Question

I need to perform a search/replace on text which contains a comma which is NOT followed by a space, to change to a comma+space.

So I can find this using:

,[^\s]

But I am struggling with the replacement; I can’t just use:

 , (space, comma)

Or

& ,

As the match originally matches two characters.

Is there a way of saying ‘&’ – 1 ? or ‘&[0]’ or something which means; ‘The Matched String, but only part of it’ in the replacement argument ?

Another way of trying to ask this:

1. Can I use Regex to IDENTIFY one part of my string.
2. But REPLACE a (slightly different,but related) part of my string.

I could just probably replace every comma with a comma+space, but this is a little more controlled and less likely to make a change I do not need….

For example:

Original:

Hello,World.

Should become:

Hello, World.

But:

Hello, World.

Should remain as :

Hello, World.

And currently, using my (bad) pattern I have:

Original:

Hello,World

After (wrong):

Hello, orld

I’m actually using Python’s (2.6) ‘re’ module for this as it happens.

Answer 1

Using parantheses to capture a part of the string is one way to do it. Another possibility is to use “lookahead assertion”:

,(?=\S)

This pattern matches a comma only if it is followed by a non-whitespace character. It does not match anything followed by comma but uses that information to decide whether or not to match the comma.

For example:

>>> re.sub(r",(?=\S)", ", ", "Hello,World! Hello, World!")
'Hello, World! Hello, World!'