I have a program that has been working for years. Today, we upgraded from SAS 9.4M3 to 9.4M7.
proc setinit
Current version: 9.04.01M7P080520
Since then, I am not able to get the same results as before the upgrade.
Please note that I am querying on Oracle databases directly.
Trying to replicate the issue with a minimal, reproducible SAS table example, I found that the issue disappear when querying on a SAS table instead of on Oracle databases.
Let's say I have the following dataset:
data have;
infile datalines delimiter="|";
input name :$8. id $1. value :$8. t1 :$10.;
datalines;
Joe|A|TLO
Joe|B|IKSK
Joe|C|Yes
;
Using the temporary table:
proc sql;
create table want as
select name,
min(case when id = "A" then value else "" end) as A length 8
from have
group by name;
quit;
Results:
name A
Joe TLO
However, when running the very same query on the oracle database directly I get a missing value instead:
proc sql;
create table want as
select name,
min(case when id = "A" then value else "" end) as A length 8
from have_oracle
group by name;
quit;
name A
Joe
As per documentation, the min() function is behaving properly when used on the SAS table
The MIN function returns a missing value (.) only if all arguments are missing.
I believe this happens when Oracle don't understand the function that SAS is passing it - the min functions in SAS and Oracle are very different and the equivalent in SAS would be LEAST().
So my guess is that the upgrade messed up how is translates the SAS min function to Oracle, but it remains a guess. Does anyone ran into this type of behavior?
EDIT: #Richard's comment
options sastrace=',,,d' sastraceloc=saslog nostsuffix;
proc sql;
create table want as
select t1.name,
min(case when id = 'A' then value else "" end) as A length 8
from oracle_db.names t1 inner join oracle_db.ids t2 on (t1.tid = t2.tid)
group by t1.name;
ORACLE_26: Prepared: on connection 0
SELECT * FROM NAMES
ORACLE_27: Prepared: on connection 1
SELECT UI.INDEX_NAME, UIC.COLUMN_NAME FROM USER_INDEXES UI,USER_IND_COLUMNS UIC WHERE UI.TABLE_NAME='NAMES' AND
UIC.TABLE_NAME='NAMES' AND UI.INDEX_NAME=UIC.INDEX_NAME
ORACLE_28: Executed: on connection 1
SELECT statement ORACLE_27
ORACLE_29: Prepared: on connection 0
SELECT * FROM IDS
ORACLE_30: Prepared: on connection 1
SELECT UI.INDEX_NAME, UIC.COLUMN_NAME FROM USER_INDEXES UI,USER_IND_COLUMNS UIC WHERE UI.TABLE_NAME='IDS' AND
UIC.TABLE_NAME='IDS' AND UI.INDEX_NAME=UIC.INDEX_NAME
ORACLE_31: Executed: on connection 1
SELECT statement ORACLE_30
ORACLE_32: Prepared: on connection 0
select t1."NAME", MIN(case when t2."ID" = 'A' then t1."VALUE" else ' ' end) as A from
NAMES t1 inner join IDS t2 on t1."TID" = t2."TID" group by t1."NAME"
ORACLE_33: Executed: on connection 0
SELECT statement ORACLE_32
ACCESS ENGINE: SQL statement was passed to the DBMS for fetching data.
NOTE: Table WORK.SELECTED_ATTR created, with 1 row and 2 columns.
! quit;
NOTE: PROCEDURE SQL used (Total process time):
real time 0.34 seconds
cpu time 0.09 seconds
Use the SASTRACE= system option to log SQL statements sent to the DBMS.
options SASTRACE=',,,d';
will provide the most detailed logging.
From the prepared statement you can see why you are getting a blank from the Oracle query.
select
t1."NAME"
, MIN ( case
when t2."ID" = 'A' then t1."VALUE"
else ' '
end
) as A
from
NAMES t1 inner join IDS t2 on t1."TID" = t2."TID"
group by
t1."NAME"
The SQL MIN () aggregate function will exclude null values from consideration.
In SAS SQL, a blank value is also interpreted as null.
In SAS your SQL query returns the min non-null value TLO
In Oracle transformed query, the SAS blank '' is transformed to ' ' a single blank character, which is not-null, and thus ' ' < 'TLO' and you get the blank result.
The actual MIN you want to force in Oracle is min(case when id = "A" then value else null end) which #Tom has shown is possible by omitting the else clause.
The only way to see the actual difference is to run the query with trace in the prior SAS version, or if lucky, see the explanation in the (ignored by many) "What's New" documents.
Why are you using ' ' or '' as the ELSE value? Perhaps Oracle is treating a string with blanks in it differently than a null string.
