Trying to extract a string of text pattern from the beginning and the end of a cell in Excel - excel

Question

I have the following data and what I would like to see on the column result:
Data
Result
PN 65011:2020text text text PN 65011:2020
PN 65011:2020, PN 65011:2020
PN 45014-1:2017text text text text PN 65014-1:2017 PN 8726-1:2017/P11:2020
PN 45014-1:2017, PN 65014-1:2017, PN 8726-1:2017/P11:2020
PN 6534:2020text text text text
PN 6534:2020
PN 65014-1:2017text text text text PN 65014-1:2017/PC1:2013
PN 65014-1:2017,PN 65014-1:2017/PC1:2013
PN ESO 67345:2019text text text PN 65018-1:2019/PC2:2020
PN ESO 67345:2019, PN 65018-1:2019/PC2:2020
PN ESO/EOC 5320:2013text text text PN ESO 27380:2019 PN 65015-1:2020/PC:2021
PN ESO/EOC 5320:2013, PN ESO 27380:2019, PN 65015-1:2020/PC:2021
I have used ="PN "&TEXTJOIN(", PN ",1,IF(ISNUMBER(SEARCH("/",TRIM(MID(SUBSTITUTE(A2,"PN ",REPT(" ",LEN(A2))),(ROW(INDIRECT("1:"&LEN(A2))))*LEN(A2)-(LEN(A2)-1),LEN(A2))))),TRIM(MID(SUBSTITUTE(A2,"PN ",REPT(" ",LEN(A2))),(ROW(INDIRECT("1:"&LEN(A2))))*LEN(A2)-(LEN(A2)-1),LEN(A2))),LEFT(TRIM(MID(SUBSTITUTE(A2,"PN ",REPT(" ",LEN(A2))),(ROW(INDIRECT("1:"&LEN(A2))))*LEN(A2)-(LEN(A2)-1),LEN(A2))),MIN(IFERROR(FIND({" "},LOWER(TRIM(MID(SUBSTITUTE(A2,"PN ",REPT(" ",LEN(A2))),(ROW(INDIRECT("1:"&LEN(A2))))*LEN(A2)-(LEN(A2)-1),LEN(A2))))),""))-1)))
And I almost get what I would like to see, except for the last row (PN ESO 5320:2013), I don't get the numbers. It stops at PN ESO. Like this:
Data
Result
PN ESO/EOC 5320:2013text text PN ESO 27380:2019 text PN 65015-1:2020/PC:2021
PN ESO/EOC, PN ESO
Any ideas on how I can get the entire reference?
Thank you very much in advance.

Answer 1

Here is an example on how you could approach this using Excel O365
Formula in B2:
=TEXTJOIN(", ",,LET(X,FILTERXML("<t><s>"&SUBSTITUTE(A2,"PN ","</s><s>PN ")&"</s></t>","//s[position() > 1]"),Y,LEFT(X,FIND("|",SUBSTITUTE(X,":","|",LEN(X)-LEN(SUBSTITUTE(X,":",""))))+4),Y))
The idea here is to first SUBSTITUTE() all instances of "PN " to a valid xpath construct. Then we using FILTERXML() to return all values as an array, obviously still with the concatenated "text text text". Therefor I used LET() to load the array as a variable and use some string manipulation on all elements.
First I substituted the last occurence of the colon in all strings into a pipe-symbol which we then FIND() and return its position. Now we have the positions we can extract the the proper substrings using LEFT(). Used TEXTJOIN() to join the resulting array back together.

Answer 2

If you can accept a VBA solution, regular expressions are well suited for this kind of problem. If your examples are all as you show:
We use the regex which will look for substrings that
start with PN
pick up the following characters until we end with a colon followed by multiple digits.
if there is a / following, then look for the next set up to colon-multiple digit pattern.
To enter this User Defined Function (UDF), <alt-F11> opens the Visual Basic Editor.
Ensure your project is highlighted in the Project Explorer window.
Then, from the top menu, select Insert/Module and
paste the code below into the window that opens.
To use this User Defined Function (UDF), enter a formula like =extrPN(cell_Ref) in some cell.
Option Explicit
Function extrPN(S As String) As String
Dim RE As Object, MC As Object, M As Object
Const sPat As String = "PN[^:]+:\d+(?:/[^:]+:\d+)?"
Dim sTemp As String
Set RE = CreateObject("vbscript.regexp")
With RE
.Global = True
.Pattern = sPat
.ignorecase = False
If .Test(S) = True Then
Set MC = .Execute(S)
For Each M In MC
sTemp = sTemp & ", " & M
Next M
extrPN = Mid(sTemp, 3)
Else: extrPN = "no match"
End If
End With
End Function
Explanation of Regex
extract PN
PN.*?:\d+(?:/[^:]+:\d+)?
Options: Case insensitive; ^$ match at line breaks
Match the character string “PN” literally PN
Match any single character that is NOT a line break character .*?
Between zero and unlimited times, as few times as possible, expanding as needed (lazy) *?
Match the colon character :
Match a single character that is a “digit” \d+
Between one and unlimited times, as many times as possible, giving back as needed (greedy) +
Match the regular expression below (?:/[^:]+:\d+)?
Between zero and one times, as many times as possible, giving back as needed (greedy) ?
Match the character “/” literally /
Match any character that is NOT the colon character [^:]+
Between one and unlimited times, as many times as possible, giving back as needed (greedy) +
Match the colon character :
Match a single character that is a “digit” \d+
Between one and unlimited times, as many times as possible, giving back as needed (greedy) +
Created with RegexBuddy