help with str_replace or preg_replace

[mstringham]

I need help understanding whether to use preg_replace or str_replace.

I have the following :

{list}

{}list item 1
{}list item 2

{/list}

I need to replace{list} with the HTML <ul> tag
I need to replace {*} with the HTML <li> tag
I need to replace{/list} with the HTML </ul> tag

I've done a little research on both functions but don't know which to use - or really how to use them correctly here.

Any examples would be helpful

Thanks

[ednark]

user [man]str_replace[/man]... [man]preg_replace[/man] or [man]ereg_replace[/man] are only good for patterns... if you know the exact string you want to replace each time than you dont need regular expressions

to be nice and end your <li> with </li> you will need a regular expression... however this in not necessary

read up on the functions... they have the ability to be used as i am using them here... compile a list of things you want to replace and then call the function once... both str and preg work the same in that respect

<?php

header('Content-type: text/plain');

$orig_string = "
{list}

{*}list item 1
{*}list item 2

{/list} 
";

/// the str method

$str_find = array(
    '{list}',
    '{/list}',
    '{*}'
);
$str_replacement = array(
    '<ul>',
    '</ul>',
    '<li>'
);

$str_string = str_replace( $str_find, $str_replacement, $orig_string );

/// the preg method

$preg_find = array(
    '/\{list\}/',
    '/\{\/list\}/',
    '/\{\*\}([^\n]*)/'
);
$preg_replacement = array(
    '<ul>',
    '</ul>',
    '<li>\1</li>'
);

$preg_string = preg_replace( $preg_find, $preg_replacement, $orig_string );

echo "
 ---------------- : original string :  ----------------
";
echo $orig_string;
echo "
 ---------------- : str_replace :  ----------------
";
echo $str_string;
echo "
 ---------------- : preg_replace :  ----------------
";
echo $preg_string;

?>

OUTPUT:
 ---------------- : original string :  ----------------

{list}

{*}list item 1
{*}list item 2

{/list} 

 ---------------- : str_replace :  ----------------

<ul>

<li>list item 1
<li>list item 2

</ul> 

 ---------------- : preg_replace :  ----------------

<ul>

<li>list item 1
</li>
<li>list item 2
</li>

</ul> 

[mstringham]

Thanks for the explaination ednark I will give them a try -

I thought about using an array - but have read different things regarding the usage of preg and str replace.

Another question I have is regarding the preg_replace -
I have seen expressions with the /\/\/\ in them and $1 or $2 that seem to used as a delimeter of sorts - what's with the funky syntax?

thanks

[Weedpacket]

Originally posted by mstringham
Thanks for the explaination ednark I will give them a try -

I thought about using an array - but have read different things regarding the usage of preg and str replace.

Well, as ednark says, if you know the exact string you're trying to replace, str_replace() is perfectly adequate, and a lot faster (and less error-prone, because preg_replace() treats a lot of characters specially) than preg_replace.

Another question I have is regarding the preg_replace -
I have seen expressions with the /\/\/\ in them and $1 or $2 that seem to used as a delimeter of sorts - what's with the funky syntax?

This is covered in the manual pages about preg_* functions (particularly the section on "Pattern Syntax". The /\/\/ you mention is commonly referred to as "falling toothpick syndrome" - since / has a special meaning in regexps, if you really want to say / (in URLs, for example), you have to escape it - and the escape character is a . Of course, that means \ is a special character as well...

As for $1 and $2, those are back references (something that regular expressions aren't really supposed to be capable of - but this is programming, not language theory). If you wrap bits of your pattern in () then you can refer back to the strings those bits match with a $1 etc. (actually, \1 etc., but apparently $1 works as well) and say "... and then a repeat of that bit".

A simple pattern, that finds bits of strings that are doubled up (in this sentence, it would find the double "t" in "pattern"). In the previous paragraph there is a double s, a double m, a double p, a double l, a double ., and in this paragraph there is a deliberate doubling deliberate doubling of the phrase "deliberate doubling":

/(.+)\1/

[mstringham]

This code -
$orig_string = "
{list}

{}list item 1
{}list item 2

{/list}
";

/// the str method

$str_find = array(
'{list}',
'{/list}',
'{*}'
);
$str_replacement = array(
'<ul>',
'</ul>',
'<li>'
);

$str_string = str_replace( $str_find, $str_replacement, $orig_string );

}

Returns
{list} {}list item 1 {}list item 2 {/list}

Am I doing something wrong here?

Thanks

MS

[ednark]

should work

1) \n characters: you should either do a header('Content-type: text/plain') at the beginning or wrap your prints in <pre></pre>
so that \n get shown as newlines

2) are you sure you are printing out $str_string and not $orig_string...