php regex-仅匹配1次匹配不匹配一次

Question

I am trying to write regex that will catch this phrase:
background: url("../images/btn-send.png") no-repeat scroll -70px center rgba(0, 0, 0, 0);
but not this:
background: url("../images/btn-send.png") no-repeat scroll -70px 10px rgba(0, 0, 0, 0);

so I should get match only on ...<number>px ... and not
...<number>px <number>px ....

I have tried to use the next regex:
background\\s*:[^;]*?((-*\\d+px|0)\\s+){1}[az]
I am using it in PHP preg_match

I had tried to find only one occurrence of px, but it's not working and it's matching both of the strings, can you please help me with that? Thank!

Answer 1

Pending a more elegant answer (I'm not entirely convinced by this one), this would work:

<?php
$str1 = 'background: url("../images/btn-send.png") no-repeat scroll -70px  center rgba(0, 0, 0, 0);';
$str2 = 'background: url("../images/btn-send.png") no-repeat scroll -70px   10px   rgba(0, 0, 0, 0);';

var_dump(preg_match('/background\s*:[^;]*?((?<!\dpx|0|\s)\s+(-*\d+px|0)\s+)[a-z]/i',$str1));
var_dump(preg_match('/background\s*:[^;]*?((?<!\dpx|0|\s)\s+(-*\d+px|0)\s+)[a-z]/i',$str2));

So, what does (?<!\\dpx|0|\\s)\\s+ do? Is says your Npx/0 should be proceded by at least one whitespace character, but maybe more, not preceded by ( (?<!...) ) either whitespace (ensuring it matches from the first whitespace character when there are several), another Npx , or 0 .