Function that can identify certain texts - php

Question

Here's my problem. I'll try to keep it concise:
I have a wab page that displays the texts in the selected language like this:
<p><?php echo $langFile->data->message; ?></p>
The actual text can look like this:
Hello, %1. You have %2 orders. If you %forgot% your password click %here% !
So I have to write a function that can replace (called one time for each replacement) the %1 and %2 with certain texts, and the words between % to be replaced with a link text. The reason it's tricky is that I can have texts that look like this:
Bla bla %1, bla %yada yada %2 yada% bla bla bla.
So here, the function has to replace %yada yada %2 yada% with a linkable text and, when called again, can replace %2 with another text.
I'm kind of new with programming, so I could use some pointers since I don't know how to get where I want if even possible. Thanks a lot in advance.
Regards

Answer 1

$whatever = explode('%', $something);
Then run through it like any other array. If you can find a common denominator and explode it, you can usually get any puzzle of regex or if/then to work how you need. The logic practically reveals itself afterwards.
As an example, I tried to find a way to guarantee removal of ALL script tags. Honestly, no one's regex worked. Then, I found the common denominator in html that I can use. I came up with this. It will guarantee to strip all script tags. Of course, it's only this simple with script tags because they won't be nested. However, it's a great example of how to use "exploding logic" or "accordion coding" to get your mind wrapped around the problem and make it happen.
$h = explode('>', $html);
foreach($h as $k => $v){
$v = trim($v);
if(preg_match('/^(<script[.*]*)/ius', $v)){
$counter = $k;//starting key of the script stored for later
}elseif(preg_match('/([.*]*<\/script$)/ius', $v)){
$script = $k - $counter;//difference between opening and closing tag
$counter = 0;//reset counter for next script
for($i = $script; $i >= 0; $i--){//now use the keys to empty everything in between
$h[$k-$i] = '';
}
}
}
for($i = 0; $i <= count($h); $i++){
if($h[$i] != ''){
$ht[$i] = $h[$i];//clean out the blanks so when we implode it works right.
}
}
$html = implode('> ', $ht);//all scripts stripped.GUARANTEED
I believe you can use this same idea with what you are trying to do. Though your internal statements will vary, but you will begin to see how it can flow more easily and actually show progress on what you are trying to do.

Answer 2

Try this: preg_replace http://php.net/manual/es/function.preg-replace.php
May be %[1-9] for the %1, %2, %3 and %[A-Za-z ]+% for the other.
There is something important... You should change the % to any other caracter for one of the cases... that make easier the search, because...
This case is easy: Hello, %1. You have %2 orders. If you %forgot% your password click %here% !
This is not: Bla bla %1, bla %yada yada %2 yada% bla bla bla.
But this, for example is easier: Bla bla #1, bla %yada yada #2 yada% bla bla bla.

Answer 3

Rather than using % you can use string variables in the text (unless you're writing some sort of a dynamic thing that will allow people to add their own vars like a CMS). You would make the variables come out of a database or something like that for the different languages. Example:
$name = 'Bob';
$salutation = "Hello, $name.";
$order_count = '2';
if ($order_count == '1'){
$ordertext = "You have $order_count order.";
} else{
$ordertext = "You have $order_count orders.";
}
$forgot = 'forgot';
$link = 'http://www.domain.com/reset/password';
$forgotphrase = "If you $forgot your password click <a href='$link'>here</a>!";
$message = $salutation.$ordertext.$forgotphrase;
Output
Hello, Bob. You have 2 orders. If you forgot your password click here!
Update
To do this on a larger scale you could have a database table called 'phrasing'.
id,salutation,ordertext_plur,ordertext_sing,forgot,link,forgotphrase,language,updated
So in MySQL might look like this:
1,
'Hello, $name.',
'You have $order_count orders.',
'You have $order_count order.',
'forgot',
'http://www.domain.com/reset/password/en',
'If you $forgot your password click <a href='$link'>here</a>!',
'EN',
'2013-04-23 00:00:00'
Your SQL statement would look like:
select * from phrasing where language='EN';
Then you would assign the appropriate vars based on the output. You would have to give more thought to how you would implement this and how large the database should be.
Other thoughts You might consider using something like the Google Translate Gadget on your pages to translate the HTML code, instead of hard-coding translations into the database. Their database of translations is updated daily. Some of my large international clients use it instead of paying the money to translate into every language. Also it makes it easier to make changes to the code in one language without having to worry about missing it in others.