class=\"content\">$r_lasttime</font> </td>";
			echo "<td bgcolor=\"$bgcolor1\" align=\"center\"><a href=\"".$admin_file.".php?op=DelReferer&ref_id=".$r_id."\"><img src=\"images/objets/wrong.gif\" border=\"0\" alt=\""._NCDELETE."\" title=\""._NCDELETE."\"></a></td>";
			echo "</tr>";
		}
} else {
	echo "<tr><td align=\"center\" bgcolor=\"$bgcolor1\" colspan=\"4\">"._NCNOREFERERS."</td></tr>";
}
echo "</table><br />";
echo "<form action=\"".$admin_file.".php\" method=\"post\">";
echo "<input type=\"hidden\" name=\"op\" value=\"DelReferer\">";
echo "<input type=\"hidden\" name=\"ref_id\" value=\"all\">";
echo "<center><input type=\"submit\" value=\""._NCDELETEALL."\"></center>";
CloseTable();

include_once("footer.php");
}

The thing is, if I do a stripslashes on the entire file contents, the stripslashes function removes all backslashes, including the ones contained in the PHP Code.

The script would also allow me to parse the file with a more complicated function, based on the htmLawed library.

Kommius · August 17, 2009

You can use preg_replace() with the e pattern modifier, treating the replacement as PHP. Or use preg_replace_callback():
<?php
$file_contents = file_get_contents($fichier);

$file_contents = preg_replace_callback(
'~^echo\s*([\'"])(.*?)\1;~sm',
create_function(
	'$matches',
	'return stripslashes($matches[2]);'
),
$file_contents
);
?>
The pattern matches echo at the start of the string or at the start of a line (m modifier). Then zero or more whitespace characters followed by either a single or a double quote. Then anything including new lines (s modifier) are matched and captured, until the opening quote (\1) followed by a semi colon is found.

But this will yield unexpected results with code like
echo "testing with a $var" . 'the end';

Your function looks quite interesting, I will have a look at it and get back to you with more feedback.. As you pointed out, it may yield unexpected result.

thebadbad · August 17, 2009

How do you want to parse the file? 'Cause if you remove the slashes you will obviously mess with the syntax. Do you want to grab the output of the script?

Kommius · August 17, 2009

How do you want to parse the file? 'Cause if you remove the slashes you will obviously mess with the syntax. Do you want to grab the output of the script?

Removing the slashes allows the htmLawed library to read and modify it. The output of the script adds the backslashes again in the xhtml parts of the PHP file.

Currently, the full script looks like this :

$file_contents = file_get_contents($list);

//-- Removing slashes
if(get_magic_quotes_gpc()) {
	$text = stripslashes($file_contents);
}

$processed = htmLawed($text);

file_put_contents($list, $processed);

For now, the script manages to transform :

#==================================================
# function HTTPReferer
# Liste des référants
#==================================================
function HTTPReferer() {
global $bgcolor1, $bgcolor2, $bgcolor3, $db, $admin_file;
include_once("header.php");
NCGraphicAdmin();

OpenTable();
echo "<center><b>"._NCWHOLINKS."</b></center><br /><br />";
echo "<table width=\"100%\" border=\"0\" cellspacing=\"1\" bgcolor=\"$bgcolor2\">";
echo "<tr>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCHITS."<b></td>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCURL."</b></td>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCUPDATED."</b></td>";
echo "<td bgcolor=\"$bgcolor3\" align=\"center\"><b>"._NCDELETE."</b></td>";
echo "</tr>";

