How do you change this to a preg_replace() equivalent (without using a loop)?
Note: $text is utf-8.
do $text = str_replace("* *", "*", $text, $totRepla); while ($totRepla);
-
Please provide a text example.koen– koen2011-04-27 18:55:05 +00:00Commented Apr 27, 2011 at 18:55
-
thaks to the one that change the subject. I was afraid about an horible google indexing top rank on a tribial and poorly formulated questionSaic Siquot– Saic Siquot2011-04-27 19:19:22 +00:00Commented Apr 27, 2011 at 19:19
Add a comment
|
4 Answers
I believe the regexp pattern to match an arbitrary long * * * * * * is
/(\* )*\*/
Please add the rest of the code yourself, I don't have PHP handy right now to provide a full code snippet.
3 Comments
Saic Siquot
thanks i will use: /(* )+*/ all other exposed solutions, do not reduce multiple ocurences to just one.
fingerman
Would you try to benchmark your initial solution against regex one? It's interesting which runs faster :D
Saic Siquot
I'm sure it depends on how match consecutive asterisks do the $text have. I bet regex is more eficient (cos compilated) with more than 10 on the input string. If you write the code I run it on my server also.
Actually, String functions are much more efficient then regex, I see no reason using regex if you have a working solution using str_replace.
http://php.net/manual/en/function.str-replace.php
see documentation:
If you don't need fancy replacing rules (like regular expressions), you should always use this function instead of preg_replace().
1 Comment
Alix Axel
str_replace() isn't necessarily more efficient if it's used in a do ... while loop.
The PCRE equivalent of that is:
$text = preg_replace('~(?:[*] )+[*]~', '*', $text);
Your code suggests that you have no interest in knowing how many replacements occurred, so you don't need the $totRepla variable. However, if I am wrong, this is the way to get that count:
$text = preg_replace('~(?:[*] )+[*]~', '*', $text, -1, $totRepla);
