Scan HTML for values with a special character before them

Question

Say I have values on my page, like #100 #246, What I want to do is scan the page for values with a # before them and then alter them to put a hyperlink on it

$MooringNumbers = '#!' . $MooringNumbers . ' | ' . '#!' . $row1["Number"];
  }  
    $viewedResult =  '<tr><td>' .$Surname.'</td><td>'.$Title.'</td><td>'.$MooringNumbers . '</td><td>'.$Telephone.'</td><td>' . '<a href="rlayCustomerUpdtForm.php?id='.$id.'">[EDIT]</a></td>'.'<td>'. '<a href="deleteCustomer.php?id='.$id.'">[x]</a></td>'. '</tr>';   preg_replace('/#!(\d\d\d)/', '<a href="update.php?id=${1}">${1}</a>', $viewedResult);
    echo $viewedResult;

This is the broken code which doesnt work.

Majki · Accepted Answer · 2010-12-31T11:39:52.320

2

I second Xoc - use PHP manual. The method next to the one he pointed is preg-replace-callback

Just call:

preg_replace_callback(
        '/#\d\d\d/',
        create_function(
            // single quotes are essential here,
            // or alternative escape all $ as \$
            '$matches',
            'return strtolower($matches[0]);' //this you replace with what you want to fetch from database
        )

EDIT: Since you want to always perform the same replacement go with Xoc's preg-replace:

preg_replace('/#!(\d\d\d)/', '<a href="update.php?id=${1}">${1}</a>', $your_input);

Note: I don't have PHP here, so I give no guarantee of this code not wiping your entire hard disk ;)

edited Dec 31 '10 at 11:39

answered Dec 31 '10 at 11:04

Majki

618
7
14

What if... html goes like that?
Normal link Normal text, Product price : 450 ™
. Here preg_replace wont work like expected... – markcial Dec 31 '10 at 11:07
Still he has to come with a unique pattern, no matter what solution he uses for replacing... – Majki Dec 31 '10 at 11:08
@Majki, welcome to Stack Overflow! =) If you have a minute, you should take a look at the [Stack Overflow Markdown help-page](http://stackoverflow.com/editing-help/) for how to format code *as* code. – David Thomas Dec 31 '10 at 11:09
But since you pointed out quite an important example, one could change pattern to /[^&](#\d\d\d)/ – Majki Dec 31 '10 at 11:10
On the other hand - if you can't come up with any unique pattern, but you know that your special code appears only in then you should probably start parsing HTML as shown in posts that Markcial pointed you to. – Majki Dec 31 '10 at 11:20
@David Thomas, thanks for pointing that function out. Hadn't noticed that one yet. – joelcox Dec 31 '10 at 11:26
All I want is for the PHP to Scan the HTML page for all bits of text labeled #!123 or any other value with #! before it. Then Automatically change it into a link like 123 – DogPooOnYourShoe Dec 31 '10 at 11:26
1

Hey! So your values are not random, but you always change them the same way. Then stick with what Xoc gave you in his first post. – Majki Dec 31 '10 at 11:35
Not totally random, but they are so many I just dont want to have to write out all of them, When I would expect there would be something to automatically send it in a GET URL. – DogPooOnYourShoe Dec 31 '10 at 11:39
Sadly, it didnt work. My code was still viewed normally. I will post the code which doesnt work so hopefully someone can fix it – DogPooOnYourShoe Dec 31 '10 at 11:46
What it looks like is: #!013223 #!233234 even after using Preg_replace – DogPooOnYourShoe Dec 31 '10 at 11:48
First: you need to assign the value back to the variable: $viewedResult = preg_replace(...); – Majki Dec 31 '10 at 11:54
Second: If your codes have 6 digits you have to change your regular expression to : '/#!(\d\d\d\d\d\d)/' or '/#!(\d{6})/' – Majki Dec 31 '10 at 11:55
I don't understand, please could you explain more? – DogPooOnYourShoe Dec 31 '10 at 11:55
The outcome of the Numbers is not upto me, they are just example ones and can have multiple amounts of digits – DogPooOnYourShoe Dec 31 '10 at 11:56
The last part of your should go like this: $viewedResult = preg_replace('/#!(\d{3,6})/', '${1}', $viewedResult); – Majki Dec 31 '10 at 11:59
the numbers {3,6} mean that there has to be at least 3 digits and not more than 6 - change as needed – Majki Dec 31 '10 at 12:03

joelcox · Answer 2 · 2010-12-31T11:23:34.387

0

You can accomplish this by using regular expressions, see PHP's preg_replace function.

$text = 'Lorem ipsum #300 dolar amet #20';

preg_match_all('/(^|\s)#(\w+)/', $text, $matches);

// Perform you database magic here for each element in $matches[2]
var_dump($matches[2]);

// Fake query result
$query_result = array ( 300 => 'http://www.example1.com', 20 => 'http://www.example2.com');

foreach($query_result as $result_key => $result_value)
{
    $text = str_replace('#'.$result_key, '<a href="'.$result_value.'">'. $result_value . '</a>', $text);
}

var_dump($text);

edited Dec 31 '10 at 11:23

answered Dec 31 '10 at 10:48

joelcox

562
9
19

My values are random, and I do not know what all of them will be as they are pulled down from a database. That page just shows how to do it if I knew what the values were. – DogPooOnYourShoe Dec 31 '10 at 10:51
IMHO this articles are relevant and completely related http://stackoverflow.com/questions/590747/using-regular-expressions-to-parse-html-why-not http://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags – markcial Dec 31 '10 at 10:53
So each #integer has a URL stored in the database? I would write a regex to find all #s, put the integers in an array, query your database and perform a str_replace() for each array element. – joelcox Dec 31 '10 at 11:03
I added a code example to my original answer that uses the algorithm I described. – joelcox Dec 31 '10 at 11:28
$query_result = array ( 300 => 'http://www.example1.com', 20 => 'http://www.example2.com'); That is the bit that I dont want to use, as my values are random and only need to be submitted like – DogPooOnYourShoe Dec 31 '10 at 11:34
Is there any way to automatically convert the Found String with # Before it, to a URL? – DogPooOnYourShoe Dec 31 '10 at 11:36

Scan HTML for values with a special character before them

2 Answers2