Why not use null in the ELSE clause?
or just leave off the ELSE clause and let it default to null?
libname mylib oracle .... ;
proc sql;
create table want as
select name
, min(case when id = "A" then value else null end) as A length 8
from mylib.have_oracle
group by name
;
quit;
Also try running the Oracle code yourself, instead of using implicit pass thru.
proc sql;
connect to oracle ..... ;
create table want as
select * from connection to oracle
(
select name,
min(case when id = "A" then value else null end) as A length 8
from have_oracle
group by name
)
;
quit;
When I try to reproduce this in Oracle I get the result you are looking for so I suspect it has something to do with SAS (which I'm not familiar with).
with t as (
select 'Joe' name, 'A' id, 'TLO' value from dual union all
select 'Joe' name, 'B' id, 'IKSK' value from dual union all
select 'Joe' name, 'C' id, 'Yes' value from dual
)
select name
, min(case when id = 'A' then value else '' end) as a
from t
group by name;
NAME A
---- ----
Joe TLO
Unrelated, if you are only interested in id = 'A' then a better query would be:
select name
, min(value) as a
from t
where id = 'A'
group by name;
I'm using APEX 20.2.
I have a checkbox group with 3 different values:
The values checked in this group are used to filter a query and I'm not sure how to write the where clause
Here is what I have:
SELECT Null as link, PERSONNE_EN_CHARGE as label, count(*) as value
FROM FLYWAY_MEP_HISTORY
WHERE STATUT_MEP='SUCCESS'
AND (DATE_MEP BETWEEN to_date(:DATE_DEBUT,'DD/MM/YYYY') AND to_date(:DATE_FIN ,'DD/MM/YYYY')+1)
OR (:DATE_DEBUT is null and :DATE_FIN is null)
AND (UPPER(ENVIRONNEMENT) is null or UPPER(ENVIRONNEMENT) IN (:ENVIRONNEMENT))
GROUP BY PERSONNE_EN_CHARGE
ORDER BY 2;
the part I'm not sure of is this part:
UPPER(ENVIRONNEMENT) IN (:ENVIRONNEMENT))
How can I use the values of the checkbox group in the where clause?
If I debug the page I can see this in the devlopper console
When you select multiple values in Apex, they are separated by colon. So, if you split that value into rows, you can use it as a subquery. In your case:
...
or UPPER(ENVIRONNEMENT) IN
(select regexp_substr(:ENVIRONNEMENT, '[^:]+', 1, level)
from dual
connect by level <= regexp_count(:ENVIRONMENT, ':') + 1
)
I have a checkbox on my page P3_Checkbox1 that is populated from the database. How can I loop through all the checked values of my checkbox in PL/SQL code?
I assume that APEX_APPLICATION.G_F01 is used only by sql generated checkboxes and I cannot use it for regular checbox as it would not populate G_Fxx array.
I think I understand now what you're saying.
For example, suppose that the P3_CHECKBOX1 checkbox item allows 3 values (display/return):
Absent: -1
Unknown: 0
Here: 1
I created a button (which will just SUBMIT the page) and two text items which will display checked values: P3_CHECKED_VALUES and P3_CHECKED_VALUES_2.
Then I created a process which fires when the button is pressed. The process looks like this (read the comments as well, please):
begin
-- This is trivial; checked (selected) values are separated by a colon sign,
-- just like in a Shuttle item
:P3_CHECKED_VALUES := :P3_CHECKBOX1;
-- This is what you might be looking for; it uses the APEX_STRING.SPLIT function
-- which splits selected values (the result is ROWS, not a column), and these
-- values can then be used in a join or anywhere else, as if it was result of a
-- subquery. The TABLE function is used as well.
-- LISTAGG is used just to return a single value so that I wouldn't have to worry
-- about TOO-MANY-ROWS error.
with
description (code, descr) as
(select -1, 'Absent' from dual union all
select 0 , 'Unknown' from dual union all
select 1 , 'Here' from dual),
desc_join as
(select d.descr
from description d join (select * from table(apex_string.split(:P3_CHECKED_VALUES, ':'))) s
on d.code = s.column_value
)
select listagg(j.descr, ' / ') within group (order by null)
into :P3_CHECKED_VALUES_2
from desc_join j;
end;
Suppose that the Absent and Unknown values have been checked. The result of that PL/SQL process is:
P3_CHECKED_VALUES = -1:0
P3_CHECKED_VALUES_2 = Absent / Unknown
You can rewrite it as you want; this is just one example.
