RegEx and Replace..

Eddie Adolfsson

I’ve a file with alot of lines in them with various data, but what I’m looking to do is to grab all text containing +99(87654321) and i understand i can just use \d{11} actually but the problem is that when i do, i dont know how to replace any non-qualifying text with empties? So say the file looks like …:
123 455 61 +9901020304 blah blah 901

I wish to blank everything but the column of numbers prefixed with +99

Alan Kilborn

This might give you a solid start:

Search for ^(?-s).*?(\+\d{10}).* and replace with \1.

Eddie Adolfsson

Thank you so much, but I forgot to mention that the same document has lines in it that doesn’t have the +99 thing at all in them, and i need those gone too, what do i do then?

AdrianHHH

For similar tasks I adopt a two step process, first to separate the text I want from the rest by putting them on separate lines, second by selecting the (un)wanted lines.

Step 1: Replace (\+\d{10}) with \r\n\1\r\n.
Step 2: Mark all lines matching ^\+\d{10}$
Step 3a: Either Menu => Search => Bookmarks => copy all marked lines, so they can be pasted into another buffer.
Step 3b: Or Menu => Search => Bookmarks => Remove unmarked lines.

Alan Kilborn

@Eddie-Adolfsson

Just informationally, you can remove lines NOT containing a specific bit of text via finding it with the following method:

Say you want to match lines that DON’T contain abc. Then your search expression might be ^(?-s)(?!.*abc).*\R. If you replace with nothing when you run a replacement you effectively remove those lines.

Obviously you can replace abc with whatever you need. It doesn’t have to be constant text, it can be a regular expression of course.