CODEWITHSUNDEEP
phpregex
fish man
fish man

Reputation: 2710

php preg_replace remove thousand separator in a string

there have a long articles, I want only remove thousand separator, not a comma.

$str = "Last month's income is 1,022 yuan, not too bad.";
//=>Last month's income is 1022 yuan, not too bad.

preg_replace('#(\d)\,(\d)#i','???',$str);

How to write the regex patterns? Thanks

Upvotes: 1

Views: 4624

Answers (3)

Tim Pietzcker
Tim Pietzcker

Reputation: 336408

If the simplified rule "Match any comma that lies directly between digits" is good enough for you, then

preg_replace('/(?<=\d),(?=\d)/','',$str);

should do.

You could improve it by making sure that exactly three digits follow:

preg_replace('/(?<=\d),(?=\d{3}\b)/','',$str);

Upvotes: 6

akirk
akirk

Reputation: 6857

The first (\d) is represented by $1, the second (\d) by $2. Therefore the solution is to use something like this:

preg_replace('#(\d)\,(\d)#','$1$2',$str);

Actually it would be better to have 3 numbers behind the comma to avoid causing havoc in lists of numbers:

preg_replace('#(\d)\,(\d{3})#','$1$2',$str);

Upvotes: 0

Martin Ender
Martin Ender

Reputation: 44279

If you have a look at the preg_replace documentation you can see that you can write captures back in the replacement string using $n:

preg_replace('#(\d),(\d)#','$1$2',$str);

Note that there is no need to escape the comma, or to use i (as there are not letters in the pattern).

An alternative (and probably more efficient) way is to use lookarounds. These are not included in the match, so they don't have to written back:

preg_replace('#(?<=\d),(?=\d)#','',$str);

Upvotes: 3

Related Questions