Keywords:
APEX_STRING.SPLIT
TABLE function
I hope it'll help.
[EDIT: loop through selected values]
Looping through those values isn't difficult; you'd do it as follows (see the cursor FOR loop):
declare
l_dummy number;
begin
for cur_r in (select * From table(apex_string.split(:P3_CHECKED_VALUES, ':')))
loop
select max(1)
into l_dummy
from some_table where some_column = cur_r.column_value;
if l_dummy is null then
-- checked value does not exist
insert into some_table (some_column, ...) valued (cur_r.column_value, ...);
else
-- checked value exists
delete from some_table where ...
end if;
end loop;
end;
However, I'm not sure what you meant by saying that a "particular combination is in the database". Does it mean that you are storing colon-separated values into a column in that table? If so, the above code won't work either because you'd compare, for example
-1 to 0:-1 and then
0 to 0:-1
which won't be true (except in the simplest cases, when checked and stored values have only one value).
Although Apex' "multiple choices" look nice, they can become a nightmare when you have to actually do something with them (like in your case).
Maybe you should first sort checkbox values, sort database values, and then compare those two strings.
It means that LISTAGG might once again become handy, such as
listagg(j.descr, ' / ') within group (order by j.descr)
Database values could be sorted this way:
SQL> with test (col) as
2 (select '1:0' from dual union all
3 select '1:-1:0' from dual
4 ),
5 inter as
6 (select col,
7 regexp_substr(col, '[^:]+', 1, column_value) token
8 from test,
9 table(cast(multiset(select level from dual
10 connect by level <= regexp_count(col, ':') + 1
11 ) as sys.odcinumberlist))
12 )
13 select
14 col source_value,
15 listagg(token, ':') within group (order by token) sorted_value
16 from inter
17 group by col;
SOURCE SORTED_VALUE
------ --------------------
1:-1:0 -1:0:1
1:0 0:1
SQL>
Once you have them both sorted, you can compare them and either INSERT or DELETE a row.
I am developing an application in Oracle APEX. I have a string with user id's that is comma deliminated which looks like this,
45,4932,20,19
This string is stored as
:P5_USER_ID_LIST
I want a query that will find all users that are within this list my query looks like this
SELECT * FROM users u WHERE u.user_id IN (:P5_USER_ID_LIST);
I keep getting an Oracle error: Invalid number. If I however hard code the string into the query it works. Like this:
SELECT * FROM users u WHERE u.user_id IN (45,4932,20,19);
Anyone know why this might be an issue?
A bind variable binds a value, in this case the string '45,4932,20,19'. You could use dynamic SQL and concatenation as suggested by Randy, but you would need to be very careful that the user is not able to modify this value, otherwise you have a SQL Injection issue.
A safer route would be to put the IDs into an Apex collection in a PL/SQL process:
declare
array apex_application_global.vc_arr2;
begin
array := apex_util.string_to_table (:P5_USER_ID_LIST, ',');
apex_collection.create_or_truncate_collection ('P5_ID_COLL');
apex_collection.add_members ('P5_ID_COLL', array);
end;
Then change your query to:
SELECT * FROM users u WHERE u.user_id IN
(SELECT c001 FROM apex_collections
WHERE collection_name = 'P5_ID_COLL')
An easier solution is to use instr:
SELECT * FROM users u
WHERE instr(',' || :P5_USER_ID_LIST ||',' ,',' || u.user_id|| ',', 1) !=0;
tricks:
',' || :P5_USER_ID_LIST ||','
to make your string ,45,4932,20,19,
',' || u.user_id|| ','
to have i.e. ,32, and avoid to select the 32 being in ,4932,
I have faced this situation several times and here is what i've used:
SELECT *
FROM users u
WHERE ','||to_char(:P5_USER_ID_LIST)||',' like '%,'||to_char(u.user_id)||',%'
ive used the like operator but you must be a little carefull of one aspect here: your item P5_USER_ID_LIST must be ",45,4932,20,19," so that like will compare with an exact number "',45,'".
When using it like this, the select will not mistake lets say : 5 with 15, 155, 55.
Try it out and let me know how it goes;)
Cheers ,
Alex
Create a native query rather than using "createQuery/createNamedQuery"
The reason this is an issue is that you cannot just bind an in list the way you want, and just about everyone makes this mistake at least once as they are learning Oracle (and probably SQL!).
When you bind the string '32,64,128', it effectively becomes a query like:
select ...
from t
where t.c1 in ('32,64,128')
To Oracle this is totally different to:
select ...
from t
where t.c1 in (32,64,128)
The first example has a single string value in the in list and the second has a 3 numbers in the in list. The reason you get an invalid number error is because Oracle attempts to cast the string '32,64,128' into a number, which it cannot do due to the commas in the string.