$req_refs = $db->sql_query("SELECT r_id, r_url, r_count, r_lasttime FROM ".TABLE_REFERER." ORDER BY r_count DESC");
$num_refs = $db->sql_numrows($req_refs);

if (!empty($num_refs)) {
	while ($row = $db->sql_fetchrow($req_refs)) {
			$r_id = intval($row['r_id']);
			$r_url = filter($row['r_url'], nohtml);
			$r_url2 = urlencode($r_url);
			$r_count = intval($row['r_count']);
			$r_lasttime = formatTimestamp($row['r_lasttime']);

			echo "<tr><td bgcolor=\"$bgcolor1\" align=\"center\"><font class=\"content\">$r_count</font></td>";
			echo "<td bgcolor=\"$bgcolor1\" width=\"100%\"><font class=\"content\"><a href=\"index.php?url=$r_url2\" target=\"_blank\">$r_url</a></font></td>";
			echo "<td bgcolor=\"$bgcolor1\" nowrap> <font class=\"content\">$r_lasttime</font> </td>";
			echo "<td bgcolor=\"$bgcolor1\" align=\"center\"><a href=\"".$admin_file.".php?op=DelReferer&ref_id=".$r_id."\"><img src=\"images/objets/wrong.gif\" border=\"0\" alt=\""._NCDELETE."\" title=\""._NCDELETE."\"></a></td>";
			echo "</tr>";
		}
} else {
	echo "<tr><td align=\"center\" bgcolor=\"$bgcolor1\" colspan=\"4\">"._NCNOREFERERS."</td></tr>";
}
echo "</table><br />";
echo "<form action=\"".$admin_file.".php\" method=\"post\">";
echo "<input type=\"hidden\" name=\"op\" value=\"DelReferer\">";
echo "<input type=\"hidden\" name=\"ref_id\" value=\"all\">";
echo "<center><input type=\"submit\" value=\""._NCDELETEALL."\"></center>";
CloseTable();

include_once("footer.php");
}

to

#==================================================
# function HTTPReferer
# Liste des référants
#==================================================
function HTTPReferer() {
global $bgcolor1, $bgcolor2, $bgcolor3, $db, $admin_file;
include_once("header.php");
NCGraphicAdmin();

OpenTable();
echo "<center><b>"._NCWHOLINKS."<b><center><br /><br />";
echo "<table width=\"100%\" border=\"0\" cellspacing=\"1\" style=\"background-color: $bgcolor2;\">";
echo "<tr>";
echo "<td align=\"center\" style=\"background-color: $bgcolor3;\"><b>"._NCHITS."<b><td>";
echo "<td align=\"center\" style=\"background-color: $bgcolor3;\"><b>"._NCURL."<b><td>";
echo "<td align=\"center\" style=\"background-color: $bgcolor3;\"><b>"._NCUPDATED."<b><td>";
echo "<td align=\"center\" style=\"background-color: $bgcolor3;\"><b>"._NCDELETE."<b><td>";
echo "<tr>";

$req_refs = $db->sql_query("SELECT r_id, r_url, r_count, r_lasttime FROM ".TABLE_REFERER." ORDER BY r_count DESC");
$num_refs = $db->sql_numrows($req_refs);

if (!empty($num_refs)) {
	while ($row = $db->sql_fetchrow($req_refs)) {
			$r_id = intval($row['r_id']);
			$r_url = filter($row['r_url'], nohtml);
			$r_url2 = urlencode($r_url);
			$r_count = intval($row['r_count']);
			$r_lasttime = formatTimestamp($row['r_lasttime']);

			echo "<tr><td align=\"center\" style=\"background-color: $bgcolor1;\"><font class=\"content\">$r_count<font><td>";
			echo "<td style=\"background-color: $bgcolor1; width: 100%;\"><font class=\"content\"><a href=\"index.php?url=$r_url2\" target=\"_blank\">$r_url<a><font><td>";
			echo "<td style=\"background-color: $bgcolor1; white-space: nowrap;\"><font class=\"content\">$r_lasttime<font><td>";
			echo "<td align=\"center\" style=\"background-color: $bgcolor1;\"><a href=\"".$admin_file.".php?op=DelRefererref_id=\"><img src=\"images/objets/wrong.gif\" alt=\"\" title=\"\" style=\"border: 0px;\" /><a><td>";
			echo "<tr>";
		}
} else {
	echo "<tr><td align=\"center\" colspan=\"4\" style=\"background-color: $bgcolor1;\">"._NCNOREFERERS."<td><tr>";
}
echo "<table><br />";
echo "<form action=\"".$admin_file.".php\" method=\"post\">";
echo "<input type=\"hidden\" name=\"op\" value=\"DelReferer\" />";
echo "<input type=\"hidden\" name=\"ref_id\" value=\"all\" />";
echo "<center><input type=\"submit\" value=\"\" /><center>";
CloseTable();

