Question

I use the following function to filter numbers from a string (including some special cases such as prices (starting with $)):

function _process_numbers($string) {
return preg_replace(
   '/(?<!\{)         # Assert no preceding {
    (?<![^\s$-])     # Assert no preceding non-whitespace except $-
    \b               # Match start of number
    (\d+(?:\.\d+)?+) # Match number (optional decimal part)
    \b               # Match end of number
    (?![^{}]*\})     # Assert that next brace is not a closing brace
    (?![^\s.!?,%])   # Assert no following non-whitespace except .!?,%
    /x', 
    '{NUMBER|\1}', $string
  );
}

echo _process_numbers("100.000.000 and -0.33 and $100.50 and 0.06%");

This works rather well, except that (1) the minus “-” should be included inside the braces of the output and (2) numbers up to 1 billion (and minus one billion) should be supported. In other words; the above returns as output:

{NUMBER|100.000}.000 and -{NUMBER|0.33} and ${NUMBER|100.50} and {NUMBER|0.06}%

But the expected output is:

{NUMBER|100.000} and {NUMBER|-0.33} and ${NUMBER|100.50} and {NUMBER|0.06}%

What should be changed?

Answer 1

should do the trick:

(-?\d+(?:\.\d+)?+) # Match number (optional decimal part)

UPDATE

remove the – from the look-behind assertion, capture the first (\b or -).

<?php

function _process_numbers($string) {
return preg_replace(
   '/(?<!\{)         # Assert no preceding {
    (?<![^\s$])     # Assert no preceding non-whitespace except $-
    (\b|-)               # Match start of number
    (\d+(?:\.\d+)*) # Match number (optional decimal part)
    \b               # Match end of number
    (?![^{}]*\})     # Assert that next brace is not a closing brace
    (?![^\s.!?,%])   # Assert no following non-whitespace except .!?,%
    /x', 
    '{NUMBER|\1\2}', $string
  );
}

echo _process_numbers("100.000.000 and -0.33 and $100.50 and 0.06%"), "\n";