A variation of this "how do I bind an in list" question has come up on here quite a few times recently.
Generically, and without resorting to any PLSQL, worrying about SQL Injection or not binding the query correctly, you can use this trick:
with bound_inlist
as
(
select
substr(txt,
instr (txt, ',', 1, level ) + 1,
instr (txt, ',', 1, level+1) - instr (txt, ',', 1, level) -1 )
as token
from (select ','||:txt||',' txt from dual)
connect by level <= length(:txt)-length(replace(:txt,',',''))+1
)
select *
from bound_inlist a, users u
where a.token = u.id;
If possible the best idea may be to not store your user ids in csv! Put them in a table or failing that an array etc. You cannot bind a csv field as a number.
Please dont use: WHERE ','||to_char(:P5_USER_ID_LIST)||',' like '%,'||to_char(u.user_id)||',%' because you'll force a full table scan although with the users table you may not have that many so the impact will be low but against other tables in an enterprise environment this is a problem.
EDIT: I have put together a script to demonstrate the differences between the regex method and the wildcard like method. Not only is regex faster but it's also a lot more robust.
-- Create table
create table CSV_TEST
(
NUM NUMBER not null,
STR VARCHAR2(20)
);
create sequence csv_test_seq;
begin
for j in 1..10 loop
for i in 1..500000 loop
insert into csv_test( num, str ) values ( csv_test_seq.nextval, to_char( csv_test_seq.nextval ));
end loop;
commit;
end loop;
end;
/
-- Create/Recreate primary, unique and foreign key constraints
alter table CSV_TEST
add constraint CSV_TEST_PK primary key (NUM)
using index ;
alter table CSV_TEST
add constraint CSV_TEST_FK unique (STR)
using index;
select sysdate from dual;
select *
from csv_test t
where t.num in ( Select Regexp_Substr('100001, 100002, 100003 , 100004, 100005','[^,]+', 1, Level) From Dual
Connect By Regexp_Substr('100001, 100002,100003, 100004, 100005', '[^,]+', 1, Level) Is Not Null);
select sysdate from dual;
select *
from csv_test t
where ('%,' || '100001,100002, 100003, 100004 ,100005' || ',%') like '%,' || num || ',%';
select sysdate from dual;
select *
from csv_test t
where t.num in ( Select Regexp_Substr('100001, 100002, 100003 , 100004, 100005','[^,]+', 1, Level) From Dual
Connect By Regexp_Substr('100001, 100002,100003, 100004, 100005', '[^,]+', 1, Level) Is Not Null);
select sysdate from dual;
select *
from csv_test t
where ('%,' || '100001,100002, 100003, 100004 ,100005' || ',%') like '%,' || num || ',%';
select sysdate from dual;
drop table csv_test;
drop sequence csv_test_seq;
Solution from Tony Andrews works for me. The process should be added to "Page processing" >> "After submit">> "Processes".
As you are Storing User Ids as String so You can Easily match String Using Like as Below
SELECT * FROM users u WHERE u.user_id LIKE '%'||(:P5_USER_ID_LIST)||'%'
For Example
:P5_USER_ID_LIST = 45,4932,20,19
Your Query Surely Will return Any of 1 User Id which Matches to Users table
This Will Surely Resolve Your Issue , Enjoy
you will need to run this as dynamic SQL.
create the entire string, then run it dynamically.
I have a scenario in oracle where i need to be able to reuse the value of a pseudo column which was calculated previously within the same select statement something like:
select 'output1' process, process || '-Output2' from Table1
I don't want to the repeat the first columns logic again in the second column for maintenance purposes, currently it is done as
select 'output1' process, 'output1' || '-Output2' name from Table1
since i have 4 such columns which depend on the previous column output, repeating would be a maintenance nightmare
Edit: i included table name and removed dual, so that no assumptions are made about that this is not a complex process, my actual statement does have 2 to 3 joins on different tables
You could calculate the value in a sub-query:
select calculated_output process, calculated_output || '-Output2' name from
(
select 'output1' calculated_output from dual
)
You can also use the subquery factoring syntax - i think it is more readable myself:
with tmp as
(
select 'output' process from dual
)
select
process, process || '-Output2' from tmp;
very similar to the previous poster, but I prefer the method I show below, it is more readable, thusly more supportable, in my opinion
select val1 || val2 || val3 as val4, val1 from (
select 'output1' as val1, '-output2' as val2, '-output3' as val3 from dual
)