In Notepad++ regex to remove varying number of spaces in multiple instances of a substring

Question

Using Notepad++, I would like to replace all occurrences of a semi-colon followed by a space ("; ") with a semi-colon (";") in a string like this

S#14_Budget.Y#2014; 2015; 2016.P{[Third Generation]}.w#Periodic.E{PRU.[Descendants]}; CNS.PRU.V#[None]; Total.A{Asset.[Base]}; {Liab.[Base]}; {NProfit.[Base]}; {Bal.[Base]}; {Stats.[Base]}; {BalSht.[Base]}; {NIncome.[Base]}.I{[ICP Top].[Base]}.C1#TotalC1.C2#TotalC2.C3#[None].C4#[None]</ns

The string can occur hundreds of times in one file, and I find it using S#.*</ns.
The string always begins with S# and always ends with </ns.
All the bits in the middle can vary, including the number of spaces.
Only the "; " needs to be changed to ";".

["Notepad++ seems to not have implemented variable-length look-behinds"](http://stackoverflow.com/questions/17286667/error-in-regular-expression-in-notepad-using-negative-lookbehind) (at least not when that was written, according to random-stranger-on-the-internet) - this might make this rather difficult / impossible (otherwise replacing `(?<=S#14.*?); (?=.*? — Bernhard Barker, Sep 18 '13 at 22:17
@HamZa Since there is a `Replace All`, you can just hit that as many times as the string containing the most `; `'s, if this isn't too great a number, but a macro isn't too difficult either. It doesn't want to replace with `\K`, but you can use grouping, posted an answer. — Bernhard Barker, Sep 18 '13 at 22:53

score 2 · Accepted Answer · edited May 23 '17 at 11:50

With a bit of help from HamZa, I came up with this:

Replace:

(S#.*?;) (?=.*?</ns)

with:

\1

Then just hit Replace All until no replacements are made any more (each find includes the S#, so you can only do one replacement for each of these strings at a time), or you can write a simple macro to find and replace (all?) and run that as many times as required.

If this string of yours is on its own line, you should also include start (^) and end ($) of line indicators:

^(S#.*?;) (?=.*?</ns$)

Explained: (source)

NODE                     EXPLANATION
--------------------------------------------------------------------------------
  (                        group and capture to \1:
--------------------------------------------------------------------------------
    S#                       'S#'
--------------------------------------------------------------------------------
    .*?                      any character except \n (0 or more times
                             (matching the least amount possible))
--------------------------------------------------------------------------------
    ;                        ';'
--------------------------------------------------------------------------------
  )                        end of \1
--------------------------------------------------------------------------------
                           ' '
--------------------------------------------------------------------------------
  (?=                      look ahead to see if there is:
--------------------------------------------------------------------------------
    .*?                      any character except \n (0 or more times
                             (matching the least amount possible))
--------------------------------------------------------------------------------
    </ns                     '</ns'
--------------------------------------------------------------------------------
  )                        end of look-ahead

If Notepad++ supported variable-length look-behinds (at least it doesn't in 6.4.5), you could've replaced (?<=S#14.*?); (?=.*?</ns) with ; (and just have done one Replace All).

doesn't work work for me at all(( "Can't find any expression" — Frankie Drake, Sep 19 '14 at 12:50

In Notepad++ regex to remove varying number of spaces in multiple instances of a substring

1 Answers1