Share a regular expression

---

One 、 Regular expressions

2022 year 7 month 14 Japan , Our group came across a regular expression , It feels quite complicated , At that time, everyone was baffled , Later, it took some time to figure it out . The regular expression is as follows ：

/^(?:[^#<]*(<[wW]+>)[^>]*$)/

Readers who are interested can think for themselves first , Then continue to look at the following content .

Two 、 The analysis process

Usually you see a regular expression , To figure out the logic , A wave of simplification is needed first .
First look from left to right , A ^, Indicates the beginning of the qualification , Last $ Indicates a qualified end .
Then I saw in () In the ?:, What does that mean ？ You need to know the following first ：
() The content in itself represents capture packets , After the content is regularly recognized , Will be saved for future use . and
If this happens (?:), It means non capture packets , The only difference from capture grouping is , The matching values of non captured groups will not be saved , So for the server , Reduce a stored procedure , Reduce the consumption of resources , It can also improve the efficiency of regularization .
So for analyzing the regular expression itself , Is there any ?: It doesn't make any difference , You can simplify it for the first time ：

/^([^#<]*(<[wW]+>)[^>]*$)/

Then look at [^#<], Don't be trapped. # Scared me. , Actually, after testing ,^#< and ^< There is no difference here , It means that they don't match <
So do the second simplification ：

/^([^<]*(<[wW]+>)[^>]*$)/

Then do the third simplification , Obviously ,* The existence of can be completely ignored , Delete directly , So as to realize the third simplification ：

/^([^<](<[wW]+>)[^>]$)/

Now look at the end $, There is no difference between putting it inside and outside the brackets , So just move it outside , Realize the fourth simplification ：

/^([^<](<[wW]+>)[^>])$/

Now simplify again , Namely + Its function is to connect , You can delete , So do the fifth simplification as follows ：

/^([^<](<[wW]>)[^>])$/

You can see , The parentheses in the middle mean , Only accept or these two situations .
The brackets on the left indicate , Can accept a non < Any character of , Note that yes , Not necessarily .
The bracket on the right is the same , Can accept a non > Any character of , Note that yes , Not necessarily .
Therefore, there can be countless people who conform to the rules , The core is to include or , The beginning and end can be followed by any character other than angle brackets .
Analysis complete , It can be used as follows php Try the code test ：

<?php 
$a=$_REQUEST['a'];
var_dump(Preg_match_all('/^(?:[^#<]*(<[wW]+>)[^>]*$)/',$a,$b));
echo '<br />';
?>

First, test the transmission parameters <w> and <W> as follows ：


Try passing a non angle bracket character at any beginning and end ：

succeed , It shows that the previous analysis is correct .

---

3、 ... and 、 Summary

This article will share with you a regular expression encountered in the work , I hope that's helpful .