[RESOLVED] Validating file link

[Scripteen]

Hi,

I'm making simple form to upload files from links and want to make sure the user input link is actually a link to a file

Valid formats will be

http://www.site.com/file.php
http://site.com/file.php
www.site.com/file.php

but anything else should be refused by the script

How can I do this ?

Thanks in advance

[devinemke]

if your server allows allow_url_fopen then you can use [man]fopen[/man] to test that the links are valid

[Scripteen]

devinemke;10904720 wrote:

if your server allows allow_url_fopen then you can use [man]fopen[/man] to test that the links are valid

Sorry you got me wrong. I don't want to test if the link is online or not. I want to verify that the string is a file link pattern. I think preg_match is used for this but dunno how.

[nrg_alpha]

You mean something along these lines?

$str = array('http://www.site.com/file.php', 'http://site.com/file.php', 'www.site.com/file.php', 'site.com/somefolder/file.php', 'subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?site\.com/file\.php$#', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
}

Granted, in this case, the site name and file is hard coded (site.com and file.php respectively). The pattern can be easily modified to be more open ended on this issue if need be.

EDIT: Also, if you want it to be case insensitive, you can simply add the i modifier after the closing delimiter.

[Scripteen]

nrg_alpha;10904754 wrote:

You mean something along these lines?

$str = array('http://www.site.com/file.php', 'http://site.com/file.php', 'www.site.com/file.php', 'site.com/somefolder/file.php', 'subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?site\.com/file\.php$#', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
}

Granted, in this case, the site name and file is hard coded (site.com and file.php respectively). The pattern can be easily modified to be more open ended on this issue if need be.

EDIT: Also, if you want it to be case insensitive, you can simply add the i modifier after the closing delimiter.

Thanks for your help but hard coded format is not usable for me as the url will be based on user input.

The user will be entering the url to a form and the script will grab this file to the server and provide local link

After some more search in google I found this

if(!preg_match('|^http(s)?://[a-z0-9-]+(\.[a-z0-9-]+)*(:[0-9]+)?(/.*)?$|i', $link1)) {
  print "Invalid image link"; 
  }

But this validates until host name but I want to make it validate full file link with extension.

[nrg_alpha]

A much looser 'quick and dirty' pattern could be:

$str = array('http://www.site.asia/file.php', 'http://some-site.co.uk/file-01.php', 'www.site.com/file.inc.php', 'site.com/somefolder/file.php', 'subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?[a-z0-9-]+\.(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/[\w-]+\.(php|x?html|asp)$#i', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
}

So with everything case insensitive:

http:// is optional.
so is www. (dot included)
Then it looks for any a-z0-9- one or more times (truth be told, I haven't really looked into what all the possible characters are allowed.. but this can be modified within the [a-z0-9-] character class.
then it looks for a dot, followed by [a-z]{2}.[a-z]{2} (think .co.uk) or alternatively, [a-z]{2,4} (ranges can be .ca to .com to .asia by example).
For the file name, [\w-]+.(php|x?html|asp) basically accepts a-zA-Z0-9_- one or more times, then the dot, followed by an alternation that contains a list of some existing file extentions..

There are issues of course.. this pattern would not allow a file like addon.inc.php by example. To solve this, we can alter the pattern to match practically anything like this:

#^(?:http://)?(www\.)?[a-z0-9-]+\.(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/(?:[\w-]+\.)+(php|x?html|asp)$#

I am assuming you want only specific file extensions.. but if you want to be even more open ended, you can change the final alternation within the pattern to just include anything occuring 3 or 4 times...

#^(?:http://)?(www\.)?[a-z0-9-]+\.(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/(?:[\w-]+\.)+[a-z]{3,4}$#i

So it really depends on how strict or how open ended you would like to go.. URL's are also tricky moving targets, so I hope my pattern (pick the one that suites you best) will work for what you have in mind.

[Scripteen]

nrg_alpha;10904770 wrote:

A much looser 'quick and dirty' pattern could be:

$str = array('http://www.site.asia/file.php', 'http://some-site.co.uk/file-01.php', 'www.site.com/file.inc.php', 'site.com/somefolder/file.php', 'subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?[a-z0-9-]+\.(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/[\w-]+\.(php|x?html|asp)$#i', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
}

So with everything case insensitive:

http:// is optional.
so is www. (dot included)
Then it looks for any a-z0-9- one or more times (truth be told, I haven't really looked into what all the possible characters are allowed.. but this can be modified within the [a-z0-9-] character class.
then it looks for a dot, followed by [a-z]{2}.[a-z]{2} (think .co.uk) or alternatively, [a-z]{2,4} (ranges can be .ca to .com to .asia by example).
For the file name, [\w-]+.(php|x?html|asp) basically accepts a-zA-Z0-9_- one or more times, then the dot, followed by an alternation that contains a list of some existing file extentions..

