Approximate Match match in the Same Field

[rvc_121985]

Hi Experts,

I need your help with regard to identifying approximate match in one filed where values are different and can be keep on changing.

For example:

Reference

ABCD1

CDEF

abcd2

abcd

cdef/1

cdef/2

CDEF.

1234

1234/1

1234/2

1234

XYZ

YZX

1xyz

As i have mentioned reference is uncountable so, i do not know how to use it in wildmatch. i am expecting result must be something like this.

Reference	Result
ABCD1	ABCD1
CDEF	CDEF
abcd2	ABCD1
abcd	ABCD1
cdef/1	CDEF
cdef/2	CDEF
CDEF.	CDEF
1234	1234
1234/1	1234
1234/2	1234
1234	1234
XYZ	XYZ
YZX	YZX
1xyz	XYZ

It is not necessary that result should have same that i have mentioned but the row in result field should have the same result(identifier) wherever it is approximate matching for grouping purpose.

So, please help me.

Thanks

[lironbaram]

hi 

how do you know the logic how to group 

for example why ABCD1,abcd2,abcd are grouped to ABCD1 and not to ABCD

if you have a list of groups you could use wildmatch with some type of loop 

but how do you know the list of results? 

[rvc_121985]

Hi @lironbaram 

As i have mentioned in my post. Need not be necessary that my result should be same. But only requirement is  to generate any value that i can club these together.

As you have mentioned "Why ABCD1,abcd2,abcd are grouped to ABCD1 and not to ABCD" , i am fine that all the rows have any one of these ABCD1 or ABCD or abcd or abcd2 or any othere numerical sequence or First value of the Duplicate.

Thanks,

RC

[Brett_Bleess]

Are you wishing to do this in the Script or in an Expression?  If the script, I would check out the following, as I think that will potentially do the trick, but for expression, I am going to leave this for someone that is a better developer than am I.

https://help.qlik.com/en-US/qlikview/April2020/Subsystems/Client/Content/QV_QlikView/Scripting/Scrip...

Apologies if I am way off, my post will kick things back up, so someone else may shout back with further ideas that will be a better fit for you.

Regards,
Brett