include_once("footer.php");
}

There are still problems though, as all encapsed values in double-double quotes return as empty values.

Kommius · August 17, 2009

You can use preg_replace() with the e pattern modifier, treating the replacement as PHP. Or use preg_replace_callback():
<?php
$file_contents = file_get_contents($fichier);

$file_contents = preg_replace_callback(
'~^echo\s*([\'"])(.*?)\1;~sm',
create_function(
	'$matches',
	'return stripslashes($matches[2]);'
),
$file_contents
);
?>
The pattern matches echo at the start of the string or at the start of a line (m modifier). Then zero or more whitespace characters followed by either a single or a double quote. Then anything including new lines (s modifier) are matched and captured, until the opening quote (\1) followed by a semi colon is found.

But this will yield unexpected results with code like
echo "testing with a $var" . 'the end';
Your function looks quite interesting, I will have a look at it and get back to you with more feedback.. As you pointed out, it may yield unexpected result.

I tried your function, it does not yield any results (any typo in there?)

Even doing a :

$test = preg_match_all('~^echo\s*([\"])(.*?)\1;~sm', $file_contents, $matches);

does not match anything in a file containing these lines :

echo "<select>";
echo "<td><tr>";

I'm looking into it..

thebadbad · August 17, 2009

That's because the caret (^) matches at the start of the string and at the start of a line. You have tabs in front of echo. You can simply remove the caret to make it match echo anywhere.

DarkendSoul · August 17, 2009

Or make it include the possibility for tabs and spaces \t|\s (with word boundaries).

'~^\b\t*|\s*\becho\s*([\'"])(.*?)\1;~sm'

But then again if you have tabs and spaces it would be \t*\s* (no need for boundaries).

thebadbad · August 17, 2009

In that case you could use \s* for any optional whitespace characters (including spaces, tabs and newlines).

'~^\s*echo\s*([\'"])(.*?)\1;~sm'

Kommius · August 17, 2009

Thank you so much for your help and your quick replies, the script works like a charm..

I can't thank you enough for your patience and your help, this has helped me understand regular expressions and the usage of preg_replace_callback, which I find is complicated to understand with the PHP manual!!

PS : Can somebody set this topic as solved please?

ignace · August 17, 2009

In order for it to work flawlessly, we would have to write our own parser, taking every thing into account. And that's not easy, as you might have guessed.

All it takes is an implementation of the Interpreter pattern

ignace · August 17, 2009

Can somebody set this topic as solved please?

You can set it as solved by clicking 'Solved' below left above 'Quick Reply'

Kommius · August 17, 2009

Can somebody set this topic as solved please?

You can set it as solved by clicking 'Solved' below left above 'Quick Reply'

Thx ::)

thebadbad · August 17, 2009

Thank you so much for your help and your quick replies, the script works like a charm..

I can't thank you enough for your patience and your help, this has helped me understand regular expressions and the usage of preg_replace_callback, which I find is complicated to understand with the PHP manual!!

PS : Can somebody set this topic as solved please?

You're welcome! Just remember that the script isn't very versatile, and won't convert every variant of echo blocks you throw at it.

Happy coding

Sign In

[SOLVED] Select certain lines for stripslashes

Recommended Posts

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Join the conversation

Important Information