CreativePro Forum

Return to Member Forum

• Author
  Posts
• • October 15, 2010 at 12:39 am #57383
    jpannier

    Member

    I’m quite new to GREP. (This is an excuse in case this question is too simple ;-) But I don’t get it.

    I want to create a find/change query to replace a regular space with an En space within an abbreviation for example: e._g. or J._K. Rowling

    Finding a space followed by any word character followed by a period is not the problem. The big issue – for me – is to exclude the case of a word followed by a period at the end of a sentence. For example: … good point. E. g. blabla

    (Maybe this example looks strange, but I’m German and we could start a sentence with an abbreviation – and we do ;-)

    If somebody knows the answer or a workaround I would be happy. If this is a problem that GREP couldn’t solve – this would be an answer too.

    Thanks

    Jeldrik

  • October 15, 2010 at 1:50 am #57384
    Theunis De Jong

    Member

    So far you seem to be doing alright :) Except — an en-space? Isn't that far too wide? Perhaps you meant a Thin Space? (If so, use ~< below where I now have ~>)

    It's fairly easy — you don't really want this:

    w. w..

    because the first w may be the last character of a longer word. So all you have to do is insert a word break b before that first w!

    b[ul]. [ul].

    (I'd rather suggest using [ul] instead of w, because the 'word character' set also includes digits.)

    With all parentheses in place, this would be your Find

    b([ul].) ([ul].)

    and this your Replace

    $1~>$2

    (Advanced:

    It's also possible to use Lookbehind/Lookahead, like this:

    (?<=b[ul].) (?=[ul].)

    and then you'd only have to use this in Replace:

    ~>

    I think the Lookbehind/Lookahead approach would be slightly faster, because ID only has to 'touch' the space itself (although the actual difference would be very minor).)

  • October 15, 2010 at 3:43 am #57385
    jpannier

    Member

    Hi,

    thanks for this respose. In my first tries I was dealing with Lookbehind/Lookahead and with [ul]. but the clou are your boundaries. Perfect. And you're right with the ~< – in the German version we would call it “Achtelgeviert” ;-).

  • October 15, 2010 at 3:45 pm #57399
    Theunis De Jong

    Member

    Good!

    (A tip, specific for this forum: as you might have seen on re-reading your own post, the forum editor can't mentally cope with single backslashes. To make them visible, you have to insert them twice: \u and . Guess how many I had to insert here.

    It's the main cause of many a frantic post-edit. An additional “ouch” is that if you forgot just one and edit your post to correct this, all of the double backslashes were already translated to a single one, and you have to hunt for all of them again!

    … As you can see, I didn't need to edit my post ;))

  • October 16, 2010 at 6:07 am #57403
    jpannier

    Member

    Thanks again.

    I thought I screwed up. But it wasn't me – it was the editor ;-).

• Author
  Posts