There are issues of course.. this pattern would not allow a file like addon.inc.php by example. To solve this, we can alter the pattern to match practically anything like this:

#^(?:http://)?(www\.)?[a-z0-9-]+\.(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/(?:[\w-]+\.)+(php|x?html|asp)$#

I am assuming you want only specific file extensions.. but if you want to be even more open ended, you can change the final alternation within the pattern to just include anything occuring 3 or 4 times...

#^(?:http://)?(www\.)?[a-z0-9-]+\.(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/(?:[\w-]+\.)+[a-z]{3,4}$#i

So it really depends on how strict or how open ended you would like to go.. URL's are also tricky moving targets, so I hope my pattern (pick the one that suites you best) will work for what you have in mind.

Thanks for your support.

Now I'm at this step

<? 	
$str = array('http://www.site.asia/file.php', 'http://some-site.co.uk/file-01.php', 'www.site.com/file.php', 'http://www.site.com/somefolder/file.php', 'site.com/somefolder/file.php', 'http://www.subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?[a-z0-9-]+\.(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/[\w-]+\.(php|x?html|asp)$#i', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
} 
 ?>

The output becomes

http://www.site.asia/file.php => valid format.
http://some-site.co.uk/file-01.php => valid format.
www.site.com/file.php => valid format.
http://www.site.com/somefolder/file.php => invalid format.
site.com/somefolder/file.php => invalid format.
http://www.subdomain.site.com/file.php => invalid format.

Now I want those "Invalid format" above to be valid so we have 6 valid formats in total.

[Jumanji]

nrg_alpha;10904754 wrote:

You mean something along these lines?

$str = array('http://www.site.com/file.php', 'http://site.com/file.php', 'www.site.com/file.php', 'site.com/somefolder/file.php', 'subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?site\.com/file\.php$#', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
}

Granted, in this case, the site name and file is hard coded (site.com and file.php respectively). The pattern can be easily modified to be more open ended on this issue if need be.

EDIT: Also, if you want it to be case insensitive, you can simply add the i modifier after the closing delimiter.

this is what im looking!
http://myfacefriends.com

[nrg_alpha]

Scripteen;10905349 wrote:

http://www.site.asia/file.php => valid format.
http://some-site.co.uk/file-01.php => valid format.
www.site.com/file.php => valid format.
http://www.site.com/somefolder/file.php => invalid format.
site.com/somefolder/file.php => invalid format.
http://www.subdomain.site.com/file.php => invalid format.

Now I want those "Invalid format" above to be valid so we have 6 valid formats in total.

So something like this then:

$str = array('http://www.site.asia/file.php', 'http://some-site.co.uk/file-01.php', 'www.site.com/file.php', 'http://www.site.com/somefolder/file.php', 'site.com/somefolder/file.php', 'http://www.subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?(?:[a-z0-9-]+\.)+(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/(?:[\w-]+/?)+\.(php|x?html|aspx?)$#i', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
}

This will validate all those formats.

[Scripteen]

nrg_alpha;10905792 wrote:

So something like this then:

$str = array('http://www.site.asia/file.php', 'http://some-site.co.uk/file-01.php', 'www.site.com/file.php', 'http://www.site.com/somefolder/file.php', 'site.com/somefolder/file.php', 'http://www.subdomain.site.com/file.php');
foreach ($str as $val) {
    echo $val . ' => ';
    echo (preg_match('#^(?:http://)?(www\.)?(?:[a-z0-9-]+\.)+(?:[a-z]{2}\.[a-z]{2}|[a-z]{2,4})/(?:[\w-]+/?)+\.(php|x?html|aspx?)$#i', $val))? "valid format.<br />\n" : "invalid format.<br />\n";
}

This will validate all those formats.

This is just what I needed. Hope you get married to preg soon and live happily ever after 😃

Thanks a ton buddy.

[nrg_alpha]

Excellent 🙂
Please don't forget to flag this thread as resolved.