Exclusion in PRXMATCH ?

[ybz12003]

Hello,

I would like to exclude the 'SUB' and 'NCV' in the PRXMATCH command.  I have the code below.   I found the texts are still in the result TEMPID.  Please help.  Thank you.

 if prxmatch ('/000|^SUB|^NCV/i',id) then TempID=id;

[PeterClemmensen]

I'm not entirely sure what you want to do here. Your code matches string that contain '000' or starts with either 'sub' or 'ncv' (case insensitive).

[novinosrin · Accepted solution]

Hi @ybz12003  Not sure your objective, 

Assuming you are saying something like below is what you want

 if id = '000' or id not in  ('SUB','NCV') then tempid=id;/*assuming you want this*/

is literally equivalent to

 if  id not in  ('SUB','NCV') then tempid=id;

which basically is equivalent to

if not prxmatch ('/SUB|NCV/i',id) then TempID=id;

[Patrick]

---

@ybz12003 wrote:

Hello,

I would like to exclude the 'SUB' and 'NCV' in the PRXMATCH command.  I have the code below.   I found the texts are still in the result TEMPID.  Please help.  Thank you.

 if prxmatch ('/000|^SUB|^NCV/i',id) then TempID=id;

---

Please provide some sample data and then show us the desired result.

For your RegEx: Do you by any chance believe the ^ has a meaning of NOT? 

^abc      Here the ^ means beginning of string so strings starting with abc will match.

[^abc]    Here the ^ has the meaning of NOT and string containing a, b, or c anywhere will not match

[ybz12003]

Below is my sample dataset.

data datain;
      infile datalines dsd;
  input Name : $300. ;
datalines;
	0001234,
	SUB0001031,
	NCV0001234,
	CARID0001070,
	;
run;

The final result I want is 0001234 and CARID0001070.

[novinosrin]

Hi @ybz12003  Please check if this works

data datain;
      infile datalines dsd;
  input Name : $300. ;
datalines;
	0001234,
	SUB0001031,
	NCV0001234,
	CARID0001070,
	;
run;
data want;
 set datain;
 where not prxmatch ('/^(SUB|NCV)/i',name);
run;

[ChrisNZ]

Like this?

where NAME not in: ('SUB','NCV') and index(NAME,'000');