Regular Expression to Search/Replace Multiple Times on Same Line

Question

I am trying to use regular expression to search and replace multiple times on a same line. I believe that Regex can do this using either positive lookahead/lookbehind feature.

I have a long list of states and the airports in them. And I am trying to put them all in the State, Aiport pairs, one line at a time.

Here is the input

State, Airports
----------------
Chicago, ORD, MDW
NY, JFK, LGA, EWR
California, LAX, JWA, LGB, BUR

Here is the output.

Chicago, ORD
Chicago, MDW  
NY, JFK
NY, LGA
NY, EWR
California, LAX
California, JWA
California, LGB
California, BUR

Can you please help suggest a regex or any other way to accomplish this? Thank you.

I use Notepad++ for the Regex, but can use any text editor for this.

Answer 1

• Ctrl+H
• Find what: ^(\w+,\h*)(\w+)(?:,\h*(\w+))?(?:,\h*(\w+))?(?:,\h*(\w+))?(?:,\h*(\w+))?(?:,\h*(\w+))?(?:,\h*(\w+))?(?:,\h*(\w+))?(?:,\h*(\w+))?(?:,\h*(\w+))?
• Replace with: $1$2\n(?3$1$3)(?4\n$1$4)(?5\n$1$5)(?6\n$1$6)(?7\n$1$7)(?8\n$1$8)(?9\n$1$9)
• TICK Match case
• TICK Wrap around
• SELECT Regular expression
• UNTICK . matches newline
• Replace all

Explanation:

^               # beginning of line
    (               # group 1
        \w+             # 1 or more word characters
        ,               # a comma
        \h*             # 0 or more horizontal spaces
    )               # end group 1
    (\w+)           # group 2, 1 or more word characters
    (?:             # non capture group
        \h*             # 0 or more horizontal spaces
        (\w+)           # group 3, 1 or more word characters
    )?              # end group, optional
(?:,\h*(\w+))?      # same as above
(?:,\h*(\w+))?      # same as above
(?:,\h*(\w+))?      # same as above
(?:,\h*(\w+))?      # ... 
(?:,\h*(\w+))?      # ... 
(?:,\h*(\w+))?      # ... 
(?:,\h*(\w+))?      # ... 
(?:,\h*(\w+))?      # ... 

Replacement:

$1              # content of group 1
$2              # content of group 2
\n              # line feed, you can use \r\n for Windows
(?3             # if group 3 exists
    $1              # content of group 1
    $3              # content of group 3
)               # endif
(?4\n$1$4)      # same as above
(?5\n$1$5)      # ...
(?6\n$1$6)      # ...
(?7\n$1$7)      # ...
(?8\n$1$8)      # ...
(?9\n$1$9)      # ...

Screenshot (before):

Screenshot (after):

Answer 2

That regex is way too complicated. By using an actual programming language, things will be much simpler.

Here I give an example in Python. Get Python here.

Say you have this input:

Chicago, ORD, MDW
NY, JFK, LGA, EWR
California, LAX, JWA, LGB, BUR

And you want to convert it to your given output:

Chicago, ORD
Chicago, MDW  
NY, JFK
NY, LGA
NY, EWR
California, LAX
California, JWA
California, LGB
California, BUR

It is simple, first split the string into lines, the split each line into list of strings by commas. Finally return the combination of the first element and every other element of the same list.

lines = """Chicago, ORD, MDW
NY, JFK, LGA, EWR
California, LAX, JWA, LGB, BUR"""

for line in lines.splitlines():
    lst = line.split(', ')
    first = lst[0]
    for e in lst[1:]:
        print(f'{first}, {e}')

Answer 3

For Notepad++, if you have to use it, you can use PythonScript also:

1. Plugins -> Plugins Admin... -> Check PythonScript and click Install.
2. Plugins -> PythonScript -> New Script -> create it in default scripts directory.
3. Add these contents and save:

import re

def split_text(match):
  # Split by comma and any number of horizontal whitespace
  parts = re.split(',[\t ]*', match.group(1))  
  results = []
  
  # Skip first item and create strings with all the combinations
  for part in parts[1:]:
    results.append("%s, %s"%(parts[0], part))
    
  # Combine all parts with separated by newline
  return "\n".join(results)

# Ensure proper matches and call split_text() handler function
# (?!\A) - don't match start of file (first line)
# [\w\t ,]+ - match only word characters, horizontal whitespace and commas
editor.rereplace('^(?!\A)([\w\t ,]+)$', split_text);

This works with the header and preserves it.

Answer 4

I know you asked for a regex to use in Notepad++, but here's a perl one-liner anyway:

perl -nle '($state,@f) = split /,\s*/; print "$state, $_" for @f;' < txt

Explanation:

• perl: invoke perl
• -n flag: will perform a loop over each line of the input, and execute the code provided for each line
• -l flag: automatically remove line endings on input and add them to outputs
• -e flag: use the code provided as argument
• ($state,@f) = split /,\s*/: split the line anywhere there is a comma, separated by optional white space, put the first value in $state, and the rest in array @f
• print "$state, $_" for @f: for each item in @f (so each airport), print a line with the state and the airport

Alternative:

perl -F',\s*' -le '$state = shift @F; print "$state, $_" for @F;' < txt

• -F',\s*' flag: auto-split using the provided separator regex, store the results in @F. Implies -n.
• $state = shift @F: remove first item from @F and store it in $state

Of course Chicago is not a state :-)