BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
AP718
Obsidian | Level 7

Using SAS hash to merge/update records. How can I use hash to only update the records that are missing on my set dataset. Which in the below example will be ID's - 1, 2, 4 and 6? And without setting ID# 3, 5, 7 and 8 to a blank as it's blank on my define dataset (doesn't need an update). Thanks!

 

 

ID# Date
1 .
2 .
3 1/1/2021
4 .
5 7/10/2020
6 .
7 3/1/2019
8 5/2/2017
1 ACCEPTED SOLUTION

Accepted Solutions
AP718
Obsidian | Level 7

I found the solution by adding a second hash. The first HASH retains the DATE values from Table 2 and the second hash updates only the DATE values on Table 2 where the DATE is not missing. Which is exactly what I wanted. Helps to be passionate or "obsessed" to solve problems and find solutions. 

 

data Want;
drop rc;
length Flag $35 Comment $35;

if _n_=1 then
do;
declare hash lookup(dataset:"Have");
lookup.defineKey("ID");
lookup.defineData("Gender"
,"Code"
,"Class"
,"Flag"
,"Comment");
lookup.definedone();
call missing(Flag,Comment, DATE);

 

declare hash lookup2(dataset:"Have (where=(not missing(DATE))");

lookup2.defineKey("ID");
lookup2.defineData("Date")
lookup2.definedone();

end;


set Need_Update;

rc=lookup.find();

if rc eq 0 then do;

rc=lookup2.find();

 

end;

run;

View solution in original post

11 REPLIES 11
data_null__
Jade | Level 19

I'm guessing an IF missing(DATE) then do; UPDATE; end;  

 

Don't you reckon?

AP718
Obsidian | Level 7
OKay----where in the hash statement will this be declared? Below is the syntax I am using. Thanks!
data ;
drop rc;
length ;

if _n_=1 then
do;
declare hash lookup(dataset:"");
lookup.defineKey("");
lookup.defineData("");
lookup.definedone();
call missing();
end;
set ;
rc=lookup.find();
run;
data_null__
Jade | Level 19
   set ;
   if (this obs needs update) then rc=lookup.find();
   run;

You have not provided enough details for me to know what "this obs needs update" actually is.  Perhaps

if missing(date) then rc=lookup.find();
AP718
Obsidian | Level 7

I have multiple vars I am attempting to update. For the Date var I only want to update where it's missing. Below is the code I am using and the result I get. How can I update all the vars and only the instance where the Date is missing? Thanks

 

data Want;
drop rc;
length Flag $35 Comment $35;

if _n_=1 then
do;
declare hash lookup(dataset:"Have");
lookup.defineKey("ID");
lookup.defineData("Date"
,"Gender"
,"Code"
,"Class"
,"Flag"
,"Comment");
lookup.definedone();
call missing(Flag,Comment);
end;


set Need_Update;

rc=lookup.find();
run;

 

Have          
         
ID# Date Gender Code Class
1 . F AB  
2 . F    
3 1/1/2021 M   MIDDLE
4 .     MIDDLE
5 7/10/2020 M AB  
6 . M   HIGH
7 3/1/2019   AB  
8 5/2/2017   AB HIGH

 

Want        
         
ID# Date Gender Code Class
1 7/1/2020 F AB LOW
2 7/2/2020 F CD LOW
3 1/1/2021 M CD MIDDLE
4 1/2/2021 F CD MIDDLE
5 7/10/2020 M AB LOW
6 7/11/2020 M DE HIGH
7 3/1/2019 F AB HIGH
8 5/2/2017 F AB HIGH

 

What I am getting      
ID# Date Gender Code Class
1 7/1/2020 F AB LOW
2 7/2/2020 F CD LOW
3   M CD MIDDLE
4 1/2/2021 F CD MIDDLE
5   M AB LOW
6 7/11/2020 M DE HIGH
7   F AB HIGH
8   F AB HIGH
AP718
Obsidian | Level 7

Correction Below is what I Have and the output I am getting on my Need_Update dataset

 

Have        
ID# Date Gender Code Class
1 . F AB LOW
2 . F CD LOW
3 1/1/2021 M CD MIDDLE
4 . F CD MIDDLE
5 7/10/2020 M AB LOW
6 . M DE HIGH
7 3/1/2019 F AB HIGH
8 5/2/2017 F AB

HIGH

 

Need_Update       
ID# Date Gender Code Class
1 7/1/2020 F AB LOW
2 7/2/2020 F CD LOW
3   M CD MIDDLE
4 1/2/2021 F CD MIDDLE
5   M AB LOW
6 7/11/2020 M DE HIGH
7   F AB HIGH
8   F AB HIGH
data_null__
Jade | Level 19
Which data set is MASTER and which is CHANGE/TRANSACTION?
AP718
Obsidian | Level 7

Sorry for the confusion. Let me try again:

Table 1 is the table I am using to update/merge the missing values in Table 2.                                

