
Alex
Staff Alumni-
Posts
2,467 -
Joined
-
Last visited
Everything posted by Alex
-
Call mysql_close() after $id = mysql_insert_id();
-
No, it's not the same content, completely different. If you try to get the contents of that page you get: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=us-ascii"> <title>Download xml.xml</title> <script type="text/javascript" language="JavaScript"><!-- function docountdown(){ s_div = document.getElementById("dldiv"); count = count - 1; if (count > 0){ s_div.innerHTML = "Your request to download xml.xml is being processed.<"+"p>Please Wait... ("+count+" seconds remaining)<"+"/p>"; setTimeout("docountdown();", 1000); } else { s_div.innerHTML = "Your file is ready for download.<"+"p><"+"a href=\"/gr/xml.xml?\">Click here to Download File<"+"/a><"+"/p>"; } } var count = 60; window.onload = docountdown; --></script> </head> <body> <div align="center"> <script type="text/javascript"><!-- google_ad_client = "pub-2927823587298693"; google_alternate_color = "FFFFFF"; google_ad_width = 468; google_ad_height = 60; google_ad_format = "468x60_as"; google_ad_channel =""; google_color_border = "336699"; google_color_bg = "FFFFFF"; google_color_link = "0000FF"; google_color_url = "008000"; google_color_text = "000000"; //--></script> <script type="text/javascript" src="http://pagead2.googlesyndication.com/pagead/show_ads.js"> </script> <br /><br /> <table border="0" width="500"> <tr> <td> <fieldset style="padding: 2"> <legend>Download xml.xml</legend> <br /> <div id="dldiv"> Your request to download xml.xml is being processed.<p>Please Wait...</p> </div> <br /> </fieldset> <p> </td> </tr> </table> <script type="text/javascript"><!-- google_ad_client = "pub-2927823587298693"; google_alternate_color = "FFFFFF"; google_ad_width = 468; google_ad_height = 60; google_ad_format = "468x60_as"; google_ad_channel =""; google_color_border = "336699"; google_color_bg = "FFFFFF"; google_color_link = "0000FF"; google_color_url = "008000"; google_color_text = "000000"; //--></script> <script type="text/javascript" src="http://pagead2.googlesyndication.com/pagead/show_ads.js"> </script> </div> </body> </html>
-
Definitely not one of the most useful features in PHP, but it can come in handy sometimes, especially when you plan on distributing your code.
-
That won't work because http://fr33.ulmb.com/gr/xml.xml prompts you for a 60 second wait before downloading the file, so you'll be getting the contents of that page, not the actual xml file.
-
The point of escaping data isn't really to find out who's trying to hack your website and ban them. It's to protect just in case, someone might typo a < or something, that's no reason to ban them. You could create a black-list of things to ban people based on (like <meta..) that would give a better indication that they were trying to post something malicious, but I don't even think that's necessary. As long as you're correctly escaping user input someone trying to post a meta refresh is no more harmful than someone spamming, which some moderation will solve.
-
You're obviously having a problem getting the contents of the file. Check to see if $myXML contains the contents of the file you expect it does.
-
If someone tries to post "Hope you feel better & get well soon" you'd take that as an attack on your website?
-
Use date_default_timezone_set. Btw, you don't need to use time() as the second parameter in date(), if the second parameter is omitted it defaults to that anyway.
-
Well, no that's not how you would do it. It doesn't matter though, it would extremely unfair to ban a user for posting something that would be converted to an htmlentity by htmlentities().. If someone posts &, <, >, ', ", (or anything else from this list) you would ban them? Doesn't seem very fair to me.
-
Uploading the xml file to a remote local myself and testing this it worked fine: $curl = curl_init(); curl_setopt($curl, CURLOPT_URL, 'http://website.com/somefile.xml'); curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1); $myXML = curl_exec ($curl); curl_close($curl); $sxml = simplexml_load_string($myXML); list($node) = $sxml->xpath('/playlist/trackList/track/location'); header("Location: $node"); I suggest you put error_reporting(E_ALL); at the top of your page and post back any errors.
-
htmlentities() doesn't convert commas..
-
That won't display the information that the OP is asking for.
-
htmlentities
-
Can you post the actual link to the website so I can try this myself?
-
As with many standards in programming (especially with OOP) interfaces aren't totally for you, they're for other programmers that might use your code. It's a good way to insure that an object will maintain certain methods.
-
Yes, except for use simplexml_load_string() inside of simplexml_load_file() now that you should have the contents of the XML file in $myXML.
-
That is not a multidimensional array. Neither $myArray["name"] nor $myArray["value"] is an array. Edit: Perhaps you meant something like this: $myArray = Array( 'name' => Array("SomeValue"), 'value' => Array("AnotherValue") ); array_push($myArray['name'], 'Added Name'); array_push($myArray['value'], 'Added Value'); echo "<pre>" . print_r($myArray, true) . "</pre>"; Output: Array ( [name] => Array ( [0] => SomeValue [1] => Added Name ) [value] => Array ( [0] => AnotherValue [1] => Added Value ) )
-
Here's an example: $curl = curl_init(); curl_setopt($curl, CURLOPT_URL, 'WEBSITE URL HERE'); curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1); $myXML = curl_exec ($curl); curl_close($curl); // .. The rest of your code
-
Lol, I forgot about that too.. I was pretty sure there was something like that, I went to test it and I must've typed it in wrong.
-
That's not my code.. You tried to add an unnecessary endforeach; to my code and put it in the wrong place which is what was making it not work. <?php foreach($_GET as $key => $val) $vars[] = "$key=$val"; echo implode('&', $vars);
-
It shouldn't do that. Are you sure you're not echoing out the first variable previously? Or have it twice in your url?..
-
Alternatively you could use httpquerystring. foreach($_GET as $key => $val) $vars[] = "$key=$val"; echo implode('&', $vars);
-
I thought you wanted to compare the $csv string to the indexes of the array. Try something like this: $arr = explode(', ', $csv); foreach($row as $key => $value) { echo '<select size="5" name="'.$key.'">'; $c = 0; foreach($subcats as $index1 => $value1) { echo '<option value="'.$index1.'"', $index1==$arr[$c] ? ' selected="selected"' : null ,'>'.$value1[1].'</option>'; $c++; } $c = 0; echo '</select>'; }
-
This is done using mod_rewrite. Here's a decent tutorial on it, http://articles.sitepoint.com/article/guide-url-rewriting
-
if($csv == implode(', ',array_keys($subcats))) { // ... } implode array_keys