Table 3 is what I get as the result. The values from Table 1 replaces the values on Table 2.  How can I only update values if missing on Table 2? In this instance ID# 1, 2, 4 and 6 will be retained after the merge.                        

Table 4 is what I want.  Thank you 

Table 1         
ID# Date Gender Code Class
1 . F AB LOW
2 . F CD LOW
3 1/1/2021 M CD MIDDLE
4 . F CD MIDDLE
5 7/10/2020 M AB LOW
6 . M DE HIGH
7 3/1/2019 F AB HIGH
8 5/2/2017 F AB HIGH

 

Table 2        
ID# Date Gender Code Class
1 7/1/2020 F AB  
2 7/2/2020 F    
3   M   MIDDLE
4 1/2/2021     MIDDLE
5   M AB  
6 7/11/2020 M   HIGH
7     AB  
8     AB HIGH

 

Table 3         
ID# Date Gender Code Class
1 . F AB LOW
2 . F CD LOW
3 1/1/2021 M CD MIDDLE
4 . F CD MIDDLE
5 7/10/2020 M AB LOW
6 . M DE HIGH
7 3/1/2019 F AB HIGH
8 5/2/2017 F AB HIGH

 

Table 4        
ID# Date Gender Code Class
1 7/1/2020 F AB LOW
2 7/2/2020 F CD LOW
3 1/1/2021 M CD MIDDLE
4 1/2/2021 F CD MIDDLE
5 7/10/2020 M AB LOW
6 7/11/2020 M DE HIGH
7 3/1/2019 F AB HIGH
8 5/2/2017 F AB HIGH
Tom
Super User Tom
Super User

I don't understand what HASH objects have to do with this problem.

A simple UPDATE statement should handle it, but you need to reverse how you are thinking about the data.  You are not updating TWO with the looked up values from ONE.  You are overriding the default values in ONE with the actual values from TWO.  So TWO is your transaction file.  Any non-missing value replace the value from ONE.  And a missing value means leave the value from ONE in place.

ata one ;
  input ID Date :mmddyy. Gender $ Code $ Class $;
  format date yymmdd10.;
cards;
1 . F AB LOW
2 . F CD LOW
3 1/1/2021 M CD MIDDLE
4 . F CD MIDDLE
5 7/10/2020 M AB LOW
6 . M DE HIGH
7 3/1/2019 F AB HIGH
8 5/2/2017 F AB HIGH
;

data two;
  input ID Date :mmddyy. Gender $ Code $ Class $;
  format date yymmdd10.;
cards;
1 7/1/2020 F AB . 
2 7/2/2020 F . .
3        . M . MIDDLE
4 1/2/2021 . . MIDDLE
5        . M AB .  
6 7/11/2020 M .  HIGH
7  .  . AB  .
8  .  . AB HIGH
;
data want ;
  update two one;
  by id;
run;

Results:

image.png

AP718
Obsidian | Level 7
I am using Table 1 to update the missing values in Table 2 as most of the actual values don't exist in Table 2. But I only want to replace the missing values in Table 2 if missing and retain the values if not missing. I am using hash for processing as it's a large table. Is there a way to do this with a hash statement? Thanks
Kurt_Bremser
Super User

Don't be obsessed with a certain tool. See if @Tom's suggestion works. If not, supply data that illustrates where the result deviates from what you expected.

 

And do supply example data in usable form (data step(s) with datalines), as demonstrated.

AP718
Obsidian | Level 7

I found the solution by adding a second hash. The first HASH retains the DATE values from Table 2 and the second hash updates only the DATE values on Table 2 where the DATE is not missing. Which is exactly what I wanted. Helps to be passionate or "obsessed" to solve problems and find solutions. 

 

data Want;
drop rc;
length Flag $35 Comment $35;

if _n_=1 then
do;
declare hash lookup(dataset:"Have");
lookup.defineKey("ID");
lookup.defineData("Gender"
,"Code"
,"Class"
,"Flag"
,"Comment");
lookup.definedone();
call missing(Flag,Comment, DATE);

 

declare hash lookup2(dataset:"Have (where=(not missing(DATE))");

lookup2.defineKey("ID");
lookup2.defineData("Date")
lookup2.definedone();

end;


set Need_Update;

rc=lookup.find();

if rc eq 0 then do;

rc=lookup2.find();

 

end;

run;

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

Register now!

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 11 replies
  • 1483 views
  • 4 likes
  • 4 in conversation